Document Emxx4LoJ27y5kxXzV3EgVz0YL

THEJOURNALOF BIOLOGICCAHLEMISTRY Vol 257, No. 20, Issue of October 25. pp. 12419-12425, 1982 Printed in U.S.A. The Metabolism of Menadione (2-Methyl-1,4-naphthoquinoneb)y Isolated Hepatocytes A STUDY OF THE IMPLICATIONS OF OXIDATIVE STRESS IN INTACT CELLS* (Received for publication, March 11, 1982) Hjordis Thort, MartynT. Smith& Pia HartzellS, Giorgio Bellomo$l(, SaAr.aJhewelll, and Sten Orreniusl From the +Department of Forensic Medicine, Karolinska Institutet, S-10.401 Stockholm, Sweden, and the $Toxicology Unit, Department of Pharmacology, The School of Pharmacy, London W C l N I A X , United Kingdom Downloaded from www.jbc.org by guest, on March 29, 2010 The cytotoxic effects of many quinones are thought conditions of oxidative stress via the oxidation of reduced to be mediated through their one-electron reduction topyridine nucleotides and the formation of active oxygen spe- semiquinone radicals, which subsequently enter redox cies, capable of inflicting damage by processes such as lipid cycles with molecular oxygen to produce active oxypgeernoxidation (9). The redox cycling of quinonoid drugs and species and oxidative stress. The two-electron reduc- other related compounds has, therefore, been widely impli- tion of quinones to diols, mediated by DT-diaphorase cated as a mechanism for their cytotoxicity (8,9). (NAD(P)H: (quinone-acceptor)oxidoreductase),may Relatively little is known, however, about the metabolism therefore represent a detoxifying pathway which pro-of quinonoid drugs in cells, or the factors which govern the tects the ceflrlom the formationof these reactive inter-formation of semiquinoneradicals and toxic oxygenspecies in mediates. By using menadione (2-methyl-1,4-naphthoquinone) and isolated hepatocytes, the relative contribution of the two pathwaystoquinonemetabolism has been studied and a protective role for DT-diaphorase dem- onstrated. Moreover,in the presence of cytotoxic con- uivo. For example, the flavoprotein (NAD(P)H: (quinone-ac- ceptor) oxidoreductase, alsoknown as DT-diaphorase (lo), catalyzes the two-electron reduction of quinones to hydroqui- nones without the formation of semiquinone radical interme- diates (2, lo), whereas NADPH-cytochrome P-450 reductase centrations of menadione rapid chainginetsracellular and NADH-ubiquinone oxidoreductase catalyze the one-elec- thiol and Ca" homeostasis were observed. These werteron reduction of quinones to semiquinone radicals (2, 11).As associated with alterations in the surface structuroef first postulated by Ernster and associates (19,DT-diaphorase the hepatocytes which may be an early indication of may, therefore, protect cells against the oxidative stress in- cytotoxicity. duced by quinonoid drugs by competing with the single elec- tron reduction pathways. In thepresent investigation this potentially pratective role of DT-diaphorase and other factors relevant to the metabo- Quinones are widely distributed in nature and many clini- lism and cytotoxicity of quinonoid drugs have been studied cally important antitumor drugs contain the quinone nucleus using menadione (2-methyl-l,4-naphthoquinonaen)d freshly (1).They form an important group of substrates for flavoen- isolated rat hepatocytes as theexperimental model. Pretreat- zymes and can undergo either two-electron reduction to the ment of the rats with either phenobarbital or 3-methylchol- hydroquinone or one-electron reduction to the semiquinone radical ( 2 ) .The antitumor and cytotoxic effects of quinonoid drugs are thought to be mediated through their one-electron reduction to semiquinone radicals (3). Most semiquinones rapidly reduce dioxygen to form superoxide anion radical (02:) and thus regenerate the quinone (4).Quinones may, therefore, enter flavoprotein-catalyzedredox cycles with dioxygen which result in the formation of large amounts of 02: and the oxidation of reduced pyridine nucleotides. The enzymatic or spontaneous dismutation of 02:yields HzOzand O2(5).02: and Hz02can react together, in a process catalyzed by certain metal ions, to form even more deleterious oxygen species such as thehydroxyl radical (OH') andsinglet oxygen ('Ag 02)(6-8). The flavoprotein-catalyzed redox cycling of quinones in cellswould, therefore, quickly lead to anthrene facilitated the alterationof specific enzymeactivities under investigation. For example, phenobarbital administration is known to induce hepatic NADPH-cytochrome P-450 reductase (13)but has littleeffect on DT-diaphorase, whereas 3C' pretreatment increases the activity of hepatic cytosolic DT-diaphorase (14)but has little or no effect on NADPHcytochrome P-450 reductase (13).It has, therefore, been possible to study the relative roles of DT-diaphorase and NADPH-cytochrome P-450 reductase in regulating the metabolism and cytotoxicity of menadione in isolated hepatocytes. The results obtained clearly show a protective role for DT-diaphorase in preventing menadione cytotoxicity and demonstrate the importance of competition between activating and detoxifyingpathways in regulating the oxidativestress caused by quinonoid drugs. Moreover, it has been possible * This work was supported by grants from the Swedish Medical Research Council (Project 03)3-2471), the Swedish Council for Planning and Coordination of Research, and theNobel Foundation. The costs of publication of this article were defrayed in part by the using techniques recently developed in our laboratory (15) to show that one of the early events in menadione-induced cytotoxicity is a change in surface structure of the isolated hepatocytes which appears to be caused by alterations in payment of page charges. This articlemusttherefore be hereby intracellular thiol and Ca2+homeostasis. marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. fi Present address, Clinica Medica 11",University of Pavia, Pavia, The abbreviations used are: 3-MC, 3-methylcholanthrene; GSH, Italy. glutathione, reduced form. 12419 12420 HepatocytesinMeMtaebnoaldisimone EXPERIMENTALPROCEDURES Materials-Collagenase (Grade 11) and carbonylcyanidep-trifluoromethoxyphenyl hydrazone were obtained from Boehringer-Mannheim GmbH, Mannheim, FRG. Horse heart cytochrome c, antimycin A, superoxidedismutase, dicoumarol, menadione, arsenazo 111,bovine serum albumin (Fraction V), NADH, NADPH, and 4-(2-hydroxyethyl)-1-piperazineethanesulfonicacid were obtained from Sigma. The Ca2+ionophore A23187 was from Calbiochem-Behring and Percoll was obtained from Pharmacia Fine Chemicals AB, Uppsala, Sweden. Acetylated cytochrome c was prepared by the method of Azzi et al. (16). Animals andPretreatments-Male Sprague-Dawley rats (200-220 g) were used for all experiments and allowedfood andwater ad libitum. Some rats were given phenobarbital (80 mg/kg of sodium phenobarbital i.p.) for 5 days and others a single i.p. injection of 3MC (50 mg/kg) as a 5 mg/ml solution in corn oil 40 h prior to use. Hepatocyte Isolation and Incubation-Hepatocytes were isolated by collagenaseperfusion of the liver as previously described (17).The yield of each preparation was 2-4 X 10' cells/liver, and immediately after isolation the hepatocytes excluded both trypanblue and NADH (90-100%).Cell viability was determined during the course of the experiments by the exclusion of trypan blue and by the NADH penetration assay (17).Krebs-Henseleit buffer, pH 7.4, containing 25 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfoniaccid was used for the incubation. The hepatocytes were incubated at IO6 cells/ml in rotating round-bottom flasks as previously described (17).Menadione was added in 10p1 of dimethyl sulfoxide. Preparation of Subcellular Fractions-The microsomal and cytosolic fractions of control rat liver were isolated as described by Ernster et al. (18). The subcellular fractions were incubated in 100 m~ Tris-HC1 buffer, pH 7.4, containing 50 mM KC1 a t aprotein concentration of 1mg/ml. Biochemical Assays-Oxygen consumption was measured polarographically a t 37 "C using a 2-ml water-jacketed cell fitted with a Clark oxygen electrode. Antimycin A (25 p ~ w)as used to inhibit mitochondrial oxygen uptake during the oxygen consumption measurements in isolated hepatocytes. Superoxide anion production was measured as the reduction of acetylated cytochrome c using the wavelength pair 550-540 nm (16, 19).Addition of superoxide dismutase (0.2 mg/ml) enabled the background reduction rate tobe subtracted. Acetylated cytochrome c (0.2 mg/ml) was added prior to theaddition of substrate. When subcellular fractions were used, NADPH or NADH (1m ~ w)as also included in the incubation. NADPH and NADH oxidation by the subcellular fractions was measured at 340 nm using incubation conditions identical to those described above in the absence of acetylated cytochrome e. DTdiaphorase was assayed in the microsomal and cytosolic fractions of rat liver using NADPH oxidation at 340 nm as previously described (10). The background rate of NADPH oxidation measured in the presence of30 p~ dicoumarol was subtracted from the total rate observed. Hepatocyte GSH level was measured as acid-soluble thiols, using the colorimetric assay of Saville (20).NAD(P)+and NAD(P)H concentrations were assayed by the spectrophotometric method de- scribed in Ref. 21. Protein was determined by the method of Lowry et al. (22). Ca2+ Compartmentation Measurements-For technical reasons (15), hepatocytes isolated from untreated rats were incubated as described above, but at a cell concentration of 6 X lo6ceUs/ml, for the measurement of intracellular Ca2+distribution. The hepatocytes were separated from the Ca*+-containingKrebs-Henseleit buffer by rapid centrifugation through a suspension of Percoll in Ca2'- and Mg2+-freeHank's solution (23) (final density: 1.06 g/ml) (15). They were then quickly resuspended in the modified Hank's medium and separated into two parts, one part was used for Ca2+ measurement and the other for counting the number of cells presentand for assaying cell viability. Intracellular Ca2+distribution was determined by dual wavelength spectrophotometry using purified arsenazo I11 (2, 2'-(1,8-dihydroxy3,6-disulfonaphtbalene-2,7-bisazo)bis-(benzeanresonic acid) and the wavelength pair 685-675 nm (15, 23, 24). The Ca2+ releasable by carbonylcyanide p-trifluoromethoxyphenyl hydrazone is thought to represent the mitochondrial Ca2+pool and Ca2' released by A23187, following carbonylcyanide p-trifluoromethoxyphenyl hydrazone release, the extramitochondrial Ca2+pool (15,23). Scanning Electron Microscopy-Samples were processed by standard procedures, involving glutaraldehyde and osmium fixation followed bycritical point drying. A Jeol model JSM35 scanning electron Downloaded from www.jbc.org by guest, on March 29, 2010 microscope was usedto visualize and photograph the samples. A large number of hepatocytes were examined and representativecells showing a typical morphology were photographed. RESULTS Effect of Dicoumarol on Menadione-stimulated Oxygen Uptake by Isolated Hepatocytes-The addition of micromolar concentrations of menadione to suspensions of isolated hepatocytes greatly stimulated oxygen consumption in the presence of antimycin A (Fig. 1). This increase occurred to approximately the same extent in hepatocytes from phenobarbital- and 3-MC-treatedrats. However,concentrations of menadione less than 15PM were without effect on oxygenuptake by hepatocytes from 3-MC-treated rats, but caused up to a 50%stimulation of oxygen consumption by hepatocytes from phenobarbital-treated rats. Fig. 1also shows the marked effect of adding dicoumarol(30 PM), prior to menadione addition, on menadione-stimulated oxygen uptake. The presence of dicoumarol produced a left shift in the dose-responsecurve (Fig. l), greatly stimulating antimycin A-independent oxygen consumption by hepatocytes isolated from both phenobarbitaland 3-MC-treatedrats. The effect of dicoumarolwas, however, slightly larger in hepatocytes from 3-MC-treated than from phenobarbital-treated rats (Fig. 1).A t menadione concentrations greater than 50 p ~ pr,ior addition of dicoumarol was without effect. Dicoumarol addition also had no effect upon antimycin A-independent oxygen uptake by hepatocytes in the absence of menadione (Fig. 1). Taken together these results suggest that inhibition of DT-diaphorase-mediated metabolism of menadione in hepatocytes by dicoumarol increases the availability of menadione for single electron reduction and redox cycling resulting in enhanced oxygen consumption. 1 I 10 20 30 10 M Menadione (pM) FIG. 1. Effects of menadione and dicoumarol on antimycin A-independent0, uptake byhepatocytes from phenobarbitaland 3-MC-treated rats. The 0 2 consumption by hepatocytes isolated from either phenobarbital(0)or-3C (.)-pretreated rats was measured polarographically in the presence of 25 p~ antimycin A and varying concentrations of menadione. The effect of adding 30 pM dicoumarol prior tothe addition of menadione is also shown: 0, phenobarbital cells, m, 3C cells. The mean values of a t least four separate experiments are shown. Downloaded from www.jbc.org by guest, on March 29, 2010 HepatocytesinMeMtaebnoaldisimone 12421 Superoxide Formation during thMe etabolism of Menadi- activity obtained with NADPH was much higher than with one in Isolated Hepatocytes-Fig. 2 shows that hepatocytes NADH (Fig. 3). In contrast to the findings with the micro- isolated from phenobarbital-treated rats did not on their own somal fraction, where superoxide dismutase addition caused liberate significant quantities of 02' intothe medium, as an almost complete inhibition of cytochrome c reduction, the measured by the reduction of acetylated cytochromec. When menadione-dependent reduction of acetylated cytochrome c menadione was added to the incubation medium, however, catalyzed by the cytosolic fraction was unaffected by addition there was amarked increase in therate of reduction of of superoxide dismutase (Fig. 3 ) , but was almost completely extracellular acetylated cytochrome c (Fig. 2). A large propor- inhibited by dicoumarol (Fig. 3). These results confirm that tion of this menadione-dependent cytochromcereduction was cytosolic DT-diaphorase reduces menadione to therelatively inhibited by addition of 0.2 mg/ml of superoxide dismutase, stable hydroquinone, menadiol, without the formation of showing that it was due to 0 2 - release from the hepatocytes. On'. Some DT-diaphorase is also present in the microsomal Interestingly, a significant proportion of the menadione-de- fraction and converts menadione to menadiol by the same pendent reduction of acetylated cytochrome c was resistant two-electron transfer process without formingO2' (Fig. 3). to superoxide dismutase inhibition. Since the fully reduced Depletion of Glutathione duringthe Metabolism of Men- form of menadione, 2-methyl-1,4-naphthohydroquinone adione in Isolated Hepatocytes-A marked decrease in intra- (menadiol),is capable of reducing artificial electron acceptors cellular GSH level occurred independently of measurable such as acetylatedcytochrome c (10-14), the reduction of changes in the pyridine nucleotide redox state when hepato- cytochrome c observed in the presence of added superoxide cytes isolated from phenobarbital-treatreadts were incubated dismutase may reflect the release of menadiol from the he- with menadione concentrations greater than 35 PM (Fig. 4). patocytes. The use of superoxide dismutase and dicoumarol in combination almost completely inhibited the reduction of cytochrome c (Fig. 2 ) showing that menadiol and Ony were the only substances reducing cytochrome c in this system. Dicoumarol alone did not increasethe menadione-dependent release of 02' from hepatocytes (Fig. 2 ) , as one would have expected from thedata shown inFig. 1. Thisapparently .55240 '0al 'i c< L o contradictory result is discussed in detail later in this report. Superoxide Production during the Metabolism of Mena- dione by Isolated SubcellularFractions-In the presence of 10 PM menadione and either NADH or NADPH both the microsomal and cytosolic fractions of rat liver catalyzed the rapid reduction of added acetylated cytochrome c (Fig. 3 ) . SF :'20 0 20 2c 5- 10 U The majority of the cytochrome c reduction produced by the microsomal fraction could be attributed to the formation of microsomes microsomes ++ On-, since it was prevented by the addition of superoxide NADHNADPH.NADHNADPH cytosol + cytosol + dismutase. NADPH-cytochrome P-450reductase also appears FIG. 3. Effect of addedsuperoxide dismutase anddicou- to be a more effective catalyst of menadione-dependent 0 2 - marol on acetylatedcytochromee reduction during menadione formation than NADH-cytochrome b5 reductase, since the metabolism by different subcellular fractions, isolated from phenobarbital-treated rats. The subcellularfractions (1 mg of protein/ml) were incubated with 10 PM menadione in 0.1 M Tris-HC1, 'I pH 7.4, containing50 mMKC1 in the presence of either 1 mM NADPH or 1 mM NADH. Reactions were performedat 25 "C.Dicoumarol(30 pM) (solid bar) or superoxidedismutase (0.2 rng/ml) (striped bar) was added separately to test the specificity of the reactions. Values are expressed as the mean f S.E. of three experiments. 0 10 50 100 Menadione ( V M ) FIG.2. Effect of added superoxide dismutase anddicoumarolon the rate of reduction of extracellular acetylated cytochromec during menadione metabolismin isolated hepatocytes. Hepatocytes isolated from phenobarbital-treatedrats were incubated (IO6 cells/ml) in Krebs-Henseleit buffer, pH 7.4, with different concentrations of menadione in the presence of acetylated cytochrome c (0.2 mg/ml). Dicoumarol (30 PM) (solid bar) or superoxide dismutase (0.2 mg/ml) (striped bar) wasadded either separately, or in combination (speckled bar),to test the specificity of the measurements. Values are expressed as the mean rt S.E. of three to five experiments. Menadione ( pM) FIG.4. Low concentrations of menadionedeplete GSH from isolated hepatocytes without affecting theredox state of pyridine nucleotides. Hepatocytes were isolated from phenobarbitaltreated rats and incubated with different concentrations of menadione. After 20 min, samples were taken and the GSH level (o"-o), the NADPH/(NADP' + NADPH) (0....O)a,nd NADH/(NAD+ + NADH) (0---0)ratios determined.The results of one experiment typical of three are shown. 12422 Menadione Metabolism in Hepatocytes Downloaded from www.jbc.org by guest, on March 29, 2010 Up to this concentration the hepatocytes were relatively re- sistant to menadione-induced GSH loss, but in the 30-40 p~ range a threshold value appears to have been reached. This could reflect saturation of the capacity of the hepatocytes to carry out two-electron reduction of menadione, since one- electron reduction resulting in the formation ofO2' and H20z is almost certainly responsiblefor the GSH loss (25). This result therefore suggests the preferential metabolism of men- adione via the DT-diaphorase-mediated two-electron reduc- tion pathway in isolated hepatocytes. Support for this idea also comes from a comparison of the apparentK , values for menadione of cytosolic DT-diaphorase and microsomal NADPH-cytochrome P-450 reductase, whichwere deter- * *mined to be 3.6 1.0 and 43.0 8.5 ,UMres,pectively. (Values *represent the mean S.E. of five separate determinations in isolated subcellular fractions.) Potentiation of the Cytotoxic Effect of Menadione by Di- coumarol and Diethyldithiocarbamate-As shown in Fig. 5, depletion of cellular GSH during menadione metabolism was followedby rapid loss of cell viability measured by trypan blue uptake. The loss of both GSH and cell viability was more Zb 10 60 i 80 100 Menodione($4) FIG. 6. Potentiation of menadione cytotoxicity by dicoumarol and diethyldithiocarbamate. Hepatocytes isolated from phenobarbital-treated rats were incubat(eldo6cells/ml) with varying concentrationsof menadione alone (O),and with dicoumarol, 30 p~ (0)or diethyldithiocarbamate,25 YM (A) alsopresent.After a 2-h incubation the viability of the hepatocytes was determined as the percentage of cells taking up trypan blue. The results of one experiment typical of three are shown. rapid and extensive with hepatocytes isolated from phenobarbital-treated rats than with those from either 3-MC-treated oruntreated (control) animals. These findings supportthe hypothesis that one-electron reduction of menadione leading to redox cycling is responsible for the effects on both cellular GSH level and viability. Further support for this hypothesis is presented in Fig. 6, whichshows that both dicoumarol and diethyldithiocarbamate, an inhibitor of superoxide dismutase (26), potentiated This suggested to us that surface blebbing was a very early event in menadione cytotoxicity,which would bien agreement with our previous findings on the toxicity of bromobenzene in isolated hepatocytes (27). Moreover, we found that dicoumarol greatly potentiated the formation of surface blebs (Fig. 70) atconcentrations of menadione which caused only minor perturbations of hepatocyte surface structure (Fig. 7 B ) .Dicoumarol(30 p ~alo) ne had no effect on the surface morphology of the hepatocytes (Fig. 7C). menadione toxicity to isolated hepatocytes. This effect was most pronounced with dicoumarol,which caused a 3- to &fold increase in the toxicity of menadione at concentrations between 30 and 100p ~Th. e potentiation of menadione toxicity by diethyldithiocarbamate was less marked, but readily observable at higher menadione concentrations (Fig. 6).Neither dicoumarol (30 e)not diethyldithiocarbamate (25 p ~ ha)d Similar alterations in the surface structure of isolated hepatocytes also occur during the metabolism of t-butylhydroperoxide, and we have recently reported (15) that these changes are associated with alterations in intracellular thiol and Ca2+homeostasis. Fig. 8 shows the effects of menadione and dicoumarol, alone and in combination, on the levels of total Ca2+,mitochondrial Ca2+(A23187-releasable, following any effect on the viability of the hepatocytes in the absence of menadione (cf zero time values in Fig. 6).The results confirm that both DT-diaphorase and superoxide dismutase protect isolated hepatocytes against the toxic effects of menadione. Changes in CellularCaz+Homeostasis during Menadione carbonylcyanide p-trifluoromethoxyphenyl hydrazone-induced release of mitochondrial Ca2+)in isolated hepatocytes. The addition of either menadione (50 p ~ or) dicoumarol (30 PM)led to a substantial decrease in total cell ea'+, but only the twoin combination caused a complete loss (Fig. a). Metabolism-Early in the investigation we noticed that high Dicoumarol caused a release of mitochondrial Ca'+ (Fig. 8B), concentrations (100-200 PM)of menadione caused the forma- but did not produce blebbing on the hepatocyte surface (Fig. tion of a great number of s m d blebs on the surface of many 7C). Thus, the level of mitochondrial Ca" does not appearto of the hepatocytes. Formation of surface blebs occurred during directly control the surface structure of isolated hepatocytes. the 1st h of the incubation and prior to anyincrease in trypan Both menadione and dicoumarol alsocaused an approxiblue uptake or other indications of a decrease in cell viability. mately 50%loss of extramitochondrial Ca2+after 30-60 min of incubation, but the two together caused a total loss of extra- mitochondrial Ca2+after only 20 min (Fig. 8C). This was hssociated with extensive blebbing of the hepatocytes which subsequently showed fragmentation and substantial loss of viability. DISCUSSION A schematic view of quinonoid drug metabolism in isolated hepatocytes is presented in Fig. 9. Although some quinones can be reduced directly by intracellular reductants, such as ascorbate and GSH, the majority are reduced enzymatically llme Ihours1 by flavoproteinsin either aone-electron or two-electrontrans- FIG. 5. Relationshipbetween GSH concentration (A)and cell fer process (2, 3, 10).The one-electron reduction of quinones viability (B) during menadione metabolism in isolated hepa- results in the formation of semiquinoneradicals,which usually tocytes. Hepatocytes isolated from control (A), phenobarbital (0)- have a high affinity for O2 and reduce it to02' (3, 9, 11). The or 3C (.)-treated rats were incubatedwith menadione (50p ~an)d dicoumarol (30 p ~fo)r up to 3 h. At specific times the GSH concentration (A)and the percentageof cells taking up trypan blu(eB )were quinone is regenerated in this process and a redox cycle is initiated forming large amounts of 02' (9, 28, 29); Od subse- determined.The values are expressedas the means of three separate quently dismutates to form H202and O2in a reaction catalyzed experiments. by the superoxide dismutases (5-8, 30). However, 02' and Menadione Metabolism in Hepatocytes 12423 FIG.7. Effect of menadioneand dicoumarol, alone and in combina- tion, on the surface structure of isolated hepatocytes. Hepatocytes from phenobarbital-treatedrats were incubated (IO"cells/ml) with ( A )no addition; ( B ) menadione, 50 p ~ (;C ) dicoumarol, 30 p ~ a;nd (D)menadione, 50 p~ + di- coumarol, 30 p~ for 20 min, and samples were processed for scanning electron microscopy. Typical hepatocyateres shown. Total magnifications: A, B, and cx4ooo:Dx2500. Downloaded from www.jbc.org by guest, on March 29, 2010 C oxidation of GSH andimportant, protein thiogl roups islargely responsible for its cytotoxic effects. In the present study we have shown that NADPH-cyto- chrome P-450 reductase is a moreefficient catalyst of the one- electron reduction of menadione to menasemiquinone than NADH-cytochrome bDreductase.However, it is likely that the mitochondrial flavoprotein, NADH-ubiquinone oxidore- ductase, may play a moreimportant role for the redox cycling of other quinones, as recently suggested by Powis and co- Minute, FIG.8. Effect of menadione and dicoumarol on the distribution of Ca2+in isolated hepatocytes. Hepatocytes from un- workers (29).The contribution of each flavoprotein will largely be controlled by the one-electron reduction potential of the treated rats were incubated (6 X IO6 cells/ml) with menadione (0)quinone (11, 29). and dicoumarol (V),alone and in combination (Ob,at concentrations The two-electron reduction of menadione is catalyzed most of 50 and 30 p ~ re,spectively. The amountsof total ( A ) ,mitochondrial ( B ) a, nd extramitochondrial( C )Ca" were then determineda t specific time points using the technique described under "Experimental Pro- cedures." The results of one experimenttypical of three are shown. effectively by the flavoprotein NAD(P)H: (quinone-acceptor) oxidoreductase,otherwise known as DT-diaphorase (2, 10). The resultsof the present studcyonfiim the findings reported in Refs. 2 and 10 that DT-diaphorase reduces quinones to HsOs may also take part in metal-catalyzed reactions to form more toxic species of active oxygen, such as hydroxyl radical (OH`)and singlet oxygen ('Ag 02)(8,9,28). Preliminary evidencesuggests that `Ag O2 is formed duringthe redox cycling of menadione in isolated hepatocytes,' and itis therefore likely that OH. is also formed during this process. Both 0,: and more active formsof oxygen are potentiallyvery toxic and are known to cause DNA strand breaks(31,32),enzyme inhibition (33), lipid peroxidation (34,35), and oxidation of thiol groups in proteins (36). However, lipid peroxidation is hydroquinones (diols) without the formation of semiquinone radical intermediates and 0 2 : (Fig. 9). Moreover, the affinity of menadione for hepaticDT-diaphorase was found to be more than 10 times greater than its affinity for microsomal NADPH-cytochrome P-450 reductase. DT-diaphorase may therefore protect against menadione cytotoxicity by competing with the potentially toxic one-electron reductionpathway.:' Measurement of antimycin A-independent oxygen consumption by isolated hepatocytes allowed us to demonstrate that addition of dicoumarol, a potent inhibitor of DT-diaphorase unlikely to be involved in the toxic effects of menadione, since Since this report was submitted for publication a paper has been this agentis a potent inhibitorof the propagation reactions of lipid peroxidation (37). Menadionedid,however, cause the rapid loss of GSH during redox cycling in isolated hepatocytes (Ref. 25, see also "Results"), and it appears likely that the 'H. Sies, personal communication. published (Lind, C., Hochstein, P., andEmster, L. (1982) Arch. Biochem. Biophys. 216,178-185) in which data obtainedwith isolated subcellular fractions andpurified enzymes are presented. Theriersults are consistent with ours in that they support theidea of DT-diaphoraseactingas acellularcontrol device against semiquinone and superoxide radical formation and hence quinonetoxicity. 12424 Menadione Metabolism in Hepatocytes Downloaded from www.jbc.org by guest, on March 29, 2010 Q release of menadiol from hepatocytes but had less than the HFPAT expected stimulatory effect on 0 2 : release. This apparently anomalous result can probably be explained by dicoumarol havingsomeeffect on the permeability of the hepatocyte plasma membrane to 02'. It has been calculated that only about 3%of the 0 2 - formed during menadione metabolism by hepatocytes will be released (29).Thus, only a small nonspe- cific effect of dicoumarol on 02' permeability could be very significant. The development of specific indicators for Oh production in intact cells is probably required to clarify this question. The addition of dicoumarol did, however,greatly potentiate the cytotoxicity of low concentrations of menadione, further indicating that DT-diaphorase plays a protective role in pre- venting the potentially toxic redox cycling of menadione in isolated hepatocytes. The idea that DT-diaphorase will com- pete with the singleelectron reduction pathway for menadione is alsosupported by our previous in vitro studies (9).However, the importance of this protective role of DT-diaphorase will depend upon the relative affinity of the quinonoid drug for DT-diaphorase and the flavoproteins catalyzing its one-elec- tron reduction. For example, the quinonoid antitumor drug Adriamycin (doxorubicin)has a similar affinity for both DT- diaphorase and NADPH-cytochrome P-450 reductase in ui- t ~ oan, d~it is therefore unlikely that DT-diaphorase plays any conjugate major role in preventing the redox cyclingof this drug. FIG. 9. A schematic representation of quinonoid drug me- Finally, dicoumarol not only potentiated the overall cytotabolism in isolated hepatocytes. Q,quinone; SQ, semiquinone toxicity of menadione,as determined by trypan blue exclusion, radical, HQ, hydroquinone (diol); FP, flavoprotein;SOD, superoxide but also markedly abrogated its early toxic effectsa, s indicated dismutase; PAPS, 3"phosphoadenosine-5'-phosphosulfate;UDPGA, by alterations in the surface structure of the hepatocytes. This uridine-5'-diphosphoglucuronicacid. potentiation appeared to be fargreaterthan the blocking (lo), greatly increased the availability of menadione for one- electron reduction. The results obtained by measurement of OZTproduction during menadione metabolism in isolated he- action of dicoumarol on DT-diaphorase could possibly cause alone. The studies on changesin intracellular Cazfdistribution during menadione metabolism reported here would also Cend patocytes were, however,slightly more complex. to suggest that dicoumarol further potentiates menadione Derivatized forms of cytochrome c are presently the indicators of choice in measuring 02' formation in biological, sytems (16,38). However, one major drawback to their use is that they can also be reduced by other compounds, such as menadiol (10, 14). This limitation is clearly shownin the present study where we introduced extracellular acetylated cytochrome c to measure 02' release by isolated hepatocytes. Incubation of hepatocytes with menadione greatly stimulated therate of reduction of acetylated cytochrome c, a large proportion of which was superoxide dismutase-inhibitable. This stimulatory effect of menadione on hepatocyte 02'' formation is in sharp contrast to its inhibitory effect on 0 2 : production by stimulated human neutrophils (39). The 0; - cytotoxicity by its mild uncoupling effect. Recent studies in our laboratory (15, 41) have shown that interference with mitochondrial ea2*homeostasis is also necessary if one is to observe a rapid onset of extensive surface blebbing following GSH depletion in isolated hepatocytes. Dicoumarol could therefore be potentiating the cytotoxicityof menadione in two ways: ( a )by inhibiting its two-electron reduction to menadiol, thereby making moreof the quinone available for potentially toxic redox cycling; and (b) by decreasing the ability of the mitochondria to control overall intracellular Ca2+homeostasis. The importance of changes in intracellular ea2+homeostasis during quinonoid drug metabolism in isolated hepatocytes and other cell types is worthy of further investigation. generating ability of neutrophils is, however,dependent upon Acknowledgment-The authors are grateful to Sten Thorold for the NAD(P)H oxidoreductase complex in the plasma mem- preparing the scanning electron micrographs. brane, which appears to have an intrinsic quinone reductase activity (40). Moreover, the quinonoid compound dichlorophenol-indophenol competitively inhibits this Ozr -generating oxidoreductase (40). To the best of our knowledge no such REFERENCES 1. Driscoll, J. S., Hazard, G. F., Wood, H. B., and Goldin, A. (1974) Cancer Chemother. Rep. Part 2 4, 1-362 2. Iyanagi, T.,and Yamazaki,I. (1970)Biochzm. Biophys. Acta 216, oxidoreductase complex is present to anysignificant extent in the plasma membrane of rat hepatocytes. The hepatocyte and neutrophil systems are therefore obviously different and care should be taken when comparing the effects of menadione on different cell types. From the results of the oxygen uptake measurements one would have expected dicoumarol to increase the production 282-294 3. Bachur, N., Gordon, S. L., and Gee, M. V. (1978) Cancer Res. 38, 1745-1 750 4. Patel. K. B.. and Willson. R. L. (1973) J. Chem. Soc. Faraday Trans. 69; 814-825 ' 5. McCord. J. M.. andFridovich. 1. (1969) J. Biol. Chem. 244, 6049-6055 6. Beauchamp, C., and Fridovich, I. (1970) J. Biol. Chem. 245, and release of OZ7by hepatocytes in the presence of menadione. This increased production should have resulted in a higher rate of extracellular cytochrome c reduction which was counterbalanced to some extent by dicoumarol preventing the 4641-4646 7. Kellogg, E. W., 111, and Fridovich, I. (1975) J. Biol. Chem. 250, 8812-8817 8. Hassan, H. M., and Fridovich, I. (1979)Arch. Biochem. Biophys. formation and release of menadiol. Careful analysis of the H. Thor, M. T. Smith, P. Hartzell, and S. Orrenius, unpublished data presented in Fig, 3 shows that dicoumarol inhibited the observations. HepatocytesinMeMtaebnoaldisimone 12425 196,385-395 9. Smith, M. T., Thor, H., and Orrenius, S. (1982) in Microsomes, Drug Oxidation and Drug Toxicity (Sato, R., and Kato, R., eds) Japan Scientific Societies Press, Tokyo, 605-612 10. Ernster, L. (1967)Methods Enzymol. 10,309-317 11. Powis, G., and Appel, P. L. (1980) Biochem.Pharmacol. 29, 2567-2572 12. Ernster, L., Lind, C., Nordenbrand, K., Thor, H., and Orrenius, S. (1982) in Oxygenases and Oxygen Metabolism (Nozaki, M., Yamamoto, P., Ishimura, Y., Coon,M. J., Ernster, L., and Estabrook, R. W., eds) Academic Press, Tokyo, in press 13. Yasukochi, U., and Masters, B. S. S. (1976) J. Biol. Chem. 251, 5337-5344 14. Lind, C., and Ernster, L. (1974)Biochem. Biophys. Res. Commun. 56,392-400 15. Bellomo, G., Jewell, S. A,, and Orrenius, S. (1982) Proc. Natl. Acad. Sci. U. S. A , , in press 16. Azzi, A,, Montecucco, C., and Richter, C. (1975) Biochem. Biophys. Res.Commun. 65,597-607 17. Moldeus, P., Hogberg, J., and Orrenius, S. (1978) Methods Enzymol. 52.60-71 18. Ernster, L., Siekevitz, P., and Palade, G. E. (1962) J. Cell Biol. 15, 541-562 19. Thayer, W. S. (1977) Chem. Biol. Interact. 19,265-275 20. Saville, B. (1958)Analyst 83,670-672 21. Klingenberg,M. (1970)inMethoden der Enzymatischen Analyze (Bergmeyer, H. V., ed) p. 1975, Verlag-Chemie,Weinheim 22. Lowry, 0.H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951)J. Biol. Chem. 193,265-275 23. Murphy, E., Coll, K., Rich, T. L., and Williamson,J. R. (1980) J. Biol. Chem. 255,6600-6608 24. Kendrick, N.C., Ratzlaff, R.W., and Blaustein, M. P. (1977) Anal. Biochem. 83,433-450 25. Eklow,L., Thor, H., and Orrenius, S. (1981) FEBS Lett. 127, 125-128 26. Heikkila, R. E., Cabbat, F. S., and Cohen, G. (1976)J. Biol. Chem. 251,2182-2185 27. Thor, H., and Orrenius, S. (1980)Arch. Toxicol. 44,31-43 28. Kappus, H., and Sies, H. (1981) Experientia (Basel) 37, 1233-1241 29. Powis, G., Svingen, B. A., and Appel, P. (1981)Mol. Pharmacol. 20,387-394 30. Fridovich, I. (1975)Annu. Rev. Biochem. 44, 147-159 31. Lesks, S. A., Lorentzen, R. J., and Ts'o, P. 0.P. (1980)Biochem- istry 19,3023-3028 32. Brawn, K., and Fridovich, I. (1981)Arch. Biochem. Biophys.206, 414-419 33. Searle, A. J., and Willson, R. L. (1980) Znt. J. Radiat. Biol. 37, 213-217 34. Bus, J. S., and Gibson, J. E. (1979) in Reviews in Biochemical Toxicology (Hodgson, E., Bend, J . R., and Philpot, R. M., eds) Vol. 1, pp. 125-149, Elsevier, Amsterdam 35. Tien, M., Svingen, B. A., and Aust, S. D. (1981)Fed. Proc. 40, 179-182 36. McCord, J. M. (1979) in Reviews in Biochemical Toxicology (Hodgson, E., Bend, J. R., and Philpot, R. M., eds) Vol. 1, pp. 109-124, Elsevier, Amsterdam 37. Wills, E.D. (1972)Biochem. Pharmacol. 21,1879-1883 38. McCord, J . M., Crapo, J . D., and Fridovich, I. (1977) in Super- oxide and Superoxide Dismutases (Michelson,A. M., McCord, J. M., and Fridovich, I., eds) pp. 11-18, Academic Press, New York 39. Gallin, J. I., Seligmann, B. E., Cramer, E. B., Schiffmann, E., and Fletcher, M. P. (1982)J. Zmmunol. 128, 1399-1408 40. Green, T. R., and Schaefer, R. E. (1981) Biochemistry 20, 7483-7487 41. Jewell, S. A., Bellomo, G., Thor, H., Orrenius, S., and Smith, M. T. (1982)Science (Wash. D.C . ) ,in press Downloaded from www.jbc.org by guest, on March 29, 2010