Protection against oxygen toxicity by intravenous injection of liposome-entrapped catalase and superoxide dismutase.

Survival of rats exposed to 100% oxygen was increased from 69.5 +/- 1.5 to 118.1 +/- 9.9 h (mean +/- SEM, P less than 0.05) when liposomes containing catalase and superoxide dismutase were injected intravenously before and during exposure. The increased survival time in 100% oxygen was also associated with significantly less fluid in the pleural cavity. Rats injected with catalase- and superoxide dismutase-containing liposomes, which had increased survival in 100% oxygen, had increased lung wet weight upon autopsy compared with saline-injected controls (2.9 +/- 0.2 g/lung vs. 4.8 +/- 0.4 g/lung, mean +/- SE, P less than 0.05). Intravenous injection of control liposomes along with catalase and superoxide dismutase in the suspending buffer decreased the mean pleural effusion volume 89% and had no significant effect on survival time. Lung catalase and superoxide dismutase activities were increased 3.1- and 1.7-fold, respectively, 2 h after a single intravenous injection of liposomes containing catalase or superoxide dismutase. Superoxide dismutase activity was also significantly greater than controls in both air- and 100% oxygen-exposed rat lungs, when enzyme activity was assayed 24 h after cessation of injection of control and oxygen-exposed rats with enzyme-containing liposomes every 12 h for 36 h. Free superoxide dismutase and catalase injected intravenously in the absence of liposomes did not increase corresponding lung enzyme activities, affect pleural effusion volume, lung wet weight, or extend the mean survival time of rats exposed to 100% oxygen. The clearance of liposome-augmented 125I-labeled catalase from lung and plasma obeyed first order kinetics according to a one-compartment model. When clearance of liposome-augmented catalase activity or radioactivity were the parameters used for pharmacokinetic studies, the half-life of augmented lung catalase was 1.9 and 2.6 h, respectively. The half-life of liposome-entrapped catalase and superoxide dismutase activity in the circulation was 2.5 and 4 h, respectively, while intravenously injected catalase and superoxide dismutase had a circulation half-life of 23 and 6 min, respectively.

Evidence for an essential role of reactive oxygen species in the genesis of late preconditioning against myocardial stunning in conscious pigs.

Conscious pigs underwent a sequence of 10 2-min coronary occlusions, each separated by 2 min of reperfusion, for three consecutive days (days 1, 2, and 3). On day 1, pigs received an i.v. infusion of a combination of antioxidants (superoxide dismutase, catalase, and N-2 mercaptopropionyl glycine; group II, n = 9), nisoldipine (group III, n = 6), or vehicle (group I [controls], n = 9). In the control group, systolic wall thickening (WTh) in the ischemic-reperfused region on day 1 remained significantly depressed for 4 h after the 10th reperfusion, indicating myocardial "stunning." On days 2 and 3, however, the recovery of WTh improved markedly, so that the total deficit of WTh decreased by 53% on day 2 and 56% on day 3 compared with day 1 (P < 0.01), indicating the development of a powerful cardioprotective response (late preconditioning against stunning). In the anti-oxidant-treated group, the total deficit of WTh on day 1 was 54% less than in the control group (P < 0.01). On day 2, the total deficit of WTh was 85% greater than that observed on day 1 and similar to that observed on day 1 in the control group. On day 3, the total deficit of WTh was 58% less than that noted on day 2 (P < 0.01). In the nisoldipine-treated group, the total deficit of WTh on day 1 was 53% less than that noted in controls (P < 0.01). On days 2 and 3, the total deficit of WTh was similar to the corresponding values in the control group. These results demonstrate that: (a) in the conscious pig, antioxidant therapy completely blocks the development of late preconditioning against stunning, indicating that the production of reactive oxygen species (ROS) on day 1 is the mechanism whereby ischemia induces the protective response observed on day 2; (b) antioxidant therapy markedly attenuates myocardial stunning on day 1, indicating that ROS play an important pathogenetic role in postischemic dysfunction in the porcine heart despite the lack of xanthine oxidase; (c) although the administration of a calcium-channel antagonist (nisoldipine) is as effective as antioxidant therapy in attenuating myocardial stunning on day 1, it has no effect on late preconditioning on day 2, indicating that the ability of antioxidants to block late preconditioning is not a nonspecific result of the mitigation of postischemic dysfunction on day 1. Generation of ROS during reperfusion is generally viewed as a deleterious process. Our finding that ROS contribute to the genesis of myocardial stunning but, at the same time, trigger the development of late preconditioning against stunning supports a complex pathophysiological paradigm, in which ROS play an immediate injurious role (as mediators of stunning) followed by a useful function (as mediators of subsequent preconditioning).

How relevant is the reoxidation of ferrocytochrome c by hydrogen peroxide when determining superoxide anion production?

In a recent publication [(1987) FEBS Lett. 210, 195-198] the authors claim the use of cytochrome c to detect superoxide anion underestimates the real rate of superoxide anion formation on the basis that: (i) the rate of uric acid formation by xanthine oxidase is about 4-fold faster than the rate of cytochrome c reduction and (ii) hydrogen peroxide formed upon dismutation of the superoxide anion generated by xanthine oxidase is capable of reoxidizing ferrocytochrome c. That paper may have been misleading for readers not very familiar with the field of oxygen radicals, since both assumptions are, in fact, incorrect. In this report we demonstrate that the build up in concentration of H2O2 during most reactions in which superoxide anion is being produced is not enough to affect the rate of cytochrome c reduction. Our results suggest that the authors may have been misled by an artifact due to exposure of the samples containing H2O2 to UV light, which generates hydroxyl radicals by photolysis.

Ascorbate- and hemoglobin-dependent brain chemiluminescence.

It has been indicated recently that ascorbic acid is responsible for the hemoglobin-mediated oxidative damage to the central nervous system (Sadrzadeh & Eaton, J. Clin. Invest. 82:1510-1515, 1988). In this paper we describe the changes in chemiluminescence accompanying hemoglobin- and ascorbate-dependent oxidative injury to brain tissue. Addition of either hemoglobin (15 microM) or ascorbate (1 or 2 mM) to rat brain homogenates stimulated spontaneous chemiluminescence in a synergistic manner. This increase in chemiluminescence was inhibited by desferrioxamine indicating that free iron was involved in the reactions leading to lipid peroxidation. Preincubation with ascorbate oxidase inhibited both spontaneous and hemoglobin-dependent chemiluminescence, suggesting that ascorbate was required for the reactions leading to lipid peroxidation. Supplementation with aminotriazole (an irreversible inhibitor of the catalase-H2O2 complex) increased chemiluminescence in a time-dependent manner, as catalase reacted with accumulated H2O2, suggesting that ascorbic acid has a dual action being involved in the production of H2O2 and also maintaining Fe in the reduced state to catalyze a Fenton-like reaction. The excited species responsible for the chemiluminescence were partially characterized by adding specific fluorescent energy acceptors: dibromoanthracene (DBA) and diphenylanthracene (DPA). Both DBA and DPA stimulated chemiluminescence several-fold indicating that triplet and singlet species are responsible for the observed chemiluminescence. Excited singlet carbonyls (identified with DPA) may be produced during the collision of two ROO.. Singlet oxygen may also be generated during the same reaction. It decays to the triplet state (emitting chemiluminescence at 634 nm) and reacts with double bonds producing dioxetanes, which may breakdown generating triplet carbonyls (identified with DBA).(ABSTRACT TRUNCATED AT 250 WORDS)

Oxygen-concentration dependence of microsomal chemiluminescence.

The effect of varying concentrations of oxygen on NADPH-dependent microsomal chemiluminescence was determined. Light emission increased as the concentration of oxygen was elevated from 0 to 10 to 20%, and then began to decrease upon further increases in oxygen concentration to 50 and 100%. This biphasic response of chemiluminescence is similar to that previously observed for microsomal generation of hydroxyl radical, however, the light emission was not sensitive to superoxide dismutase, catalase or benzoate confirming the lack of a role for .OH in the light emission. The biphasic nature of the response of chemiluminescence is similar to that reported for exhalation of ethane and pentane but not that of malondialdehyde as a measure of lipid peroxidation, although the concentrations of O2 to reach the maximum effect differ. Activity of NADPH-cytochrome P450 reductase was decreased at the elevated concentrations of O2. The biphasic response of chemiluminescence to O2 appears to reflect the need for a critical amount of O2 to generate the initiating oxidizing species, and the effect of O2 on the appropriate redox state of the iron catalyst.

Roles of catalase and cytochrome c in hydroperoxide-dependent lipid peroxidation and chemiluminescence in rat heart and kidney mitochondria.

A recent report (Radi et al., J. Biol. Chem. 266:22028-22034, 1991) showed that rat heart mitochondria contain catalase. The protective role of mitochondrial catalase was tested by exposing heart or kidney mitochondria and mitoplasts to two oxidants (H2O2) or tert-butyl hydroperoxide, t-BOOH), estimating lipid peroxidation (as thiobarbituric acid-reactive substances, TBARS) and overall oxidative stress (as chemiluminescence). Additional controls included heart and kidney preparations from aminotriazole-treated (catalase-depleted) rats. Both oxidants increased TBARS in catalase-free preparations to similar extents over their respective controls (between 200 to 350%). In catalase-containing preparations, H2O2 lipid peroxidation increased by only 40 to 96% over controls. Similar qualitative results were obtained when measuring chemiluminescence. The catalytic role of cytochrome c in mitochondrial lipid peroxidation was investigated by exposing either control or cytochrome-c-depleted kidney mitoplasts (catalase free) to either H2O2 or t-BOOH. Hydrogen-peroxide-dependent mitochondrial lipid peroxidation varied with cytochrome c concentration, remaining close to controls when cytochrome c concentration decreased by 66%, even though there was no catalase present. Tert-butyl hydroperoxide-dependent lipid peroxidation was less affected by cytochrome c remaining 2.3-fold above controls under the same conditions, suggesting that organic peroxides are more likely to remain in the less polar membrane environment being decomposed by heme or nonheme iron imbedded in the inner mitochondrial membrane. Chemiluminescence was less affected by cytochrome c depletion. Comparing control and cytochrome-c-deficient mitochondria, chemiluminescence was 1.7-fold and 2.8-fold higher when control preparations were challenged with t-BOOH or H2O2, respectively.

Spontaneous lung chemiluminescence upon paraquat administration.

In vivo rat lung chemiluminescence was measured at different times after a single injection of either 30 or 60 mg paraquat/kg b.w. The lungs were isolated to determine myeloperoxidase (index of polymorphonuclear leukocytes), lung wet weight (lung edema) and malondialdehyde (lipid peroxidation). The highest chemiluminescence was reached 30 hours after injection of 30 mg/kg or 6 hours after a 60 mg/kg dose. The peak chemiluminescence was coincident with the maximum concentration of myeloperoxidase and lung wet weight suggesting that most chemiluminescence was the consequence of polymorphonuclear activation after migration to the injured areas.

Antioxidant enzyme status of ischemic and postischemic liver and ischemic kidney in rats.

The specific activity of seven enzymes involved in protecting tissue from oxidative stress was determined in rat kidneys subjected to 0, 2, 4, or 8 h of normothermic ischemia and in isolated rat livers during control perfusion, after 2 h ischemia, and after 2 h ischemia plus 1 h of reperfusion. In general, none of the antioxidant enzymes measured showed any consistent variation throughout the ischemic period even though mitochondrial function was significantly decreased, indicating substantial cell injury. Glutathione peroxidase (Se-GSH-Px) activity remained constant during 8 h of ischemia, although a small (29%) increase above control activity was noted at 4 h of ischemia. Se-independent GSH-Px activity (non-Se-GSH-Px) and glutathione reductase (GSSG-Red) remained constant up to 8 h of ischemia, when we measured an increase of 158% above controls in non-Se-GSH-Px and a decrease of 35% relative to controls in GSSG-Red. In perfused livers, the only change in enzyme activity after 2 h of ischemia was an increased GSSG-Red activity of 21% above control. This increase persisted into the reperfusion phase (35% above control activity) and was accompanied by decreases in both forms of GSH-Px (28% Se-GSH-Px and 44% non-Se-GSH-Px).

Purification of cytochrome c peroxidase for monitoring H2O2 production.

One of the most precise methods of determining hydrogen peroxide (H2O2) formation by biological systems is based on measuring the rate of enzyme-substrate complex formation between H2O2 and cytochrome c peroxidase (CCP). The main problem with this method is that CCP is not commercially available and has to be prepared in the laboratory. We have modified some currently available methods for purifying a highly active preparation of CCP in about 4 d. It includes a batch extraction of protein using DEAE-sepharose followed by concentration either by lyophilization or by passing the extract through a small DEAE-sepharose column instead of by ultrafiltration. The concentrated preparation is passed through a Sephadex G-75 column and the final CCP crystallized against water. The final preparations had a purity index (PI, ratio of absorbance at 408 nm/280 nm, equivalent to heme/protein ratio) above 1.2. These changes make the overall procedure very simple, preserving enzyme activity and spectral properties. In addition, we point out that special care has to be taken to eliminate cytochrome c from crude CCP extracts. Cytochrome c not only introduces an artifact when determining PI, but is also may act as a hydrogen donor for CCP when monitoring H2O2 formation, thus decreasing the sensitivity of this method.

Possible role of the NADH-fumarate reductase in superoxide anion and hydrogen peroxide production in Trypanosoma brucei.

Mitochondrial membranes from Trypanosoma brucei procyclic trypomastigotes generated superoxide anion and hydrogen peroxide in a 2:1 ratio when supplemented with NADH. Fumarate inhibited hydrogen peroxide formation (Ki = 16 microM) with the same affinity as it stimulated NADH-fumarate reductase activity. Superoxide anion production was also 65% inhibited by fumarate (Ki = 20 microM). The KM for NADH of the NADH-fumarate reductase (60 microM) was also similar to that for hydrogen peroxide generation in the absence of fumarate (30 microM). These results suggest that the NADH-fumarate reductase is involved as a source for free radical generation in T. brucei mitochondria.

Inhibitory action of the antitumor agent lonidamine on mitochondrial respiration of Trypanosoma cruzi and T. brucei.

The antitumor and antispermatogenic agent lonidamine inhibits Trypanosoma cruzi epimastigotes growth in culture with an ID50 around 80 microM. The main site of action appears to be the mitochondria, where the rate of uncoupled respiration was inhibited in 50% at a similar lonidamine concentration (50 microM). Hexokinase (the other point where lonidamine inhibits tumor energy metabolism) was not sensitive to this drug. Lonidamine also inhibited uncoupled respiration in T. brucei procyclic trypomastigotes, suggesting a common mechanism of action with T. cruzi. When lonidamine was added to T. brucei trypomastigotes, there was little effect on the CN-insensitive respiration, demonstrating that at least in T. brucei glycolysis is not affected by the drug.

Chemiluminescence enhancement by trypanocidal drugs and by inhibitors of antioxidant enzymes in Trypanosoma cruzi.

The spontaneous emission of chemiluminescence by Trypanosoma cruzi epimastigotes was 133 +/- 5 counts s-1 (mg protein)-1. The measured intracellular steady state concentration of hydrogen peroxide in the same cells was 1.5 +/- 0.5 microM. These two values are about 12- and 15-times higher than the corresponding ones for isolated rat hepatocytes. The intracellular steady state concentrations of superoxide radical and hydrogen peroxide were apparently increased by inhibiting superoxide dismutase (with diethyldithiocarbamate or KCN addition) and by the addition of two different trypanocidal agents (beta-lapachone and nifurtimox) capable of intracellular redox cycling and in each case an increased chemiluminescence was observed. Depletion of intracellular reduced non-protein SH groups by 80% increased 3-fold the chemiluminescence of T. cruzi cells. It is apparent that both an increase in the intracellular steady state concentration of superoxide anion or hydrogen peroxide and a decrease in the level of reduced SH groups lead to an increase in the level of peroxy radicals which are the precursor species for light emission.

Inhibitors of the mitochondrial cytochrome b-c1 complex inhibit the cyanide-insensitive respiration of Trypanosoma brucei.

The cyanide-insensitive respiration of bloodstream trypomastigote forms of Trypanosoma brucei (75 +/- 8 nmol O2 min-1(mg protein)-1) is completely inhibited by the mitochondrial ubiquinone-like inhibitors 2-hydroxy-3-undecyl-1,4-naphthoquinone (UHNQ) and 5-n-undecyl-6-hydroxy-4,7-dioxobenzothiazole (UHDBT). The Ki values for UHDBT (30 nM) and UHNQ (2 microM) are much lower than the reported Ki for salicylhydroxamic acid (SHAM) (5 microM), a widely used inhibitor of the cyanide-insensitive oxidase. UHNQ also stimulated the glycerol-3-phosphate-dependent reduction of phenazine methosulfate, demonstrating that the site of UHNQ inhibition is on the terminal oxidase of the cyanide-insensitive respiration of T. brucei. These results suggest that a ubiquinone-like compound may act as an electron carrier between the two enzymatic components of the cyanide-insensitive glycerol-3-phosphate oxidase.

Fumarate reductase and other mitochondrial activities in Trypanosoma cruzi.

Subcellular fractions obtained from Trypanosoma cruzi epimastigotes broken by freezing and thawing were assayed for fumarate reductase activity with reduced methyl viologen as electron donor and fumarate as electron acceptor under anaerobic conditions. Two distinct activities were detected: one in the mitochondrial membranes, 115 mU(mg protein)-1, accounting for 96% of the total and the other in the cytosol, 3 mU(mg protein)-1, accounting for 3% of the total. The activity of membrane-bound fumarate reductase correlated statistically with either the activity or the amount of mitochondrial markers such as succinate and NADH dehydrogenases, cytochromes b + c558, cytochrome a611 and 5,7-diene sterols in the obtained subcellular fractions (580 X g, 12 000 X g, and 105 000 X g sediments and supernatant). Mitochondrial fumarate reductase was inhibited by succinate, malonate, cyanide, and 2-thenoyltrifluoroacetone (TTFA); whereas the soluble enzyme was inhibited by succinate and not by TTFA. The 12 000 X g sediment (mitochondrial membranes) showed after dithionite addition, absorption maxima at 611, 560 and 530 nm accounting for the presence of cytochrome b560, c558 and a611. A CO-binding cytochrome o was also detected. A scheme of the T. cruzi mitochondrial respiratory chain is presented.

Succinate-dependent metabolism in Trypanosoma cruzi epimastigotes.

Trypanosoma cruzi epimastigotes permeabilized with digitonin (65 micrograms (mg protein)-1) to measure mitochondrial respiration were exposed to different substrates. Although none of the NADH-dependent substrates stimulated respiration, succinate supported not only oxygen consumption but also oxidative phosphorylation (respiratory control ratio of 1.9 +/- 0.3) indicating that the mitochondria were coupled. The rate of NADH-dependent oxygen consumption by membrane fractions (9.4 +/- 0.7 nmol min-1 (mg protein)-1) was reduced by 50% upon addition of catalase indicating that the electrons from NADH oxidation reduced oxygen to H2O2. NADH-dependent H2O2 production (16 +/- 1 nmol min-1 (mg protein)-1) was confirmed using cytochrome c peroxidase. This activity was inhibited by fumarate by 70%, suggesting a competition between fumarate and oxygen for the electrons from NADH, probably at the fumarate reductase level. The respiratory chain inhibitor antimycin blocked both respiration by intact cells and succinate-dependent cytochrome c by isolated membranes. No inhibition by antimycin was observed when NADH replaced succinate as an electron donor, indicating that the electrons from NADH oxidation reduced cytochrome c through a different route. Malonate blocked not only succinate-cytochrome c reductase and fumarate reductase, but also intact cell motility. These results suggest that succinate has a central role in the intermediate metabolism of i. cruzi, as it may be used for respiration or excreted to the extracellular space under anaerobic conditions. In addition, 2 potential sources of H2O2 were tentatively identified as: (a) the enzyme fumarate reductase; and (b) a succinate-dependent site, which may be the semiquinone form of Coenzyme Q9, as in mammalian mitochondria.

Biochemical and ultrastructural alterations produced by miconazole and econazole in Trypanosoma cruzi.

Miconazole and econazole, two fungicide imidazole derivatives, completely inhibited growth of Trypanosoma cruzi (Tulahuen strain) at concentrations of about 20 muM. Culturing of T. cruzi in the presence of lower doses of imidazole derivatives produced: decrease of 5,7-diene sterol content in epimastigotes (including ergosterol); disappearance of the nuclear chromatin, vacuolization and decrease in the electron density of the cytoplasm; selective surface alterations as revealed by an increased response to wheat-germ- and phytohemagglutinin. At variance with the effect of miconazole on Candida (De Nollin et al. (1977) Antimicrobial. Agents Chemother. 11, 500-513), miconazole and econazole, under the experimental conditions used, did not increase the rate of hydrogen peroxide generation by T. cruzi.

Negligible prevalence of antibodies against Trypanosoma cruzi among blood donors in the southeastern United States.

Trypanosoma cruzi, a hemoflagellate, causes Chagas' disease and is endemic throughout Latin America. Increasing Latin American immigration to the United States has enhanced concern about transmission of Chagas' disease by infected donor blood. The insect vector and parasites also have been found in the southeastern United States. Autochthonous infection of several species of wild and domesticated mammals suggests that the general human population also may be at risk. To assess the prevalence of antibodies to T cruzi in humans, randomly selected donor blood was screened. Initial screening was performed by indirect hemagglutination (1:4 initial serum dilution) and at least one of three different enzyme immunoassays. All samples testing positive by at least one screening method were tested by radioimmunoprecipitation and indirect immunofluorescence supplemental methods, which were used for confirmation and calculation of specificity. Of the 6,013 serum samples evaluated, 85 tested positive by one screening method. Only 10 of the samples tested positive by more than one method. The percentages of positive screening tests are 0.05% by indirect hemagglutination and 0.06%, 0.91%, 3.97% by Abbott Laboratories (Abbott Park, Ill), Gull (Gull Laboratories, Salt Lake City, Utah), and Polychaco (Polychaco S.A.I.C., Buenos Aires, Argentina) enzyme immunoassays, respectively. All samples were negative by radioimmunoprecipitation and indirect immunofluorescence. These results suggest that although parasite and vector are found in the southeastern United States and both infect mammals, the risk of natural infection to humans in this region seems to be negligible. There was variation in positivity among different screening methods. The highest percentage of positive results was with the enzyme immunoassay, in which the binding of serum antibodies to antigens is amplified by enzymatic reactions.

Activated polymorphonuclear leukocytes increase low-level chemiluminescence of isolated perfused rat lungs.

Low-level chemiluminescence was measured in isolated perfused rat lungs subjected to different types of oxidative stress: perfusion with tert-butyl hydroperoxide (t-BOOH) or stimulation of polymorphonuclear cells (PMN). The time required for t-BOOH-dependent lung chemiluminescence to return to background levels was proportional to the concentration of t-BOOH. From the half times of the decay at different t-BOOH concentrations, we estimated that the lungs metabolize organic peroxides at a rate of 0.045 mM/min. Use of a high dose of t-BOOH (3 mM)or pretreatment of lungs with 1,3-bis(2-chloroethyl)-nitrosourea (100 micrograms/ml) to inhibit glutathione reductase produced chemiluminescence that was much greater and did not decay. Stimulation of 5 x 10(7) PMN with 1 micrograms of phorbol myristate acetate resulted in significant increases in chemiluminescence that occurred in the absence of a significant lung weight gain or measurable lipid peroxidation. Perfusion of isolated lungs with superoxide dismutase (100 U/ml) completely inhibited the chemiluminescence response to PMN activation, whereas treatment with 100 microM U-74389F, a lipid-soluble antioxidant, also significantly decreased PMN-dependent chemiluminescence. Neither catalase (2,000 U/ml) nor 100 microM U-78518F, a water-soluble antioxidant, decreased chemiluminescence after PMN activation. These results indicate that low-level chemiluminescence is a sensitive indicator of oxidative stress in the isolated perfused rat lung and provides a tool for devising and characterizing the effectiveness of antioxidant interventions.

Separation of NADH-fumarate reductase and succinate dehydrogenase activities in Trypanosoma cruzi.

A recent review suggested that the activity of NADH-fumarate reductase from trypanosomatids could be catalyzed by succinate dehydrogenase working in reverse (Tielens and van Hellemond, Parasitol. Today 14, 265-271, 1999). The results reported in this study demonstrate that the two activities can easily be separated without any loss in either activity, suggesting that fumarate reductase and succinate dehydrogenase are separate enzymes.

Mercaptopyridine-N-oxide, an NADH-fumarate reductase inhibitor, blocks Trypanosoma cruzi growth in culture and in infected myoblasts.

The enzyme NADH-fumarate reductase is not found in mammalian cells but it is present in several parasitic protozoa including Trypanosoma cruzi, the parasite that causes Chagas' disease. This study shows that the drug 2-mercaptopyridine-N-oxide (MPNO) inhibits NADH-fumarate reductase purified from T. cruzi (ID50 = 35 microM). When added to intact cells, MPNO inhibited the growth of T. cruzi epimastigotes in culture (ID50 = 0.08 microM) as well as the infection of mammalian myoblasts by T. cruzi trypomastigotes (ID50 = 20 microM). At a concentration of 2.4 microM, MPNO also inhibited the growth of amastigotes (intracellular dividing forms) in cultured mammalian myoblasts. Supplementation of culture media with 5 mM succinate, the product of fumarate reductase, partially protected against the inhibition of the growth of epimastigotes by MPNO. Moreover, MPNO inhibited the accumulation of succinate in cultures of epimastigotes, as measured by high performance liquid chromatography. Although MPNO may have other intracellular targets in addition to fumarate reductase, these results support the hypothesis that compounds which inhibit the enzyme fumarate reductase may be potential chemotherapeutic agents against Chagas' disease.