Vanadate-stimulated oxidation of NAD(P)H

Vanadate-Induced Renal cAMP and Malondialdehyde Accumulation Suppresses Alpha 1 Sodium Potassium Adenosine Triphosphatase Protein Levels

It has been demonstrated that vanadate causes nephrotoxicity. Vanadate inhibits renal sodium potassium adenosine triphosphatase (Na, K-ATPase) activity and this is more pronounced in injured renal tissues. Cardiac cyclic adenosine monophosphate (cAMP) is enhanced by vanadate, while increased cAMP suppresses Na, K-ATPase action in renal tubular cells. There are no in vivo data collectively demonstrating the effect of vanadate on renal cAMP levels; on the abundance of the alpha 1 isoform (α₁) of the Na, K-ATPase protein or its cellular localization; or on renal tissue injury. In this study, rats received a normal saline solution or vanadate (5 mg/kg BW) by intraperitoneal injection for 10 days. Levels of vanadium, cAMP, and malondialdehyde (MDA), a marker of lipid peroxidation were measured in renal tissues. Protein abundance and the localization of renal α₁-Na, K-ATPase was determined by Western blot and immunohistochemistry, respectively. Renal tissue injury was examined by histological evaluation and renal function was assessed by blood biochemical parameters. Rats treated with vanadate had markedly increased vanadium levels in their plasma, urine, and renal tissues. Vanadate significantly induced renal cAMP and MDA accumulation, whereas the protein level of α₁-Na, K-ATPase was suppressed. Vanadate caused renal damage, azotemia, hypokalemia, and hypophosphatemia. Fractional excretions of all studied electrolytes were increased with vanadate administration. These in vivo findings demonstrate that vanadate might suppress renal α₁-Na, K-ATPase protein functionally by enhancing cAMP and structurally by augmenting lipid peroxidation.

Vanadium-induced kappaB-dependent transcription depends upon peroxide-induced activation of the p38 mitogen-activated protein kinase

Activation of nuclear factor (NF)-kappaB and subsequent proinflammatory gene expression in human airway epithelial cells can be evoked by oxidative stress. In this study we examined signal transduction pathways activated by vanadyl sulfate (V(IV))-induced oxidative stress in normal human bronchial epithelial cells. Both nuclear translocation of NF-kappaB and enhanced kappaB-dependent transcription induced by V(IV) were inhibited by overexpression of catalase, but not Cu,Zn superoxide dismutase (Cu,Zn-SOD), indicating that peroxides rather than superoxides initiated signaling. Catalase selectively blocked the response to V(IV) because it inhibited neither NF-kappaB translocation nor kappaB-dependent transcription evoked by the proinflammatory cytokine tumor necrosis factor (TNF)-alpha. The V(IV)-induced kappaB-dependent transcription was dependent upon activation of the p38 mitogen-activated protein kinase because overexpression of dominant-negative mutants of the p38 MAPK pathway inhibited V(IV)-induced kappaB-dependent transcription. This inhibition was not due to suppression of NF-kappaB nuclear translocation because NF-kappaB DNA binding was unaffected by the inhibition of p38 activity. Overexpression of catalase, but not Cu,Zn-SOD, inhibited p38 activation, indicating that peroxides activated p38. Catalase failed to block V(IV)- induced increases in phosphotyrosine levels, suggesting that the catalase-sensitive signaling components were independent of V(IV)-induced tyrosine phosphorylation. The data demonstrate that V(IV)-induced oxidative stress activates at least two distinct pathways, NF-kappaB nuclear translocation and p38-dependent transactivation of NF-kappaB, both of which are required to fully activate kappaB-dependent transcription. Moreover, V(IV)-induced oxidative stress activated these pathways in bronchial epithelial cells by upstream signaling cascades that were distinct at some level from those used by the proinflammatory cytokine TNF-alpha.

Sensitive superoxide detection in vascular cells by the new chemiluminescence dye L-012

The detection superoxide production in vascular cells is usually limited by a low sensitivity of available assays. We tested the applicability of the luminol derivate L-012 ¿8-amino-5-chloro-7-phenylpyridol¿3,4-dpyridazine-1,4(2H,3H)dione to measure superoxide production in cultured endothelial cells (human umbilical vein endothelial cells) and rat aortic segments. Following stimulation with the protein kinase stimulator phorbol 12-myristate 13-acetate (PMA, 1 microM) there was an 2.8-fold increase of L-012 chemiluminescence, whereas incubation with angiotensin II (100 nM) did not result in a measurable increase. Addition of vanadate (100 microM) considerably increased the chemiluminescence (up to 17-fold) after PMA and made possible the detection of an enhanced superoxide production after stimulation with angiotensin II (by 1.7-fold). This was due to a approximately 9-fold increase in signal intensity of L-012 in the presence of vanadate. Prolonged incubation with vanadate also led to a tyrosine phosphorylation-dependent increase in superoxide formation which was predominantly produced by an NAD(P)H oxidase. Short-term vanadate-enhanced L-012 chemiluminescence represents a highly sensitive assay making it possible to detect small changes of superoxide formation in intact vascular cells.

Lipid peroxidation and antioxidant enzymes in vanadate-treated rats

Male Wistar rats received an aqueous solution of ammonium metavanadate (AMV) of 0.15 mg/V/ml concentration instead of water for 14 days. The erythrocyte count and haemoglobin level in blood were not changed; the haematocrit index was slightly increased. The spontaneous lipid peroxidation in kidney and liver homogenates was increased. The Fe(II)- or ascorbate-induced lipid peroxidation was more pronounced in the kidney than in the liver. No changes in lipid peroxidation were observed in erythrocytes after AMV treatment. The AMV treatment resulted in a decrease in the activity of the antioxidant enzymes, catalase and glutathione peroxidase in the kidney and liver; the cytosolic Cu,Zn-SOD and mitochondrial Mn-SOD were unchanged. The activity of the enzymes in blood was not changed. The results are discussed with a view to the participation of lipid peroxidation in vanadium toxicity.

Haematological effects of vanadium on living organisms

Although vanadium has been of great interest for many researchers over a number of years, its biochemical and physiological role is not yet fully clear. There are many papers describing the haematological consequences of its excess in living organisms and most of their data are quoted in this mini-review. The authors of these papers used various laboratory animals, different vanadium compounds, frequently different routes of administration and duration of intoxication. Hence a checklist and comparison of the results are rather difficult. Vanadium reduces the deformability of erythrocytes, and such cells are rather frequently retained in the reticuloendothelial system of the spleen and eliminated faster from the blood stream (Kogawa et al., 1976). Vanadium produces peroxidative changes in the erythrocyte membrane, this leading to haemolysis. Therefore, the depressed erythrocyte count in animals intoxicated with vanadium may be the consequence of both the haemolytic action of vanadium and the shortened time of survival of erythrocytes. Changes of the haem precursor level in blood serum and urine observed in humans exposed occupationally to vanadium suggest an influence of this element on haem synthesis. This problem requires, however, further studies and observations. Changes occurring under the influence of vanadium on the leukocyte system of animals suggest the influence of this element on the resistance of the organism, but the mechanism of the action of vanadium still requires elucidation.(ABSTRACT TRUNCATED AT 250 WORDS)

Superoxide-independent reduction of vanadate by rat liver microsomes/NAD(P)H: vanadate reductase activity

It has been reported that vanadate-stimulated oxidation of NAD(P)H by microsomal systems can proceed anaerobically, in contrast to the general notion that the oxidation proceeds exclusively by an O(2-)-dependent free radical chain mechanism. The current study indicates that microsomal systems are endowed with a vanadate-reductase property, involving a NAD(P)H-dependent electron transport cytochrome P450 system. Our ESR measurements demonstrated the formation of a vanadium(IV) species in a mixture containing vanadate, rat liver microsomes, and NAD(P)H. This vanadium(IV) species was identified as the vanadyl ion (VO2+) by comparison with the ESR spectrum of VOSO4. The initial rate of vanadium(IV) formation depends linearly on the concentration of microsomes. The Michaelis-Menten constants were found to be: km = 1.25 mM and Vmax = 0.066 mumol (min)-1 (mg microsomes)-1, respectively. Pretreatment of the microsomes with carbon monoxide or K3Fe(CN)6 reduced vanadium(IV) generation, suggesting that the NAD(P)H-dependent electron transport cytochrome P450 system plays a significant role in the microsomal reduction of vanadate. Measurements under argon or in the presence of superoxide dismutase caused only minor (less than 10%) reductions in vanadium(IV) generation. The VO2+ species was also detected in NAD(P)H oxidation by fructose plus vanadate, a reaction known to proceed via an O(2-)-mediated chain mechanism. However, the amount of vanadium(IV) generated by this reaction was an order of magnitude smaller than that by the microsomal system and was inhibitable by superoxide dismutase, affirming the conclusion that the microsomal/NAD(P)H system is endowed with the (O(2-)-independent) vanadium(V) reductase property.

Superoxide generated by glutathione reductase initiates a vanadate-dependent free radical chain oxidation of NADH

Vanadate V(V) markedly stimulated the oxidation of NADPH by GSSG reductase and this oxidation was accompanied by the consumption of O2 and the accumulation of H2O2. Superoxide dismutases completely eliminated this effect of V(V), whereas catalase was without effect, as was exogenous H2O2 added to 0.1 mM. These effects could be seen equally well in phosphate- or in 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid-buffered solutions. Under anaerobic conditions there was no V(V)-stimulated oxidation of NADPH. Approximately 4% of the electrons flowing from NADPH to O2, through GSSG reductase, resulted in release of O2-. The average length of the free radical chains causing the oxidation of NADPH, initiated by O2- plus V(V), was calculated to be in the range 140-200 NADPH oxidized per O2- introduced. We conclude that GSSG reductase, and by extension other O2(-)-producing flavoprotein dehydrogenases such as lipoyl dehydrogenase and ferredoxin reductase, catalyze V(V)-stimulated oxidation of NAD(P)H because they release O2- and because O2- plus V(V) initiate a free radical chain oxidation of NAD(P)H. There is no reason to suppose that these enzymes can act as NAD(P)H:V(V) oxidoreductases.

Vanadium redox cycling, lipid peroxidation and co-oxygenation of benzo(a)pyrene-7,8-dihydrodiol

Mechanism of lipid peroxidation triggered by vanadium in human term placental microsomes was reinvestigated in vitro. Production of lipid peroxyl radicals was estimated from co-oxygenation of benzo(a)pyrene and benzo(a)pyrene-7,8-dihydrodiol. Vanadyl(IV), but not vanadate(V) caused a dose-dependent co-oxygenation. Vanadate(V) required the presence of reduced nicotinamide adenine dinucleotide phosphate to trigger co-oxygenation of benzo(a)pyrene-7,8-dihydrodiol. To determine the role of pre-formed lipid hydroperoxides, the results obtained with partially peroxidized linoleic acid were compared with those of fresh linoleate. Superoxide dismutase inhibited the co-oxygenation of reaction when fresh linoleic acid was used. To further characterize the role of superoxide anion-radical in the vanadium redox cycling, the increase of optical density of vanadate(V) dissolved in Tris buffer was measured at 328 nm during the addition of KO2. The rate of this reaction producing peroxy-vanadyl complex was decreased by superoxide dismutase, especially, in the presence of catalase. It is suggested that vanadium catalyzes two separate processes, both leading to enhanced lipid peroxidation: (i) initiation, dependent on superoxide and triggered by peroxy-vanadyl; (ii) propagation, dependent on pre-formed lipid hydroperoxide not sensitive to superoxide dismutase. It is postulated that the vanadium-triggered initiation of lipid peroxidation may be crucial for toxicity in organs with limited endogenous lipid peroxidation.

The roles of O2-, HO(.), and secondarily derived radicals in oxidation reactions catalyzed by vanadium salts

V(IV) decomposed H2O2, with evolution of O2, in a free radical chain process involving O2- and HO(.). When V(IV) was limiting, the presence of V(V) augmented O2 evolution because it allowed production of additional V(IV) from the reduction of V(V) by O2-. Gradual addition of V(IV) increased the yield of O2 evolved, per V(IV) added, to greater than 1--a clear indication of a free radical chain reaction. Reductants such as ethanol, Hepes, and NADH imposed a phase of O2 consumption because of HO.-initiated oxidation reactions. The radical produced from the reaction of HO. with ethanol was unable to directly oxidize NADH, whereas that produced from Hepes was able to do so. Ethanol consequently inhibited the oxidation of NADH by anaerobic V(IV) + H2O2, whereas Hepes did not. These results, and others reported herein, are explained on the basis of a coherent set of reactions. Data already in the literature are also clarified on the basis of these reactions.

Flavoenzymes reduce vanadium(V) and molecular oxygen and generate hydroxyl radical

ESR spectroscopic evidence is presented for the formation of vanadium(IV) in the reduction of vanadium(V) by three typical, NADPH-dependent, flavoenzymes: glutathione reductase, lipoyl dehydrogenase, and ferredoxin-NADP+ oxidoreductase. The vanadium(V)-reduction mechanism appears to be an enzymatic one-electron reduction process. Addition of superoxide dismutase (SOD) showed that the generation of vanadium(IV) does not involve the superoxide (O2-) radical significantly. Measurements under anaerobic atmosphere showed, however, that the enzymes-vanadium-NADPH mixture can cause the reduction of molecular oxygen to generate H2O2. The H2O2 and vanadium(IV) thus formed react to generate hydroxyl (.OH) radical. The .OH formation is inhibited strongly by catalase and to a lesser degree by SOD, but it is enhanced by exogenous H2O2, suggesting the occurrence of a Fenton-like reaction. The inhibition of vanadium(IV) formation by N-ethylmaleimide indicates that the SH group on the flavoenzyme's cystine residue plays an important role in the enzyme's vanadium(V) reductase function. These results thus reveal a new property of the above-mentioned, NADPH-dependent flavoenzymes--their function as vanadium(V) reductases, as well as that as generators of .OH radical in the vanadium(V) reduction mechanism.

Vanadate-induced toxicity towards isolated perfused rat livers: the role of lipid peroxidation

The toxic potential of sodium orthovanadate towards isolated perfused rat livers was investigated at a dose of 2 mmol/l. In livers from fasted rats, vanadate led to a release of cytosolic (glutamate-pyruvate-transaminase (GPT) and lactate dehydrogenase (LDH] and mitochondrial (glutamate dehydrogenase (GLDH] enzymes, an accumulation of calcium in the liver, a marked depletion of hepatic glutathione and an enhanced release of it into the perfusate, as well as an augmented formation and release of thiobarbituric acid-reactive material by the liver. Furthermore, a marked inhibition of oxygen consumption was observed. Vanadate-induced vasoconstriction resulted in a progressive decrease in perfusate flow rate. Control experiments with similarly reduced flow rates led to a comparable reduction in oxygen consumption. GPT and LDH release and hepatic glutathione depletion were also evident, though to a lesser extent than in the presence of vanadate, but no increase in GLDH release, in tissue calcium content or TBA-reactive material in the liver or the perfusate were observed. Thus, indirect toxic effects due to a reduced flow rate contribute only partly to vanadate hepatotoxicity and do not affect mitochondrial integrity. Omission of calcium from the perfusate did not prevent hepatotoxic responses to vanadate, although less calcium was present in the treated livers than in the control organs, indicating that calcium influx is not involved in vanadate-induced hepatotoxicity in the intact organ, in contrast to isolated hepatocytes. Feeding the animals, resulting in an activation of anaerobic energy conservation reactions, strongly attenuated vanadate hepatotoxicity indicating that the energetic status of the liver is the main target of vanadate. Superoxide dismutase did not affect the hepatotoxic responses of livers from fasted rats towards vanadate, while allopurinol and deferrioxamine inhibited lipid peroxidation and hepatotoxicity due to vanadate. The strong correlation between induction of lipid peroxidation and hepatotoxicity and the inhibition of both processes in parallel by antioxidants are suggestive of a causative role for lipid peroxidation in vanadate-induced hepatotoxicity.

Effect of antioxidants on vanadate-induced toxicity towards isolated perfused rat livers

The effect of trolox C, a water soluble vitamin E analogue, propyl gallate and ascorbate on vanadate hepatotoxicity was investigated in vitro. In isolated perfused livers from fasted rats, sodium orthovanadate (2 mmol/l) led to toxic responses including reduction of oxygen consumption, release of cytosolic (glutamate-pyruvate-transaminase (GPT) and lactate dehydrogenase (LDH)) and mitochondrial (glutamate-dehydrogenase (GLDH)) enzymes, intracellular accumulation of calcium, a marked depletion of glutathione (GSH) and an enhanced formation and release of thiobarbituric acid- (TBA) reactive material. Trolox C and propyl gallate inhibited the release of GPT and LDH partially and that of GLDH totally, but had no influence on vanadate-induced calcium accumulation or on the reduction of oxygen consumption. Both agents suppressed vanadate-induced lipid peroxidation (LPO) and partially prevented GSH depletion. Ascorbate failed to provide any protection probably due to the interference of its pro-oxidant potential with its antioxidant activity. The protection, mainly of mitochondria, afforded by those agents which also inhibited LPO substantiates our previous findings that the pro-oxidant activity of vanadate is mainly responsible for its direct hepatotoxic actions [2]. Besides, reduction of organ perfusion rate due to vasoconstriction also contributes to vanadate toxicity, but oxidative stress is not involved in this indirect toxic activity.

Characterization of vanadate-dependent NADH oxidation stimulated by Saccharomyces cerevisiae plasma membranes

Plasma membrane-stimulated vanadate-dependent NADH oxidation has been characterized in Saccharomyces cerevisiae. This activity is specific for vanadate, because molybdate, a similar metal oxide, did not substitute for vanadate in the reaction. Vanadate-dependent plasma membrane-stimulated NADH oxidation activity was dependent on the concentrations of vanadate, NADH, and NADPH and required functional plasma membranes; no stimulation occurred in the presence of boiled membranes or bovine serum albumin. The dependence of membrane-stimulated vanadate-dependent NADH oxidation was not linearly dependent on added membrane protein. The activity was abolished by the superoxide anion scavenger superoxide dismutase and was stimulated by paraquat and NADPH. These data are consistent with the previously proposed chain reaction for vanadate-dependent NADH oxidation. The role of the plasma membrane appears to be to stimulate superoxide radical formation, which is coupled to NADH oxidation by vanadate. 51V-nuclear magnetic resonance studies are consistent with the hypothesis that a phosphovanadate anhydride is the stimulatory oxyvanadium species in the phosphate buffers used at pHs 5.0 and 7.0. In phosphate buffers, compared with acetate buffers, the single vanadate resonance was shifted upfield at both pH 5.0 and pH 7.0, which is characteristic of the phosphovanadate anhydride. Since the cell contains an excess of phosphate to vanadate, the phosphovanadate anhydride may be involved in membrane-mediated vanadate-dependent NADH oxidation in vivo.

Reaction of vanadium(V) with thiols generates vanadium (IV) and thiyl radicals

The in vivo toxicity of vanadium(V) has been found to correlate with the depletion of cellular glutathione and related non-protein thiols. With a view to understanding the mechanism for this observation, we have investigated the oxidation of glutathione, cysteine N-acetylcysteine and penicillamine by vanadium(V), using electron spin resonance (ESR) and ESR spin trapping methodology. The spin trap used was 5,5-dimethyl-1-pyrroline 1-oxide (DMPO). It is found that the oxidation of these thiols by vanadium(V) generates the corresponding thiyl radicals and vanadium- (IV) complexes. The results suggest that free radical reactions play a significant role in the depletion of cellular thiols by vanadium(V) and hence in vanadium(V) toxicity.

Evidence for a Fenton-type mechanism for the generation of .OH radicals in the reduction of Cr(VI) in cellular media

Electron spin resonance (ESR) spectroscopy has been employed to examine the role of tetraperoxochromate (V) ions (CrO3-8) and other Cr(V) species in the generation of hydroxyl (OH) radicals in the reaction of Cr(VI) with H2O2 in biological media. In contrast to earlier suggestions, the present ESR studies using crystalline K3CrO8 as a source of CrO3-8 show that decomposition of CrO3-8 in water or in H2O2 does not generate significant amounts of OH radicals. Addition of NADH to a solution containing CrO3-8 yields a Cr(V)-NADH complex, which readily reacts with H2O2 to generate OH radicals. Similar results obtained from several other biological reductants, including vitamin B2, indicate that a vacant coordination site on a Cr(V) complex facilitates its reaction with H2O2 to generate .OH radicals. We thus suggest that in biological media, reaction [3] instead of [1] or [2] is the major pathway for the .OH radical generation: [formula, see text]

Role of glutathione in vanadate reduction in young and mature rats: evidence for direct participation of glutathione in vanadate inactivation

The influence of renal glutathione content, modulated by the glutathione synthesis inhibitor buthionine sulphoximine or by glutathione infusion, on the polyuric and natriuretic effects of i.v. administered vanadate was investigated in 20- and 55-day-old rats. The modulation of renal glutathione content led to significant changes in urine volume and sodium excretion, independently of age. A decrease in the renal glutathione level led to intensification and prolongation of the diuretic effects of vanadate in 55-day-old animals. Treatment with glutathione abolished and treatment with buthionine sulphoximine increased the polyuric effect of vanadate. These observations indicate a role for renal glutathione in vanadate inactivation. Age-dependent differences in the polyuric and natriuretic effectiveness of vanadate are caused by differences in renal glutathione content during maturation of the kidney.

Vanadate-catalyzed, conformationally specific photocleavage of the Ca2(+)-ATPase of sarcoplasmic reticulum

Vanadate-sensitized photocleavage of the Ca2(+)-ATPase of rabbit sarcoplasmic reticulum was observed upon illumination of sarcoplasmic reticulum vesicles or the purified Ca2(+)-ATPase by ultraviolet light in the presence of 1 mM monovanadate or decavanadate. The site of the photocleavage is influenced by the Ca2+ concentration of the medium. When the [Ca2+] is maintained below 10 nM by EGTA, the vanadate-catalyzed photocleavage yields fragments of approximately equal to 87 and approximately equal to 22 kDa, while in the presence of 2-20 mM Ca, polypeptides of 71 and 38 kDa are obtained as the principal cleavage products. These observations indicate that the site of the vanadate-catalyzed photocleavage is altered by changes in the conformation of Ca2(+)-ATPase. Selective tryptic proteolysis, at Arg-505-Ala-506, combined with covalent labeling of Lys-515 by fluorescein 5'-isothiocyanate and with the use of anti-ATPase antibodies of defined specificity, permitted the tentative allocation of the sites of photocleavage to the A fragment near the T2 cleavage site in the absence of Ca2+, and to the B fragment between Lys-515 and Asp-659 in the presence of 2-20 mM Ca2+. The loss of ATPase activity during illumination is accelerated by calcium in the presence of vanadate. The vanadate-catalyzed photocleavage in the presence of Ca2+ is consistent with the existence of an ATPase-Ca2(+)-vanadate complex (Markus et al. (1989) Biochemistry 28, 793-799).

Glutathione reductase functions as vanadate(V) reductase

The oxidation of NADPH by vanadate(V) in the presence of glutathione reductase showed typical enzymatic kinetics. The oxidation was inhibited by N-ethylmaleimide, a glutathione reductase inhibitor. Superoxide dismutase had no significant effect on the oxidation, indicating noninvolvement of the superoxide radical. The vanadate(V) reduction was found to be a one-electron transfer process. These results suggest a new pathway for vanadate(V) metabolism and a new function of glutathione reductase.

On the hydroxyl radical formation in the reaction between hydrogen peroxide and biologically generated chromium(V) species

Electron spin resonance (ESR) measurements on solutions and isolated powders provide direct evidence for the involvement of Cr(V) species in the reduction of Cr(VI) by NAD(P)H. ESR analysis of an isolated Cr(V)-NAD(P)H solid yields g parallel = 1.9831 and g perpendicular = 1.9772, indicating that the unpaired electron occupies the dz2 orbital of the Cr(V) ion, with square-pyramidal geometry. Addition of hydrogen peroxide (H2O2) to the NAD(P)H-Cr(VI) reaction mixtures suppresses the Cr(V) species and generates hydroxyl (.OH) radicals. The .OH radicals were detected via ESR spin trapping, employing 5,5-dimethyl-1-pyrroline-N-oxide and alpha-(4-pyridyl-1-oxide)-N-tert-butylnitrone as spin traps. The dependence of Cr(V) and .OH radical formation on the H2O2 and Cr(VI) concentrations indicates that the Cr(V) species react with H2O2 to generate the .OH radicals. Similar results were obtained by using various diols (arabinose, cellobiose, FAD, fructose, glyceraldehyde, ribose, and tartaric acid), alpha-hydroxycarboxylic acids, and glutathione. Investigations with superoxide dismutase showed no significant participation of O2- in the generation of .OH radicals. These results thus indicate that the Cr(V) complexes, produced in the reduction of Cr(VI) by cellular reductants, react with H2O2 to generate .OH radicals, which might be initiators of the primary events in the Cr(VI) cytotoxicity.