Competition between paraquat and putrescine for uptake by suspensions of rat alveolar type II cells.

Paraquat and the structurally similar polyamines, such as putrescine and spermidine, are accumulated actively and selectively by the alveolar type II cells via the polyamine uptake system. We report the uptake kinetics of paraquat and putrescine and their mutual inhibition in freshly isolated rat type II cell suspensions. The uptake of paraquat by type II cells exhibited saturation kinetics and could be inhibited in a concentration-dependent manner by putrescine. By applying enzyme kinetic analysis to our experimental data it was demonstrated that the uptake of paraquat or putrescine is inhibited in a partially competitive manner by the respective inhibitor. Thus, we postulate that the polyamine uptake pathway in type II cells for paraquat and putrescine has two separate sites, one for each substrate, and that binding of one leads to a conformational change in the other.     

Mechanism of protection of alveolar type II cells against paraquat-induced cytotoxicity by deferoxamine

Paraquat toxicity has been associated with the generation of free radicals in alveolar epithelial cells in which paraquat specifically accumulates via a polyamine uptake system. In the present study we investigated whether deferoxamine (DF), an iron chelator that has antioxidant capacity and that also has a polyamine-like structure, could protect alveolar type II cells (ATTC) against injury by paraquat. Radiolabeled [3H]adenine ATTC were incubated in a medium containing 75 microM paraquat in the absence or presence of DF (500 microM). After 3 hr of incubation paraquat-mediated cytotoxicity of ATTC, as measured by [3H]adenine release, was significantly (P less than 0.005) decreased by addition of DF (26.6 +/- 2.6% vs 7.4 +/- 1.7%). Accumulation of radiolabeled [14C]paraquat at a concentration of 75 microM was also decreased (70%) by 500 microM DF from 94.8 +/- 2.1 to 28.9 +/- 6.7 nmoles paraquat/2.5 x 10(5) ATTC. This effect of DF was dose dependent and comparable with the protective effect of equimolar concentrations of putrescine. However, per cent uptake of paraquat at a concentration of 500 microM was not significantly inhibited by DF (1 mM), whereas paraquat-induced injury was still markedly reduced (36.2 +/- 2.5% vs 2.6 +/- 4.2%). This indicated that the protective effect of DF could not be explained by its competition with paraquat on uptake alone. In the same series of experiments using another iron chelator, pyridoxal benzoyl hydrazone (PBH), which has antioxidant properties similar to DF but does not show its polyamine-like structure, ATTC lysis was also prevented although paraquat uptake was not reduced. These in vitro data indicate that the mechanism of protection by DF against paraquat toxicity in lung epithelial type II cells is two-fold: inhibition of paraquat uptake through its compliance with the structural requirements necessary for transport, and inhibition of paraquat-induced iron-catalysed free radical generation.     

Kinetics of paraquat in the isolated rat lung: Influence of sodium depletion

Paraquat accumulates in the lung through a characteristic polyamine uptake system. It has been previously shown that paraquat uptake can be significantly prevented if extracellular sodium (Na+) is reduced, although the available data correspond to experiments performed using tissue slices or incubated cells. This type of in vitro study fails to give information on the actual behaviour occurring in vivo since the anatomy and physiology of the studied tissue is disrupted. Accordingly, the aim of the present study was to explore the usefulness of the isolated rat lung model when applied to characterize the kinetic behaviour of paraquat in this tissue after bolus injection under standard experimental conditions as well as to evaluate the influence of iso-osmotic replacement of Na+ by lithium (Li+) in the perfusion medium. The obtained results show that the present isolated rat lung model is useful for the analysis of paraquat toxicokinetics, which is reported herein for the first time. It was also observed that Na+ depletion in the perfusion medium leads to a decreased uptake of paraquat in the isolated rat lung, although it seems that this condition does not contribute to improve the elimination of paraquat once the herbicide reaches the extravascular structures of the tissue, since the paraquat tissue wash-out phase is similar under both experimental conditions assayed.     

Uptake and efflux of 14C-paraquat by rat lung slices: the effect of imipramine and other drugs.

Paraquat accumulation by rat lung slices incubated at 10 or 100 μm concentration was linear with time and the accumulated paraquat was “noneffluxable.” Imipramine (100 or 500 μm) inhibited paraquat (10 μm) uptake by 38 and 85%, respectively, and 500 μm imipramine enhanced paraquat efflux by 40%. The combination of impaired uptake and enhanced efflux suggested the possibility that imipramine might reduce the toxicity of paraquat in intact animals. However, at the doses used, imipramine did not alter paraquat toxicity in vivo. Eleven other drugs were shown to inhibit uptake, but only five enhanced paraquat efflux.     

Putrescine and 5-hydroxytryptamine accumulation in rat lung slices: cellular localization and responses to cell-specific lung injury

The cellular localization of putrescine (1,4-diaminobutane) and 5-hydroxytryptamine (5HT) following the accumulation of tritium-labeled putrescine (2.5 microM) or 5HT (0.5 microM) into rat lung slices was determined by autoradiography at the light microscope level. Putrescine labeling was found to occur in type II alveolar epithelial cells and in branchiolar nonciliated (Clara) cells, and possibly also in type I alveolar epithelial cells. The pattern of 5HT labeling was clearly different from that with putrescine, since the parenchyma was diffusely labeled with no preferential location in type II cells, but with strong labeling of the endothelium of large vessels and also the pleural mesothelium. The apparent kinetic parameters for the tissue uptake of [3H]putrescine (2.5 to 80 microM) and [14C]5HT (0.5 to 16 microM; both being simultaneously present in a 5 to 1 molar ratio) were studied in lung slices from normal rats and rats pretreated with O,S,S-trimethyl phosphorodithioate (OSSMe, 11 to 95 mg/kg, po), with paraquat (20 mg/kg, ip), or with alpha-naphthylthiourea (ANTU, 5 or 10 mg/kg, ip). OSSMe and paraquat were used as models for pulmonary epithelium-damaging agents, and ANTU was taken as a model for a pulmonary endothelium-damaging agent. The Vmax for the uptake of 5HT was significantly increased (without change in Km) following treatment with OSSMe and paraquat. Following ANTU treatment the Vmax for the uptake of 5HT was unchanged (5 mg/kg) or increased (10 mg/kg, Km also increased). These results indicate that in lung slices the response to lung injury may be associated with an increased accumulation of 5HT. The Vmax for the uptake of putrescine was significantly decreased (without change in Km) following treatment with OSSMe and paraquat. Following ANTU treatment the Vmax for the uptake of putrescine was unchanged (5 mg/kg) or decreased (10 mg/kg, no change in Km). These results suggest that a decreased putrescine uptake is a sensitive index of pulmonary epithelial damage.     

The accumulation and localisation of putrescine, spermidine, spermine and paraquat in the rat lung. In vitro and in vivo studies

Putrescine was accumulated into the isolated perfused rat lung by a temperature dependent process. The uptake obeyed saturation kinetics for which an apparent Km of 14 microM and Vmax of 48 nmol/g wet wt/hr was derived. After rats were dosed subcutaneously with [14C]putrescine, it was accumulated in the lung to concentrations greater than that in the plasma with the highest amount found between 3 and 12 hr. From 3 hr after dosing until 24 hr, there was a progressive increase in 14C label incorporated into spermidine, indicating that putrescine was converted to spermidine. Using autoradiographic techniques in lung slices the [3H]oligoamines were found in the alveolar epithelial type II. Clara and very probably the alveolar type I cells. With [3H]paraquat, the presence was detected only in the alveolar type II cells. Likewise, in the isolated perfused rat lung or following s.c. dosing of rats with [3H]putrescine the radiolabel was located only in the alveolar type II cell. We have suggested that the most likely explanation for the differences in localisation of label between in vitro and in vivo studies resulted from the use of [3H] label of different specific activity. Consequently we have concluded that the cell types with the ability of accumulate paraquat and oligoamines were the alveolar epithelial type I and type II cells and Clara cells.     

Captopril inhibits the pulmonary toxicity of paraquat in rats

Paraquat (PQ) is a herbicide that is very toxic to all living organisms. It generates free radicals and leads to acute or chronic lung injury. Free radicals are often associated with fibrogenesis, which occurs in various disease states. The purpose of this study was to determine whether captopril prevents paraquat toxicity in lung tissue. Paraquat alone increased the level of lipid peroxidation (LPO) and the activity of superoxide dismutase (SOD) after 4, 12, 24 and 72 h of administration. Also, the level of hydroxyproline showed an increase after 24 h of paraquat administration. However, paraquat also decreased the level of glutathione (GSH) and the activity of glutathione peroxidase (GSH-Px). Captopril (50 mg/kg i.p.) and paraquat were simultaneously injected (40 mg/kg i.p.), and the captopril injection 1 h after paraquat ameliorated the biochemical toxicity induced by paraquat. This was evidenced by a significant reduction in LPO and balancing the endogenous antioxidant capacity by normalizing the activities of SOD and GSH-Px and the GSH content in the lung tissue. Moreover, captopril injection prevented the increase of hydroxyproline content as an index of lung fibrosis. From these results, the beneficial effects of captopril on paraquat toxicity appear to be through enhancement of the endogenous antioxidant system preventing the lung fibrosis.     

The relationship between 5-hydroxytryptamine and paraquat accumulation into rat lung

The uptake of 5-hydroxytryptamine (5HT) into rat lung slices has been shown to obey saturation kinetics and to be inhibited by imipramine, metabolic inhibitors and a sodium-free medium. The apparent K_m for the uptake process was found to be 3.3 μMwith a V_max of 6 nmoles/g wet wt/min. Lung slices taken from rats given a dose of paraquat known to damage type I and type II lung epithelial cells showed inhibition of paraquat uptake but no inhibition of 5HT uptake. This together with the stimulation of paraquat accumulation into rat lung slices in a sodium-free medium leads to the conclusion that the uptake of paraquat and 5HT into the lung does not occur in the same cell type.     

Structure-activity correlations of amines inhibiting active uptake of paraquat (Methyl viologen) into rat lung slices

Analysis of amine structure with respect to inhibitory potency utilized a new method for determining equipotent inhibitor concentrations of paraquat uptake by lung slices. Fifteen N-alkyl homologues of paraquat (viologens) were tested and inhibition of lung uptake of paraquat was found to be a function of the inductive effect and steric bulk of groups attached to the nitrogens of 4,4′-bipyridyl. Several classes of amine inhibitors were examined. Polyamines were generally more potent than compounds containing only one quaternizable nitrogen at pH 7.4. α, ω-Diaminoalkanes were the most potent inhibitors of paraquat accumulation by lung slices.     

Metabolic alterations in hepatocytes promoted by the herbicides paraquat,dinoseb and 2,4-D

The cytotoxic effects of the herbicides paraquat (1,1′-dimethyl-4,4′-bipyridylium dichloride), dinoseb (2-sec-butyl-4,6-dinitrophenol) and 2,4-D (2,4-dichlorophenoxyacetic acid) on freshly isolated rat hepatocytes were investigated. Paraquat and 2,4-D (1–10 mM) caused a dose and time dependent cell death accompanied by depletion of intracellular glutathione (GSH) and mirroring increase of oxidized glutathione (GSSG). Dinoseb, the most effective cytotoxic compound under study (used in concentrations 1000 fold lower than paraquat and 2,4-D), exhibited moderate effects upon the level of GSH and GSSG. These limited effects are at variance with significant effects upon the adenine and pyridine nucleotide contents. ATP and NADH levels are rapidly depleted by herbicide metabolism. This depletion is observed in the millimolar range for paraquat and 2,4-D and in the micromolar range for dinoseb. 2,4-D completely depletes cellular ATP, with subsequent cell death, as detected by LDH leakage. Paraquat rapidly depletes NADH, according to the redox cycling of the herbicide metabolism. The most effective compound is dinoseb since it exerts similar effects as described for paraquat and 2,4-D at concentrations 1000 fold lower. Simultaneously with NADH and ATP depletion, the levels of ADP, AMP and NAd+ increase in hepatocytes incubated in the presence of the herbicides. In contrast to NADH, the time course and extent of ATP depletion and fall in energy charge correlate reasonably with the time of onset and rate of cell death. It is concluded that the herbicides, paraquat and 2,4-D are hepatotoxic and initiate the process of cell death by decreasing cellular GSH. As a consequence of this primary disturbance, alteration of adenine and pyridine nucleotides contents is a critical event in the induction of irreversible cell injury.     

辅助性T细胞17活化在百草枯致小鼠肺纤维化模型中的作用

目的 观察辅助性T细胞17(Th17)活化在百草枯致小鼠肺纤维化模型中的作用.方法 将30只C57BL/6小鼠随机分为对照组(10只)和百草枯中毒组(20只).百草枯中毒组经腹腔注射百草枯35 mg/kg建立小鼠中毒模型;对照组经腹腔注入等体积生理盐水.观察2组小鼠肺脏组织病理变化情况及肺纤维化评分,检测外周血Th17细胞、肺组织中维甲酸相关核孤儿受体γt(RORγt)基因及蛋白的表达情况.结果 百草枯中毒组肺平均内衬间隔、肺泡腔与肺总面积比高于对照组,平均肺泡数低于对照组,外周血Th17细胞阳性率高于对照组,肺组织中ROR-γtmRNA和蛋白表达水平高于对照组(P＜0.01).结论 百草枯致小鼠肺纤维化模型外周血及肺组织中Th17细胞明显活化,提示Th17细胞活化在百草枯致肺纤维化过程中起着重要作用.     

The relevance of pentose phosphate pathway stimulation in rat lung to the mechanism of paraquat toxicity

Paraquat and diquat have been shown to stimulate the production of ¹⁴CO₂ from [1-¹⁴C]-glucose by slices of rat lung, but not the production of ¹⁴CO₂ from [6-¹⁴C]glucose. This indicates stimulation of the pentose phosphate pathway. Paraquat was effective at concentrations as low as 7.5 × 10^?7 M whilst a concentration of diquat of 10^?5 M was required for comparable stimulation. Maximal stimulation occurred with approximately 10^?5 M paraquat and approximately 10^?4 M diquat. The stimulation of pentose phosphate pathway in lung slices by paraquat has been shown to be related to paraquat accumulation.Lung slices from rats dosed intravenously with 65 μmoles of either paraquat or diquat/kg body wt had increased pentose phosphate pathway activity compared with slices from saline injected controls. At all times studied from 0.5 to 18 hr after injection, pentose phosphate pathway activity in slices from diquat poisoned rats was equal to or greater than that observed in slices from paraquat poisoned rats. Since only rats dosed intravenously with paraquat subsequently develop lung damage, it is concluded that there is no simple relationship between stimulation of the pentose phosphate pathway in lung and the production of lung damage.     

Molecular features necessary for the uptake of diamines and related compounds by the polyamine receptor of rat lung slices.

The influence of 17 putrescine analogues on the uptake of putrescine and/or paraquat by rat lung slices has been determined. Most of these compounds are competitive inhibitors of putrescine and/or paraquat uptake, but three show no inhibiting activity. Apparent Ki values of the putrescine derivatives increase, and thus the inhibitory effects decrease, with increasing N-methylation. Comparison of N-methyl-1,4-diaminobutane (Ki = 8 microM) with N,N'-bis-methyl-1,4-diaminobutane (Ki = 25.5 microM) shows that a single primary amino group is desirable for high inhibiting activity. Dimethylation at one amino function does not greatly decrease inhibitory potential (thus N,N-dimethyl-1,4-diaminobutane has Ki = 11.5 microM). Increasing the size of N-alkyl substituents in putrescine derivatives, decreased their inhibitory action on the uptake of putrescine. Investigation of the effect of conformationally-restricted analogues of putrescine shows that both (E) and (Z) isomers of 1,4-diaminobut-2-ene are poor inhibitors of putrescine uptake. Analogues of putrescine with bulky substituents on the butyl chain, i.e. the meso- and rac-isomers of 1,1-dichloro-2,3-diaminomethylcyclopropane, do not inhibit putrescine uptake. Inhibiting putrescine derivatives which contain aziridine groups are competitive inhibitors of putrescine and paraquat uptake. Surprisingly, N-(4-aminobutyl)aziridine is the most effective inhibitor of putrescine uptake studied, and is a better inhibitor of paraquat uptake than the endogenous polyamine, putrescine. N-(4-Aminobutyl)aziridine binds reversibly to the polyamine transporter and its inhibitory effects do not appear to be due to any cytotoxic activity of the aziridine. The parameter A (mM)-1 defined as 1000/Ki (where Ki units are microM) was taken as a measure of the affinity of a compound for the polyamine receptor in this paper.     

Paraquat induces cyclooxygenase-2 (COX-2) implicated toxicity in human neuroblastoma SH-SY5Y cells

Paraquat produces dopaminergic pathologies of Parkinson’s disease, in which cyclooxygenase-2 (COX-2) is implicated. However, it is unclear whether paraquat induces toxicity within dopaminergic neurons through COX-2. To address this, human neuroblastoma SH-SY5Y cells were treated with paraquat and then the involving mechanism of COX-2 was investigated. We initially examined the involvement of COX-2 in paraquat-induced toxicity. Data suggest that COX-2 is implicated in paraquat-induced reduction of viability in SY5Y cells. Then, to confirm the presence of COX-2 in SY5Y cells, we examined COX-2 mRNA and protein levels, which are regulated by NF-κB. Data indicate that paraquat activates NF-κB and up-regulates COX-2. We then checked quinone-bound proteins as quinones produced by COX-2 bind to intracellular proteins. Paraquat obviously forms quinone-bound proteins, in particular, quinone-bound DJ-1 and this formation is attenuated by meloxicam. Finally, we investigated antioxidant system including nuclear factor erythroid-related factor 2 (Nrf2), gamma glutamylcysteine synthetase (γGCS), and glutathione (GSH) as DJ-1 is linked to Nrf2 and Nrf2 regulates γGCS expression and γGCS is a GSH synthesis enzyme. Paraquat decreases protein levels of Nrf2 and γGCS and intracellular GSH level and these decreases are alleviated by meloxicam. Therefore, collectively, our data indicate that paraquat induces COX-2 implicated toxicity in SY5Y cells. In conclusion, current findings support the idea that paraquat might produce toxicity in dopaminergic neurons through COX-2.     

Effects of 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) on the levels of glutathione and lipid peroxidation and the activity of glutathione reductase in liver and lung

After subcutaneous injection of 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU) to rats, glutathione reductase activity in lung and liver diminished rapidly. The restoration of enzyme activity occurred more slowly in the lung than in the liver. The pattern for the time-course of total glutathione (GSH) levels was similar between lung and liver, except for a marked depression of hepatic levels 6 h after drug administration. The level of malondialdehyde (MDA) in lung was not affected by BCNU throughout the experimental period (3 days). However, the level in liver had increased significantly by 6 h after drug administration. These observations indicate that lipid peroxidation in lung was not induced by BCNU even when glutathione reductase activity was markedly diminished. In contrast, the lipid peroxidation in liver was induced by BCNU and was preceded by an early marked depression in total GSH.     

Interconversion of NAD(H) to NADP(H). A cellular response to quinone-induced oxidative stress in isolated hepatocytes

Quinones may be toxic by a number of mechanisms, including oxidative stress caused by redox cycling and arylation. This study has compared the cytotoxicity of four quinones, with differing abilities to arylate cellular nucleophiles and redox cycle, in relation to their effects on cellular pyridine nucleotides and ATP levels in rat hepatocytes. Non-toxic concentrations (50 microM) of menadione (redox cycles and arylates), 2-hydroxy-1,4-naphthoquinone (neither arylates nor redox cycles via a one electron reduction) and 2,3-dimethoxy-1,4-naphthoquinone (a pure redox cycler) all caused markedly similar changes in cellular pyridine nucleotides. An initial decrease in NAD+ was accompanied by a small, transient increase in NADP+ and followed by a larger, prolonged increased in NADPH and total NADP+ + NADPH. At toxic concentrations (200 microM), the quinones caused an extensive depletion of NAD(H), an increase in levels of NADP+ and an initial rise in total NADP+ + NADPH, prior to a decrease in ATP levels and cell death. Nucleotide changes were not observed with non-toxic (20 microM) or toxic (100 microM) concentrations of p-benzoquinone (a pure arylator) and ATP loss accompanied or followed cell death. A novel mechanism for the activation of 2-hydroxy-1,4-naphthoquinone has been implicated. Our findings also suggest that a primary event in the response of the cell to redox cycling quinones is to bring about an interconversion of pyridine nucleotides, possibly mediated by an NAD+ reduction, in an attempt to combat the effects of oxidative stress.     

The relationship between paraquat accumulation and covalent binding in rat lung slices

The accumulation and covalent binding of paraquat in rat lung slices were both linear for 6 hr in room air incubations. Binding continued to increase in slices transferred to paraquat-free buffer after 3 hr of incubation in paraquat although accumulated paraquat decreased. Binding in 100% O2 was decreased slightly. Active accumulation in 100% N2 did not occur, but binding proceeded at one-third the rate observed in room air. Ascorbate decreased accumulation in room air, although binding was unaffected. Reductants had no effect on binding in 100% nitrogen. Paraquat binding in slices of various organs was in the order of lung greater than liver greater than heart greater than kidney cortex. Mitochondrial proteins were found to have the highest concentration of bound paraquat in lung slices followed in order by microsomal protein greater than nuclear protein = cytosolic protein. The binding of paraquat is postulated to involve a reduced species, presumably the monovalent radical.     

In vitro analysis of the renal handling of paraquat.

Accumulation of paraquat by mouse renal cortical slices was related to the concentration of paraquat in the medium and the duration of incubation. Paraquat accumulation was depressed by incubation of slices under nitrogen or by addition of metabolic inhibitors. Accumulation of a second organic base, N-methylnicotinamide (NMN), was depressed by a concentration of paraquat which failed to influence accumulation of the organic acid, p-aminohippurate (PAH). The uptake component of NMN accumulation was inhibited by paraquat. The data suggest that paraquat is accumulated by an energy-requiring process and that this accumulation occurs via the organic base transport system. In addition, an apparently toxic effect of paraquat on cortical slice function was observed when the incubation temperature was raised from 25 to 37°C. At this temperature, 10^?3m paraquat depressed not only NMN accumulation but PAH accumulation and slice oxygen consumption as well. Thus, paraquat can be toxic to the function of kidney slices and this effect appears to be temperature-dependent.     

Effects of paraquat on development of preimplantation embryos in vivo and in vitro

Paraquat can cause oxidative stress through redox cycling, and preimplantation embryos are sensitive to oxidative stress in vitro. In this study, the effects of paraquat on preimplantation embryo development were examined. Exposure of preimplantation embryos (collected on the day after ovulation) to paraquat in vitro for 24 h at concentrations as low as 8 microM caused a significant decrease in the percentage of 8-cell embryos and an increase in the percentage of compacted morulae, but the content of reduced glutathione (GSH) in embryos was not changed. Altered embryo development was most likely due to premature compaction because a 42% decrease in cell number per compacted morulae was observed in embryos exposed to paraquat at 1 mM. Exposure of preimplantation embryos to paraquat in vitro for 4 days at 200 microM or higher eliminated development beyond the blastocyst stage. Exposure of bred female mice to paraquat at 30 mg/kg on day 2 after ovulation led to a small but significant decrease in the percentage of 8-cell embryos on day 3 without a detectable increase in the percentage of compacted morulae. No detectable change in preimplantation embryo development was found following paraquat exposure on the day of ovulation (day 0), although a significant decrease in embryo GSH was found on day 1. These data indicate that paraquat can adversely impact the development of preimplantation embryos in vitro and in vivo without consistent modulation of GSH level.     

Relevance of NADPH depletion and mixed disulphide formation in rat lung to the mechanism of cell damage following paraquat administration

We have examined the possibility that the mechanism of paraquat toxicity in the lung involves both the formation of mixed disulphides (the amount of NPSH or GSH involved in protein disulphide formation) and the prolonged oxidation of NADPH, leading to NADPH depletion. We have compared the oxidation-reduction status of the lung, 2, 8 and 24 hr after dosing rats subcutaneously with 20 mg paraquat/kg (a dose which causes extensive lung damage 24 hr after dosing) or 20 mg diquat/kg (a chemically related bipyridyl which only causes very minimal lung damage 24 hr after dosing). Lung NADP⁺ levels were not affected 2, 8 or 24 hr after dosing with either bipyridyl. However, although NADPH levels were unchanged 2 hr after dosing with paraquat, and 2, 8 and 24 hr after dosing with diquat, there was a significant decrease in NADPH levels by 8 and 24 hr after dosing with paraquat. The changes in NADPH levels were coincident with the lung damage (characterized previously) caused by these treatments. In contrast with these effects on NADPH levels, there was an increase in NPSH and GSH levels in the lung by 8 and 24 hr after dosing with paraquat or diquat. Thus, there was no simple relationship between lung NADPH levels and lung sulphydryl levels.Lung mixed disulphide levels (the amounts of NPSH or GSH involved in disulphide formation) were increased 2, 8 and 24 hr after dosing with paraquat or diquat, although oxidized glutathione levels remained normal. Thus, an early and persistent biochemical effect of paraquat and diquat in the lung involves an increase in mixed disulphide levels, which is probably a consequence of the lungs' response to an increase in the oxidation of NADPH and GSH. As suggested previously, the increase in mixed disulphide levels appears to be a mechanism for regulating the normal redox state of the lung. However, despite this regulatory mechanism, NADPH depletion occurs 8 and 24 hr after dosing with paraquat, but not diquat, coincident with the development of lung damage.In conclusion, we suggest that mixed disulphide formation is not only a regulatory mechanism, but in some circumstances also may cause changes in essential biosynthetic and regulatory functions of the lung. This may ultimately lead to the drop in NADPH levels, which we propose is a critical biochemical event in the development of alveolar epithelial cell damage following the administration of paraquat.