- Email: [email protected]
Mechanism of cytotoxicity of paraquat
Exp Toxic Patho11993 ; 45: 375-380 Gustav Fischer Verlag lena
Department of Environmental Medicine, Shimane Medical University, lzumo, Japan
Mechanism of cytotoxicity of paraquat II. Organ specificity of paraquat-stimulated lipid peroxidation in the inner membrane of mitochondria K. YAMADA and T. FUKUSHIMA With 3 figures and 3 tables Received: August 10, 1992; Revised: November 15, 1992; Accepted: November 27, 1992
Address for correspondence: K. YAMADA, Department of Environmental Medicine, Shimane Medical University, Enya-cho 891, Izumo, 693. Japan. Fax No. 0853-25-0792. Phone No. 0853-23-2111 Ext. 2444.
Key words: Paraquat cytotoxicity; Cytotoxicity, paraquat; Mitochondria, inner membrane; Membrane mitochondria; Lipid peroxidation; Peroxidation, lipid; Complex I ; Free radicals, paraquat; Superoxide anion; Organ specificity, paraquat.
Summary The production of superoxide anion (02-) statistically increased, stimulated by paraquat (1,1' dimethyl-4,4' -bipyridylium dichloride) in lung, liver, kidney and heart submitochondrial particles (SMP) isolated from rats given paraquat intravenously. Paraquat also stimulated 02- production in bovine liver NADHubiquinone oxidoreductase (complex I). The reaction mixture used in these assays turned its color into blue proving the occurrence of paraquat free radicals. The pH optimum for NADH dependent 02- production with paraquat was 9.5. ° 2- production was stimulated by paraquat even in the presence of rotenone, one of the mitochondrial respiratory chain inhibitors. The lipid peroxidation increased in lung SMP but not in heart SMP of paraquat-treated rats. These results may suggest that paraquat was reduced by complex I, but there was difference in the lipid peroxidation by the paraquat radical between rat lung and heart.
Introduction GAGE (1968) reported that paraquat (PQ), 1,1' -dimethyl4,4' bipyridylium dichloride, caused a slight increase in the oxygen uptake of rat liver mitochondria, while a similar increase was found by NAGATA et al. (1987) in beef heart and yeast mitochondria, and by THAKAR and HASSEN (1988) in rat striatum, cortex, and liver mitochondria. Some studies reported that PQ affected not only micro somes but also mitochondria. KOPACZYK-LoCKE (1977) described that PQ caused mitochondrial swelling in the presence of oxidizable substrate and SYKES et al. (1977) demonstrated that in rats 8 hours after intravenous injection of PQ mitochondria of alveolar type II cells were swollen and became less election dense with dilated cristae. HIRAI et al. (1985) reported that a single intravenous injection of 40 mg/kg wt PQ caused selecti ve mitochondrial swelling 6-12 hours later and loss of intramitochondrial granules 24 hours later in alveolar
type II cells, and he suggested that PQ might affect primarily alveolar type II cells, initially injuring mitochondria. The toxicity of PQ involved multiple organs (BULLIVANT 1966). Similar results were obtained in laboratory animals (CLARK et al. 1966). The clinically most serious injury appears in the lung, where an exposure to PQ of over a certain density causes an irreversible fibrosis , and there is no effective treatment for it (BULLIVANT 1966, MALONE et al. 1971). Reports on injuries in heart or brain caused by PQ are very few (PARKINSON 1980). The present study researches the organ specificity in the clinical appearance of PQ toxication.
Material and methods
Chemicals Adrenaline, antimycin A, catalase, paraquat (l,1' -dimethyl4,4' bipyridylium dichloride), NADH and superoxide dismutase (SOD) were purchased from Shigma Chern. Co. (Saint Louis, MO, USA), rotenone and sodium cyanide from Nakarai Inc. (Kyoto, Japan), and 0.9 % NaCl, sulfate, thiobarbiturate acid and n-butanol from Wako Pure Chern. Industries Ltd. (Osaka, Japan).
Preparation of rat submitochondrial particles (SMP) and bovine complex I Experiments were performed on 8 weeks old Sprague-Dawley male rats (200-300 g body WL). The animals were decapitated and their lungs, livers, kidneys and hearts were quickly exised and placed in ice-cold physiological saline. Rat mitochondria were isolated by the procedure of HATEFI and RlESKE (1967). SMP were prepared by the method of TURRENS and BOVERIS ExpToxic Pathol45 (1993) 5-6
375
HADHPO~ 485nml-
l
sr____
Hver SMP
SOD lung SMP
l _--
HADHpQ~
USnm
I_ I
I I ..l
84
T 575nm
10
S75nm
I
HADH,~ ' --
l_
,---
SDO
SOD
k l dn.~SMP
C85nm
T
1.
84
£\1.=0 •
h""SIOP. / NAOHPO~
'850m
~_ l
152
T
T 575 nm
575nm
1.
17
SOD
comp. l . / ,' HADH PO
485 nm
l_ 1
1
574 3
T 575nm
236' 1 min ~
Fig. 1. Production of 02- by SMP or copmlex I with treated PQ. For measurements adrenochrome formation (L1A48S-L1As7S) was utilized after two min. incubation at 30°C. Assay mixture included I mM adrenaline, 211M catalase, 230 mM mannitol, 70 mM sucrose, 30 mM Tris CI buffer, pH 7.4. The reaction started by the addition of 0.5 mM NADH, 0.24 mg rat SMPs, 10!-'g bovine complex I, 0.5 mM PQ and 0.3 11M SOD were supplemented as indicated. Values given on the traces indicate the initial velocities expressed as L1485/min. or L1575/min. Table 1. Effects of PQ on production of 02- in SMP and complex I. Production of 02- (nmol/min per mg of protein) Bovine complex I
RatSMP
Control Paraquat
Lung
Liver
Kidney
Heart
Liver
1.0 ± 0.2 12.7 ± 0.8')
1,1±0.2 13.5±1.l"i
1.2 ± 0.3 12.0 ± 0.8")
1.2 ± 0.2 26.0 ± 3.0')
35.5 ± 6.4 378.4 ± 34.5')
Measurements of NADH dependent 02- production was measured by utilizing adrenochrome formation with the addition of 0.5 mM NADH and 0.5 mM PQ after two min. incubation at 30 C. The assay mixture included 0.24 mg rat SMP or 10 !-'g, bovine complex I, 211M catalase, 230 mM mannitol, 70 mM sucrose, 30 mM Tris CI buffer, pH 7.4. Specific activity of rat SMP (complex I): Lung 1.43 U/mg; Liver 1.33 U/mg; Kidney 0.92 U/mg; Heart 2,95 U/mg. Specific activity of bovine liver complex I: 26.4 U/mg. Figures are means ± SD from five measurements of rat SMP and bovine liver complex I. ,) P < 0.001 with respect to control.
(1980) as follows: Mitochondria were suspended in the mixture of 250 mM sucrose, I mM EDTA, 5 mM Tris-HCI buffer, pH 7.4 was sonicated twice for 20 sec with 2 min interval in a Branson sonifier cell disruptor 200. SMP were obtained by cen376
ExpToxic Pathol45 (1993) 5-6
trifugation at 27000 g x 15 min and successively at 77000 g x 60 min at4 C. Bovine liver was purchased from the Shimane Meat Supply Center (Ohda, Japan). Its complex I was isolated by the proce-
dure ofHATEFI and RIESKE (1967). Enzyme activities were measured by the reduction of ferricyanide with complex J. All the operations were performed at 0-4 C. Protein was detennined by the biuret method (GORNALL et at. 1949).
250
Detennination of 02- production in SMP
200
02- production was determined by the SOD-sensitive adrenochrome formation (MISRA and F'RIDovICH 1972) measuring the absorption change at 485 nm minus 575 nm with a Hitachi U-3210 spectrophotometer after preincubation for 2 min at 30 C. The absorption coefficient used was 2.96 litre x mmoY' xcm-I (GREEN et al. 1956). The reaction mixture consisted of 0.5 mM NADH, 0.5 mM PQ, rat lung, liver, kidney, heart SMP (0.24 mg) and bovine liver complex I (0.01 mg) were suspended in 230 mM mannitol, 70 mM sucrose and 30 mM Tris-C 1 buffer, pH 7.4. The absorbance change at 485 nm minus 575 nm was recorded by adding 0.5 mM NADH for 6 min. All the assays mentioned above were carried out at 30 C.
Detennination of lipid peroxidation Lipid peroxidation was measured by the method of malondialdehyde (OHKAWA et al. 1979) with a Hitachi 850 fluorescence spectrophotometer. The reaction mixture containing 0.12 mg SMP, 0.4 mM NADH, 0.4 mM PQ in 1 ml of 100 mM Tris-CI buffer, pH 7.4 was incubated at room temperature for 15 min before measurement of lipid peroxidation.
Results 02- production stimulated by PQ in SMP or complex I Adrenochrome formation from adrenaline was measured in lung, liver, kidney and heart SMP with NADH as electron donor. The formation was stimulated by PQ and was completely inhibited by the addition of exogenous superoxide dismutase (fig. 1), indicating a contribution of 02- to the reaction. The rates of NADH dependent 02- production in lung, liver and kidney SMP of rats given PQ were about 10-12 fold higher than the control and that in heart SMP was 21.7 fold higher than the control (table 1). Color of the reaction mixture including PQ turned macroscopically into blue. Bovine liver complex I was purified to indentify the position in the mitochondrial respiratory chain, where the influence of PQ appeared. Adrenochrome formation by bovine liver complex I was increased by PQ and was inhibited by superoxide dismutase (fig. 1). The rate of NADH dependent 02- production in bovine liver complex I exposed to PQ was about 10.7 fold higher than the control (table 1). The reaction mixture of bovine liver complex T and PQ also turned its color into blue.
pH -dependence Effects of pH on NADH dependent 02-production in rat lung SMP are shown in fig. 2. The pH optimum of NADH dependent 02- reaction was 9.5 . The pH optimum for 02-
• •
c
.2 c
III
0
•
-~ c
E 100
c ... "
.2 ea
•
":>
~
0
Ii: 50
o
•
• pH
9
10
Fig. 2. Effect of pH on 0rProduction by SMP For measurements adrenochrome formation was utilized after two min. incubation at 30 DC with treated NADH and PQ: Assay mixture included 0.24 mg SMP, 0.5 mM NADH, 0.5 mM PQ, 21JM catalase in 1 m1 of 30 mM Tris CI buffer. Ce) with PQ; (A) without PQ.
production without PQ was also 9.5. The same results were observed in the NADH dependent Of production in rat liver, kidney and heart SMP.
Effects of respiratory chain inhibitors The site where PQ receives an electron from the respiratory chain was further investigated by measuring the effects of the respiratory chain inhibitors on 02- production stimulated by PQ. The 02-production was influenced by various respiratory chain inhibitors (table 2). Rotenone inhibited slightly the 02- production except in kidney SMP. Anitmycin A increased the 02- production stimulated by PQ in SMP of all the assayed organs.
Effects of PQ on lipid peroxidation Malondialdehyde formation was measured. As shown in table 3, lipid peroxidation was increased by PQ in lung SMP, but not in heart SMP. The reaction mixture turned its color into blue by PQ in both lung and heart SMP. Exp Toxic Pathol45 (1993) 5-6
377
Table 2. Effects of respiratiory-chain inhibitors on 02 - production stimulated by PQ. Production of 02- (nmol/min per mg of protein)
Control Rotenone Antimycin A
Lung
Liver
Kidney
Heart
12.7 ± 0.8 10.1 ± 0.9 bl 30.3 ± 2.2"
13,5 ± 1.1 12.8 ± 1.0 24.7 ± 2.3"
12.0 ± 0.8 13.5 ± OSI 20.6± 1.2')
26.0 ± 3.0 20.8 ± 3.1') 37.7 ± 4.3 b)
°minproduction stimulated by PQ was measured by utilizing adrenochrome formation after two incubation at C with added NADH and respiratouy chain inhibitor. The reaction 2-
30 medium consisted of 0.24 mg SMP or 10 j..Ig complex J, 0.5 mM NADH, 0.5 mM PQ, 2 11M catalase, 230 mM mannitol, 70 mM sucrose, 30 mM Tris Cl buffer, pH 7.4, respriatory chain inhibitors: 1.5 11M rotenone, 2 j..Ig/ml antimycin A, I mM sodium cyanide. Complex I specific activity of SMP: Lung 1.43 U/mg; Liver 1.33 U/mg; Kidney 0.92 U/mg; Heart 2.95 U/mg. Figures are means ± SD from five measurements. ,) p < 0.001; b) P < 0.01; 'I P < 0.05 with respect control.
Table 3. Effects of PQ on lipid peroxidation. Lipid production (x 10-2 nmol per mg of protein)
SMP SMP+ NADH SMP+PQ SMP + NADH + PQ
Lung
Heart
73.8 ± 6.4 79.3 ± 5.1 122.3 ± 13.8" 208.0 ± 12.3 b '
31.6 ± 3.1 34.1 ± 5.7 39.4 ± 4.1 38.2 ± 3.9
Lipid peroxidation was estimated by measuring the accumulation of malondialdehyde after 15 min. with treated NADH and/or PQ. The assay mixture included 121 j..Ig SMP. 0.4 mM NADH, 0.4 mM PQ. 100 mM Tris Cl buffer pH 7.4. Complex J specific activity of SMP: Lung 4.64 U/mg; Heart 5.49 U/mg. Figures are means ± SD from five measurements. ,) p < 0.01 with respect to SMP only; b) p < 0.01 with respect to treated NADH.
Discussion NADH dependent 02- production in rat SMP increased by the effect of treated PQ and the existence of PQ radicals was macroscopically identified by the color of the medium turning macroscopically into blue. PQ+e02 + PQ·
-t -t
PQ·
°2 +
(1)
PQ
(2)
This phenomenon explains that in the electron transport system of mitochondria, PQ radicals were derived from PQ reducing one electron (reaction (1», and 02 was generated from PQ radicals (reaction (2». NADH dependent 02- production in bovine complex I also increased with the treatment of PQ and the existence of PQ radicals was identified by the color of the medium. It is possible that reactions (1) and (2) may occur also on the site of complex I in the electron transport system of mitochondria. As generally accepted, adrenochrome formation which indicates autoxidation in the electron transport pathway of mitochondria, increases with the increase of the pH range 378
Exp Toxic Pathol45 (1993) 5-6
between 7.8 and 10.2 (MISRA and FRIDOVICH 1972). TAKESHIGE and MINAKAMI (1979) also reported that the optimal pH for the NADH dependent formation of superoxide anions in bovine heart SMP was 9.0. In the present study the optimal pH for the adrenochrome formation with NADH in rat lung SMP was 9.5 (fig. 2) and it corresponded with the pH peak obtained when PQ was administered. In the SMP of rat liver, kidney and heart, the optimal pH was similar to that in lung SMP. Based on these results we suggest two hypotheses on the 02- production in the electron transport pathway of mitochondria: first, PQ plays the role of an electron recipient, receiving one electron directly from the electron transport pathway and giving it to singlet oxygen, and second, PQ enhances the 02- production by autoxidation in the electron transport pathway of mitochondria. The site where PQ derives electrons from the transport pathway may be located before the rotenone-sensitive part, considering that in bovine liver complex I, PQ radicals were identified and 02- production was increased by PQ, and that in rat SMP 2- production was increased by PQ, even when rotenone was added (fig. 3). The effect of PQ mentioned in
°
complex III
complex I
c-1 a,a ,Jcomplex IV
3
Antimycin A
I
H20
cyanide
lipid
Fig. 3. Proposed scheme for effect of PQ on lipid peroxidation in rat lung SMP.
-~-
--1 lipid
peroxidation
the second hypothesis was assumed from the fact that the pH peak for adrenochrome formation did not vary in our experiment, whether PQ was treated or not. However, it is unclear if this hypothesis can be applied to complex I. Involvement of complex III in the PQ-stimulated 02- production in the electron transport pathway of mitochondria has not been reported so far. The sites of autoxidation in the electron transport pathway of mitochondria are reported to be complex I and complex III (TURRENS and BOVERIS 1980, TURRENS et al. 1985). In complex I, the generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria decreased with rotenone (TURRENS and BOVERIS 1980) and in complex III, it increased with antimycin A (TURRENS et al. 1985). In our experiment we could not determine whether PQ was related to the electron transport from these enzyme complexes to 02- or PQ only enhanced the electron transport. When antimycin A was added, PQ stimulated the 02- production in the electron transport pathway of mitochondria increased in all the experimented organs. This result indicates that PQ affects the 02 production, influenced in some way by the autoxidation of complex III.
her in air or in an atmosphere of 90 %-95 % oxygen was effective in extending the lifetime of these rats. RAFFIN et al. (1980), however, indicated that SOD had no significant effect on PQ-exposed cells and that the 02 tension increased with the enhancement of the PQ toxicity. Two important points of the controversy are as follows: 02 itself caused an increase in the rate of biosynthesis of intrinsic SOD in ESCHERICHIA COLI (HASSAN and FRIDOVICH 1977); after 7 days of exposure to 90 % 02 (1 atm). SOD activities in rat mitochondria cytosolic fractions increased (KIMBALL et al. 1976). It should be further discussed whether the produced radical species induce radical scavengers such as intrinsic SOD or not. In our experiment PQ-stimulated, the 02- production in rat lung, liver, kidney, heart SMP, and bovine liver complex I were inhibited completely with the treated exogeneous SOD, though the induction of intrinsic SOD activities was not investigated. The difference in the production of lipid peroxidation between rat lung and heart may be associated with the difference in the mechanism of radical scavengers such as intrinsic SOD induced by PQ.
There are many studies reporting lung injury caused by PQ (CLARK et al. 1966, SYKES et al. 1977, HIRAI et al. 1985) but only a few dealt with the injury in heart (PARKINSON 1980). Table 3 shows that the lipid peroxidation was increased by PQ in rat lung SMP but not in rat heart SMP. This result suggests that the increase of lipid peroxidation indicates an organ specificity corresponding with the organ specificity in clinical phenomena. It is generally accepted that the production of lipid peroxidation is enhanced in the presence of Fe-ADP (TYLER 1975). SATA et al. (1983) reported that in the presence of Fe-ADP PQ enhanced the NADH dependent lipid peroxidation of bovine heart SMP. In the present research Fe-ADP was not treated and this may be the cause of the difference between our result and SATA'S (1983). In our experiment, the reaction mixture turned its color into blue in both the lung and heart SMP of PQ-treated rats showing the presence of PQ radicals. Thus, the organ specificity distinguishing between lung and heart may be associated iwth some difference in the pathway from PQ radicals to lipid peroxidation. Effects of exogenous SOD for PQ toxicity were reported but aroused much controversy. AUTOR (1974) reported that the intravenously administered SOD in rats which were given PQ and kept eit-
References AUTOR AP: Reduction of paraquat by superoxide dismutase. Life Science 1974; 14: 1309-1319. BULLIV ANT CM: Accidental poisoning by paraquat: Report of two cases in man. Brit Med J 1966; 1: 1272-1273. CLARK DJ, McELLIGOTT TF, HURST EW: The toxicity of paraquat. Brit J Ind Med 1966; 23: 126-132. GAGE JC: The action of paraquat and diquat on the respiration ofliver cell fractions. Biochem J 1968; 109: 757-761. GORNALL AG, BARDAWILL CS, DAVID MN: Determination of serum proteins by means of the biuret reaction. J BioI Chern 1949; 177:751-766. GREEN S, MAZUR A, SHORR E: Mechanism of the catalytic oxidation of adrenaline by ferritin. J BioI Chern 1956; 220: 237-255. HASSAN HM, FRIDOVICH I: Regulation of synthesis of superoxide dismutase in Escherichia coli. Induction by methylviologen. J BioI Chern 1977; 252: 7667-7672. HATEFl Y, RIESKE JS: Preparation and properties of DPNHcoenzyme Q reductase (complex I of the respiratory chain). In. ESTABROOK RW, PULLMAN ME (eds.) Methods in Enzymology, vol. 10, Academic Press, New York 1967; pp. 235-239. HIRAI K, WITS CHI H, COTE MG: Mitochondrial injury of pulExp Toxic Pathol45 (1993) 5-6
379
monary alveolar epithelial cells in acute paraquat intoxication. Exp Mol Patho11985; 43: 242-252. KIMBALL RE, REDDY K, Peirece TH, et al.: Oxygen toxicity: Augmentation of autoxidant defense mechanisms in rat lung. Am I Physiol1976; 230: 1425-1431. KOPACZYK-LOCKE: In vitro and in vivo effects of paraquat on rat liver mitochondria. In: AUTOR AP (ed) Biochemical mechanisms of paraquat toxicity. Academic Press, New York 1977; pp. 93-99. MALONE ID, CARMODY M, KEOGH B: Paraquat poisoning; a review of nineteen cases. I Ir Med Assoc 1971; 64: 59-68. MISRA HP, FRIDOVICH I: The role of supreoxide anion in the autoxidation of epinephrine and a simple assay for superoxide dismutase. I BioI Chern 1972; 247: 3170-3175. NAGATA S, GUNTHER H, BADER I, et al.: Mitochondria catalyze the reduction of NAD by reduced methylviologen. FEBS Lett 1987; 210: 66-70. OHKAWA H, OHNISHI N, YAGI K: Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal Biochem 1979; 95: 351-358. PARKINSON C: The changing pattern of paraquat poisoning in man. Histopathology 1980; 4: 171-183. RAFFIN TA, SIMON LM, DOUGLAS WHI, et al.: The effects of variable O 2 tension and of exogenous superoxide dismutase on Type II pneumocyte exposed to paraquat. Lab Invest 1980:42:295-208.
380
Exp Toxic Pathol45 (1993) 5-6
SATA T, TAKESHIGE K, TAKAYANAGI R, et al.: Lipid peroxidation by bovine heart submitochondrial particles stimulated by 1,1' -dimethyl-4,4' -bipyridylium dichloride (paraquat). Biochem Pharmacol1983; 32: 13-19. SYKES BI, PURCHASE IFH, SMITH LL: Pulmonary ultrastructure after oral and intravenous dosage of paraquat to rats. I Pathol 1977;121:233-241. TAKESHIGE K, MINAKAMI S: NADH-and NADPH-dependent formation of superoxide anions by bovine heart submitochondrial particles and HADH-ubiquinone reductase preparation. Biochem I 1979; 180: 120-135. TYLER DD: Role of superoxide radicals in the lipid peroxidation of intracellular membranes. FEBS Lett 1975; 51: 180-183. THAKAR I, HASSAN MN: Effects of I-methyl-4-phenyl-l, 2, 3, 6-tetrahydropyridine (MPTP), cyperquat (MPP'), and paraquat on isolated mitochondria from rat striatum, cortex and liver. Life Science 1988; 43: 143-149. TURRENS IF, BOVERIS A: Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochem I 1980; 191: 421-427. TURRENS IF, ALEXANDER A, LEHNINGER A: Ubiserniquinone is the electron donor for superoxide formation by complex III of heart mitochondria. Arch Biochem Biophys 1985; 237: 408-414.
Department of Environmental Medicine, Shimane Medical University, lzumo, Japan
Mechanism of cytotoxicity of paraquat II. Organ specificity of paraquat-stimulated lipid peroxidation in the inner membrane of mitochondria K. YAMADA and T. FUKUSHIMA With 3 figures and 3 tables Received: August 10, 1992; Revised: November 15, 1992; Accepted: November 27, 1992
Address for correspondence: K. YAMADA, Department of Environmental Medicine, Shimane Medical University, Enya-cho 891, Izumo, 693. Japan. Fax No. 0853-25-0792. Phone No. 0853-23-2111 Ext. 2444.
Key words: Paraquat cytotoxicity; Cytotoxicity, paraquat; Mitochondria, inner membrane; Membrane mitochondria; Lipid peroxidation; Peroxidation, lipid; Complex I ; Free radicals, paraquat; Superoxide anion; Organ specificity, paraquat.
Summary The production of superoxide anion (02-) statistically increased, stimulated by paraquat (1,1' dimethyl-4,4' -bipyridylium dichloride) in lung, liver, kidney and heart submitochondrial particles (SMP) isolated from rats given paraquat intravenously. Paraquat also stimulated 02- production in bovine liver NADHubiquinone oxidoreductase (complex I). The reaction mixture used in these assays turned its color into blue proving the occurrence of paraquat free radicals. The pH optimum for NADH dependent 02- production with paraquat was 9.5. ° 2- production was stimulated by paraquat even in the presence of rotenone, one of the mitochondrial respiratory chain inhibitors. The lipid peroxidation increased in lung SMP but not in heart SMP of paraquat-treated rats. These results may suggest that paraquat was reduced by complex I, but there was difference in the lipid peroxidation by the paraquat radical between rat lung and heart.
Introduction GAGE (1968) reported that paraquat (PQ), 1,1' -dimethyl4,4' bipyridylium dichloride, caused a slight increase in the oxygen uptake of rat liver mitochondria, while a similar increase was found by NAGATA et al. (1987) in beef heart and yeast mitochondria, and by THAKAR and HASSEN (1988) in rat striatum, cortex, and liver mitochondria. Some studies reported that PQ affected not only micro somes but also mitochondria. KOPACZYK-LoCKE (1977) described that PQ caused mitochondrial swelling in the presence of oxidizable substrate and SYKES et al. (1977) demonstrated that in rats 8 hours after intravenous injection of PQ mitochondria of alveolar type II cells were swollen and became less election dense with dilated cristae. HIRAI et al. (1985) reported that a single intravenous injection of 40 mg/kg wt PQ caused selecti ve mitochondrial swelling 6-12 hours later and loss of intramitochondrial granules 24 hours later in alveolar
type II cells, and he suggested that PQ might affect primarily alveolar type II cells, initially injuring mitochondria. The toxicity of PQ involved multiple organs (BULLIVANT 1966). Similar results were obtained in laboratory animals (CLARK et al. 1966). The clinically most serious injury appears in the lung, where an exposure to PQ of over a certain density causes an irreversible fibrosis , and there is no effective treatment for it (BULLIVANT 1966, MALONE et al. 1971). Reports on injuries in heart or brain caused by PQ are very few (PARKINSON 1980). The present study researches the organ specificity in the clinical appearance of PQ toxication.
Material and methods
Chemicals Adrenaline, antimycin A, catalase, paraquat (l,1' -dimethyl4,4' bipyridylium dichloride), NADH and superoxide dismutase (SOD) were purchased from Shigma Chern. Co. (Saint Louis, MO, USA), rotenone and sodium cyanide from Nakarai Inc. (Kyoto, Japan), and 0.9 % NaCl, sulfate, thiobarbiturate acid and n-butanol from Wako Pure Chern. Industries Ltd. (Osaka, Japan).
Preparation of rat submitochondrial particles (SMP) and bovine complex I Experiments were performed on 8 weeks old Sprague-Dawley male rats (200-300 g body WL). The animals were decapitated and their lungs, livers, kidneys and hearts were quickly exised and placed in ice-cold physiological saline. Rat mitochondria were isolated by the procedure of HATEFI and RlESKE (1967). SMP were prepared by the method of TURRENS and BOVERIS ExpToxic Pathol45 (1993) 5-6
375
HADHPO~ 485nml-
l
sr____
Hver SMP
SOD lung SMP
l _--
HADHpQ~
USnm
I_ I
I I ..l
84
T 575nm
10
S75nm
I
HADH,~ ' --
l_
,---
SDO
SOD
k l dn.~SMP
C85nm
T
1.
84
£\1.=0 •
h""SIOP. / NAOHPO~
'850m
~_ l
152
T
T 575 nm
575nm
1.
17
SOD
comp. l . / ,' HADH PO
485 nm
l_ 1
1
574 3
T 575nm
236' 1 min ~
Fig. 1. Production of 02- by SMP or copmlex I with treated PQ. For measurements adrenochrome formation (L1A48S-L1As7S) was utilized after two min. incubation at 30°C. Assay mixture included I mM adrenaline, 211M catalase, 230 mM mannitol, 70 mM sucrose, 30 mM Tris CI buffer, pH 7.4. The reaction started by the addition of 0.5 mM NADH, 0.24 mg rat SMPs, 10!-'g bovine complex I, 0.5 mM PQ and 0.3 11M SOD were supplemented as indicated. Values given on the traces indicate the initial velocities expressed as L1485/min. or L1575/min. Table 1. Effects of PQ on production of 02- in SMP and complex I. Production of 02- (nmol/min per mg of protein) Bovine complex I
RatSMP
Control Paraquat
Lung
Liver
Kidney
Heart
Liver
1.0 ± 0.2 12.7 ± 0.8')
1,1±0.2 13.5±1.l"i
1.2 ± 0.3 12.0 ± 0.8")
1.2 ± 0.2 26.0 ± 3.0')
35.5 ± 6.4 378.4 ± 34.5')
Measurements of NADH dependent 02- production was measured by utilizing adrenochrome formation with the addition of 0.5 mM NADH and 0.5 mM PQ after two min. incubation at 30 C. The assay mixture included 0.24 mg rat SMP or 10 !-'g, bovine complex I, 211M catalase, 230 mM mannitol, 70 mM sucrose, 30 mM Tris CI buffer, pH 7.4. Specific activity of rat SMP (complex I): Lung 1.43 U/mg; Liver 1.33 U/mg; Kidney 0.92 U/mg; Heart 2,95 U/mg. Specific activity of bovine liver complex I: 26.4 U/mg. Figures are means ± SD from five measurements of rat SMP and bovine liver complex I. ,) P < 0.001 with respect to control.
(1980) as follows: Mitochondria were suspended in the mixture of 250 mM sucrose, I mM EDTA, 5 mM Tris-HCI buffer, pH 7.4 was sonicated twice for 20 sec with 2 min interval in a Branson sonifier cell disruptor 200. SMP were obtained by cen376
ExpToxic Pathol45 (1993) 5-6
trifugation at 27000 g x 15 min and successively at 77000 g x 60 min at4 C. Bovine liver was purchased from the Shimane Meat Supply Center (Ohda, Japan). Its complex I was isolated by the proce-
dure ofHATEFI and RIESKE (1967). Enzyme activities were measured by the reduction of ferricyanide with complex J. All the operations were performed at 0-4 C. Protein was detennined by the biuret method (GORNALL et at. 1949).
250
Detennination of 02- production in SMP
200
02- production was determined by the SOD-sensitive adrenochrome formation (MISRA and F'RIDovICH 1972) measuring the absorption change at 485 nm minus 575 nm with a Hitachi U-3210 spectrophotometer after preincubation for 2 min at 30 C. The absorption coefficient used was 2.96 litre x mmoY' xcm-I (GREEN et al. 1956). The reaction mixture consisted of 0.5 mM NADH, 0.5 mM PQ, rat lung, liver, kidney, heart SMP (0.24 mg) and bovine liver complex I (0.01 mg) were suspended in 230 mM mannitol, 70 mM sucrose and 30 mM Tris-C 1 buffer, pH 7.4. The absorbance change at 485 nm minus 575 nm was recorded by adding 0.5 mM NADH for 6 min. All the assays mentioned above were carried out at 30 C.
Detennination of lipid peroxidation Lipid peroxidation was measured by the method of malondialdehyde (OHKAWA et al. 1979) with a Hitachi 850 fluorescence spectrophotometer. The reaction mixture containing 0.12 mg SMP, 0.4 mM NADH, 0.4 mM PQ in 1 ml of 100 mM Tris-CI buffer, pH 7.4 was incubated at room temperature for 15 min before measurement of lipid peroxidation.
Results 02- production stimulated by PQ in SMP or complex I Adrenochrome formation from adrenaline was measured in lung, liver, kidney and heart SMP with NADH as electron donor. The formation was stimulated by PQ and was completely inhibited by the addition of exogenous superoxide dismutase (fig. 1), indicating a contribution of 02- to the reaction. The rates of NADH dependent 02- production in lung, liver and kidney SMP of rats given PQ were about 10-12 fold higher than the control and that in heart SMP was 21.7 fold higher than the control (table 1). Color of the reaction mixture including PQ turned macroscopically into blue. Bovine liver complex I was purified to indentify the position in the mitochondrial respiratory chain, where the influence of PQ appeared. Adrenochrome formation by bovine liver complex I was increased by PQ and was inhibited by superoxide dismutase (fig. 1). The rate of NADH dependent 02- production in bovine liver complex I exposed to PQ was about 10.7 fold higher than the control (table 1). The reaction mixture of bovine liver complex T and PQ also turned its color into blue.
pH -dependence Effects of pH on NADH dependent 02-production in rat lung SMP are shown in fig. 2. The pH optimum of NADH dependent 02- reaction was 9.5 . The pH optimum for 02-
• •
c
.2 c
III
0
•
-~ c
E 100
c ... "
.2 ea
•
":>
~
0
Ii: 50
o
•
• pH
9
10
Fig. 2. Effect of pH on 0rProduction by SMP For measurements adrenochrome formation was utilized after two min. incubation at 30 DC with treated NADH and PQ: Assay mixture included 0.24 mg SMP, 0.5 mM NADH, 0.5 mM PQ, 21JM catalase in 1 m1 of 30 mM Tris CI buffer. Ce) with PQ; (A) without PQ.
production without PQ was also 9.5. The same results were observed in the NADH dependent Of production in rat liver, kidney and heart SMP.
Effects of respiratory chain inhibitors The site where PQ receives an electron from the respiratory chain was further investigated by measuring the effects of the respiratory chain inhibitors on 02- production stimulated by PQ. The 02-production was influenced by various respiratory chain inhibitors (table 2). Rotenone inhibited slightly the 02- production except in kidney SMP. Anitmycin A increased the 02- production stimulated by PQ in SMP of all the assayed organs.
Effects of PQ on lipid peroxidation Malondialdehyde formation was measured. As shown in table 3, lipid peroxidation was increased by PQ in lung SMP, but not in heart SMP. The reaction mixture turned its color into blue by PQ in both lung and heart SMP. Exp Toxic Pathol45 (1993) 5-6
377
Table 2. Effects of respiratiory-chain inhibitors on 02 - production stimulated by PQ. Production of 02- (nmol/min per mg of protein)
Control Rotenone Antimycin A
Lung
Liver
Kidney
Heart
12.7 ± 0.8 10.1 ± 0.9 bl 30.3 ± 2.2"
13,5 ± 1.1 12.8 ± 1.0 24.7 ± 2.3"
12.0 ± 0.8 13.5 ± OSI 20.6± 1.2')
26.0 ± 3.0 20.8 ± 3.1') 37.7 ± 4.3 b)
°minproduction stimulated by PQ was measured by utilizing adrenochrome formation after two incubation at C with added NADH and respiratouy chain inhibitor. The reaction 2-
30 medium consisted of 0.24 mg SMP or 10 j..Ig complex J, 0.5 mM NADH, 0.5 mM PQ, 2 11M catalase, 230 mM mannitol, 70 mM sucrose, 30 mM Tris Cl buffer, pH 7.4, respriatory chain inhibitors: 1.5 11M rotenone, 2 j..Ig/ml antimycin A, I mM sodium cyanide. Complex I specific activity of SMP: Lung 1.43 U/mg; Liver 1.33 U/mg; Kidney 0.92 U/mg; Heart 2.95 U/mg. Figures are means ± SD from five measurements. ,) p < 0.001; b) P < 0.01; 'I P < 0.05 with respect control.
Table 3. Effects of PQ on lipid peroxidation. Lipid production (x 10-2 nmol per mg of protein)
SMP SMP+ NADH SMP+PQ SMP + NADH + PQ
Lung
Heart
73.8 ± 6.4 79.3 ± 5.1 122.3 ± 13.8" 208.0 ± 12.3 b '
31.6 ± 3.1 34.1 ± 5.7 39.4 ± 4.1 38.2 ± 3.9
Lipid peroxidation was estimated by measuring the accumulation of malondialdehyde after 15 min. with treated NADH and/or PQ. The assay mixture included 121 j..Ig SMP. 0.4 mM NADH, 0.4 mM PQ. 100 mM Tris Cl buffer pH 7.4. Complex J specific activity of SMP: Lung 4.64 U/mg; Heart 5.49 U/mg. Figures are means ± SD from five measurements. ,) p < 0.01 with respect to SMP only; b) p < 0.01 with respect to treated NADH.
Discussion NADH dependent 02- production in rat SMP increased by the effect of treated PQ and the existence of PQ radicals was macroscopically identified by the color of the medium turning macroscopically into blue. PQ+e02 + PQ·
-t -t
PQ·
°2 +
(1)
PQ
(2)
This phenomenon explains that in the electron transport system of mitochondria, PQ radicals were derived from PQ reducing one electron (reaction (1», and 02 was generated from PQ radicals (reaction (2». NADH dependent 02- production in bovine complex I also increased with the treatment of PQ and the existence of PQ radicals was identified by the color of the medium. It is possible that reactions (1) and (2) may occur also on the site of complex I in the electron transport system of mitochondria. As generally accepted, adrenochrome formation which indicates autoxidation in the electron transport pathway of mitochondria, increases with the increase of the pH range 378
Exp Toxic Pathol45 (1993) 5-6
between 7.8 and 10.2 (MISRA and FRIDOVICH 1972). TAKESHIGE and MINAKAMI (1979) also reported that the optimal pH for the NADH dependent formation of superoxide anions in bovine heart SMP was 9.0. In the present study the optimal pH for the adrenochrome formation with NADH in rat lung SMP was 9.5 (fig. 2) and it corresponded with the pH peak obtained when PQ was administered. In the SMP of rat liver, kidney and heart, the optimal pH was similar to that in lung SMP. Based on these results we suggest two hypotheses on the 02- production in the electron transport pathway of mitochondria: first, PQ plays the role of an electron recipient, receiving one electron directly from the electron transport pathway and giving it to singlet oxygen, and second, PQ enhances the 02- production by autoxidation in the electron transport pathway of mitochondria. The site where PQ derives electrons from the transport pathway may be located before the rotenone-sensitive part, considering that in bovine liver complex I, PQ radicals were identified and 02- production was increased by PQ, and that in rat SMP 2- production was increased by PQ, even when rotenone was added (fig. 3). The effect of PQ mentioned in
°
complex III
complex I
c-1 a,a ,Jcomplex IV
3
Antimycin A
I
H20
cyanide
lipid
Fig. 3. Proposed scheme for effect of PQ on lipid peroxidation in rat lung SMP.
-~-
--1 lipid
peroxidation
the second hypothesis was assumed from the fact that the pH peak for adrenochrome formation did not vary in our experiment, whether PQ was treated or not. However, it is unclear if this hypothesis can be applied to complex I. Involvement of complex III in the PQ-stimulated 02- production in the electron transport pathway of mitochondria has not been reported so far. The sites of autoxidation in the electron transport pathway of mitochondria are reported to be complex I and complex III (TURRENS and BOVERIS 1980, TURRENS et al. 1985). In complex I, the generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria decreased with rotenone (TURRENS and BOVERIS 1980) and in complex III, it increased with antimycin A (TURRENS et al. 1985). In our experiment we could not determine whether PQ was related to the electron transport from these enzyme complexes to 02- or PQ only enhanced the electron transport. When antimycin A was added, PQ stimulated the 02- production in the electron transport pathway of mitochondria increased in all the experimented organs. This result indicates that PQ affects the 02 production, influenced in some way by the autoxidation of complex III.
her in air or in an atmosphere of 90 %-95 % oxygen was effective in extending the lifetime of these rats. RAFFIN et al. (1980), however, indicated that SOD had no significant effect on PQ-exposed cells and that the 02 tension increased with the enhancement of the PQ toxicity. Two important points of the controversy are as follows: 02 itself caused an increase in the rate of biosynthesis of intrinsic SOD in ESCHERICHIA COLI (HASSAN and FRIDOVICH 1977); after 7 days of exposure to 90 % 02 (1 atm). SOD activities in rat mitochondria cytosolic fractions increased (KIMBALL et al. 1976). It should be further discussed whether the produced radical species induce radical scavengers such as intrinsic SOD or not. In our experiment PQ-stimulated, the 02- production in rat lung, liver, kidney, heart SMP, and bovine liver complex I were inhibited completely with the treated exogeneous SOD, though the induction of intrinsic SOD activities was not investigated. The difference in the production of lipid peroxidation between rat lung and heart may be associated with the difference in the mechanism of radical scavengers such as intrinsic SOD induced by PQ.
There are many studies reporting lung injury caused by PQ (CLARK et al. 1966, SYKES et al. 1977, HIRAI et al. 1985) but only a few dealt with the injury in heart (PARKINSON 1980). Table 3 shows that the lipid peroxidation was increased by PQ in rat lung SMP but not in rat heart SMP. This result suggests that the increase of lipid peroxidation indicates an organ specificity corresponding with the organ specificity in clinical phenomena. It is generally accepted that the production of lipid peroxidation is enhanced in the presence of Fe-ADP (TYLER 1975). SATA et al. (1983) reported that in the presence of Fe-ADP PQ enhanced the NADH dependent lipid peroxidation of bovine heart SMP. In the present research Fe-ADP was not treated and this may be the cause of the difference between our result and SATA'S (1983). In our experiment, the reaction mixture turned its color into blue in both the lung and heart SMP of PQ-treated rats showing the presence of PQ radicals. Thus, the organ specificity distinguishing between lung and heart may be associated iwth some difference in the pathway from PQ radicals to lipid peroxidation. Effects of exogenous SOD for PQ toxicity were reported but aroused much controversy. AUTOR (1974) reported that the intravenously administered SOD in rats which were given PQ and kept eit-
References AUTOR AP: Reduction of paraquat by superoxide dismutase. Life Science 1974; 14: 1309-1319. BULLIV ANT CM: Accidental poisoning by paraquat: Report of two cases in man. Brit Med J 1966; 1: 1272-1273. CLARK DJ, McELLIGOTT TF, HURST EW: The toxicity of paraquat. Brit J Ind Med 1966; 23: 126-132. GAGE JC: The action of paraquat and diquat on the respiration ofliver cell fractions. Biochem J 1968; 109: 757-761. GORNALL AG, BARDAWILL CS, DAVID MN: Determination of serum proteins by means of the biuret reaction. J BioI Chern 1949; 177:751-766. GREEN S, MAZUR A, SHORR E: Mechanism of the catalytic oxidation of adrenaline by ferritin. J BioI Chern 1956; 220: 237-255. HASSAN HM, FRIDOVICH I: Regulation of synthesis of superoxide dismutase in Escherichia coli. Induction by methylviologen. J BioI Chern 1977; 252: 7667-7672. HATEFl Y, RIESKE JS: Preparation and properties of DPNHcoenzyme Q reductase (complex I of the respiratory chain). In. ESTABROOK RW, PULLMAN ME (eds.) Methods in Enzymology, vol. 10, Academic Press, New York 1967; pp. 235-239. HIRAI K, WITS CHI H, COTE MG: Mitochondrial injury of pulExp Toxic Pathol45 (1993) 5-6
379
monary alveolar epithelial cells in acute paraquat intoxication. Exp Mol Patho11985; 43: 242-252. KIMBALL RE, REDDY K, Peirece TH, et al.: Oxygen toxicity: Augmentation of autoxidant defense mechanisms in rat lung. Am I Physiol1976; 230: 1425-1431. KOPACZYK-LOCKE: In vitro and in vivo effects of paraquat on rat liver mitochondria. In: AUTOR AP (ed) Biochemical mechanisms of paraquat toxicity. Academic Press, New York 1977; pp. 93-99. MALONE ID, CARMODY M, KEOGH B: Paraquat poisoning; a review of nineteen cases. I Ir Med Assoc 1971; 64: 59-68. MISRA HP, FRIDOVICH I: The role of supreoxide anion in the autoxidation of epinephrine and a simple assay for superoxide dismutase. I BioI Chern 1972; 247: 3170-3175. NAGATA S, GUNTHER H, BADER I, et al.: Mitochondria catalyze the reduction of NAD by reduced methylviologen. FEBS Lett 1987; 210: 66-70. OHKAWA H, OHNISHI N, YAGI K: Assay for lipid peroxides in animal tissues by thiobarbituric acid reaction. Anal Biochem 1979; 95: 351-358. PARKINSON C: The changing pattern of paraquat poisoning in man. Histopathology 1980; 4: 171-183. RAFFIN TA, SIMON LM, DOUGLAS WHI, et al.: The effects of variable O 2 tension and of exogenous superoxide dismutase on Type II pneumocyte exposed to paraquat. Lab Invest 1980:42:295-208.
380
Exp Toxic Pathol45 (1993) 5-6
SATA T, TAKESHIGE K, TAKAYANAGI R, et al.: Lipid peroxidation by bovine heart submitochondrial particles stimulated by 1,1' -dimethyl-4,4' -bipyridylium dichloride (paraquat). Biochem Pharmacol1983; 32: 13-19. SYKES BI, PURCHASE IFH, SMITH LL: Pulmonary ultrastructure after oral and intravenous dosage of paraquat to rats. I Pathol 1977;121:233-241. TAKESHIGE K, MINAKAMI S: NADH-and NADPH-dependent formation of superoxide anions by bovine heart submitochondrial particles and HADH-ubiquinone reductase preparation. Biochem I 1979; 180: 120-135. TYLER DD: Role of superoxide radicals in the lipid peroxidation of intracellular membranes. FEBS Lett 1975; 51: 180-183. THAKAR I, HASSAN MN: Effects of I-methyl-4-phenyl-l, 2, 3, 6-tetrahydropyridine (MPTP), cyperquat (MPP'), and paraquat on isolated mitochondria from rat striatum, cortex and liver. Life Science 1988; 43: 143-149. TURRENS IF, BOVERIS A: Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochem I 1980; 191: 421-427. TURRENS IF, ALEXANDER A, LEHNINGER A: Ubiserniquinone is the electron donor for superoxide formation by complex III of heart mitochondria. Arch Biochem Biophys 1985; 237: 408-414.