Contribution of increased glutathione content to mechanisms of oxidative stress resistance in hydrogen peroxide resistant hamster fibroblasts

An H₂O₂-resistant variant (OC14) of the HA1 Chinese hamster fibroblast cell line, which demonstrates cross resistance to 95% O₂ and a 2-fold increase in total glutathione content, was utilized to investigate mechanisms responsible for cellular resistance to H₂O₂- and O₂-toxicity. OC14 and HA1 cells were pretreated with buthionine sulfoximine (BSO) to deplete total cellular glutathione. Following BSO pretreatment, cells were either placed in 250 μM BSO to maintain the glutathione depleted condition and challenged with 95% O₂, or challenged with hydroged peroxide in the absence of BSO. Total glutathione and the activities of CuZn superoxide dismutase, Mn superoxide dismutase, catalase, glutathione peroxidase, and glutathione transferase were evaluated immediately following the BSO pretreatment as well as following 39 to 42 hr of exposure to 250 μM BSO. BSO treatment did not cause significant decreases in any cellular antioxidant tested, except total glutathione depletion resulted in significant (P < 0.05) sensitization to O₂-toxicity and H₂O₂-toxicity in both cell lines at every time point tested. However, glutathione depletion did not completely abolish the resistance to either O₂- or H₂O₂-toxicity demonstrated by OC14 cells, relative to HA1 cells. Also, glutathione depletion did not effect the ability of OC14 cells to metabolize extracellular H₂O₂. These data indicate that glutathione dependent processes significantly contribute to cellular resistance to acute H₂O₂- and O₂-toxicity, but are not the only determinants of resistance in cell lines. The contribition of aldehydes formed by lipid peroxidation in mechanisms involved with the sensitization to O₂-toxicity in glutathione depleted cells was tested by measuring the lipid peroxidation byproduct, 4-hydroxy-2-nonenal (4HNE), bound in Schiff-base linkages or in its free form in cell homogenates at 49 hr of 95% O₂-exposure. No significant increase in 4HNE was detected in glutathione depleted cells relative to glutathione competent cells, indicating that glutathione depletion does not sensitize these cells to O₂-toxicity by altering the intracellular accumulation of free or Schiff-base bound 4HNE. © 1995 Wiley-Liss Inc.     

Pulmonary oxygen toxicity: Lack of tolerance of mice of various ages

Prolonged continuous exposure of adult (3–4 months) and old (21 months) mice to hyperoxia did not lead to significant changes in the activities of superoxide dismutase and catalase in liver or blood. Lung superoxide dismutase activity increased by 25% during initial exposure to 100% O₂, but then fell progressively to below control level. Exposure of mice to 60% or 80% O₂ increased their susceptibility to further exposure to 100% O₂. The results clearly show that both adult and old mice are incapable of coping with the high oxygen environment and that antioxidant enzyme induction and the associated partial protection from pulmonary O₂ toxicity are not the general rule in mammalian lung exposed to subtoxic oxygen levels.     

Responses of Drosophila melanogaster to atypical oxygen atmospheres

We examined physiological phenotypes of Drosophila melanogaster in hypoxic to hyperoxic atmospheres. We performed measurements on life span or behavioural function in 5, 21, 40, 60, and 80% O₂, and combined this with literature data for 2% and 100% O₂. O₂ incubation resulted in a concentration-dependent reduction of life span in both hypoxia and hyperoxia, though different measures of life span were affected differently. We also examined how behavioural and metabolic functions were affected by exposure to hyperoxia (up to 60% O₂). Climbing behaviour was measured as a fast (4 s) and slow (55 s) response in a negative geotaxis assay. In normoxia, both measures of climbing response declined exponentially until disappearing completely. Interestingly, survivorship was very high until the loss of climbing ability, after which it dropped rapidly. This pattern appeared accelerated in 40% O₂. However, while flies in 60% O₂ also apparently lost their fast climbing ability immediately prior to the drop in survivorship, they maintained considerable climbing ability over the longer trial. Metabolism, measured by CO₂ release, did not change with age in normoxic flies, but was significantly lower in flies exposed to hyperoxia, particularly as the flies aged. There was, however, a slight increase in water loss rate with age in normoxia, while in hyperoxia, water loss was reduced. Uniquely, the water loss rates of flies in 60% O₂ doubled immediately prior to the end of their life span. Because ageing results in generally irreversible functional declines, we examined if functional declines in hyperoxia (60% O₂) were also irreversible, or whether some functioning could recover after a return to normoxia. After 7 days of recovery, water loss rates decreased, CO₂ exhalation slightly increased, and climbing ability was partially recovered. Therefore, the effect of O₂ on D. melanogaster function is non-linear, may be reversible, and may include unique phenotypes that arise at some O₂ concentrations, and not others.     

Effects of hyperoxia on biomarkers of oxidative stress in closed-circuit oxygen military divers

Oxygen toxicity is a problem in diving which can have fatal consequences in the water. When divers use closed-circuit oxygen rebreathing apparatus they are taking only oxygen 100% and this hyperoxic exposure increases the generation of reactive oxygen species (ROS) in biological tissues. The objective of the present study is to evaluate the effects of hyperoxia on biomarkers of oxidative stress in closed-circuit oxygen military divers. Fifteen professional divers of Spanish Navy Diving Center participated in a training program which consisted of one-hour immersion at seven metres of depth breathing oxygen 100% with closed-circuit oxygen rebreathing apparatus. The training went on two or three times per week for the first six weeks and once a week for the last six weeks. Serum total antioxidant status (TAS), levels of glutathione peroxidase (GPx), nitrates (NO₃ ^?) and urinary concentrations of 15-isoprostane F_2t were measured. The results show that TAS decreased significantly after 6 weeks (mean 1.38 versus 1.23 mmol/l), with a slight increase at the end (mean 1.31 mmol/l). GPx and F₂-isoprostanes were significantly lower after 6 and 12 weeks and NO₃ ^? was significantly lower after 6 weeks and remained unchanged until the end. In summary, professional divers who use closed-circuit apparatus and therefore breathe oxygen 100%, do not suffer an important oxidative hyperoxia-induced stress, probably due an adaptive process after hyperoxia. The age and good physical form of the subjects studied could probably enhance the adaptive process to hyperoxia.     

Hyperoxia increases oxygen radical production in rat lung homogenates

Lung damage during hyperoxia has been postulated to be due to increased rates of local organ oxygen radical production. Lung homogenate respiration was inhibited with cyanide, and residual respiration was used as an indicator of electron diversion to O₂^? and H₂O₂. Cyanide-resistant respiration in lung homogenates, supplemented with 1 mm NADH, increased linearly with oxygen tension, and accounted for 7% of total respiration in air and for 17% of total respiration when homogenates were incubated in 80% oxygen. Exposure of rats to 85% oxygen for 7 days induces tolerance to the lethal effects of 100% oxygen. Rats which previously breathed 85% oxygen for 7 days had a greater CN^?-resistant respiration than control rats. This implies that adaptation to hyperoxia does not include decreased lung tissue oxygen radical production as indicated by CN^?-resistant respiration. One possible explanation for the increased CN^?-resistant respiration in oxygen tolerant rat lungs is that they contain increased cell mass. Lung homogenates of rats exposed to 85% oxygen for 7 days also had 2.5 times greater thiobarbituric acid positive material than controls, indicating that increased lung lipid peroxidation occurs as a consequence of hyperoxia. Incubation of normal rat lung homogenates under hyperoxic conditions also acutely increased lipid peroxidation, which could be inhibited by both superoxide dismutase and catalase. This confirms that hyperoxia enhances cellular production of O₂^? and H₂O₂ and implies an essential role for both O₂^? and H₂O₂ in hyperoxic lung damage.     

Effects of oxygen fraction in inspired air on force production and electromyogram activity during ergometer rowing

Six male rowers rowed maximally for 2500 m in ergometer tests during normoxia (fractional concentration of oxygen in inspired air, F _IO₂ 0.209), in hyperoxia (F _IO₂ 0.622) and in hypoxia (F _IO₂ 0.158) in a randomized single-blind fashion. Oxygen consumption (V˙O₂), force production of strokes as well as integrated electromyographs (iEMG) and mean power frequency (MPF) from seven muscles were measured in 500-m intervals. The iEMG signals from individual muscles were summed to represent overall electrical activity of these muscles (sum-iEMG). Maximal force of a stroke (F _max) decreased from the 100% pre-exercise maximal value to 67 (SD 12)%, 63 (SD 15)% and 76 (SD 13)% (P<0.05 to normoxia, ANOVA) and impulse to 78 (SD 4)%, 75 (SD 14)% and 84 (SD 7)% (P<0.05) in normoxia, hypoxia and hyperoxia, respectively. A strong correlation between F _max and V˙O₂ was found in normoxia but not in hypoxia and hyperoxia. The mean sum-iEMG tended to be lower (P<0.05) in hypoxia than in normoxia but hyperoxia had no significant effect on it. In general, F _IO₂ did not affect MPF of individual muscles. In conclusion, it was found that force output during ergometer rowing was impaired during hypoxia and improved during hyperoxia when compared with normoxia. Moreover, the changes in force output were only partly accompanied by changes in muscle electrical activity as sum-iEMG was affected by hypoxic but not by hyperoxic gas. The lack of a significant correlation between F _max and V˙O₂ during hypoxia and hyperoxia may suggest a partial uncoupling of these processes and the existence of other limiting factors in addition to V˙O₂. Accepted: 2 June 1997     

Effect of reactive oxygen and carbonyl species on crucial cellular antioxidant enzymes

Numerous reactive oxygen species (ROS) and reactive carbonyl species (RCS) issuing from lipid and sugar oxidation are known to damage a large number of proteins leading to enzyme inhibition and alteration of cellular functions. Whereas studies in literature only focus on the reactivity of one or two of these compounds, we aimed at comparing in the same conditions of incubations (4 and 24 h at 37 °C) the effects of both various RCS (4-hydroxynonenal, 4-hydroxyhexenal, acrolein, methylglyoxal, glyoxal, malondialdehyde) and ROS (H₂O₂, AAPH) on the activity of key enzymes involved in cellular oxidative stress: superoxide dismutase (Cu,Zn-SOD), glutathione peroxidase (GPx), glutathione S-transferase (GST) and glucose-6-phosphate dehydrogenase (G6PDH). This was realized both in vitro on purified proteins and MIAPaCa-2 cells. Incubation of these enzymes with RCS resulted in a significant time- and concentration-dependent inhibition for both pure enzymes and in cell lysates. Among all RCS and ROS, hydroxynonenal (HNE) was observed as the most toxic for all studied enzymes except for SOD and is followed by hydrogen peroxide. At 100 μM, HNE resulted in a 50% reduction of GPx, 56% of GST, 65% of G6PDH, and only 10% of Cu,Zn-SOD. Meanwhile it seems that concentrations used in our study are closer to biological conditions for ROS than for RCS. H₂O₂ and AAPH-induced peroxyl radicals may be probably more toxic towards the studied enzymes in vivo.     

Effect of 4-hydroxynonenal on superoxide anion production from primed human neutrophils

HNE (4-hydroxy-2,3-trans-nonenal), an aldehydic product of lipid peroxidation, has been reported to modulate different functional parameters of human and rat neutrophils (PMNs), such as chemiluminescence, migration and some enzymatic activities, thus exerting effects that varied according to the concentration tested. Experiments were done to evaluate the effects of HNE on superoxide anion (O₂^?.) production from human PMNs, isolated from healthy volunteers. After having tested that HNE by itself was not able to activate the cells, comparisons were made between its effects on PMNs, stimulated by either a single stimulus, N-formyl-methionyl-leucyl-phenylalanine (FMLP), or a combination of stimuli, such as FMLP and the neuropeptide substance P (SP; primed PMNs). In the concentration range tested (10^?12–10^?4 M ), HNE inhibited FMLP-evoked O₂^?. production with an IC₅₀ of 11·6 ± 1·5 × 10^?6 M ; at concentrations ≤10^?6 M , HNE enhanced O₂^?. production elicited by FMLP + SP, while higher concentrations were inhibitory. There was a bell-shaped dose–response curve to the enhancing effects of HNE, depending on the incubation time being recorded after only short periods (≤5 min) of the exposure of the cells to HNE; this was not shown by structurally-related aldehydes, such as 2-nonenal and nonanal. These results suggest that low concentrations of HNE may participate in the evolution of the inflammatory process, by contributing to the activation of PMNs. The effects of high concentrations of the aldehyde may represent a mechanism which contributes to the regulation of the extent of the inflammatory response.     

Hyperoxia Induces S-Phase Cell-Cycle Arrest and p21-Independent Cdk2 Inhibition in Human Carcinoma T47D-H3 Cells

Little is known about cell-cycle checkpoint activation by oxidative stress in mammalian cells. The effects of hyperoxia on cell-cycle progression were investigated in asynchronous human T47D-H3 cells, which contain mutated p53 and fail to arrest at G1/S in response to DNA damage. Hyperoxic exposure (95% O₂, 40–64 h) induced an S-phase arrest associated with acute inhibition of Cdk2 activity and DNA synthesis. In contrast, exit from G2/M was not inhibited in these cells. After 40 h of hyperoxia, these effects were partially reversible during recovery under normoxic conditions. The inhibition of Cdk2 activity was not due to degradation of Cdk2, cyclin E or A, nor impairment of Cdk2 complex formation with cyclin A or E and p21^Cip1. The loss of Cdk2 activity occurred in the absence of induction and recruitment of cdk inhibitor p21^Cip1 or p27^Kip1 in cyclin A/Cdk2 or cyclin E/Cdk2 complexes. In contrast, Cdk2 inhibition was associated with increased Cdk2-Tyr15 phosphorylation, increased E2F-1 recruitment, and decreased PCNA contents in Cdk2 complexes. The latter results indicate a p21^Cip1/p27^Kip1-independent mechanism of S-phase checkpoint activation in the hyperoxic T47D cell model investigated.     

Effect of oxygen radicals and hyperoxia on rat heart ornithine decarboxylase activity

Rat heart ornithine decarboxylase activity from isoproterenol-treated rats was inactivated in vitro by reactive species of oxygen generated by the reaction xanthine/xanthine oxidase. Reduced glutathione, dithiothreitol and superoxide dismutase had a protective effect in homogenates and in partially purified ornithine decarboxylase exposed to the xanthine/xanthine oxidase reaction, while diethyldithiocarbamate, which is an inhibitor of superoxide dismutase, potentiated the damage induced by O₂^? on enzyme activity. Dithiothreitol at concentrations above 1.25 mM had an inhibitory effect oupon supernatant ornithine decarboxylase activity, while at 2.5 mM it was most effective in the recovery of ornithine decarboxylase activity, after the purification of the enzyme by the ammonium sulphate precipitation procedure. The ornithine decarboxylase inactivated by the xanthine/xanthine oxidase reaction showed a higher value of K_m and a reduction of V_max with respect to control activity. The exposure of rates to 100% oxygen for 3 h reduced significantly the isoproterenol-induced heart ornithine decarboxylase activity. The injection with diethyldithiocarbamate 1 h before hyperoxic exposure further reduced heart ornithine decarboxylase activity.     

Hyperbaric oxygen enhances apoptosis in hematopoietic cells

Hyperbaric oxygen (HBO) is 100% oxygen administered at elevated atmospheric pressure. In this study, we examined the effect of HBO on hematopoietic cell apoptosis. Cells exposed to HBO were incubated in a chamber containing 97.9% O₂ and 2.1% CO₂ at 2.4 atmospheres absolute (ATA). HBO enhanced spontaneous HL-60 cell apoptosis in a time-dependent manner; a 12 h exposure increased apoptosis by 42%. Exposing these cells to hyperoxia at standard atmospheric pressure (95% O₂, 5% CO₂ at 1 ATA) or increased pressure alone (8.75% O₂, 2.1% CO₂ at 2.4 ATA) had minimal effect on apoptosis. HBO also enhanced stimulus-induced apoptosis. HL-60 cells stimulated to die using radiation underwent 33% more apoptosis than cells exposed to radiation alone. HBO enhanced melphalan, camptothecin, and chlorambucil-induced apoptosis by 22%, 13%, and 8%, respectively. Jurkat cells stimulated to die with anti-Fas antibody underwent 44% more apoptosis when exposed to HBO. Spontaneous apoptosis was increased by 15% in HBO-exposed murine thymocytes. HBO's effect on apoptosis did not require new protein synthesis. As expected, HBO exposure increased the intracellular concentration of H₂O₂. Incubating HL-60 cells in the presence of dehydroascorbic acid partially abrogated HBO-induced increases in intracellular H₂O₂ and apoptosis. In summary, HBO enhances spontaneous and stimulus-induced apoptosis in hematopoietic cells, at least in part, by enhancing the intracellular accumulation of H₂O₂.     

Responses of type II pneumocyte antioxidant enzymes to normoxic and hyperoxic culture

Summary Cultured type II pneumocyte responses to in vitro normoxia (95% air: 5% CO₂) or hyperoxia (95% O₂:5% CO₂) were quantified. Normoxic culture (0 to 96 h) of rabbit type II cells resulted in enhanced cell-monolayer protein and DNA content. During this same time, cellular activities of superoxide dismutase (SOD), catalase, and glutathione peroxidase (GSH Px) decreased. Compared to cultures maintained in normoxia, hyperoxic exposure of cultures resulted in decreased cell-associated protein and DNA content. Exposure to hyperoxia also resulted in cytotoxicity as demonstrated by elevated cellular release of DNA, lactate dehydrogenase (LDH), and preincorporated 8-[¹⁴C]adenine. Cellular catalase and GSH Px activities in hyperoxic cells decreased similarly to normoxic controls. In contrast, cellular SOD activity in hyperoxic cells decreased less than in normoxic cultures. Cellular SOD activity in hyperoxic cultures, when normalized for cellular protein, but not DNA, was greater than normoxic values after 24 to 96 h of exposure. Unlike the decrease in cellular antioxidant enzymes during normoxic and hyperoxic culture, cellular LDH activity increased during both these exposures. Cellular LDH activity in 24 to 96 h hyperoxia-exposed cells increased to a lesser extent than normoxic controls. The extent of depression in LDH activity was dependent on whether the activity was normalized for cellular protein or DNA. Type II pneumocytes, which normally undergo hyperplasia and hypertrophy during hyperoxia in vivo, exhibited oxygen sensitivity in vitro. Exposure of type II cells to hyperoxia in vitro resulted in alterations in cellular SOD and LDH activities, but recognition of such changes were dependent on whether enzymatic activities were normalized for cellular DNA or protein. This work was supported by a grant from the Health Effects Institute, grant HL40458 from the National Institutes of Health, Bethesda, MD, and a grant from the American Lung Association, New York, NY.     

Responses of superoxide dismutase and glutathione reductase activities in cotton leaf tissue exposed to an atmosphere enriched in oxygen

Responses of superoxide dismutase (EC 1.15.1.1) and glutathione reductase (EC 1.6.4.2) activities were evaluated in leaf tissue from intact cotton plants (Cotton Branch 1697) which were exposed to 75% O₂, 350 microliters per liter CO₂ for 48 hours. Soluble protein was extracted from O₂-treated and control tissue, and enzyme levels were determined. Superoxide dismutase activity in cotton leaf tissue was high (26 units per milligram protein) under normal conditions of 21% O₂, saturating light, and limiting CO₂, and neither qualitative nor quantitative differences in the cyanide-sensitive or -insensitive forms of the enzyme occurred in response to hyperoxic conditions. Glutathione reductase activity, however, was 2- to 3-fold higher in extracts from tissue exposed to 75% O₂. No increase in activity was observed for the peroxisomal enzymes, glycolate oxidase (EC 1.1.3.1) and catalase (EC 1.11.1.6). Results are consistent with an integrated pathway involving superoxide dismutase and glutathione reductase for protection of sensitive leaf components against detrimental effects of intermediate reduction products of O₂.     

Lipid peroxidation product 4-hydroxy-2-nonenal modulates base excision repair in human cells

Oxidative-stress-driven lipid peroxidation (LPO) is involved in the pathogenesis of several human diseases, including cancer. LPO products react with cellular proteins changing their properties, and with DNA bases to form mutagenic etheno-DNA adducts, removed from DNA mainly by the base excision repair (BER) pathway.One of the major reactive aldehydes generated by LPO is 4-hydroxy-2-nonenal (HNE). We investigated the effect of HNE on BER enzymes in human cells and in vitro. K21 cells pretreated with physiological HNE concentrations were more sensitive to oxidative and alkylating agents, H₂O₂ and MMS, than were untreated cells. Detailed examination of the effects of HNE on particular stages of BER in K21 cells revealed that HNE decreases the rate of excision of 1,N⁶-ethenoadenine (ɛA) and 3,N⁴-ethenocytosine (ɛC), but not of 8-oxoguanine. Simultaneously HNE increased the rate of AP-site incision and blocked the re-ligation step after the gap-filling by DNA polymerases. This suggested that HNE increases the number of unrepaired single-strand breaks (SSBs) in cells treated with oxidizing or methylating agents. Indeed, preincubation of cells with HNE and their subsequent treatment with H₂O₂ or MMS increased the number of nuclear poly(ADP-ribose) foci, known to appear in cells in response to SSBs. However, when purified BER enzymes were exposed to HNE, only ANPG and TDG glycosylases excising ɛA and ɛC from DNA were inhibited, and only at high HNE concentrations. APE1 endonuclease and 8-oxoG-DNA glycosylase 1 (OGG1) were not inhibited. These results indicate that LPO products exert their promutagenic action not only by forming DNA adducts, but in part also by compromising the BER pathway.     

Effect of hypoxia and reoxygenation on the formation and release of reactive oxygen species by porcine pulmonary artery endothelial cells

Endothelial cells are critical targets in both hypoxia-and reoxygenation-mediated lung injury. Reactive O₂ species (ROS) have been implicated in the pathogenesis of hypoxic and reoxygenation lung injury, and xanthine dehydrogenase/oxidase (XDH/XO) is a major generator of the ROS. Porcine pulmonary artery endothelial cells (PAEC) have no detectable XDH/XO. This study was undertaken to examine (1) ROS production by hypoxic porcine PAEC and their mitochondria and (2) ROS production and injury in reoxygenated PAEC lacking XDH/XO activity. Intracellular H₂O₂ generation and extracellular H₂O₂ and O/₂ release were measured after exposure to normoxia (room air-5% CO₂), hypoxia (0% O₂ -95% N-5% CO₂), or hypoxia followed by normoxia or hyperoxia (95% O₂-5% CO₂). Exposure to hypoxia results in significant reductions in intracellular H₂ O₂ formation and extracellular release of H₂ O₂ and O₂ by PAEC and mitochondria. The reductions occur with as little as a 2 h exposure and progress with continued exposure. During reoxygenation, cytotoxicity was not observed, and the production of ROS by PAEC and their mitochondria never exceeded levels observed in normoxic cells. The absence of XDH/XO may prevent porcine PAEC from developing injury and increased ROS production during reoxygenation. © 1995 Wiley-Liss, Inc.     

Optimal oxygen tension for human lymphocytes in culture

Human lymphocytes stimulated with phytohaemagglutinin in vitro show maximal ¹⁴C-thymidine incorporation when the gas phase of the cultures contains 20% O₂. The rate of DNA synthesis is more affected by hyperoxic conditions than by hypoxia. Cell viability, defined by dye exclusion, is considerably less dependent on pO₂ than is cellular replication.     

Influence of glutathione on zinc-mediated cellular toxicity

The effect of zinc on various pulmonary cell lines has been studied by measuring the depletion of total cellular glutathione after exposure to zinc(II) chloride at different concentrations. Total cellular glutathione (cGS) was measured at 31 ± 3 nmol/mg, 3.8 ± 0.6 nmol/mg, and 3.7 ±1.2 nmol/mg protein in A549, L2, and 11Lu cells, respectively. After treatment with buthionine sulfoximine (BSO), the cGS levels decreased by 20% in A549 cells and below 0.2 nmol/mg in L2 and 11Lu cells. Exposure of A549 cells to 25–200 μM ZnCl₂ for 4 h alone decreased the cGS content to 60–80%. There was little additional effect in BSO-pretreated cells. In L2 and 11Lu cells, the decrease of cGS was 70–85% following exposure to 15–150 μM ZnCl₂ for 2 h. If BSO was also used, the decrease in cGS was 85–95% in L2 cells and 75–85% in 11Lu cells. Exposure to 25–250 μM ZnCl₂ for 2 h diminished protein synthesis as determined by radiolabeled methionine incorporation, with half-maximum inhibition (EC₅₀) from 40–160 μM ZnCl₂. To attain similar EC₅₀ values in BSO-pretreated cells, only about half the zinc concentrations were required as compared to cells without pretreatment. The decrease of cGS was accompanied by an increased ratio of oxidized : reduced glutathione that was more pronounced in cells with low glutathione content.     

Acrolein toxicity: Comparison with reactive oxygen species

The toxicity of acrolein was compared with that of reactive oxygen species using a mouse mammary carcinoma FM3A cell culture system. Complete inhibition of cell growth was accomplished with 10 μM acrolein, 100 μM H₂O₂, and 20 μM H₂O₂ plus 1 mM vitamin C, which produce OH, suggesting that toxicity of acrolein is more severe than H₂O₂ and nearly equal to that of OH, when these compounds were added extracellularly. Acrolein toxicity was prevented by N-acetyl-l-cysteine and N-benzylhydroxylamine, and attenuated by putrescine and spermidine. Toxicity of H₂O₂ was prevented by glutathione peroxidase plus N-acetyl-l-cysteine, pyruvate, catalase, and reduced by polyphenol, and toxicity of OH was prevented by glutathione peroxidase plus N-acetyl-l-cysteine, pyruvate, catalase and reduced by N-acetyl-l-cysteine. The results indicate that prevention of cell toxicity by N-acetyl-l-cysteine was more effective with acrolein than with OH. Protein and DNA synthesis was damaged primarily by acrolein and reactive oxygen species, respectively.