Purification and properties of milk xanthine dehydrogenase.

Milk xanthine oxidase (XO) has been prepared in a dehydrogenase form (XDH) by purifying the enzyme in the presence of 2.5 mM dithiothreitol. Unlike XO, which reacts rapidly only with oxygen and not with NAD, the XDH form of the enzyme reacts rapidly with NAD. XDH has a turnover number for the NAD-dependent conversion of xanthine to urate of 380 mol/min/mol at pH 7.5, 25 degrees C, with a Km = < or = 1 microM for xanthine and a Km = 7 microM for NAD, but has very little O2-dependent activity. There is evidence that the two forms of the enzyme have different flavin environments: XDH stabilizes the neutral form of the flavin semiquinone and XO does not. Further, XDH binds the artificial flavin 8-mercapto-FAD in its neutral form, shifting the pK of this flavin by 5 pH units, while XO binds 8-mercapto-FAD in its benzoquinoid anionic form. XDH can be converted back to the XO form by the addition of three to four equivalents of the disulfide-forming reagent 4,4'-dithiodipyridine, suggesting that, in the XDH form of the enzyme, disulfide bonds are broken; this may cause a conformational change which creates a binding site for NAD and changes the protein structure near the flavin.     

Sulfhydryl oxidase-catalyzed conversion of xanthine dehydrogenase to xanthine oxidase

Xanthine oxidase may be isolated from various mammalian tissues as one of two interconvertible forms, viz., a dehydrogenase (NAD⁺ dependent, form D) or an oxidase (O₂ utilizing, form O). A crude preparation of rat liver xanthine dehydrogenase (form D) was treated with an immobilized preparation of crude bovine sulfhydryl oxidase. Comparison of the rates of conversion of xanthine dehydrogenase to the O form in the presence and absence of the immobilized enzyme indicated that sulfhydryl oxidase catalyzes such conversion. These results were substantiated in a more definitive study in which purified bovine milk xanthine oxidase, which had been converted to the D form by treatment with dithiothreitol, was incubated with purified bovine milk sulfhydryl oxidase. Comparison of measured rates of conversion (in the presence and absence of active sulfhydryl oxidase and in the presence of thermally denatured sulfhydryl oxidase) revealed that sulfhydryl oxidase enzymatically catalyzes the conversion of type D activity to type O activity in xanthine oxidase with the concomitant disappearance of its sulfhydryl groups. It is possible that the presence or absence of sulfhydryl oxidase in a given tissue may be an important factor in determining the form of xanthine-oxidizing activity found in that tissue.     

Proteolytic conversion of xanthine dehydrogenase to xanthine oxidase: evidence against a role for calcium-activated protease (calpain)

The present study tested the hypothesis that calpain is responsible for the limited proteolytic conversion of xanthine dehydrogenase (XD) to xanthine oxidase (XO). We compared the effects of various proteases on the activity and molecular weight of a purified preparation of xanthine dehydrogenase from rat liver. In agreement with previous reports, trypsin treatment produced a complete conversion of XD to XO accompanied by a limited proteolysis of XDH from an Mr of 140 kD to an Mr of 90 kD. Treatment with calpain I or calpain II did not produce a conversion from XD to XO nor did it result in partial proteolysis of the enzyme. Similarly, trypsin treatment partially degraded a reversibly oxidized form of xanthine dehydrogenase while calpain I or calpain II were ineffective. The possibility that an endogenous inhibitor prevented the proteolysis of XDH by calpain I or II was excluded by verifying that brain spectrin, a known calpain substrate, was degraded under the same incubation conditions. The results indicate that calpain is not likely to be responsible for the in vivo conversion of XD to XO under pathological conditions.     

Kinetic isotope effect studies on milk xanthine oxidase and on chicken liver xanthine dehydrogenase

The effect of isotopic substitution of the 8-H of xanthine (with 2H and 3H) on the rate of oxidation by bovine xanthine oxidase and by chicken xanthine dehydrogenase has been measured. V/K isotope effects were determined from competition experiments. No difference in H/T(V/K) values was observed between xanthine oxidase (3.59 +/- 0.1) and xanthine dehydrogenase (3.60 +/- 0.09). Xanthine dehydrogenase exhibited a larger T/D(V/K) value (0.616 +/- 0.028) than that observed for xanthine oxidase (0.551 +/- 0.016). Observed H/T(V/K) values for either enzyme are less than those H/T(V/K) values calculated with D/T(V/K) data. These discrepancies are suggested to arise from the presence of a rate-limiting step(s) prior to the irreversible C-H bond cleavage step in the mechanistic pathways of both enzymes. These kinetic complexities preclude examination of whether tunneling contributes to the reaction coordinate for the H-transfer step in each enzyme. No observable exchange of tritium with solvent is observed during the anaerobic incubation of [8-3H]xanthine with either enzyme, which suggests the reverse commitment to catalysis (Cr) is essentially zero. With the assumption of adherence to reduced mass relationships, the intrinsic deuterium isotope effect (Dk) for xanthine oxidation is calculated to be 7.4 +/- 0.7 for xanthine oxidase and 4.2 +/- 0.2 for xanthine dehydrogenase. By use of these values and steady-state kinetic data, the minimal rate for the hydrogen-transfer step is calculated to be approximately 75-fold faster than kcat for xanthine oxidase and approximately 10-fold faster than kcat for xanthine dehydrogenase. This calculated rate is consistent with data obtained by rapid-quench experiments with XO. A stoichiometry of 1.0 +/- 0.3 mol of uric acid/mol of functional enzyme is formed within the mixing time of the instrument (5-10 ms). The kinetic isotope effect data also permitted the calculation of the Kd values [Klinman, J. P., & Mathews, R. G. (1985) J. Am. Chem. Soc. 107, 1058-1060] for substrate dissociation, including all reversible steps prior to C-H bond cleavage. Values calculated for each enzyme (Kd = 120 microM) were found to be identical within experimental uncertainty.     

Myocardial xanthine oxidase/dehydrogenase

High-energy phosphates in heart muscle deprived of oxygen are rapidly broken down to purine nucleosides and oxypurines. We studied the role of xanthine oxidase/dehydrogenase (EC 1.2.3.2/EC 1.2.1.37) in this process with novel high-pressure liquid chromatographic techniques. Under various conditions, including ischemia and anoxia, the isolated perfused rat heart released adenosine, inosine and hypoxanthine, and also substantial amounts of xanthine and urate. Allopurinol, an inhibitor of xanthine oxidase, greatly enhanced the release of hypoxanthine. From the purine release we calculated that the rat heart contained about 18 mU xanthine oxidase per g wet weight. Subsequently, we measured a xanthine oxidase activity of 9 mU/g wet wt. in rat-heart homogenate. When endogenous low molecular weight inhibitors were removed by gel-filtration, the activity increased to 31 mU/g wet wt. Rat myocardial xanthine oxidase seems to be present mainly in the dehydrogenase form, which upon storage at -20 degrees C is converted to the oxidase form.     

Mammalian xanthine oxidoreductase - mechanism of transition from xanthine dehydrogenase to xanthine oxidase

Reactive oxygen species are generated by various biological systems, including NADPH oxidases, xanthine oxidoreductase, and mitochondrial respiratory enzymes, and contribute to many physiological and pathological phenomena. Mammalian xanthine dehydrogenase (XDH) can be converted to xanthine oxidase (XO), which produces both superoxide anion and hydrogen peroxide. Recent X-ray crystallographic and site-directed mutagenesis studies have revealed a highly sophisticated mechanism of conversion from XDH to XO, suggesting that the conversion is not a simple artefact, but rather has a function in mammalian organisms. Furthermore, this transition seems to involve a thermodynamic equilibrium between XDH and XO; disulfide bond formation or proteolysis can then lock the enzyme in the XO form. In this review, we focus on recent advances in our understanding of the mechanism of conversion from XDH to XO.     

Mechanism of transition from xanthine dehydrogenase to xanthine oxidase: effect of guanidine-HCL or urea on the activity

Mammalian xanthine oxidoreductase can be converted from the dehydrogenase to the oxidase form, either reversibly by formation of disulfide bridges or irreversibly by proteolytic cleavage within the xanthine oxidoreductase protein molecule. A tightly packed amino acid cluster stabilizes the dehydrogenase form, and disruption of this cluster is accompanied with rearrangement of the active site loop. Here, we show that the conversion occurs in the presence of guanidine-HCl or urea. We propose that xanthine dehydrogenase and oxidase are in a thermodynamic equilibrium that can be shifted by disruption of the amino acid cluster with a denaturant.     

Reoxygenation injury in isolated hepatocytes: cell death precedes conversion of xanthine dehydrogenase to xanthine oxidase

Reoxygenation of isolated hepatocytes from fed rats after 3 h of anaerobic incubation led to a significantly enhanced loss of cell viability. No evidence for the participation of reactive oxygen species generated by xanthine oxidase in this reoxygenation injury was found. Conversion of xanthine dehydrogenase to xanthine oxidase occurred at a time when almost all of the hepatocytes had lost their viability. Furthermore, xanthine dehydrogenase was first released from the severely injured cells and then converted to the oxidase form. The results suggest that in the intact organ participation of reactive oxygen species, generated by xanthine oxidase, in reoxygenation injury may only occur when, upon reoxygenation, hypoxic cell injury in part of the tissue has progressed to such an extent that there is a significant conversion of xanthine dehydrogenase to xanthine oxidase.     

Effects of hypoxia and ethanol on xanthine oxidase of isolated rat hepatocytes: conversion from D to O form and leakage from cells.

The combined effects of ethanol and hypoxia on the conversion of xanthine dehydrogenase (D form) to xanthine oxidase (O form) and on the leakage of the enzyme from isolated rat hepatocytes was studied. Time-dependent death of cells occurred during incubation in hypoxic conditions. Ethanol (40 mM) had only a moderate effect on viability in aerobiosis, but accelerated the loss of hypoxic cells, which was 96% after 3 h of incubation. In hypoxic conditions, the xanthine oxidase was gradually converted from D into O form. The conversion was complete in 3 h, and was accelerated by 1 mM xanthine or by ethanol, in a concentration-related manner. Hypoxia brought about a progressive leakage of xanthine oxidase from hepatocytes, which was accelerated by ethanol in a concentration-dependent manner. The enzyme found outside hepatocytes was mostly in its O form. The xanthine oxidase of hepatocytes cytosol was converted from D into O form by human plasma or serum. In all cases the conversion could be completely reverted by treatment of the extract with dithiothreitol.     

Proteolytic conversion of xanthine dehydrogenase from the NAD-dependent type to the O2-dependent type. Amino acid sequence of rat liver xanthine dehydrogenase and identification of the cleavage sites of the enzyme protein during irreversible conversion by trypsin

The primary structure of rat liver xanthine dehydrogenase (EC 1.1.1.204) was determined by sequence analysis of cDNA and purified enzyme. The enzyme consists of 1,319 amino acid residues with a calculated molecular mass of 145,034 Da, including initiation methionine, and is homologous to the previously reported Drosophila melanogaster enzyme (Lee, C. S., Curtis, D., McCarron, M., Love, C., Gray, M., Bender, W., and Chovnick, A. (1987) Genetics 116, 55-66; Keith, T. P., Riley, M. A., Kreitman, M., Lewontin, R. C., Curtis, D., and Chambers, G. (1987) Genetics 116, 67-73) with an identity of 52%. The enzyme exists originally as the NAD-dependent type in a freshly prepared sample. When the purified NAD-dependent type enzyme was digested with trypsin, it cleaved into three fragments with molecular masses of 20, 40, and 85 kDa and was irreversibly converted to the O2-dependent type. Comparison of the amino-terminal sequences of the three peptide fragments with the cDNA-deduced sequence reveals that the 20-, 40-, and 85-kDa peptide fragments correspond residues to 1-184, 185-539, and 540-1319 of the enzyme, respectively. Comparison of the 5'-p-fluorosulfonylbenzoyladenosine-labeled peptide sequence of the chicken enzyme (Nishino, T., and Nishino, T. (1989) J. Biol. Chem. 264, 5468-5473) reveals that the NAD binding site is associated with the 40-kDa fragment portion of the enzyme. Hydropathy analysis around the cysteine residues suggests that the 2Fe/2S sites are associated with the 20-kDa fragment portion of the enzyme.     

Recombinant Rhodobacter capsulatus xanthine dehydrogenase,a useful model system for the characterization of protein variants leading to xanthinuria I in humans

Rhodobacter capsulatus xanthine dehydrogenase (XDH) forms an (alphabeta)2 heterotetramer and is highly homologous to homodimeric eukaryotic XDHs. The crystal structures of bovine XDH and R. capsulatus XDH showed that the two proteins have highly similar folds. We have developed an efficient system for the recombinant expression of R. capsulatus XDH in Escherichia coli. The recombinant protein shows spectral features and a range of substrate specificities similar to bovine milk xanthine oxidase. However, R. capsulatus XDH is at least 5 times more active than bovine XDH and, unlike mammalian XDH, does not undergo the conversion to the oxidase form. EPR spectra were obtained for the FeS centers of the enzyme showing an axial signal for FeSI, which is different from that reported for xanthine oxidase. X-ray absorption spectroscopy at the iron and molybdenum K-edge and the tungsten LIII-edge have been used to probe the different metal coordinations of variant forms of the enzyme. Based on a mutation identified in a patient suffering from xanthinuria I, the corresponding arginine 135 was substituted to a cysteine in R. capsulatus XDH, and the protein variant was purified and characterized. Two different forms of XDH-R135C were purified, an active (alphabeta)2 heterotetrameric form and an inactive (alphabeta) heterodimeric form. The active form contains a full complement of redox centers, whereas in the inactive form the FeSI center is likely to be missing.     

Effects of hypoxia and reoxygenation on the conversion of xanthine dehydrogenase to oxidase in Chinese hamster V79 cells

The effects of hypoxia and reoxygenation on the conversion of xanthine dehydrogenase to the free radical-producing xanthine oxidase in Chinese hamster V79 cells have been investigated using a newly developed fluorimetric enzyme assay. Hypoxia caused an increase in xanthine oxidase activity from 25% to 80% of the total activity of xanthine oxidase and dehydrogenase. The ratio returned to normal levels within 24 h of aerobic incubation. Hypoxia caused the release of xanthine oxidase in the medium of V79 cells and an increase in total protein concentration in the medium. There was an early change induced in lipid peroxidation markers and this was inhibited by allopurinol. The effects of glucose deprivation and calcium blockers were also investigated. Fura-2 AM was found to interact with V79 cells, making it impossible to determine intracellular calcium levels in V79 cells by this reagent.     

Kinetics and stability of immobilized chicken liver xanthine dehydrogenase.

Xanthine dehydrogenase (EC 1.2.1.37) was isolated from chicken livers and immobilized by adsorption to a Sepharose derivative, prepared by reaction of n-octylamine with CNBr-activated Sepharose 4B. Using a crude preparation of enzyme for immobilization it was observed that relatively more activity was adsorbed than protein, but the yield of immobilized activity increased as a purer enzyme preparation was used. As more activity and protein were bound, relatively less immobilized activity was recovered. This effect was probably due to blocking of active xanthine dehydrogenase by protein impurities. The kinetics of free and immobilized xanthine dehydrogenase were studied in the pH range 7.5-9.1. The Km and V values estimated for free xanthine dehydrogenase increase as the pH increase; the K'm and V values for the immobilized enzyme go through a minimum at pH 8.1. By varying the amount of enzyme activity bound per unit volume of gel, it was shown that K'm is larger than Km are result of substrate diffusion limitation in the pores of the support material. Both free and immobilized xanthine dehydrogenase showed substrate activation at low concentrations (up to 2 microM xanthine). Immobilized xanthine dehydrogenase was more stable than the free enzyme during storage in the temperature range of 4-50 degrees C. The operational stability of immobilized xanthine dehydrogenase at 30 degrees C was two orders of magnitude smaller than the storage stability, t 1/2 was 9 and 800 hr, respectively. The operational stability was, however, better than than of immobilized milk xanthine oxidase (t 1/2 = 1 hr). In addition, the amount of product formed per unit initial activity in one half-life, was higher for immobilized xanthine dehydrogenase than for immobilized xanthine oxidase. Unless immobilized milk xanthine oxidase can be considerable stabilized, immobilized chicken liver xanthine dehydrogenase is more promising for application in organic synthesis.     

Reaction between sheep liver mitochondrial aldehyde dehydrogenase and various thiol-modifying reagents.

Sheep liver mitochondrial aldehyde dehydrogenase reacts with 2,2'-dithiodipyridine and 4,4'-dithiodipyridine in a two-step process: an initial rapid labelling reaction is followed by slow displacement of the thiopyridone moiety. With the 4,4'-isomer the first step results in an activated form of the enzyme, which then loses activity simultaneously with loss of the label (as has been shown to occur with the cytoplasmic enzyme). With 2,2'-dithiodipyridine, however, neither of the two steps of the reaction has any effect on the enzymic activity, showing that the mitochondrial enzyme possesses two cysteine residues that must be more accessible or reactive (to this reagent at least) than the postulated catalytically essential residue. The symmetrical reagent 5,5'-dithiobis-(1-methyltetrazole) activates mitochondrial aldehyde dehydrogenase approximately 4-fold, whereas the smaller related compound methyl l-methyltetrazol-5-yl disulphide is a potent inactivator. These results support the involvement of mixed methyl disulphides in causing unpleasant physiological responses to ethanol after the ingestion of certain antibiotics.     

The regulation of rat liver xanthine oxidase. Involvement of thiol groups in the conversion of the enzyme activity from dehydrogenase (type D) into oxidase (type O) and purification of the enzyme

1. The ;xanthine oxidase' activity of rat liver supernatant, most of which behaves as an NAD(+)-dependent dehydrogenase (type D) can be rapidly converted into an oxidase (type O) by thiol reagents such as tetraethylthiuram disulphide, copper sulphate, 5,5'-dithiobis-(2-nitrobenzoic acid), N-ethylmaleimide and p-hydroxymercuribenzoate. Treatment with copper sulphate, if prolonged, leads to almost complete inactivation of the enzyme. The effect of these reagents is prevented by dithioerythritol, and in all cases but that of N-ethylmaleimide is reversed by the same thiol. 2. Dithioerythritol prevents and reverses the conversion of xanthine oxidase from type D into type O brought about by storage of rat liver supernatant at -20 degrees C, preincubation under anaerobic conditions, treatment with carbon or with diethyl ether, and reverses, but does not prevent, the conversion obtained by preincubation of the whole liver homogenate. 3. Conversion of the enzyme from type D into type O is effected by preincubation of rat liver supernatant with the sedimentable fraction from rat liver but not from chick or pigeon liver. The xanthine dehydrogenase activity of chick liver supernatant is not changed into an oxidase by preincubation with the sedimentable fraction from rat liver. 4. The enzyme activity of rat liver supernatant is converted from type D into type O during purification of the enzyme: the purified enzyme can be reconverted into type D by dithioerythritol. 5. The enzyme appears as an oxidase in the supernatant of rat heart, intestine, spleen, pancreas, lung and kidney. The enzyme of all organs but intestine can be converted into a dehydrogenase by dithioerythritol.     

Selenium-containing xanthine dehydrogenase from Eubacterium barkeri.

A specific dehydrogenase, different from nicotinic acid hydroxylase, was induced during growth of Eubacterium barkeri on xanthine. The protein designated as xanthine dehydrogenase was enriched 39-fold to apparent homogeneity using a three-step purification scheme. It exhibited an NADP-dependent specific activity of 164 micromol xanthine oxidized per min and per mg of protein. In addition it showed an NADPH-dependent oxidase and diaphorase activity. A molecular mass of 530 kDa was determined for the native enzyme and SDS/PAGE revealed three types of subunits with molecular masses of 17.5, 30 and 81 kDa indicating a dodecameric native structure. Molybdopterin was identified as the molybdenum-complexing cofactor using activity reconstitution experiments and fluorescence measurements after KI/I2 oxidation. The molecular mass of the cofactor indicated that it is of the dinucleotide type. The enzyme contained iron, acid-labile sulfur, molybdenum, tungsten, selenium and FAD at molar ratios of 17.5, 18.4, 2.3, 1.1, 0.95 and 2.8 per mol of native enzyme. Xanthine dehydrogenase was inactivated upon incubation with arsenite, cyanide and different purine analogs. Reconstitution experiments of xanthine dehydrogenase activity by addition of selenide and selenite performed with cyanide-inactivated enzyme and with chloramphenicol-treated cells, respectively, indicated that selenium is not attached to the protein in a covalently bound form such as selenocysteine.     

Comparative study of chicken liver xanthine dehydrogenase and bovine liver xanthine oxidase. dehydrogenase activity of xanthine oxidase (author's transl)]

A method to purify bovine liver xanthine oxidase in described, with which samples of 256-fold specific activity with respect to the initial homogenate are obtained. Bovine liver xanthine oxidase and chicken liver xanthine dehydrogenase with oxygen as electron acceptor exhibit similar profile in pKM and log V versus pH plots. With NAD+ as electron acceptor a different profile in the pKM xanthine plot is obtained for chicken liver xanthine dehydrogenase. However three inflection points at the same pH values appear in all plots. Both enzymes are irreversibly inhibited by pCMB and reversibly by N-ethylmaleimide and by iodoacetamide, with competitive and uncompetitive type inhibitions respectively. These results suggest that NAD+ alters the enzymatic action since its binding to the enzyme antecedes the binding of xanthine to the xanthine oxidase molecule, without undergoing itself any modification. 0.15 M DDT of DTE treatment of bovine liver xanthine oxidase gives to the enzyme a permanent activity with NAD+ without modifying its activity with oxygen. The enzyme thus treated produces parallel straight lines in Lineweaver-Burk plots.     

Multiple molecular forms of xanthine dehydrogenase and related enzymes. IV. The relationship of aldehyde oxidase to xanthine dehydrogenase

Variants of the enzyme aldehyde oxidase in Drosophila melanogaster are described. In addition to electrophoretic variants, a mutant that causes low levels of the enzyme has been found by screening more than 80 strains for aldehyde oxidase levels. The locus of the mutation maps on the third chromosome near lpo and aldox. The existence of the ry, lpo, and aldox mutants and of the new mutant indicates that xanthine dehydrogenase, pyridoxal oxidase, and aldehyde oxidase are under a separate genetic control, in addition to a common genetic control by ma-l and lxd. The genetic separation is shown to be accompanied by physical separation of the enzymes with DEAE-cellulose column chromatography and (NH ₄)₂SO₄fractionation. Further data on the metabolism of aldehydes by xanthine dehydrogenase and aldehyde oxidase are presented. Although xanthine dehydrogenase requires NAD or a similar cofactor to metabolize purine and pteridine substrates, aldehyde oxidase oxidizes salicylaldehyde to salicylic acid without dissociable cofactors and with the uptake of oxygen.This work was supported in part by Research Grant GM-08202, by a Predoctoral Fellowship (J.C.) and a Genetics Training Grant (J.C. and E.D.), and by a Research Career Development Award (E.G.), all from the National Institutes of Health. Part of this work was submitted by J.C. to the University of North Carolina at Chapel Hill in partial fulfillment of the degree of Doctor of Philosophy.     

A conserved cysteine in molybdenum oxotransferases

The amino acid sequences of peptides derived from rat hepatic sulfite oxidase have been determined by a combination of amino acid analysis and Edman degradation of the purified protein. The data obtained showed the rat liver enzyme contained 3 cysteine residues which was confirmed by thiol modification studies using 4,4'-dithiodipyridine of the native enzyme. Combining these data with that previously published for chicken liver sulfite oxidase (Neame, P. J., and Barber, M. J. (1989) J. Biol. Chem. 264, 20894-20901) indicates that 2 cysteines (Cys186 and Cys430, based upon the numbering for the chicken sequence) are conserved in both chicken and rat liver enzymes with all the cysteine residues being present in the molybdenum-containing domain. Further comparison of the sequences of the molybdenum domains of rat and chicken liver sulfite oxidase with the amino acid sequences published for the molybdenum domains of a variety of assimilatory nitrate reductases suggests that only a single cysteine residue (Cys186) is conserved in all these enzymes, indicating that it may play a role in the binding of Mo-pterin to the protein.     

The active site cysteine of the proapoptotic protein glyceraldehyde-3-phosphate dehydrogenase is essential in oxidative stress-induced aggregation and cell death

Recent studies have revealed that the redox-sensitive glycolytic enzyme, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), is involved in neuronal cell death that is triggered by oxidative stress. GAPDH is locally deposited in disulfide-bonded aggregates at lesion sites in certain neurodegenerative diseases. In this study, we investigated the molecular mechanism that underlies oxidative stress-induced aggregation of GAPDH and the relationship between structural abnormalities in GAPDH and cell death. Under nonreducing in vitro conditions, oxidants induced oligomerization and insoluble aggregation of GAPDH via the formation of intermolecular disulfide bonds. Because GAPDH has four cysteine residues, including the active site Cys(149), we prepared the cysteine-substituted mutants C149S, C153S, C244A, C281S, and C149S/C281S to identify which is responsible for disulfide-bonded aggregation. Whereas the aggregation levels of C281S were reduced compared with the wild-type enzyme, neither C149S nor C149S/C281S aggregated, suggesting that the active site cysteine plays an essential role. Oxidants also caused conformational changes in GAPDH concomitant with an increase in beta-sheet content; these abnormal conformations specifically led to amyloid-like fibril formation via disulfide bonds, including Cys(149). Additionally, continuous exposure of GAPDH-overexpressing HeLa cells to oxidants produced disulfide bonds in GAPDH leading to both detergent-insoluble and thioflavin-S-positive aggregates, which were associated with oxidative stress-induced cell death. Thus, oxidative stresses induce amyloid-like aggregation of GAPDH via aberrant disulfide bonds of the active site cysteine, and the formation of such abnormal aggregates promotes cell death.