Interaction of 1-hydroxyethyl radical with antioxidant enzymes.

There is considerable interest in the role of the 1-hydroxyethyl radical (HER) in the toxic effects of ethanol. The goal of this study was to evaluate the effects of HER on classical antioxidant enzymes. The interaction of acetaldehyde with hydroxylamine-o-sulfonic acid has been shown to produce 1, 1'-dihydroxyazoethane (DHAE); this compound appears to be highly unstable, and its decomposition leads to the generation of HER. Addition of DHAE into a solution of PBN led to the appearance of the typical EPR spectra of PBN/HER adduct. No PBN/HER spin adduct was detected when DHAE was incubated with 0.1 M PBN in the presence of GSH. In the absence of PBN, DHAE oxidized ascorbic acid to semidehydroascorbyl radical, presumably via an ascorbate-dependent one-electron reduction of HER back to ethanol. Catalase was progressively inactivated by exposure to DHAE-generated HER in a time and HER concentration-dependent manner. Ascorbic acid and PBN gave full protection to catalase against HER-dependent inactivation. The antioxidants 2-tert-butyl-4-methylphenol, propylgallate, and alpha-tocopherol-protected catalase against inactivation by 84, 88, and 39%, respectively. Other antioxidant enzymes were also sensitive to exposure to HER. Glutathione reductase, glutathione peroxidase, and superoxide dismutase were inactivated by 46, 36, and 39%, respectively, by HER. The results reported here plus previous results showing HER interacts with GSH, ascorbate, and alpha-tocopherol suggest that prolonged generation of HER in cells from animals chronically exposed to ethanol may lower the antioxidant defense status, thereby contributing to mechanisms by which ethanol produces a state of oxidative stress and produces toxicity.

Production of reactive oxygen species by microsomes enriched in specific human cytochrome P450 enzymes.

Few studies have evaluated the production of reactive oxygen intermediates by human microsomes, especially the influence of the specific form of cytochrome P450. Experiments were carried out to evaluate the ability of CYP1A1, 1A2, 2B6, and 3A4 to consume NADPH, reduce iron, and catalyze production of reactive oxygen species. Microsomes enriched in each of these CYPs were obtained from commercial +/- lymphoblast cells that had been transfected with cDNA encoding the specific human CYP. On a per nanomole cytochrome P450 basis, CYP3A4 was the most active P450 evaluated in catalyzing NADPH oxidation, production of superoxide anion radical, NADPH-dependent chemiluminescence, oxidation of dichlorofluorescein diacetate, and reduction of either ferric-EDTA or ferric-citrate. CYP1A1 was the next most reactive CYP, whereas CYP1A2 and 2B6 displayed a comparable, lower activity. Nitric oxide, which reacts with and inactivates hemoproteins, inhibited superoxide production by all the CYPs to a similar extent. Because CYP3A4 is present in high amounts in human liver microsomes and is active in catalyzing the formation of reactive oxygen species, this CYP may make an important contribution in the overall ability of human liver microsomes to generate active oxygen species.

Inhibition of ferritin-stimulated microsomal production of reactive oxygen intermediates by nitric oxide.

Experiments were carried out to evaluate the effect of nitric oxide exposure on the ability of NADPH-dependent microsomal electron transfer to mobilize iron from ferritin. Such interactions could play a role in potential antioxidant actions of nitric oxide (NO). Preincubation of the microsomes from phenobarbital-treated rats with NO donors such as S-nitroso-D,L-N-acetyl penicillamine (SNAP), S-nitroso-L-glutathione, SIN-1, and DETANONOate followed by centrifugation, washing, and resuspension of the microsomes resulted in a decrease in the ferritin-dependent oxidation of 2',7'-dichlorofluorescein diacetate (DCFDA) or ferritin-catalyzed chemiluminescence compared to microsomes pretreated with buffer. The ferritin-stimulated rate of oxidation of DCFDA or of chemiluminescence was completely restored if the microsomal preincubation with NO donors was performed in the presence of hemoglobin. In contrast to results with ferritin, ferric-stimulated oxidation of the dye was not affected by any of the tested NO donors. The microsomal oxidation of aminopyrine was inhibited after SNAP treatment, indicating that NO inhibited cytochrome P450 catalyzed activity. Inhibition of cytochrome P450 also resulted in an inhibition of microsomal production of superoxide. Similar results were obtained using microsomes from a cloned cell line which express the CYP2E1 isoform. Since superoxide is required for the mobilization of iron from ferritin by microsomes, inhibition of superoxide production as a consequence of NO interaction with cytochrome P450 is likely to be responsible for the prevention of ferritin-catalyzed formation of reactive oxygen species by NO donors. The results suggest that NO could exhibit an antioxidant capacity through its ability of decreasing the activity of iron-heme compounds, such as cytochrome P450, preventing the release of catalytically active iron from ferritin, and thus decreasing the ability to generate oxygen free radicals involved in cytotoxicity.

Role of cytochrome P-450 in the stimulation of microsomal production of reactive oxygen species by ferritin.

Microsomes can remove iron from ferritin by a superoxide-dependent reaction. The released iron can then catalyse formation of a variety of reactive oxygen species. Experiments were carried out to evaluate the role of cytochrome P-450 in the release of iron from ferritin, and whether induction of certain P-450 isoforms alters ferritin-dependent reactive oxygen radical production. Rats were treated with phenobarbital, 3-methylcholanthrene, 4-methylpyrazole, or saline to produce microsomes with varying P-450 content and composition. Oxidation of 2,7'-dichlorofluorescein diacetate to a fluorescent product and chemiluminescence were used as indices of production of reactive oxygen species. The extreme sensitivity of these reactions to trolox, a potent chain-breaking oxidant, indicates the involvement of lipid peroxidation products in these reactions. In the absence of ferritin, formation of reactive oxygen species was higher in microsomes from the treated rats compared to saline controls when results were expressed on a per mg protein basis but not per nmol P-450, suggesting that the increased content of total P-450 (2-fold increases) rather than the population of isoforms was responsible for the increase. Superoxide dismutase had no effect on the non-ferritin catalyzed reactions. Ferritin increased production of reactive oxygen species with all the microsomal preparations; the increase by ferritin was completely prevented by superoxide dismutase. The net increase by ferritin was higher in microsomes from the treated rats compared to saline controls, but this, again, largely reflected the increased content, rather than the type of isoforms of P-450 present. Similar results were obtained with either NADPH or NADH as microsomal reductants, although NADPH was much more effective in supporting ferritin-dependent reactive oxygen formation. In microsomes from phenobarbital-treated rats, anti-CYP2B1/B2 IgG completely prevented the NADPH- and NADH-dependent increases in reactive oxygen formation produced by ferritin. Anti-cytochrome b5 IgG produced partial inhibition of the ferritin-stimulation. These results indicate that P-450, and to a lesser extent, cytochrome b5, play a role in the ferritin-dependent increase in formation of reactive oxygen species with either NADPH or NADH, most likely reflecting the requirement of these enzymes for microsomal production of superoxide anion.

Ferritin-dependent inactivation of microsomal glucose-6-phosphatase.

Glucose-6-phosphatase (G6Pase) is a microsomal enzyme which is very sensitive to inactivation by lipid peroxidation. Experiments were carried out to evaluate whether ferritin, which is the major storage form of iron within cells, could catalyze inactivation of G6Pase and to determine the mechanism responsible for this effect of ferritin. Incubation of microsomes with NADPH in the absence of ferritin led to decreased activity of G6Pase. Ferritin stimulated this inactivation of G6Pase in a time- and concentration-dependent manner. Ferritin did not stimulate G6Pase inactivation when NADH replaced NADPH as the microsomal reductant. Superoxide dismutase but not catalase or DMSO prevented the ferritin-stimulated inactivation of G6Pase suggesting a role for superoxide, but not H2O2 or hydroxyl radical, in the overall mechanism. Trolox, at concentrations which prevent lipid peroxidation, also prevented the ferritin-catalyzed inactivation of G6Pase. Inhibition of G6Pase by ferritin was further enhanced in the presence of ATP but was inhibited in the presence of EDTA or desferrioxamine; ferric-ATP stimulates, whereas ferric-EDTA inhibits microsomal lipid peroxidation. The redox cycling agent paraquat increased the ability of ferritin to inactivate G6Pase by a reaction prevented by superoxide dismutase, trolox, EDTA, and desferrioxamine, but not by catalase or DMSO. Ferritin stimulated microsomal light emission, a reaction reflecting lipid peroxidation, with time and concentration dependence, and sensitivity to scavengers (trolox, superoxide dismutase), iron chelators and paraquat, identical to the inactivation of G6Pase. These results indicate that one possible toxicological consequence of ferritin-catalyzed lipid peroxidation is inhibition of microsomal enzymes such as G6Pase.

Stimulation of microsomal chemiluminescence by ferritin.

The ability of ferritin to catalyze rat liver microsomal chemiluminescence was determined in the absence and presence of the redox cycling agent paraquat, and with either NADPH or NADH as reductant. Microsomal chemiluminescence was used as a index of lipid peroxidation. In the absence of added ferritin, NADPH-dependent microsomal light emission was 4-fold greater than the NADH-dependent reaction, and was not sensitive to superoxide dismutase, catalase or DMSO. Ferritin stimulated NADPH-, but not NADH-dependent chemiluminescence in a time- and concentration-dependent manner. The stimulation by ferritin was completely sensitive to superoxide dismutase, but not to catalase or DMSO, suggesting the requirement for superoxide to mobilize iron from ferritin. An iron ligand was not required for the stimulation by ferritin; the addition of certain ligands such as EDTA, DETAPAC or desferrioxamine resulted in inhibition of the stimulation by ferritin. Paraquat potentiated the effect of ferritin on microsomal chemiluminescence with NADPH as cofactor and was weakly stimulatory with NADH. The potentiation by paraquat plus ferritin was prevented by superoxide dismutase and was further elevated by ligands such as ATP. Chemiluminescence proved to be a more sensitive parameter than production of thiobarbituric acid-reactive components to evaluate the stimulation of oxygen radical production by iron released from ferritin, in the absence or in the presence of paraquat.

Inhibition of the oxidation of hydroxyl radical scavenging agents after alkaline phosphatase treatment of rat liver microsomes.

Treatment of rat liver microsomes with alkaline phosphatase results in a loss in the FMN but not the FAD flavin prosthetic group of NADPH-cytochrome P-450 reductase (Taniguchi, H. and Pyerin, W. (1987) Biochim. Biophys. Acta 912, 295-307). Experiments were carried out to evaluate the effect of preventing electron transfer from the FADH2 to FMN component of the reductase, and subsequent mixed function oxidase activity, on reduction of ferric chelates, production of H2O2, and the generation of .OH-like species by microsomes. Treatment with alkaline phosphatase was confirmed to decrease NADPH-cytochrome c, but not NADPH-ferricyanide, reductase activity by microsomes and by purified NADPH cytochrome P-450 reductase. The oxidation of hydroxyl radical scavenging agents by microsomes and reductase was decreased by the alkaline phosphatase treatment in accordance with the decline in cytochrome c reductase activity. This decrease in hydroxyl radical production occurred in the presence of various ferric chelate catalysts. Rates of microsomal reduction of the ferric chelates were also inhibited after alkaline phosphatase treatment. Production of H2O2 was decreased in accordance to the fall in cytochrome c reductase activity and .OH production. Rates of H2O2 production appeared to be rate-limiting for the overall generation of .OH as the addition of an external H2O2-generating system stimulated .OH production as well as prevented the decline in .OH production caused by the alkaline phosphatase treatment. These results suggest that both the FAD and FMN flavin prosthetic groups of the reductase contribute towards the reduction of various ferric chelates. However, loss of the FMN component and activities dependent on electron transfer from this prosthetic group result in a decrease in H2O2 production, which appears to be responsible for the decline in the generation of .OH-like species by microsomes after treatment with alkaline phosphatase.