Bipyridylium quaternary salts and related compounds. V. Pulse radiolysis studies of the reaction of paraquat radical with oxygen. Implications for the mode of action of bipyridyl herbicides

Effects of oxygen pressure and medium volume on the toxicity of paraquat in rat and human type II pneumocytes

The herbicide, paraquat is highly toxic for mammals, with the lungs being the main target organ, because of the active accumulation of the compound in this organ. The cellular toxicity of paraquat has been shown to be an O2-driven process and hyperoxia is known to increase the lethality of paraquat. In this study we have examined the effect of various O2 concentrations on the toxicity of paraquat in rat and human type II pneumocytes in culture, and we have tested whether the thickness of the liquid layer above the cells would influence the toxicity of paraquat. Type II pneumocytes were isolated from rat or human lung tissue using trypsin digestion, percoll density gradient centrifugation and differential attachment. Adherent cells (day 2) were incubated for 20 h in different volumes of culture medium (thickness of liquid layer), whether or not in the presence of paraquat, in the presence of different O2 tensions. The viability of the cells was assessed by the release of LDH in the culture medium. In both rat and human type II pneumocytes the toxicity of paraquat was independent of the thickness of the liquid layer (2.5 to 10 mm height). The toxicity of paraquat in rat type II pneumocytes decreased from a TC50 value of 28 microM paraquat at 21% O2 to 107 microM at 10% O2 and increased to 12 microM and 8 microM at 60% and 85% O2, respectively. For human type II pneumocytes the TC50 values were 7 microM; 25 microM and > 1000 microM paraquat at 60%, 21% and 10% O2, respectively. In this study we have shown that the diffusion of O2 through a liquid layer does not limit the toxicity of paraquat and that, as in vivo, increasing O2 partial pressure enhances the toxicity of paraquat.

Paraquat-induced formation of leukotriene B4 in rat lungs: modulation by N-acetylcysteine

The present work is focused on the formation of the inflammatory mediator leukotriene B4 (LTB4) in the lungs of paraquat (PQ)-intoxicated rats. The levels of LTB4 and the number of neutrophils in lung lavages of PQ-intoxicated rats, measured 12 h after 30 mg/kg PQ, increased significantly compared with those of control animals; administration of 50 mg/kg IP N-acetylcysteine (NAC), 8 h after PQ, inhibited this effect. The release of LTB4 from alveolar macrophages (AM) or alveolar epithelial type II cells from healthy animals incubated with PQ and/or NAC did not offer' an explanation for the effect of these chemicals on LTB4 in the bronchoalveolar lavage fluid (BALF). The PQ-enhanced, NAC-inhibited release of arachidonic acid (AA) by alveolar epithelial type II cells did, however, explain our in vivo results, when one assumes that the AM synthesize their 5-lipoxygenase products from alveolar epithelial cell-derived AA, an hypothesis demonstrated already by other researchers.

Isoprostanes--prostaglandin-like compounds formed in vivo independently of cyclooxygenase: use as clinical indicators of oxidant damage

F2-isoprostanes are prostanoids produced independently of cyclooxygenase by free radical-catalyzed peroxidation of arachidonic acid-containing lipids. Quantification of F2-isoprostanes from biologic fluids and tissues represents an important advance in the detection and measurement of lipid peroxidation in vivo. In addition, efforts to understand both the biophysical effects of isoprostane containing lipids and the biologic effects of free isoprostanes should lead to a better understanding of the mechanisms responsible for oxidant stress-related alterations in homeostasis. Continued application of F2-isoprostane measurement in experimental models of free radical-induced injury and human disease may allow better design and evaluation of antioxidant therapeutic strategies.

N-acetylcysteine increases the glutathione content and protects rat alveolar type II cells against paraquat-induced cytotoxicity

A protective effect of N-acetylcysteine in oxidative lung damage was reported by a number of workers; however, the mechanism underlying this effect was not thoroughly elucidated. The present research investigates the protection by N-acetylcysteine against paraquat-induced cytotoxicity to alveolar type II cells, which are known to be specific targets of paraquat toxicity in vivo. We found that addition of 1 mM N-acetylcysteine to alveolar type II cells incubated with 1 mM paraquat reduced the cytotoxic index from 17.4 +/- 1.3% to 9.3 +/- 1.5%. This effect could not be explained by the interference of N-acetylcysteine with the active uptake of paraquat by type II cells. Incubation of these cells with N-acetylcysteine enhances their glutathione content, thus reducing the paraquat- induced depletion of glutathione in these cells. These results suggest that N-acetylcysteine exerts its protective effect by acting as a precursor for glutathione in alveolar type II cells.

The role of the neurohormone melatonin as a buffer against macromolecular oxidative damage

This paper summarizes the recent findings which show that the neural hormone melatonin is a free radical scavenger and general antioxidant. When compared with other antioxidants melatonin seems to have greater efficacy in protecting against cellular oxidative stress. These findings illustrate that melatonin preserves macromolecules including DNA, protein and lipid from oxidative damage following the administration of the chemical carcinogen, safrole, after exposure to ionizing radiation, following glutathione depletion, and after administration of the free radical generating herbicide, paraquat. In vitro evidence shows that melatonin is a potent scavenger of the highly toxic hydroxyl radical and in vitro evidence suggests that melatonin is an important and powerful antioxidant. Considering its high lipophilicity and its non-toxic nature as well as its ability to readily cross the blood-brain barrier, the neurohormone melatonin may prove to be an effective and important molecule in the antioxidative defense system, especially in the central nervous system. Besides the ease with which melatonin enters the brain, neurons seem to accumulate readily this hormone.

Chilling-enhanced photooxidation: The production, action and study of reactive oxygen species produced during chilling in the light

Chilling-enhanced photooxidation is the light- and oxygen-dependent bleaching of photosynthetic pigments that occurs upon the exposure of chilling-sensitive plants to temperatures below approximately 10 °C. The oxidants responsible for the bleaching are the reactive oxygen species (ROS) singlet oxygen ((1)O2), superoxide anion radical (O 2 (∸) ,hydrogen peroxide (H2O2), the hydroxyl radical (OH·), and the monodehydroascorbate radical (MDA) which are generated by a leakage of absorbed light energy from the photosynthetic electron transport chain. Cold temperatures slow the energy-consuming Calvin-Benson Cycle enzymes more than the energy-transducing light reactions, thus causing leakage of energy to oxygen. ROS and MDA are removed, in part, by the action of antioxidant enzymes of the Halliwell/Foyer/Asada Cycle. Chloroplasts also contain high levels of both lipid- and water-soluble antioxidants that act alone or in concert with the HFA Cycle enzymes to scavenge ROS. The ability of chilling-resistant plants to maintain active HFA Cycle enzymes and adequate levels of antioxidants in the cold and light contributes to their ability to resist chilling-enhanced photooxidation. The absence of this ability in chilling-sensitive species makes them susceptible to chilling-enhanced photooxidation. Chloroplasts may reduce the generation of ROS by dissipating the absorbed energy through a number of quenching mechanisms involving zeaxanthin formation, state changes and the increased usage of reducing equivalents by other anabolic pathways found in the stroma. During chilling in the light, ROS produced in chilling-sensitive plants lower the redox potential of the chloroplast stroma to such a degree that reductively-activated regulatory enzymes of the Calvin Cycle, sedohepulose 1,7 bisphosphatase (EC 3.1.3.37) and fructose 1,6 bisphosphatase (EC 3.1.3.11), are oxidatively inhibited. This inhibition is reversible in vitro with a DTT treatment indicating that the enzymes themselves are not permanently damaged. The inhibition of SBPase and FBPase may fully explain the inhibition in whole leaf gas exchange seen upon the rewarming of chilling-sensitive plants chilled in the light. Methods for the study of ROS in chilling-enhanced photooxidation and challenges for the future are discussed.

The role of free radicals in paraquat-induced corneal lesions

Paraquat is a synthetic bipyridylium salt widely used as herbicide and defoliant. Enzyme-catalyzed redoxcycling of paraquat generates oxygen radicals. The toxic, even lethal, effects of paraquat are due to free radical-mediated tissue injury. Ocular lesions, sometimes quite severe, have been observed following accidental splashing of paraquat solutions onto the eyes. These studies were designed to document the generation of paraquat free radicals in corneal tissue, and to describe the histological nature of the corneal injuries in experimental animals (rabbits and monkeys). The EPR spectrum of rabbit corneas, 30 min. after intrastromal injection of paraquat, showed the signal of the free radical of paraquat. Ultrastructural studies of corneas 8 days after intrastromal injections (100 microliters) of paraquat solutions showed that the initial lesions occur at the epithelium/basement membrane interface. In rabbit cornea, dose dependent lesions were observed, i.e. whereas 50 mM paraquat caused only minimal damage to the epithelial basement membrane, 75 mM caused complete dissolution to the basement membrane with some damage to stromal collagen, and loss of epithelium with stromal ulceration and severe inflammatory response were observed with 150 mM paraquat. Monkey corneas were less susceptible than those of rabbits to the effects of paraquat. No lesions were observed following intrastromal injections of 50 mM or 75 mM paraquat. With higher concentrations of paraquat (100 mM and 150 mM) the primary injuries were to the proximal and lateral plasma membranes of basal epithelial cells; basement membrane alterations were detected only adjacent to areas of significant plasma membrane damage. The underlying Bowman's membrane and stroma were not affected. Anatomical differences between the corneas of rabbit and monkeys as well as possible biochemical differences may account for the species differences observed.

A review of the evidence supporting melatonin's role as an antioxidant

This survey summarizes the findings, accumulated within the last 2 years, concerning melatonin's role in defending against toxic free radicals. Free radicals are chemical constituents that have an unpaired electron in their outer orbital and, because of this feature, are highly reactive. Inspired oxygen, which sustains life, also is harmful because up to 5% of the oxygen (O2) taken in is converted to oxygen-free radicals. The addition of a single electron to O2 produces the superoxide anion radical (O2-.); O2-. is catalytic-reduced by superoxide dismutase, to hydrogen peroxide (H2O2). Although H2O2 is not itself a free radical, it can be toxic at high concentrations and, more importantly, it can be reduced to the hydroxyl radical (.OH). The .OH is the most toxic of the oxygen-based radicals and it wreaks havoc within cells, particularly with macromolecules. In recent in vitro studies, melatonin was shown to be a very efficient neutralizer of the .OH; indeed, in the system used to test its free radical scavenging ability it was found to be significantly more effective than the well known antioxidant, glutathione (GSH), in doing so. Likewise, melatonin has been shown to stimulate glutathione peroxidase (GSH-Px) activity in neural tissue; GSH-PX metabolizes reduced glutathione to its oxidized form and in doing so it converts H2O2 to H2O, thereby reducing generation of the .OH by eliminating its precursor. More recent studies have shown that melatonin is also a more efficient scavenger of the peroxyl radical than is vitamin E. The peroxyl radical is generated during lipid peroxidation and propagates the chain reaction that leads to massive lipid destruction in cell membranes. In vivo studies have demonstrated that melatonin is remarkably potent in protecting against free radical damage induced by a variety of means. Thus, DNA damage resulting from either the exposure of animals to the chemical carcinogen safrole or to ionizing radiation is markedly reduced when melatonin is co-administered. Likewise, the induction of cataracts, generally accepted as being a consequence of free radical attack on lenticular macromolecules, in newborn rats injected with a GSH-depleting drug are prevented when the animals are given daily melatonin injections. Also, paraquat-induced lipid peroxidation in the lungs of rats is overcome when they also receive melatonin during the exposure period. Paraquat is a highly toxic herbicide that inflicts at least part of its damage by generating free radicals.(ABSTRACT TRUNCATED AT 400 WORDS)

The role of free radicals and antioxidants: how do we know that they are working?

This review briefly discusses how free radicals are formed and the possible participation of free radicals in disease. The review describes the basic radical reactions and the types of products that are formed from the free-radical reactions of cellular constituents. In many cases, in vivo free-radical oxidation can be detected by measuring products that were derived from radical reactions. Since aerobic organisms generate oxygen-containing free radicals during oxygen metabolism, they carry chemicals and enzymes that reduce the threat posed by these radicals. The more common sources of in vivo free radicals are described in the article as well as the methods used by cells to protect themselves from free-radical damage. Generation of free radicals in vivo also may be the result of exposure to certain chemical agents present in the environment. Many of these agents cause pathologic changes to the exposed tissues and organs by initiating free-radical reactions.

The NADPH: sulfite reductase of Escherichia coli is a paraquat reductase

The NADPH:sulfite reductase of Escherichia coli is a soluble enzyme that has a subunit structure alpha 8 beta 4, where the alpha subunit is a flavoprotein and the beta subunit is a metalloprotein. Overexpression of the holoenzyme in E. coli has greatly simplified the purification of this enzyme. Under aerobic conditions, recombinant sulfite reductase catalyzes the reduction of 1,1'-dimethyl-4,4'-bipyridinium dichloride (paraquat) by NADPH, with Km values for paraquat and NADPH of approximately 70 microM and 80 microM, respectively. Since pure flavoprotein alpha subunit, encoded by the cysJ gene, has similar catalytic activities, it is suggested that paraquat receives electrons directly from the alpha subunit. A mutant strain lacking an active cysJ gene is resistant to paraquat. The NADPH:ferredoxin reductase of E. coli is also a paraquat reductase but with much higher Km values for paraquat and lower enzyme activities. These results suggest that the sulfite reductase is a major paraquat reductase in E. coli and is responsible for the toxic activation of the drug.

CHO.K1 cell mutants sensitive to active oxygen-generating agents. I. Isolation and genetic studies

Nine mutants isolated from CHO.K1 cells with increased sensitivity to the lethal effect of plumbagin (PG), a powerful superoxide generator, were classified into five groups, A-E, according to their sensitivity to PG and methyl viologen (MV). Two mutants of group B (Pa13 and Pb4) were sensitive to both drugs, and two mutants of group C (Pa14 and Pa15) were moderately sensitive to PG and extremely sensitive to MV. To mitomycin C (MMC) these mutants showed cross-sensitivity; especially Pa13 and Pb4 (group B) were highly sensitive to MMC. Genetic complementation analyses of these four mutants were carried out using MV sensitivity. Sensitivity group B was divided into two complementation group, I and II. Pa14 and Pa15 belonged to the same complementation group III. These four mutants were also classified into three complementation groups for MMC sensitivity. Because Pa13 and Pb4 were also sensitive to cis-diamminedichloroplatinum(II), they may have a defect in the repair of DNA crosslinks induced by these agents. A complementation group IV (Pa2 and Pa8) was also suggested based on the studies of MMC sensitivity.

Neurodegeneration produced by intrahippocampal injection of paraquat is reduced by systemic administration of the 21-aminosteroid U74389F in rats

The behavioural, electrocortical (ECoG) and neurodegenerative effects of intrahippocampal injection of paraquat, a well-known free radical producing agent, were studied in rats. Injection of paraquat (100 nmol) into one dorsal hippocampus produced limbic motor and ECoG seizures. These effects were accompanied at 24 h by severe damage to CA1, CA3 and CA4 hippocampal pyramidal neurones and dentate gyrus granule cells. In comparison to the cell number counted in control, untreated, side of the hippocampus, significant (P < 0.05) neuronal loss was observed in the CA1 and CA3 pyramidal cell layers of the treated hippocampus. A lower dose of the herbicide (10 nmol) did not produce consistent motor and ECoG effects and in no instance was significant neuronal loss observed. A pretreatment with U74389F [21-4-(2,6-di-l-pyrrodinyl-4-pyridinyl)-1-piperazinyl-pregna-1,4,9 (11)triene-3,20-dione monomethansulfonate] (30 mg/kg i.p., 15 min before paraquat) completely protected rats from motor and ECoG epileptogenic effects induced by intrahippocampal paraquat (100 nmol). This dose of U74389F also reduced the hippocampal damage typically produced by paraquat and no significant neuronal loss was reported in the CA1 and CA3 pyramidal cell layers. A lower dose of U74389F (10 mg/kg i.p.) did not afford any protection against the epileptogenic effects produced by paraquat (100 nmol); in these animals hippocampal damage was still evident though neuronal loss did not reach statistical significance. In conclusion, the present data show that systemic administration of U74389F possesses neuroprotective effects against seizures and neurodegeneration typically elicited by intrahippocampal injection of paraquat.

Paraquat diaphorases in Escherichia coli

Extracts of E. coli contain at least three easily separable NAD(P)H:paraquat diaphorases. One of these is identified as thioredoxin reductase, which accounts for most of the PQ++ diaphorase in a thioredoxin reductase overproducer but is only 25% of this activity in a wild type. NADP+, but not NAD+, inhibited the diaphorase activity of thioredoxin reductase. All of the soluble PQ++ diaphorases of E. coli are stable during fractionation by HPLC and none depend upon the cooperative action of components separable by this technique. GSSG reductase is inhibited by PQ++ and is not, to any significant degree, a contributor to the diaphorase activity of E. coli.

Effect of paraquat on the malondialdehyde level in rat liver microsomes (in vitro)

Toxicosis due to paraquat, a redox cycling xenobiotic, is still a subject of much debate. In the present study on lipid peroxidation, paraquat had a biphasic effect on the malondialdehyde (MDA) level in rat liver microsomes; stimulation at the initial stage (within 10 min) and depression at the later stage. Although paraquat increased the initial rate of NADPH oxidation dose-dependently, the rate was not necessarily parallel with the increase in the MDA level. The MDA level increased linearly up to 0.1 mM paraquat added, but then it attained a plateau. The stimulation obtained by paraquat within 10 min was absolutely dependent on exogenous Fe2+ ion and NADPH, and the stimulation was entirely SOD sensitive, while the iron-driven increase in MDA was 20% sensitive. Thus, there were different mechanisms between iron-driven lipid peroxidation and paraquat-modified peroxidation. Catalase increased the level, but mannitol, a scavenger of OH, had no effect. EPR spectra showed that superoxide was formed dose-dependently up to 0.1 mM paraquat and that it attained a plateau at the same as MDA level described above. From these results, we concluded that paraquat stimulates lipid peroxidation through a mechanism dependent on the superoxide complex involving Fe2+ ion.

The pneumotoxicant paraquat induces IL-8 mRNA in human mononuclear cells and pulmonary epithelial cells

Paraquat (PQ) is a herbicide which is highly pneumotoxic by generating reactive oxygen intermediates (ROI). Pro-inflammatory cytokines, particularly IL-1 and TNF, have been implicated in some ROI-mediated pathologies, including bleomycin toxicity and ischaemia/reperfusion injury. We have studied the effect of PQ on the expression of the neutrophil chemotactic cytokine, IL-8, by human peripheral blood mononuclear cells (PBMC). While almost no IL-8 mRNA was detected in unstimulated cells, PQ (100 microM) induced high mRNA expression with a maximum at 24 h of incubation. While PQ did stimulate the appearance of IL-8 mRNA, no significant production of IL-8 protein was detected. However, PQ potentiated the production of IL-8 in the presence of 1 ng/ml of endotoxin (lipopolysaccharide, LPS). This was paralleled by an increased production of chemotactic activity for neutrophils, indicating that the IL-8 was actually bioactive. Stimulation of IL-8 mRNA by PQ was suppressed by IL-4 and by free radical scavengers (dimethylsulfoxide, mannitol). Increased IL-8 expression by PQ was also observed in the human pulmonary epithelial cell line A549 indicating that the effect of PQ was not specific for PBMC. These findings suggest that IL-8 might be involved in the pulmonary effects of PQ and that its production might be stimulated following an oxidative insult, and might clarify the pathogenetic mechanisms of adult respiratory distress syndrome (ARDS) or oxidant-induced pulmonary fibrosis.

Effects of carbon tetrachloride, menadione, and paraquat on the urinary excretion of malondialdehyde, formaldehyde, acetaldehyde, and acetone in rats

Excretions of the lipid peroxidation products, formaldehyde (FA), acetaldehyde (ACT), malondialdehyde (MDA), and acetone (ACON), were simultaneously identified and quantitated in the urine of female Sprague-Dawley rats by gas chromatography-mass spectroscopy (GC-MS) and high pressure liquid chromatography (HPLC) following the acute administration of carbon tetrachloride, a model alkylating agent that does not induce glutathione depletion, and the redox cycling compounds paraquat and menadione. All three xenobiotics are well-known inducers of oxidative stress. Oxidative stress was induced by oral administration of single doses of 2.5 mL of carbon tetrachloride/kg, 60 mg menadione/kg, and 75 mg paraquat/kg. These doses are approximately 50% of the LD50's for the three xenobiotics. Urinary excretion of FA, ACT, MDA, and ACON increased relative to control animals following treatment with all xenobiotics. Over the 48 hours of the study, the greatest increases in the excretion of MDA, FA, ACT, and ACON occurred after paraquat administration, with increases of approximately 2.7-, 2.6-, 4.3-, and 11.0-fold, respectively. This technique may have wide-spread applicability as an effective biomarker for investigating altered lipid metabolism in disease states and exposure to environmental pollutants/xenobiotics.

Induction of hepatic metallothionein by paraquat

Paraquat, a frequently used contact herbicide, produces oxidative stress by undergoing redox cycling and generating reactive oxygen species. Paraquat is also effective at increasing hepatic levels of metallothionein (MT). The mechanism(s) by which agents that induce oxidative stress produce increases in MT concentrations is not yet known. Therefore, the goal of the current study was to characterize the elevation in hepatic MT produced by paraquat administration to mice and to examine potential mechanism(s) of this increase. A dose-response study for increases in MT showed that administration of 0.1 to 0.5 mmol/kg of paraquat, sc, increased hepatic MT with a maximal increase of 36-fold. Subsequent studies were carried out with paraquat at a dose (0.3 mmol/kg, sc) that caused oxidative stress, as shown by a 35-fold increase in the biliary excretion of oxidized glutathione. There were coordinate elevations of both hepatic MT-I and MT-II mRNA of approximately 5-fold with peaks at both 6 and 24 hr after paraquat. The time course for the elevation in hepatic MT protein following paraquat treatment showed that MT levels had a maximal increase of 18-fold obtained at 36 hr. Paraquat appears to be an indirect MT inducer, in that there were no elevations in MT when cultured mouse hepatocytes were exposed to paraquat. No rise in liver Zn was observed prior to the increase in hepatic MT, thus, a Zn redistribution to the liver did not cause the increase in hepatic MT following paraquat administration. Adrenalectomy did not abolish the increase in MT produced by paraquat, suggesting that adrenal gland products are not required for the increase in MT produced by paraquat. In conclusion, the chemical mediator responsible for the increase in hepatic MT after paraquat was not determined, but the elevation in MT concentration appears to be due to increased transcription.

A novel pharmacological approach for paraquat poisoning in rat and A549 cell line using ambroxol, a lung surfactant synthesis inducer

Paraquat (PQ) is a widely used herbicide that causes acute adult respiratory distress syndrome (ARDS) and chronic lung damage (diffuse fibrosis). One of the earliest biochemical effects induced by PQ is damage to type II pneumocytes with consequent depletion of surfactant. With the aim of counteracting the toxic effects of PQ, a series of investigations were performed into the possible protective effect of the drug ambroxol, which induces the synthesis of surfactant in lung alveolar type II cells. The number of survivors and survival time of rats treated ip with 35 mg PQ/kg was significantly increased by 3 days of ambroxol pretreatment and by ambroxol treatment 30 min or 2 hr after PQ. Total phospholipid content in lung and bronchoalveolar lavage fluid (BALF) was significantly reduced 30 hr after treatment with PQ alone. The association of ambroxol with PQ significantly antagonized this reduction. In BALF the ratio between palmitic acid and stearic acid concentrations was significantly lower in animals treated with PQ alone but was returned to normal by the association with ambroxol. The cell line A549, exposed in vitro to PQ concentrations from 0.5 x 10(-4) to 2 x 10(-3) M, showed a significant dose-dependent loss of viability. Cells pretreated with ambroxol (10 mg/ml) were more resistant to PQ and their viability started to decrease significantly only from a PQ concentration of 0.8 x 10(-3) M. Membrane microviscosity was measured on the same cells. Cells treated with PQ alone showed a reduction of membrane microviscosity, which was significantly counteracted by ambroxol pretreatment. The curves of modification of membrane microviscosity of cells treated with PQ and with ambroxol plus PQ paralleled those of cell viability, indicating that the stimulation of surfactant synthesis in vitro may be a prerequisite for counteracting some of the early effects of PQ.

Pulmonary hemodynamics and lung function during chronic paraquat poisoning in sheep. Possible role of reactive oxygen species

The purpose of this study was to establish a chronic model of paraquat-induced lung injury. To examine the role of reactive oxygen species in this form of lung injury, we measured malondialdehyde (MDA) and superoxide dismutase (SOD) activity in the lungs. Paraquat (5 mg/kg intramuscularly) caused a significant decrease in dynamic lung compliance from 128.5 +/- 9.2 to 63.3 +/- 11.8 ml/cm H2O (p less than 0.05), with a significant increase in AaPO2 3 wk after paraquat. Histologic findings in the lungs showed a gradual increase in the number of granulocytes and alveolar wall thickening with proliferation of reticular fibers and were coincident with the changes in physiology. A transient decrease in pressor responses to hypoxia was observed 1 wk after paraquat, although pulmonary hemodynamics did not change. The amount of lung MDA 3 wk after paraquat increased from the baseline value of 0.73 +/- 0.04 to 1.12 +/- 0.10 nmol/mg protein (p less than 0.05). SOD activity in the lung tissue significantly decreased from 6.47 +/- 0.20 to 4.82 +/- 0.25 U/mg protein (p less than 0.05) 1 wk after paraquat and remained at low levels for 3 wk. These findings suggest that a small dose of paraquat causes chronic lung injury characterized by granulocyte infiltration and lung fibrosis. Reactive oxygen species may play an important role in this chronic lung injury, and the inability to increase antioxidant defense may contribute to the reaction.

Toxicity of 1-methyl-4-phenylpyridinium derivatives in Escherichia coli

Several derivatives of 1-methyl-4-phenylpyridinium (MPP+), i.e., 1-methyl-4-(4'-nitrophenyl)pyridinium (1), 1-methyl-4-(4'-cyanophenyl)pyridinium (2), 1-methyl-4-(3'-nitrophenyl)pyridinium (3), 1-methyl-4-(4'-chlorophenyl)pyridinium (4), 1-methyl-4-(4'-acetamidophenyl)pyridinium (5), and 1-methyl-4-(4'-aminophenyl)pyridinium (6), were synthesized in order to compare their toxicity with that of paraquat (PQ2+) in Escherichia coli. Addition of compounds 1, 2, and 3 to aerobic E. coli cell suspensions caused extracellular ferricytochrome c reduction, which was inhibited by superoxide dismutase in the same manner as that in the case of PQ2+. The rate of the ferricytochrome c (cyt. c) reduction was in the order of PQ2+ greater than 1 greater than 2 greater than 3, which is the same as that of the redox potentials of these compounds. On the other hand, MPP+, 4, 5, and 6, which have more negative potentials, had no effect on the cyt. c reduction. Compound 1 inhibited the growth of E. coli under aerobic conditions, but not under anaerobic conditions. The results show that compound 1 can act as a mediator for production of superoxide (O2-.), which seriously injures E. coli cells. However, though compounds 2 and 3 catalyzed the production of O2-. in E. coli cells, their activity of O2-. production was much lower than that of compound 1 or PQ2+. Thus, compound 3 had no effect on growth or survival of E. coli at 1 mM, while compounds 2 and 4 had both bacteriostatic and bacteriocidal effects which were independent of dioxygen (O2). The results show that the toxic mechanism is different from that of compound 1. MPP+, 5, and 6 had no effect on growth of E. coli. This paper shows that compound 1 is a novel enhancer of intracellular superoxide production, though the mechanism of toxicity of compounds 2 and 4 is not clear yet. The results suggest that the redox potential is a crucial factor for manifestation of the activity.

Urate-null rosy mutants of Drosophila melanogaster are hypersensitive to oxygen stress

It has been proposed that uric acid is an important scavenger of deleterious oxygen radicals in biological systems [Ames, B. N., Cathcart, R., Schwiers, E. & Hochstein, P. (1981) Proc. Natl. Acad. Sci. USA 78, 6858-6852]. We report here an in vivo investigation of the oxygen defense role of uric acid through an analysis of mutants of the rosy (ry) gene of Drosophila melanogaster. The ry gene is the structural gene for the molybdoenzyme, xanthine dehydrogenase; xanthine dehydrogenase-null ry mutants are therefore unable to synthesize urate. The rationale of our approach was to measure the response of urate-null ry mutants to extraordinary oxygen stress as imposed by exposure to radical-generating agents and as conferred by a genetic defect in superoxide dismutase, an established oxygen defense function. We show that urate-null mutants of the ry locus are hypersensitive to paraquat, ionizing radiation, and hyperoxia. Furthermore, compound mutants doubly deficient for uric acid and Cu/Zn-containing superoxide dismutase are synthetic lethals, which are unable to complete metamorphosis under normal growth conditions. These experiments demonstrate unambiguously the importance of urate in oxygen defense in vivo and support our earlier proposal that the molybdoenzyme genetic system plays a critical role in oxygen defense in Drosophila. They also form the basis for our proposal that metamorphosis in Drosophila imposes a crisis of oxygen stress on the developing imago against which uric acid plays an important organ-specific defense. Finally, the results provide a basis for understanding the syndrome of phenotypes, including the hallmark dull brown eye color, which characterizes mutants of this classic genetic system of Drosophila.

Response of Pseudomonas aeruginosa to pyocyanin: mechanisms of resistance, antioxidant defenses, and demonstration of a manganese-cofactored superoxide dismutase

Pseudomonas aeruginosa produces a blue pigment, pyocyanin. Pyocyanin is a redox-active phenazine compound that kills mammalian and bacterial cells through the generation of reactive oxygen intermediates. We examined the mechanisms by which P. aeruginosa resists pyocyanin. [14C]pyocyanin was taken up by both Escherichia coli and P. aeruginosa, though more slowly by the latter. Cyanide-insensitive respiration, used as an indicator of intracellular superoxide and/or hydrogen peroxide production, was 50-fold less in pyocyanin-treated P. aeruginosa than in E. coli. P. aeruginosa showed less cyanide-insensitive respiration than E. coli upon exposure to other redox-active compounds (paraquat, streptonigrin, and plumbagin). Electron paramagnetic resonance spectrometry and spin trapping showed that P. aeruginosa generated less pyocyanin radical and superoxide than E. coli. Cell extracts from E. coli contained an NADPH:pyocyanin oxidoreductase which increased the rate of reduction of pyocyanin by NADPH. Conversely, cell extracts from P. aeruginosa contained no NADPH:pyocyanin oxidoreductase activity and actually decreased the rate of pyocyanin-mediated NADPH oxidation. Antioxidant defenses could also reduce the sensitivity of P. aeruginosa to pyocyanin. Under culture conditions of limited phosphate, both pyocyanin production and catalase activity were enhanced. Superoxide dismutase activity was also increased under low-phosphate conditions. When cells were grown in a high-phosphate succinate medium, P. aeruginosa formed a previously described iron-superoxide dismutase as well as a manganese-cofactored superoxide dismutase. These results demonstrate that P. aeruginosa resists pyocyanin because of limited redox cycling of this compound and that under conditions favoring pyocyanin production, catalase and superoxide dismutase activities increase.

Cellular response to oxidative damage in lung induced by the administration of dimethylarsinic acid, a major metabolite of inorganic arsenics, in mice

Oral administration of dimethylarsinic acid (DMAA), a major metabolite of inorganic arsenics, induces DNA damage in the mouse and rat lung due to both active oxygens and dimethylarsenic peroxyl radical produced in the metabolism of DMAA. Our paper describes the cellular response to DMAA in the mouse lung. In male ICR mice given a single po dose (1500 mg/kg) of DMAA-Na, the activities of mitochondrial superoxide dismutase, glutathione peroxidase, and glucose-6-phosphate dehydrogenase significantly increased at 6 hr or longer after dosing, while cytosolic superoxide dismutase and catalase did not. With regard to cellular sulfhydryls after DMAA dosing, levels of reduced glutathione and nonprotein sulfhydryl decreased, while mixed disulfides significantly increased. Further, NADPH markedly decreased at 6-9 hr after DMAA dosing. These cellular variations suggest that the mouse pulmonary cell produced active oxygens, i.e., superoxide anion radical, hydrogen peroxide, and subsequent radicals in the metabolism of DMAA and that these and also the dimethylarsenic peroxyl radical were responsible for pulmonary DNA damage.

Inverse correlation between resistance towards copper and towards the redox-cycling compound paraquat: a study in copper-tolerant hepatocytes in tissue culture

The essential mediatory role of copper or iron in the manifestation of paraquat toxicity has been demonstrated (Kohen and Chevion (1985) Free Rad. Res. Commun. 1, 79-88; Korbashi, P. et al. (1986) J. Biol. Chem. 261, 12472-12476). Several liver cell lines, characterized by their resistance to copper, were challenged with paraquat and their cross-resistance to paraquat and copper was studied. Cell growth and survival data showed that copper-resistant cells, containing elevated copper, are more sensitive towards paraquat than wild type cells. Copper-deprived resistant cells did not have this sensitivity. Paraquat was also shown to cause a marked degradation of cellular glutathione in all cell lines. Albeit the fact that the basal glutathione levels are higher in copper-resistant than in wild type cells, there is more paraquat-induced degradation of cellular glutathione (GSH + GSSG) in resistant cells. It is suggested that in copper-resistant cells which contain elevated levels of copper, paraquat-induced cellular injury is potentiated even where glutathione levels are elevated. Additionally, in vitro experiments are presented that support the in vivo findings demonstrating a role for copper in glutathione degradation.

Effect of some environmental pollutants on the superoxide dismutase activity in Lemna

Exposure of Lemna sp. to SO2 resulted in an increased activity of superoxide dismutase. About 3 to 4 fold increase in the activity was observed within 30 minutes after the plants were fumigated with 10 ml/l of SO2. Paraquat, a well known superoxide generator, doubled the enzyme activity after 1 hour of treatment with 0.1 mM paraquat. Superoxide dismutase activity was also enhanced by cadmium treatment but the response was not immediate. Optimum increase in the activity of enzyme was observed after 4 days of treatment with 40 mg/l of cadmium in the medium. Treatment with H2O2 very clearly inhibited the activity of superoxide dismutase in Lemna.

Induction of superoxide dismutases in Photobacterium leiognathi

We investigated the induction of Cu,Zn-SOD (bacteriocuprein) and Fe-SOD in Photobacterium leiognathi DK-A1 which was isolated from the light organ of the squid, Droteuthis kensaki. The induction of superoxide dismutases depended on the addition of paraquat to the medium. Induction of SOD by paraquat was attributed mostly to the bacteriocuprein by measuring of the activities of both SODs by using densitometry of isoelectrofocusing gel. When paraquat was added to the culture at various times in the early log phase of growth, the most efficient induction of the SODs, which was measured at the time of harvesting the cells (17 hours after inoculation), was observed when paraquat was added at 60 min after the inoculation. Catalase was not significantly induced by the addition of paraquat or increasing of oxygen concentration. We developed an assay of SOD by modification of a cytochrome c-xanthine oxidase method using a computer equipped absorption spectrophotometer.

An evaluation of the redox cycling potencies of paraquat and nitrofurantoin in microsomal and lung slice systems

The redox cycling abilities of the pulmonary toxins paraquat and nitrofurantoin have been compared with those of the potent redox cyclers, diquat and menadione in lung and liver microsomes by using the oxidation of NADPH and consumption of oxygen. The relative potencies of these compounds to undergo redox cycling were in the order: diquat approximately menadione much greater than paraquat congruent to nitrofurantoin. This was partly attributed to the much lower affinity (Km) of lung and liver microsomes for paraquat and nitrofurantoin than for diquat and menadione. The potential to redox cycle was assessed in an intact cellular system by determining the oxygen consumption of rat lung slices in the presence (10(-6), 10(-5) and 10(-4) M) or absence of each of the four substrates. At concentrations of paraquat (10(-5) M) known to be accumulated by lung slices, a small but significant stimulation of lung slice oxygen uptake was observed. Nitrofurantoin (10(-4)-10(-6) M) did not affect lung slice oxygen uptake in lung slices, an observation consistent with its being a poor redox cycling compound, which is not actively accumulated into lung cells. This data has important implications in assessing the risk of exposure to paraquat. Low levels of paraquat would not be expected to cause lung damage because insufficient compound is present in the lung to exert its toxicity by redox cycling (due to the high Km observed).

Overexpression of Cu-Zn superoxide dismutase in Drosophila does not affect life-span

Aging and disease processes may be due to deleterious and irreversible changes produced by free radical reactions. The enzyme copper-zinc superoxide dismutase (Cu-Zn SOD; superoxide:superoxide oxidoreductase, EC 1.15.1.1) performs a protective function by scavenging superoxide radicals. The Cu-Zn SOD gene (Sod) cloned from Drosophila melanogaster was introduced via P element-mediated transformation into the germ line. Homozygous lines carrying additional copies of the Sod gene were recovered and characterized. Increases in Sod transcripts and enzyme activity were observed in the transformed lines, indicating that all of the sequence information required for gene expression is contained on the inserted gene fragment. The effects of additional SOD on oxygen free radical metabolism and longevity were investigated. Additional SOD did not markedly affect oxygen metabolism or longevity.

Direct and respiratory chain-mediated redox cycling of adrenochrome

Adrenochrome is reduced by ascorbate in a reaction accompanied by a large and rapid oxygen uptake. The rates of adrenochrome reduction and the concomitant oxygen uptake are decreased in the presence of superoxide dismutase or catalase. The species formed on the one-electron reduction of adrenochrome (i.e., the semiquinone) was shown by pulse radiolysis to rapidly react with oxygen (9.10(8) M-1.s-1), indicating the occurrence of a redox cycling in a system formed by adrenochrome, a reducing agent, and oxygen. Adrenochrome is also reduced to the corresponding semiquinone by complex I of beef heart submitochondrial particles supplemented with NADH, while succinate is unable to support this reduction. The o-semiquinone is the intermediate species in the superoxide-generating cycle resulting from both non-enzymatic and enzymatic reduction. The toxic effects of adrenochrome and its pathophysiological role can be explained, at least in part, on the basis of the demonstrated cycle.

Reactions of the NAD radical with higher oxidation states of horseradish peroxidase

The reactions of the NAD radical (NAD.) with ferric horseradish peroxidase and with compounds I and II were investigated by pulse radiolysis. NAD. reacted with the ferric enzyme and with compound I to form the ferrous enzyme and compound II with second-order rate constants of 8 X 10(8) and 1.5 X 10(8) M-1 s-1, respectively, at pH 7.0. In contrast, no reaction of NAD. with native compound II at pH 10.0 nor with diacetyldeutero-compound II at pH 5.0-8.0 could be detected. Other reducing species generated by pulse radiolysis, such as hydrated electron (eaq-), superoxide anion (O2-), and benzoate anion radical, could not reduce compound II of the enzyme to the ferric state, although the methylviologen radical reduced it. The results are discussed in relation to the mechanism of catalysis of the one-electron oxidation of substrates by peroxidase.

Effects of paraquat on cultures of Escherichia coli: turbidity versus enumeration

The dioxygen-dependent toxicity of paraquat has been studied both in terms of its effects on growth, monitored as increases in turbidity, and on viability, measured by plating and counting of colonies. In the absence of paraquat, turbidity and enumeration increased in parallel. However, in the presence of paraquat, turbidity increased for several hours even while enumeration indicated a marked decrease in viability. The basis for this apparent discrepancy is continued increase in size of individual cells, which have stopped dividing and are losing viability under the influence of paraquat. It can evidently be misleading to study the effects of paraquat on microorganisms in terms of changes in turbidity.

Correlation between lipid peroxidation and morphological manifestation of paraquat-induced lung injury in rats

Biochemical and morphological studies of rat lung were performed to determine the role of lipid peroxidation in the in vivo lung toxicity of paraquat. Two injections of 20 mg/kg paraquate were administered intraperitoneally every other day. While notable epithelial damage in the lungs was observed on the day after the second paraquat injection and progressed through the 5th day, the concentration of lipid peroxides in the rat lungs did not increase by the 3rd day after the injection. The lipid peroxide concentrations increased after the 5th day post-injection, and reached the maximum concentrations on the 7th day, when the damaged alveolar surface had been mostly repaired by regenerative pneumocytes. On the other hand, the delayed increase of lung lipid peroxides in paraquat-treated rats paralleled the increased number of macrophages in the lung, which reached maximum numbers on the 7th day. Glutathione peroxidase activity in the lungs also increased with a similar time course. Macrophages from the lungs contained a large amount of engulfed degradation products and cellular debris, and immunohistochemical study showed high glutathione peroxidase content on the 5th and 7th days. These results suggest that lipid peroxidation is a relatively late event in the in vivo paraquat-treated lung and that the delayed increase of lipid peroxides in the lungs occurs from the phagocytic activities of macrophages rather than from toxic cell injury.

Oxygen radicals in lung pathology

Pulmonary tissue can be damaged in different ways, for instance by xenobiotics (paraquat, butylated hydroxytoluene, bleomycin), during inflammation, ischemia reperfusion, or exposure to mineral dust or to normobaric pure oxygen levels. Reactive oxygen species are partly responsible for the observed pulmonary tissue damage. Several mechanisms leading to toxicity are described in this review. The reactive oxygen species induce bronchoconstriction, elevate mucus secretion, and cause microvascular leakage, which leads to edema formation. Reactive oxygen species even induce an autonomic imbalance between muscarinic receptor-mediated contraction and the beta-adrenergic-mediated relaxation of the pulmonary smooth muscle. Vitamin E and selenium have a regulatory role in this balance between these two receptor responses. The autonomic imbalance might be involved in the development of bronchial hyperresponsiveness, occurring in lung inflammation. Finally, several antioxidants are discussed which may be beneficial as therapeutics in several lung diseases.

Mechanistic aspects of paraquat toxicity in E. coli. A spin trapping study

Mechanistic aspects of paraquat monocation radical (PQ.+) and copper involvement in paraquat toxicity have been examined using E. coli B cells. Electron spin resonance (ESR) spectrometry combined with cell survival studies were used to explore the correlation between radical production and biological damage. The line broadening agent oxalato-chromiate (CrOx) was used to characterize the anoxic partition of PQ.+ inside and outside the cell. In the presence of CrOx the ESR signal was totally eliminated, indicating that intracellular species were undetectable and that, contrary to previous reports, PQ.+ exclusively accumulates outside the cell. The PQ.+ radical does not react with H2O2 but disappears in the presence of H2O2 when catalytic traces of Cu(II) are present. Spin-trapping studies using DMPO showed that in aerobic environment paraquat-induced O2 radicals are detectable exclusively in the extracellular compartment. The correlation between PQ.+ appearance and the biological damage is not simple. PQ.+ non-toxically accumulates, in the absence of oxygen and either Cu(II) or H2O2. By contrast, with both H2O2 and Cu(II) the cells are rapidly killed but PQ.+ was undetectable. These results reconfirm the key catalytic mediatory function of transition metals in paraquat toxicity.

Exposure to paraquat through skin absorption: clinical and laboratory observations of accidental splashing on healthy skin of agricultural workers

Data concerning 15 consecutive cases of single exposures of the skin or eyes during work to paraquat solutions are presented. Urine and serum were analysed for paraquat in all these cases at the laboratory of the Israel Poison Information Center. From these data it is apparent that a single exposure of healthy skin to paraquat solutions caused only local lesions. No systemic effect was detected in these patients.

Synergistic interactions between NADPH-cytochrome P-450 reductase, paraquat, and iron in the generation of active oxygen radicals

The toxicity associated with paraquat is believed to involve the generation of active oxygen radicals and the production of oxidative stress. Paraquat can be reduced by NADPH-cytochrome P-450 reductase to the paraquat radical; this results in consumption of NADPH. A variety of ferric complexes, including ferric-ATP, -citrate, -EDTA, ferric diethylenetriamine pentaacetic acid and ferric ammonium sulfate, produced a synergistic increase in the paraquat-mediated oxidation of NADPH. This synergism could be observed with very low concentrations of iron, e.g. 0.25 microM ferric-ATP. Very low rates of hydroxyl radical were generated by the reductase with paraquat alone, or with ferric-citrate or -ATP or ferric ammonium sulfate in the absence of paraquat; however, synergistic increases in the rate of hydroxyl radical generation occurred when these ferric complexes were added together with paraquat. Ferric-EDTA and -DTPA catalyzed some production of hydroxyl radicals, which was also synergistically elevated in the presence of paraquat. Ferric desferrioxamine was essentially inert in the absence or presence of paraquat. This enhancement of hydroxyl radical generation was sensitive to catalase and competitive scavengers but not to superoxide dismutase. The interaction of paraquat with NADPH-cytochrome P-450 reductase and ferric complexes resulted in an increase in oxygen radical generation, and various ferric complexes increased the catalytic effectiveness and potentiated significantly the toxicity of paraquat via this synergistic increase in oxygen radical generation by the reductase.

Null mutation of copper/zinc superoxide dismutase in Drosophila confers hypersensitivity to paraquat and reduced longevity

The role of copper/zinc-containing superoxide dismutase (cSOD; superoxide:superoxide oxidoreductase, EC 1.15.1.1) in metabolic defense against O2 toxicity in Drosophila is examined through the properties of a mutant strain carrying a cSOD-null mutation, cSODn108. Homozygotes are viable as larvae, which indicates that cSOD is not essential for cell viability per se. cSODn108 confers recessive sensitivity to the superoxide anion (O2-)-generator paraquat and to the transition metal compound CuSO4, which indicates that the cSOD-null condition in fact leads to impaired O2- metabolism. The primary biological consequences of the reduced O2- dismutation capacity of cSODn108 Drosophila are realized in the adult as infertility and reduction in life-span. We conclude that the infertility and reduced life-span of cSODn108 adults arise as a consequence of the reduced capacity of embryos, larvae, and pupae to adequately protect developing preimaginal cells from O2- -initiated cytotoxic damage.

Effects of vanadate on the oxidation of NADH by xanthine oxidase

Vanadate (V(V)) stimulates the oxidation of NADH by xanthine oxidase and superoxide dismutase eliminates the effect of V(V). Paraquat stimulates both the oxidation of NADH by xanthine oxidase and the V(V) enhancement of that oxidation. Xanthine, which is a better substrate for xanthine oxidase than is NADH, causes a V(V)-dependent co-oxidation of NADH which is transient and eliminated by SOD. Urate inhibits the V(V)-stimulated oxidation of NADH by xanthine oxidase or by Rose Bengal plus light. Measurement of rates of both O2- production and V(V)-stimulated NADH oxidation showed that many molecules of NADH were oxidized per O2-. These chain lengths were an inverse function of overall reaction rate. Minimum chain lengths, calculated on the basis of 100% univalent reduction of O2 to O2-, were smaller than measured average chain lengths by a factor of five. All of these results are in accord with the view that V(V) does not directly affect the activity of the enzyme, but rather catalyzes the free radical chain oxidation of NADH by O2-. It was further shown that phosphate was not involved and that the active form of V(V) was orthovanadate, rather than decavanadate.

Three models of free radical-induced cell injury

Three models of free radical-induced cell injury are presented in this review. Each model is described by the mechanism of action of few prototype toxic molecules. Carbon tetrachloride and monobromotrichloromethane were selected as model molecules for alkylating agents that do not induce GSH depletion. Bromobenzene and allyl alcohol were selected as prototypes of GSH depleting agents. Paraquat and menadione were presented as prototypes of redox cycling compounds. All these groups of toxins are converted, during their intracellular metabolism, to active species which can be radical species or electrophilic intermediates. In most cases the activation is catalyzed by the microsomal mixed function oxidase system, while in other cases (e.g. allyl alcohol) cytosolic enzymes are responsible for the activation. Radical species can bind covalently to cellular macromolecules and can promote lipid peroxidation in cellular membranes. Of course both phenomena produce cell damage as in the case of CCl4 or BrCCl3 intoxication. However, the covalent binding is likely to produce damage at the molecular site where it occurs; lipid peroxidation, on the other hand, besides causing loss of membrane structure, also gives rise to toxic products such as 4-hydroxyalkenals and other aldehydes which in principle can move from the site of origin and produce effects at distant sites. Electrophilic intermediates readily reacts with cellular nucleophiles, primarily with GSH. The result is a severe GSH depletion as in the case of bromobenzene or allyl alcohol intoxication. When the depletion reaches some threshold values lipid peroxidation develops abruptly and in an extensive way. This event is accompanied by cellular death. The reason for which lipid peroxidation develops in a cell severely depleted of GSH remains to be clarified. Probably the loss of the defense systems against a constitutive oxidative stress is not compatible with cellular life. Some free radicals generated by one-electron reduction can react with oxygen to give superoxide anions which can be converted to other more dangerous reactive oxygen species. This is the case of paraquat and menadione. Damage to cellular macromolecules is due to the direct action of these oxygen radicals and, at least in the menadione-induced cytotoxicity, lipid peroxidation is not involved. All these initial events affect the protein sulfhydryl groups in the membranes. Since some protein thiols are essential components of the molecular arrangement responsible for the Ca2+ transport across cellular membranes, loss of such thiols can affect the calcium sequestration activity of subcellular compartments, that is the capacity of mitochondria and microsomes to regulate the cytosolic calcium level.(ABSTRACT TRUNCATED AT 400 WORDS)

Dioxathiadiaza-heteropentalenes mediate superoxide and hydrogen peroxide production in Escherichia coli

The dioxathiadiaza-heteropentalenes, HEP-I (4,4-dimethyl-1,7-dioxa-2,6-diaza- 7 alpha lambda 4-thia-3H,5H-benzo[cd]pentalene), HEP-II (1,7-dioxa-2, 6-diaza-4, 7 alpha lambda 4-dithia-3H, 5H-benzo[cd]pentalene), HEP-III (1,7-dioxa-2,6-diaza-4, 7 alpha lambda 4-dithia-3H, 5H-benzo[cd]pentalene-4-oxide), and HEP-IV (1,7-dioxa-2,6-diaza-4,7 alpha lambda 4-dithia-3H, 5H-benzo[cd]pentalene-4,4-dioxide), inhibited growth of Escherichia coli in a simple glucose-salt medium, with their toxicities following the order of HEP-IV greater than HEP-III greater than HEP-II greater than HEP-I. These toxicities could be suppressed by yeast extract added to the glucose-salt medium. Yeast extract also facilitated maximal induction of superoxide dismutase (SOD) and catalase. The redox potentials of HEP-I-HEP-IV and the rates of oxygen uptake dependent on heteropentalenes in cyanide-resistant respiration of E. coli were correlated with the induction of SOD and catalase. Thus, the higher the redox potential of the compounds, the more potent they were for induction of enzyme production. Under anaerobic conditions, HEP-IV did not inhibit E. coli growth. These results indicate that HEP-I-HEP-IV can be reduced within the cell of E. coli and then reoxidized by molecular oxygen, generating O2- and H2O2. The toxicities of the heteropentalenes depend largely upon superoxide and/or hydrogen peroxide toxicity, and SOD and catalase provide a defense against the potential cytotoxicity of these species.