53

Chem.-BioL Interactions, 73 (1990) 53-76 Elsevier Scientific Publishers Ireland Ltd.

EFFECT OF SUPEROXIDE DISMUTASE ON THE AUTOXIDATION OF SUBSTITUTED HYDRO- AND SEMI-NAPHTHOQUINONES

KARIN 0LLINGER',GARY D.BUFFINTON',LARSERNSTERband ENRIQUE CADENAS~*

•Department of Pathology II, University of Link~ping, S-581 85 Link~ping, and bDepartment of Biochemistry, University of Stockholm S-106 91 Stockholm (Sweden) (Received April 13th, 1989) (Revision received May 26th, 1989) (Accepted May 29th, 1989)

SUMMARY

The effect of superoxide dismutase on the autoxidation of hydro- and semi1,4-naphthoquinones with different substitution pattern and covering a oneelectron reduction potential range from 95 to - 415 mV was examined. The naphthoquinone derivatives were reduced via one or two electrons by purified NADPH-cytochrome P-450 reductase or DT-diaphorase, respectively. Superoxide dismutase did not alter or slightly enhanced the initial rates of enzymic reduction, whereas it affected in a different manner the following autoxidation of the semi- and hydroquinones formed. Autoxidation was assessed as NADPH oxidation in excess to the amounts required to reduce the quinone present, H202formation, and the redox state of the quinones. Superoxide dismutase enhanced 2--8-fold the autoxidation of 1,4-naphthosemiquinones, following the reduction of the oxidized counterpart by NADPH-cytochrome P-450 reductase, except for the glutathionyl-substituted naphthosemiquinones, whose autoxidation was not affected by superoxide dismutase. Superoxide dismutase exerted two distinct effects on the autoxidation of naphthohydroquinones formed during DT-diaphorase catalysis: on the one hand, it enhanced slightly the autoxidation of 1,4-naphthohydroquinones with a hydroxyl substituent in the benzene ring: 5-hydroxy-l,4-naphthoquinone and the corresponding derivatives with methyl- and/or glutathionyl substituents at C2 and C3, respective|y. On the other hand, superoxide dismutase inhibited the autoxidation of naphthohydroquinones that were either unsubstituted or with glutathionyl-, methyl-, methoxyl-, or hydroxyl substituents (the latter in the quinoid ring). The inhibition of hydroquinone autoxidation was reflected as a decrease of NADPH oxidation, -

*To whom all correspondence should be addressed at: Institute for Toxicology, University of Southern California, 1985 Zonal Avenue, Los Angeles, CA 90033, U.S.A. Abbreviations: CHs, methyl substituent; - O C H 8, methoxyl substituent; - O H , hydroxyl substituent; -- SG, glutathionyl substituent; SOD, superoxide dismutase. 000%2797/90/$03.50 © 1990 Elsevier Scientific Publishers Ireland Ltd. Printed and Published in Ireland

54 suppression of H2O2 production, and accumulation of the reduced form of the quinone. The enhancement of autoxidation of 1,4-naphthosemiquinones by superoxide dismutase has been previously rationalized in terms of the rapid removal of O] by the enzyme from the equilibrium of the autoxidation reaction (Q~ + 02 "~ Q + 0]), thus displacing it towards the right. The superoxide dismutase~lependent inhibition of H202 formation as well as NADPH oxidation during the autoxidation of naphthohydroquinones - except those with a hydroxyl substituent in the benzene ring - seems to apply to those organic substrates which can break down with simultaneous formation of a semiquinone and 0~. Inhibition of hydroquinone autoxidation by superoxide dismutase can be interpreted in terms of suppression by the enzyme of O~dependent chain reactions or a direct catalytic interaction with the enzyme that might involve reduction of the semiquinone at expense of 0~. It is suggested that the complimentary activities of DT-diaphorase and superoxide dismutase may represent a detoxication mechanism in the cellular disposal of several quinones.

Key words: Superoxide dismutase -- DT-diaphorase - NADPH-cytochromeP450 r e d u c t a s e

--

Autoxidation -- Naphthosemiquinones

INTRODUCTION

The cellular activation of quinoid compounds is mainly centered on redox processes including (a) electron transfer from cellular flavoproteins to the electrophilic quinones that subsequently reduce 02, a process termed redox cycling, and (b) 1,4-reductive addition reactions with cellular nucleophiles, such as GSH. The relative contribution of redox cycling and sulphydryl arylation to cellular cytotoxicity upon quinone metabolism has been recently discussed [1]. The glutathionyl-quinone conjugates participate in redox cycling as the parent quinones do, for they are reduced by cellular flavoproteins [ 2 - 4] and autoxidize in one-electron transfer steps yielding O~and glutathionyl-semiquinone intermediates [5]. Although the reduction potential of the quinone is only slightly altered upon glutathionyl substitution [6], the two-electron reduced thioether derivatives of un- and methyl-substituted naphthoquinones autoxidize at rates far faster than the parent compounds lacking a - S G substituent [4]. Both processes, redox cycling and 1,4-reductive addition with cellular nucleophiles, appear to be interrelated and can represent an initial critical bimolecular reaction within the cell leading to propagation of alterations in other biochemical pathways and ultimately to cell impairment and death [7--9]. Certain quinones can increase the cellular levels of superoxide dismutase, thus pointing to a relationship between O~-generating quinones and superoxide dismutase activity [10]. The observed inhibition of superoxide dismutase by quinones [11], along with the superoxide dismutase-mediated decrease in autoxidation of hydroxy-p-benzohydroquinones [12], cannot be entirely explained in

55

terms of the classical disproportionation of O] [13] or semiquinones [14] catalyzed by the enzyme, the latter reaction proceeding at a rate three orders of magnitude slower (3 × 10e M-1 s -1) than the disproportionation of O] (2.4 × 109 M-1 s-l). Superoxide dismutase displaces the equilibrium of semiquinone autoxidation Q: + 02.~- Q + 0]) upon removal of O] during its classical superoxidedismutating activity [15]. Within an effort to correlate kinetic and thermodynamic factors involved in the electron-transfer reactions of quinoid compounds, we have undertaken to evaluate the effect of Cu-Zn superoxide dismutase on the autoxidation of semiand hydroquinones of the naphthoquinone series covering a wide range of oneelectron reduction potential [E(Q/Q~)], from - 9 5 mV to - 4 1 5 mV. The substitution pattern of these naphthoquinones included - O H , - C H s, - O C H s and - SG groups. Following the one- or two-electron reduction of these naphthoquinones by purified NADPH-cytochrome P-450 reductase or DT~liaphorase, respectively, autoxidation of semi- or hydro-naphthoquinones was evaluated in the absence and presence of superoxide dismutase in terms of NADPH oxidation in amounts in excess to that required to reduce the quinone present, H202 formation, and the redox state of the quinone ([hydroquinone]/[quinone]). MATERIALSANDMETHODS

Chemicals and biochemicals 1,4-naphthoquinone, 2-hydroxy-l,4-naphthoquinone, 5-hydroxy-l,4-naphthoquinone, 2-methyl-5-hydroxy-naphthoquinone, p-hydroxyphenylacetic acid, and H202 were from Aldrich-Chemie (Steinheim, F.R.G.). NADPH, superoxide dismutase, and horseradish peroxidase (grade I) were from Boehringer (Mannheim, F.R.G.). 2,3-Dimethyl- and 2,3-dimethoxyl-l,4-naphthoquinones were gifts from Dr. M. Threadgill (Medical Research Council, Chilton, U.K.) and Dr. G.M. Cohen (Department of Pharmacology, University of London, U.K.). 2-Methyl1,4-naphthoquinone was from Merck (Darmstadt, F.R.G.). Glutathionyl conjugates of 1,4-naphthoquinone, 2-methyl-l,4-naphthoquinone, 5-hydroxy-l,4naphthoquinone, and 2-methyl-5-hydroxy-l,4-naphthoquinone were prepared as described before [4]. Purified DT-diaphorase [16] and NADPH-cytochrome P-450 reductase [17] were isolated from rat liver and provided by Dr. C. Lind (Department of Biochemistry, University of Stockholm, Sweden) and Dr. M. IngelmanSundberg (Department of Chemistry, Karolinska Institute, Stockholm, Sweden), respectively. The specific activity of DT-diaphorase was 2100 ~mol NADPHo~d~ed × min -1 × mg protein -1, measured with 2-methyl-l,4-naphthc~ quinone as electron acceptor. The specific activity of NADPH-cytochrome P450 reductase was 12.6 pmol cytochrome Creduced × min -1 X mg protein "1. Superoxide dismutase activity was 2880 units × mg lyophilsate -1, determined according to McCord and Fridovich [18] and calculated from linear plots of 1/ AE6s0am × min-1 vs. superoxide dismutase concentration. Concentrations of CuZn superoxide dismutase were calculated assuming a molecular weight of 32 000. The superoxide dismutase used was not contaminated with catalase.

56 Standard assay conditions The standard reaction mixture consisted of 20 ~M quinoid compound and 200 }~M N A D P H in air-saturated 0.25 M sucrose/0.1 M potassium phosphate buffer (pH 7.55) for assays using DT-diaphorase, or 0.2 M potassium phosphate buffer (pH 7.55) for assays using NADPH-cytochrome P-450 reductase. The reaction was started upon addition of 45.2 ng DT-diaphorase × m1-1 or 7.6 ~g NADPHcytochrome/)-450 reductase x m1-1 (for the case of DT-diaphorase, the enzyme suspension contained 0.1% bovine serum albumin before dilution to maintain its catalytic activity) to the reaction mixture. These concentrations of enzymes (approx. 0.095 units x m1-1) were within the linear range of activity and gave rates of N A D P H oxidation of approximately 3--50 ~M × min -1, these values being dependent on the type of naphthoquinone derivative. The concentration of superoxide dismutase, when present in the reaction mixture, was generally 1 ~M (= 95 units x ml -~ or 33 ~g x ml-i). Assay temperature was 37°C. The cuvette holder was equipped with an electronic stirrer (model 200; Rand Bros Ltd., Cambridge, U.K.) in order to ensure a homogeneous gas and reactants distribution in the reaction mixture. Spectrophotometric assays NAD(P)H oxidation was followed at 340 nm (e = 6.22 mM -1 cm -1) and cytochrome c reduction at 550 nm (~ = 21.1 mM -1 cm -1) with a Varian DMS 100 (Varian AB, Solna, Sweden) U.V.-visible spectrophotometer. HeO2 formation was determined fluorometrically coupled to p-hydroxyphenylacetic acid dimerization (~ex¢itation = 315 nm; Aemission = 410 rim) [19,20]. Fluorescence was measured with a Shimadazu RF-540 spectrofluorometer (Shimadzu Corporation, Kyoto, Japan). H.P.L. C. with electrochemical detection Quinoid compounds were analyzed with electrochemical detection using a Ag/AgCl reference electrode as previously described [4]. The mobile phase consisted of 35% propan-2-ol/65% H~O/50 mM sodium phosphate (pH 6.5). The retention times and half-wave potentials of the quinoid compounds have been reported previously [4]. The redox status of the above naphthoquinones was evaluated with oxidative (applied potential, + 0.7 V) or reductive (applied potential, - 0.7 V) electrochemical detection for the analysis of reduced or oxidized forms, respectively. Flow rate: 0.35 ml x min -1. RESULTS AND DISCUSSION The 1,4-naphthoquinone derivatives examined here are listed in Table I and cover a wide range (approx. 320 mV) of one-electron reduction potentials [E(Q/ Q:)], the two extreme values being represented by 1,4-naphthoquinones with a --OH substituent in the benzene ring (5-hydroxy-l,4-naphthoquinone; E(Q/Q:) = - 9 5 mV) [21] and the quinoid ring (2-hydroxy-l,4-naphthoquinone; E(Q/Q:) = - 415 mV)[22]. All the quinoid compounds were reduced by NADPH-cytochrome P-450 reductase and DT-diaphorase [4,23]. The initial rates of quinone enzymic reduc-

57 TABLE

I

ONE-ELECTRON DERIVATIVES

REDUCTION

POTENTIAL

OF SEVERAL

NAPHTHOQUINONE

O

R3

0 RI

R2

Rs

[E{Q/Q~).

- H

- H

- H

- 140

- CH s - CH s

- H - CH 3

-- H - H

- 203 - 240

- H

- 183 b

Unsubstituted 1,4-naphthoquinone

Methyl-substituted 2-Methyl-l,4-naphthoquinone 2,3-Dimethyl-l,4-naphthoquinone

Methoxyl-substituted 2,3-Dimethoxy-l,4-naphthoquinone

--

O C H

3

--

OCH s

Glutathionyl-substituted 3-Glutathionyl-l,4-naphthoquinone 2-Methyl-3-glutathionyl-l,4-naphthoquinone

- H - CH s

-SG

-H

- 132 c

-

-

-

-OH

-H

-H

5-Hydroxy-l,4-naphthoquinone

-- H

-H

-OH

-95

2-Methyl-5-hydroxy-l,4-naphthoquinone 3-Glutathionyl-5-hydroxy-l,4-naphthoquinone 2-Methyl-3-glutathionyl-5-hydroxy-l,4naphthoquinone

- CH 3 -- H -- CH s

- H

- O H

__d

--SG -SG

--OH -OH

_d d

SG

H

195

Hydrozybsubstituted (quinoid ring) 2-Hydroxy-l,4-naphthoquinone

-415

Hydroxyl-substituted (benzene ring/

* D a t a e x p r e s s e d in m V a n d f r o m R e f s . 6 , 1 4 , 2 1 , 2 2 , 2 5 , 3 2 - - 3 4 . bT.W. G a n t a n d G . M . C o h e n , p e r s . c o m m u n . cA. B r u n m a r k a n d E . C a d e n a s , u n p u b l i s h e d r e s u l t s . d__, d a t a n o t a v a i l a b l e .

tion were not strictly related to the reduction potential of the compound, as previously reported for the case of the one- [24] and two- [4] electron transfer flavoproteins. With every quinone - and independent of the flavoenzyme catalyzing its initialreduction - - N A D P H oxidation proceeded beyond the actual amounts of N A D P H required to reduce the quinone present. This additional N A D P H oxidation is attributed to the redox involving enzymic reduction autoxidation and its rate varied with the different naphthoquinones studied. During NADPH-cytochrome P-450 reductase catalysis, the one-electron reduction potential [E(Q/Q-)]of these naphthoquinone derivatives is the critical factor influencing the rate of subsequent electron-transfer reactions, thereby affecting strongly the equilibrium of reaction 1. At the p H utilizedin this study, 7.5, most semiquinones (having p K values between 4 and 5 [25])are present in their anionic form and participate readily in electron-transfer reactions.

58 O" R1 + 0 2 ~ R2

~,

~

O

+ 02"-

R3 O-

RI (1)

R2 R3 0

For the case of DT-diaphorase, the two-electron reduced naphthoquinone is subjected to a stepwise oxidation, i.e. one-electron transfer steps with formation of semiquinone intermediates. Therefore, both the one-electron reduction potential [E(Q/Q:)]and the reduction potential of the intermediate step [E(Q:/Q2-)] are important factors regulating the rate of electron transfer to 02 and determining the variations in the equilibrium of reaction 2. Unfortunately, the reduction potential of the intermediate step [E(Q-/Q2-)] mostly unavailable, though it can be calculated from the difference between the two-electron reduction potential of the quinone [E(Q/Q2-)]and the one-electron reduction potential [EQ/Q-)] [25]. O-

O"

O (2)

R2 R3 O-

+ 02"-

R2 R3 O-

+ O2"-

R2 R~t O

At pH 7.5, most reduced quinones will be in their protonated form - hence slowing electron-transfer reactions [26] -- and only a small fraction in their anionic form. These quinol anions will be mainly involved in the autoxidation following the two-electron reduction of the quinones by DT-diaphorase.

I. A utoxidation of semi- and hydronaphthoquinones Figures 1 and 2 show typical time courses of semi- and hydronaphthoquinone autoxidation, in terms of NADPH oxidation and H202 formation, following the reduction of the oxidized counterparts by NADPH-cytochrome P-450 reductase and DT-diaphorase, respectively. In both cases, NADPH is consumed in excess to the amount of quinone present in the reaction mixture and this is accompanied by H202 formation, thus reflecting the redox cycling linked to semi- or hydroquinone autoxidation. Reduction of 1,4-naphthoquinone by NADPH-cytochrome P-450 reductase (Fig. 1) under anaerobic conditions proceeds at a rate of 14.1 ~M x min-: and it is followed by a residual rate of NADPH oxidation (0.3 ~M × rain-:), due to decay of the semiquinone by pathways other than autoxidation, i.e. disproportionation. Under aerobic conditions ([02]in:tiaI ~ 220 ~M), the initial rate of NADPH oxidation increases slightly (16.1 t t M × rain'1), whereas the following rate is substantially higher (3.5 ~M x min -~) than in anaerobiosis. The higher rate of enzymic reduction of quinones in aerobic conditions has already been reported [24]. In the presence of 0e, H202 accumulates at a rate of 7.5 ~M X min -~.

0~

Tm~ (rain)

I

5

I

0

~-----e J

l0

" t)

'0

°

b

20 L

2°°I

1oo

5

) Timc (rain)

i

L

I

10

~'ig. 1. Time course of NADPH oxidation and H~O~ formation during the autoxidation of 1,4-naphthosemiquinone. Assay conditions: 20 ~ 1,4laphthoquinone in 0.2 M potassium phosphate buffer (pH 7.55), containing 200 p_M NADPH, was supplemented with 7.6 pg × ml-' NADPH-cyto:hrome P-450 reduetase to initiate the reaction (indicated by the arrows). (----) NADPH oxidation in anaerobic conditions. (. . . . ) and ( - - ) , NADPH ~xidation in aerobic conditions ([02~,,., ~ 220 ~ ) in the absence and presence of superoxide dismutase (1/AM), respectively. (O) and (e), H~O2 forma;ion in the absence and presence of superoxide dismutase, respectively. Inset: semilogarithmic plot of [NADPH] oxidation against time obtained ~-om the traces in the main figure; (O) and ( • ) assays in the absence and presence of superoxide dismutase, respectively.

:t

P

i,

:D

60

+ 200

~

O CH3 O

x

? L 100 .¢ Z 5.0

© 2.5~ b v

+ ;

v

v

.

.

.

.

0

v

I

i

5

10

Time (min) Fig. 2. Time course of NADPH oxidation and H2O2 formation during the autoxidation of 2methyl-l,4-naphthohydroquinone. Assay conditions: 20 #M 2-methyl-l,4-naphthoquinone in airsaturated 0.25 M sucrose/0.1 M potassium phosphate buffer (pH 7.55) containing 200 ~M NADPH, was supplemented with 42.5 ng x m1-1 DT-diaphorase to initiate the reaction (indicated by the arrows}, (. . . . ) and ( - - ) , NADPH oxidation in the absence and presence of superoxide dismutase, respectively. (O) and (e), H202 formation in the absence and presence of superoxide dismutase, respectively. Superoxide dismutase concentration, 1 ~M.

The semilogarithmic plot of [ N A D P H ] c o n s u m p t i o n a g a i n s t time (Fig. 1, inset; open symbols) s h o w s a biphasic linear relationship, which can be described in t e r m s of t w o c o m p o n e n t s : first, a rapid N A D P H oxidation, similar to t h a t o b s e r v e d u n d e r anaerobiosis and a t t r i b u t e d to the enzymic r e d u c t i o n of the quinone. Second, a slower N A D P H consumption, which is virtually a b s e n t in anaerobiosis and a t t r i b u t e d to the a u t o x i d a t i o n of the semiquinone within

61 the redox cycling process. The NADPH-cytochrome P-450 reductase-dependent metabolism of all naphthoquinones examined here showed these two phases of NADPH oxidation, except for the glutathionyl-substituted naphthoquinones. In the latter instances, two distinct effects were observed (Table II): (a) a uniform oxidation rate of NADPH during the redox process involving the glutathionyl derivatives of 1,4-naphthoquinone and 2-methyl-l,4-naphthoquinone. (b) A higher oxidation of NADPH following the initial rate of enzymic reduction during the metabolism of glutathionyl-substituted naphthoquinones with a - O H group in the benzene ring.

TABLE II AUTOXIDATION OF 1,4-NAPHTHOQUINONES DERIVATIVES FOLLOWING THEIR REDUCTION BY NADPH-CYTOCHROME P-450 REDUCTASE OR DT-DIAPHORASE Values were obtained after linear relationships of NADPH oxidation against enzyme protein concentration. Other assay conditions as in the Materials and Methods section. When present, superoxide dismutase concentration, 1 ~n~I. First and second in control experiments with NADPH~ytochrome P-450 reductase are the rates for enzymic reddction and autoxidation of the quinone, respectively, calculated as in the inset in Fig. 1. NADPH oxidation values for DT-diaphorase correspond to the second phase (autoxidation) shown in the time course in Fig. 2. Quinone

P~50 reductase°

DTdiap h°rasea

Control

+ SOD

Control

+ SOD

1st

2nd

2.12

0.46

3.76

52

16

2.63 0.73

0.59 0.39

2.89 0.97

38 46

10

2.11

0.86

3.09

891

69

1.29 0.81

1.29 0.81

1.47 0.85

550 202

103 29

0.15

0.16

0.15

30

12

3.29 3.29 2.99 1.91

0.72 1.11 5.11 4.74

3.64 4.54 5.41 5.00

1548 1349 447 621

1615 1417 487 645

Unsubstituted 1,4-Naphthoquinone

Methyl-substituted 2-CHs-l,4-naphthoquinone 2,3-diCHs-l,4-naphthoquinone

11

Methozyl-substituted 2,3-diOCHs-1,4-naphthoquinone

Glutathionyl-substituted 3-SG-1,4-naphthoquinone 2-CH3-3-SG-1,4-naphthoquinone

Hydroxyl-substituted Quinoid~n~ 2-0H-1,4-naphthoquinone

Hydroxyl-substituted fbenzene ringJ 5-OH-1,4-naphthoquinone 2-CHs-5-OH-1,4-naphthoquinone 3-SG-5-OH-1,4-naphthoquinone 2-CHs-3-SG-5-0H-1,4-naphthoquinone

"Values are nmol × min -~ × /~g protein -I.

62 The rate of N A D P H oxidation associated with hydroquinone autoxidation (Fig. 2) shows distinctively two phases, a rapid one, reflecting NADPH consumption during enzymic reduction of the quinone, and a slow phase, indicating autoxidation of the hydroquinone during redox cycling. He02 accumulates slowly, at a rate of 0.51 ~M × min -1. The low autoxidation of 2-methyl-l,4naphthohydroquinone {Fig. 2) has been previously reported [27] and this effect can be accounted for in terms of (a) the one-electron reduction potential [E(Q-/Q2-)] of the intermediate step (Q2- ,~ Q.) (reaction 2) and (b) the charge of the two-electron reduced quinone: (a) The E(Q-/Q e-) values for 1,4-naphthoquinone and 2-methyl-l,4-naphthoquinone are + 212 and + 193 mV, respectively (calculated from the E(Q/Q e-) [28] and E(Q/Q-) [25] in the literature). Given the E(02/0 ~) value of - 155 mV, very little electron transfer -- in terms of autoxidation -- can be expected and the rate of the backward reaction at the intermediate step should be several orders of magnitude higher than the forward reaction. This situation changes in the case of autoxidation of semiquinones (reaction 1), for the E(Q/Q-) values for the above quinones are - 140 and - 203 mV, respectively [14,25]. Therefore, the socalled stability of hydroquinones formed during DT-diaphorase catalysis seems to be partly supported by the more positive values of E(Q-/Q e-) related to the first-electron transfer step. (b) Because hydride transfer by DT-diaphorase is the generally accepted mechanism [29], the reduced quinone is released in its protonated form, thereby hindering dramatically electron-transfer reactions [26]. At pH 7.5, only a small fraction of the reduced quinone will be in its anionic form, which could participate in electron-transfer processes. The two-phase time course (Fig. 2), apparently characteristic of DT-diaphorase, was consistently observed with un- and methyl-substituted 1,4-naphthoquinones, but it was absent with quinones with - 0 C H 3, - O H (regardless of their location in the quinone ring), or --SG substituents. In these instances, autoxidation proceeded rapidly and, within minutes, NADPH was completely exhausted, this correlating with a much higher production of H202. This effect is illustrated in Fig. 3B for the case of the methoxyl-substituted 1,4-naphthoquinone, which showed rates of N A D P H oxidation and He02 formation of 40.3 and 25.0 ~M × min -1, respectively. Although the E{Q-/Q 2-) values for these substituted 1,4-naphthoquinones are largely unavailable, it cannot be expected that they will deviate dramatically from the general trend observed with the un- and methyl-substituted 1,4naphthoquinones, that is, E{Q:/Q 2-) more positive than E(Q/Q-). Thus, 2-hydroxy1,4-naphthoquinone, with a two-electron reduction potential [E(Q/Qe-)] of - 154 mV [28] is poorly reduced by DT-diaphorase [4,23] in comparison to 2-methyl-l,4naphthoquinone [E(Q/Q 2-) = - 5 mV] [28]. However, the following autoxidation (relative to the rate of enzymic reduction) is much higher for the hydroxy-substituted naphthoquinone than for the methyl-substituted compound. This can not be explained in terms of large changes in the reduction potential of the intermediate step, for the E(Q:/Q 2-) value for 2-hydroxy-l,4-naphthoquinone is

lOO

I

5

I

0

I

10 Time (rain)

B

t

0

i

5

O

o

I

10

OCH3

~OCH3

0

7o

P_..

2: Q

140

15

[S~x~

a i s m ~ ] (riM)

I

I00

I

50

Fig. 3. Autoxidation of the semi- and hydroquinone forms of 2,3-dimethoxy-l,4-naphthoquinone following its reduction by NADPH~hytoehrome P~I50 reduetase and DT-diaphorase, respectively. Assay conditions: 20 gM 2,3-dimethoxy-l,4-naphthoquinone in air-saturated buffer containing 200 gM NADPH, was supplemented with 7.6 ~g NADPH~ytoehrome P-450 reduetase x ml -~ (traees A) or 45.2 ng DTatiaphorase x m1-1 (traces B) to initiate the reaetion. The buffers used for either enzyme are described in the Materials and Methods section and in Figs. 1 and 2. (. . . . ) and ( - - ) , NADPH oxidation in the absence and presence of superoxide dismutase (1 ~4), respectively. (O) and (e), H202 formation in the absence and presence of superoxide dismutase (1 gM), respectively.

E

L

200

A

K

"7

64 + 107 mV and, therefore, electron transfer to 02[E(0JO ~) = - 155 mV] could still be expected to be negligible. Inspection of Table II shows that with the enzyme concentrations used in these studies, the rate of quinone reduction and subsequent autoxidation was 10--300-fold higher for DT-diaphorase than for NADPH-cytochrome P-450 reductase (in terms of specific activity). This was in agreement with the higher rate of two-electron reduction observed with quinone-imines [24,30]. Although significant autoxidation is found in connection with both quinone reducing flavoproteins, NADPH-cytochrome P-450 reductase and DT-diaphorase, the effect of --CH s, --OCH 3, --SG and --OH substituents on the autoxidation rate differs for either enzyme. Further, it can be seen that there is no correlation between the rate of autoxidation of semi- or hydroquinones of the naphthoquinone series (Table II) and their reduction potential (Table I). However, the lack of correlation between reduction potential and the rate of electron transfer to 02 (reactions 1 and 2) may be only apparent, for the rate of autoxidation is obviously influenced by the rate of enzymic reduction of the quinone. In this regard, it should be considered that, although reduction potentials are a critical factor in this metabolism of quinones by one- or two-electron transfer flavoproteins, no statistical correlation was found between E(Q/Q-) or E(Q/Q 2-) of the quinone and its rate of reduction by NADPH-cytochrome P-450 reductase [23] or DT-diaphorase [4], respectively. It is interesting, however, that H202 formation was detected with quinones with a one-electron reduction potential more positive than - 155 mV, the reduction potential of the 02/0 ~couple at 1 M concentration [31].

II. Effect of superoxide dismutase on the autoxidation of semi- and hydronaphthoquinone derivatives The autoxidation of semi- (reaction} 1) or hydro- (reaction 2) naphthoquinones was affected in a different manner by superoxide dismutase. Effect of superoxide dismutase on the autoxidation of naphthosemiquinone derivatives. Inset in Fig. 1 illustrated the biphasic relationship connected to the NADPH-cytochrome P-450 reductase-dependent reduction of naphthoquinones; a first phase ascribed to the enzymic reduction of the quinone and a second slower phase corresponding to the autoxidation of the semiquinone during redox cycling. In the presence of superoxide dismutase, a single phase corresponding to a rapid N A D P H oxidation is observed. By relating the values in the presence of superoxide dismutase to those in its absence involving the first and second phase, the effect of dismutase on enzymic reduction rate and autoxidation rate can be described (Fig. 1, inset}. Thus, for the case of 1,4naphthoquinone, the initial (enzymic) rate of N A D P H oxidation (16.1 ~M × min -1) is enhanced to 28.5 ~M × min -1 in the presence of superoxide dismutase, whereas the slower rate following enzymic reduction (3.5 ~ M × rain -~) -ascribed to autoxidation -- is enhanced 8.1-fold by the enzyme. Figure 1 also shows the enhancement of H202 formation by superoxide dismutase. Table II lists the effects of superoxide dismutase on the first and second phase of quinone reduction and autoxidation and illustrated in the inset of Fig.

65 1. Overall, superoxide dismutase enhanced slightly (1.1--1.8-fold) the rate of NADPH oxidation associated with enzymic reduction, wherease it increased significantly (2.5--8.0-fold) the subsequent rate of autoxidation. An exception were the glutathionyl-substituted naphthoquinones without or with a - O H group in the benzene ring. In these instances, superoxide dismutase enhanced only slightly or had no effect on the rate of autoxidation. The autoxidation of naphthosemiquinones involves a single electron-transfer step (reaction 1). For the range of reduction potentials ( - 95 mV - - 415 mV) covered with the naphthoquinones examined here in air-saturated buffer, superoxide dismutase enhanced the autoxidation of the semiquinones. This means that for this reduction potential range and a t [02]initia I ~ 220 ~M, the equilibrium of Q- + 02 ~ Q + 05 almost invariably favours the formation of 05 • The enhancement of 1,4-naphthosemiquinone autoxidation can be interpreted as a rapid removal of 05 from the equilibrium reaction (reaction 1) by superoxide dismutase [15]. However, because the rate of NADPH-cytochrome P450 reductase-catalyzed reduction of quinones does not correlate primarily with the E(Q/Q-) of the quinones [24], the equilibrium of the autoxidation reaction (Q- + 03 ~ Q + 05) cannot be predicted and, consequently, how superoxide dismutase will affect this equilibrium. Thus, the equilibriuim constants, K = [Q-][02]/ [Q][O]], for 1,4-naphthoquinone and 5-hydroxy-l,4-naphthoquinone, are 1.8 [ 3 2 34] and 11.2 [21], respectively. Accordingly, the rate of autoxidation of the former should have been substantially higher than that of the latter; however, following the initial reduction by NADPH-cytochrome P-450 reductase, the rate of autoxidation of the latter was higher than that of the former (Table II). Effect of superoxide dismutase on autoxidation of naphthohydroquinone derivatives. The presence of superoxide dismutase had no effect on the initial rates of DT-diaphorase-catalyzed NADPH oxidation (data not shown). However, superoxide dismutase affected the following autoxidation in a distinct fashion which depended on the substitution pattern of the quinone (Table II): on the one hand, it prevented autoxidation of the unsubstituted 1,4-naphthoquinone as well as that of those derivatives with either - C H 3, - O C H 3, - S G , or - OH (in the quinoid ring) substituents. On the other, it stimulated autoxidation of 1,4-naphthohydroquinones with a --OH substituent in the benzene ring, regardless of the presence of other substituents such as - C H 3 and --SG. Figures 2 and 3B show the former effect, that is, inhibition of naphthohydroquinone autoxidation by superoxide dismutase, in terms of NADPH oxidation and H~O~ formation. 2,3-Dimethoxy-l,4-naphthoquinone showed the greatest decrease in NADPH oxidation in the presence of superoxide dismutase, from 40.3 ~M × min -1 to 3.1 pM × min -1, whereas H202formation was not detectable (Fig. 3B). The inhibition of the autoxidation of 1,4naphthoquinone, 2-methyl-l,4-naphthoquinone and 2,3-dimethyl-l,4-naphthoquinone by superoxide dismutase was difficult to assess accurately due to the intrinsically low rate of autoxidation. In general, the rates of NADPH oxidation decreased 3-12-fold when superoxide dismutase was included in the reaction mixture before the addition of DT-diaphorase, whereas H202 formation was in every case not detectable. When superoxide dismutase was added to an ongo-

66 ing autoxidation reaction (2 min after the addition of DT-diaphorase), the oxidation of NADPH was immediately halted for all quinones (data not shown), except those with a - - OH group in the benzene ring. The residual low rate of NADPH oxidation, on the one hand, and the lack of H202 formation, on the other, in the presence of superoxide dismutase, imply that a fraction of the metabolized quinone decays by pathway(s) other than electron transfer to 02. Inhibition of H202accumulation cannot be attributed to its decomposition by a route involving quinones or transition metals, because this was not observed during the autoxidation of semiquinones (Figs. 1, 3A) and the reaction was not affected by EDTA and desferrioxamine (not shown). The typical effect of superoxide dismutase on the redox status of end molecular products in the spent reaction mixture is illustrated in Fig. 4 for the case of 2,3-dimethoxy-l,4-naphthoquinone. On the other hand, samples withdrawn from the spent reaction mixture (as in the experimental conditions of Fig. 3B in the absence of superoxide dismutase), showed that most of the end molecular product was 2,3-dimethoxy-l,4-naphthoquinone, detected with reductive applied potential (Fig. 4A), thus supporting the oxidative character of the reaction. Analysis of the spent reaction mixture in the presence of superoxide dismutase showed that a large fraction of the product was in the reduced state, that is, 2,3-dimethoxy-l,4-naphthohydroquinone, analyzed with oxidative-electrochemical detection (Fig. 4B). In summary, superoxide dismutase altered the redox status of the end molecular products in the reaction mixture, i.e. the product distribution [hydroquinone]/[quinone], by favouring the accumulation of the hydroquinone. The inset in Fig. 3 shows the inhibition of 2,3-dimethoxy-l,4-naphthohydroquinone autoxidation by increasing concentrations of superoxide dismutase; the concentration of the enzyme required to produce half-maximal inhibition of autoxidation was 2.1 nM. There was not statistical correlation between the amount of superoxide dismutase required to produce half-maximal inhibition of autoxidation (Ki) and the extent of autoxidation observed. The concentrations of superoxide dismutase required to produce half-maximal inhibition of autoxidation varied between 1 and 6 nM (Table III), with the exception of 2-hydroxy1,4-naphthoquinone. In the latter case, the amounts of superoxide dismutase needed were considerably higher (0.5 ~M) and, therefore, probably without biological significance. This might be explained because of the poor reduction of this quinone by DT-diaphorase, probably due to its very negative reduction potential. As indicated previously [15], the catalytic amounts of superoxide dismutase required will depend on the position of the equilibrium of the autoxidation reaction (reactions 1 and 2). When the semiquinone/O~ equilibrium constant decreases, as in the case of those quinones which autoxidize rapidly, small amounts of dismutase are required; conversely, for the instances in which the semiquinone reaction is favoured, larger amounts of superoxide dismutase are required. However, similar to what discussed above for the case of NADPH-cytochrome P-450 reductase, there is no correlation between the twoelectron reduction potential of the quinones and their rate of reduction by DTdiaphorase [4]. Therefore, it is not possible to determine the position of the equilibrium in reaction 2 and, consequently, no correlation between the amount

67 100

~

nA

O

A

OCH3 OCH 3 O

5G

~x//"

I I I t I

25

I I

~

OCH 3

,i II

OCH 3

,I

OH

i I iI

B

OH

50 I

I

I

0

5 Time (min)

10

Fig. 4. H.P.L.C. with electrochemical detection analysis of quinone redox state during the autoxidation of 2,3-dimethoxy-l,4-naphthoquinone following its reduction by DT-diaphorase in the absence and presence of superoxide dismutase. Assay conditions as in Fig. 3B. Samples were taken at 10 rain incubation, gassed with argon for 3 min, and injected into the H.P.L.C. column. Injection volume = 20 ~1; total quinone injected = 0.4 nmol. Analysis was performed with (At reductive applied potential ( - 0.7 volts) and (B) oxidative applied potential (+ 0.7 volts) for the identification of the oxidized and reduced form of the quinone, respectively. ( - - ) and (. . . . ) Assays carried out in the absence and presence of 1 pM superoxide dismutase, respectively. Chromatograms for control experiments ( - - ) show a signal detected with reduetive applied potential and corresponding to 2,3-dimethoxy-l,4-naphthoquinone and a negligible amount of the reduced counterpart analyzed with oxidative applied potential. In the presence of superoxide dismutase, a decreased peak is detected with reduetive-applied potential, whereas a large fraction of the quinone is found in the reduced form (detected with oxidative-applied potential).

68 of superoxide dismutase required for half-maximal inhibition of autoxidation (Table III) and the equilibrium constant can be obtained. The inhibition of autoxidation of naphthohydroquinones by superoxide dismutase cannot be argued as being a non-specific protein effect on following accounts: (a) the concentration of superoxide dismutase required to inhibit autoxidation were within catalytic values, i.e. in the nM range. (b) Superoxide dismutase exerted a dual effect in similar model systems: on the one hand, its expected activity during the autoxidation of semiquinones and 5-hydroxy-l,4naphthohydroquinones and, on the other, an inhibition of autoxidation of naphthohydroquinones. Superoxide dismutase was not contaminated with cata-

TABLE

III

EFFECT OF SUPEROXIDE NAPHTHOHYDROQUINONE

DISMUTASE ON THE RATE AUTOXIDATION

O F H202 F O R M A T I O N

DURING

Assay conditions: reaction mixture contained 20 ~ M quinoid compound and 200 ~ M N A D P H in air-saturated 0.25 M sucrose/0.1 M potassium phosphate buffer, p H 7.55. The reaction was started upon addition of 45.2 ng DT-diaphorase × ml -I (except for 2-hydroxy-l,4-naphthoquinone: 0.45 ~g enzyme × ml-1).Samples were taken at different intervals and assayed for H202 formed according to the Materials and Methods section. W h e n present, superoxide dismutase concentration was 1 ~ M to achieve maximal effect. + d(H202)/dt is expressed in nmol × min -I × ~g DTdiaphorase -I. Quinone

+ d(H202)/dt Control

Ki(nM ) + SOD a

Unsubstituted 1,4-Naphthoquinone

20

n.d.

2.1

11 19

n.d. n.d.

6.2 2.9

553

n.d.

2.1

376 199

n.d. n.d.

5.2 0.9

17

n.d.

540.8

Methyl-substituted 2-Methyl-l,4-naphthoquinone 2,3-Dimethyl-l,4-naphthoquinone

Methoxyl-substituted 2,3-Dimethoxyl-l,4-naphthoquinone

Glutathionyl-substituted 3-Glutathionyl-lA-naphthoquinone 2-Methyl-3-glutathionyl-l,4-naphthoquinone

Hydroxyl-substituted/quinoid ring) 2-Hydroxy-l,4-naphthoquinone

Hydroxyl-substituted (benzene ring) 5-Hydroxy-l,4-naphthoquinone 2-Methyl-5-hydroxy-l,4-naphthoquinone 3-Glutathionyl-5-hydroxy-l,4-naphthoquinone 2-Methyl-3-glutathionyl-5-hydroxy-l,4-naphthoquinone •n.d., not detectable.

1172 1084 343 409

1617 1137 475 608

69 lase. (c) Inhibition of autoxidation is accompanied by a decrease in I-I202 formation (Figs. 2 and 3B) and by an increase in the [hydroquinone]/[quinone] ratio (Fig. 4). (d) Inhibition of autoxidation by superoxide dismutase was prevented by CN- (not shown). The dual effect of superoxide dismutase to either prevent the autoxidation of two-electron reduced quinoid compounds (Figs. 2 and 3B) or to increase HsO2 formation (Figs. 1 and 3A) argues against a direct inhibitory action of the quinones towards the superoxide dismutase activity. Furthermore, the apparent inhibition of superoxide dismutase by quinones [11,14] (observed at quinone concentrations higher than those in this study) cannot account for the suppression of H202 production observed. The superoxide dismutase-mediated enhancement of autoxidation of benzene-ring, -OH-substituted 1,4-naphthohydroquinones is illustrated at the bottom of Tables II and III. Both NADPH oxidation and H~O2 formation rates were enhanced by superoxide dismutase, in a fashion similar to that observed when the quinones were reduced by the one-electron transfer flavoprotein NADPH-cytochrome P-450 reductase. Another aspect of the 5-hydroxy-naphthoquinone derivatives is their high rate of autoxidation following their reduction by NADPH-cytochrome P-450 reductase or DT-diaphorase. This is rather surprising considering the reduction potential of, for example, 5-hydroxy-l,4naphthoquinone, which is more positive than that of the 02/05 couple. Mechanistic considerations on the effect of superoxide dismutase on the autoxidation of semi- and hydro-naphthoquinones. Autoxidation of hydroquinones proceeds as in the sequence Q2- ~ Q. ,~ Q illustrated in reaction 2. Both steps are comprised following the reduction of 1,4-naphthoquinones by DT-diaphorase, and the observed inhibition of hydroquinone autoxidation by superoxide dismutase may involve a change in the electron transfer represented at the intermediate step (Q2- ,~ Q.). This inhibition cannot be interpreted in terms of the disproportionation of semiquinones [14] catalyzed by the enzyme. If this activity were operative, no inhibition of NADPH oxidation would be expected, because the disproportionation of semiquinones (2Q- ~ Q + Q2-)would produce new substrate for the enzymic reaction linked to further NADPH oxidation. The superoxide dismutase-dependent inhibition cannot involve a disproportionation of oxygenated biradical complexes [(R0... 02)-] as proposed for certain polycyclic phenols [35], for this reaction proceeds with accumulation of the oxidized form of the phenol, formation of H202, and stimulation of 02 consumption 2[RO" ... 0~]" + 2H ÷-* R = 0 + RO- + H2O 2 -t- 0 2. The inhibitory effect of superoxide dismutase on naphthohydroquinone autoxidation can be explained in terms of a chain reaction in which 05 functions as the chain propagating species. Autoxidation of hydroquinones results in 05 production (reaction 2). The first electron-transfer step (Q2- ~.~ Q; ) is expected to proceed slowly due to the thermodynamic restrictions [E(Q-/Q2-) > E(0~/O~)] discussed above. Once initiated, 05 can promote efficiently hydroquinone autoxidation according to reaction 3, which, on thermodynamic considerations, is expected to proceed faster than the first step in reaction 2. 05 has been shown to act as an initiator for the autoxidation of catechols and ascorbic acid by a common mechanism via a sequential proton-

70 hydrogen transfer [36] analogus to that illustrated in reaction 3. OH

O" 4-02"

-~

P"

+ H202

R2

(3)

R2

R3 OH

R3 O-

The overall hydroquinone autoxidation can be summarized as in reactions 2 and 3, in which 0~ acts as the propagating species in the chain reaction: O~ is generated during semiquinone autoxidation (Q- + 02 ~- Q + 0~; reaction 2) and consumed during hydroquinone autoxidation (QH~ + 0~ ~ Q~ + H~0z; reaction 3). As a result of this, the steady-state concentration of 0~ in the system is low and its removal by superoxide dismutase suppresses the radical chain process, this being expressed as an inhibition of H202 and semiquinone formation and accumulation of hydroquinone. This interpretation is in agreement with the results shown in Figs. 2, 3B and 4 and in Tables II and III. Superoxide dismutase-dependent inhibition of autoxidation of leucoflavin [37], trihydroxybenzene [12], pyrogallol [38], dialuric acid [39,40], and 6-hydroxy-dopamine [41] has been previously reported. However, the enhancement of autoxidation of aromaticring, --OH-substituted naphthohydroquinones (Tables II and III, bottom part) by superoxide dismutase cannot be explained in terms of reactions 2 and 3. An alternative explanation which could account for the lack of H202 formation and the enhancement of the [hydroquinone]/[quinone] ratio by superoxide dismutase during naphthohydroquinone autoxidation could imply (a) the initial reduction of superoxide dismutase by 0~ (reaction 4) S O D - C u 2÷ + 0 ~ - - S O D - C u

÷ +

02

(4)

linked to the primary autoxidation step (reaction 5) Q2- + 02 ~-~ Q. + O~

(5)

and followed by (b) the oxidation of superoxide dismutase by the semiquinone (reaction 6). SOD-Cue÷ + Q- + 2H 2÷- SOD.Cu e- + QH2

(6)

Reaction 6 is apparently thermodynamically unlikely: considering the E(Cu2÷/Cu÷) value for superoxide dismutase of + 260 mV [42], the reduction of semiquinones at expense of O~ catalyzed by the enzyme would require E(Q~-/ Q2-) values more positive than that of the Cu2÷/Cu÷ couple, in order to compete efficiently with the reduction of O~ to H202 by the enzyme [E(O~,2H÷/ H202) --- + 940 mV] [42]. This thermodynamic restriction might be overcome

71 by considering that Cu in superoxide dismutase is not alternatively reduced and oxidized in the catalytic cycle, but (according to quantum mechanical simulations [43,44]) it forms a SOD-Cu2÷ --0~-complex that might further interact with the semiquinone to form a ternary complex, e.g. - Q - Cu2* - 0 ~ . In this context, the inhibition of adrenaline autoxidation by Cu,Zn-superoxide dismutase has been attributed to the formation of a ternary complex between the activated substrate and the enzyme [45]. At variance with the native enzyme, the superoxide dismutase activity of copper-tyrosine complexes accelerated the formation of adrenochrome, thus indicating the requirement of both protein and coordinated copper to account for inhibition of adrenaline autoxidation [45]. Preliminary experiments with Mn-superoxide dismutase (not shown) indicated that this enzyme (like Cu,Zn-superoxide dismutase) inhibited the autoxidation of the hydroquinones listed in Table II, except for those bearing a - O H substituent in the benzene ring. This finding would make unnecessary to invoke a distinct mechanism of superoxide dismutase as superoxide-semiquinone oxidoreductase to account for the observed inhibition of hydroquinone autoxidation. However, the enhancement of autoxidation of aromatic ring, -OH-substituted naphthoquinones by superoxide dismutase remains to be explained. Although it was proposed [14] that reaction 6 could take place with semiquino_nes with a very positive E(Q-/Q~-) value, no NADPH consumption due to autoxidation is observed in the reaction in which methyNp-benzohydroquinone [E(Q-/Q2-) = + 437 mV] [25] is formed during DT-diaphorase catalysis, because the forward reaction rate in equation 5 is negligible small. The use of superoxide dismutase (up to a concentration of 2 ~M) to eliminate the backward reaction, did not affect the minute rate of autoxidation (not shown). The lack of autoxidation of 2-methyl-p-benzohydroquinone is in agreement with the observation that this quinone is toxic to isolated hepatocytes by means of a mechanism other than oxidative stress, e.g. arylation of cellular thiols [46].

CONCLUDINGREMARKS This study has shown some aspects of the effect of superoxide dismutase on the autoxidation of semi- and hydroquinones derivatives of the 1,4-naphthoquinone series. The following conclusions can be drawn: (1) The autoxidation of 1,4-naphthosemiquinones, following the one-electron reduction of the oxidized counterpart by NADPH-cytochrome P-450 reductase, is enhanced by superoxide dismutase. This is understood as a consequence of the rapid removal of O~ by the enzyme, hence displacing the equilibrium of the autoxidation reaction (reaction 1) towards the right. This interpretation has been offered by Winterbourn [15] (though applying to a different experimental model involving the one-electron reduction of different quinones by xanthine oxidase) and analyzed by Wardman and Wilson [47] in a pulse radiolysis model using low 02 concentrations. (2) The autoxidation of 1,4-naphthohydroquinones formed during DT-dia-

72 phorase catalysis is inhibited by superoxide dismutase. This applies to 1,4naphthoquinone derivatives bearing no substituent or with --CH 3, -OCH~, - - SG, or - HO substituents (the latter in the quinoid ring). An exception is constituted by 5-hydroxy-l,4-naphthoquinones, i.e. with a - - O H substituent in the benzene ring, whose autoxidation is enhanced by superoxide dismutase, resembling that observed during NADPH-cytochrome P-450 reductase catalysis. The less pronounced autoxidation of un- and methyl-substituted naphthohydroquinones (reaction 2) than that of naphthosemiquinones (reaction 1), following DT-diaphorase and NADPH-cytochrome P-450 reductase catalysis, respectively, has already been reported in complex systems as rat liver microsomes and perfused organs [2,3,27,48,49]. Studies on the toxicity to isolated hepatocytes of two structurally related naphthoquinones, 5-hydroxy-l,4naphthoquinone and 2-hydroxy-l,4-naphthoquinone, indicated that the former (by a mechanism involving arylation of cellular thiols and redox cycling) was much more cytoxic than the latter [22]. The present investigations are in agreement with these studies, for 5-hydroxy-l,4-naphthoquinone was reduced effectively by both one- and two electron transfer flavoproteins and no apparent protection can be expected to originate from its catalysis by DT-diaphorase, whether connected to a superoxide dismutase activity or not (Tables II and III). Conversely, 2-hydroxy-l,4-naphthoquinone, which was much less toxic to isolated hepatocytes [22] is poorly reduced by either electron-transfer flavoprotein, probably due to its very negative reduction potential. The tentative mechanism(s) for inhibition of naphthohydroquinone autoxidation by superoxide dismutase can be analyzed on similar grounds as those proposed by Willson [50] for the removal of peroxyl radicals by the enzyme. These mechanisms consider (a) a free radical scavenging effect, (b} a rapid removal by superoxide dismutase of O~ from a radical chain reaction, and (c) a direct catalytic interaction with copper in the enzyme. The first possibility is untenable, for inhibition of autoxidation by superoxide dismutase was prevented by CN- and albumin was present in the assay reaction mixture at concentrations higher than superoxide dismutase. The second possibility considers O~ as a propagating species in the chain reaction [40,41] (summarized in reactions 2 and 3) and can explain the observed inhibition of naphthohydroquinone autoxidation by superoxide dismutase, mainly in terms of inhibition of HeO2 formation (Figs. 2 and 2B and Tables II and III) and accumulation of the reduced form of the quinone (Fig. 4). However, this mechanism cannot explain the superoxide dismutase-mediated enhancement of autoxidation of aromatic-ring, -OH-substituted naphthobydroquinones. The third possibility, a direct catalytic interaction of superoxide dismutase with the radicals formed during naphthohydroquinone autoxidation, differs from the disproportionation of semiquinones [14] or peroxyl radicals [50] previously suggested. The mechanism illustrated in reactions 4 and 6 is a mixed function of superoxide dismutase, whereby the Cu in the enzyme is reduced by O~ (as in the classical reaction) and oxidized by a semiquinone. The observed inhibition of hydroquinone autoxidation by Mn-superoxide dismutase, however, would make unrealistic an interaction of the semiquinone species with superoxide dismutase.

73

SOD: Enhancement of semiquinone autoxidation , ~,

2"-

02"Rt

o-

o2"-

Rt

o-

o2

o2"-

o

o2

SOD: Inhibition of hydroquinone autoxidatice

Fig. 5. Effect of superoxide dismutase on semi- and hydro-naphthoquinone autoxidation as formed during NADPH~ytoehromeP-450 reductase and DT-diaphorase catalysis, respectively. Overall, the different effect of superoxide dismutase on semi- and hydronaphthoquinone autoxidation seems to indicate activities of a prooxidant and antioxidant character, respectively. In the former case, superoxide dismutase enhances semiquinone autoxidation by displacing the equilibrium of the autoxidation reaction towards the right; in the latter case, interruption of a chain reaction in which O~ functions as a chain propagating species could account for the inhibition of hydroquinone autoxidation observed here, except for the arc~ matic-ring, H0-substituted naphthoquinones. These relationships are summarized in Fig. 5. The present studies indicate a complimentary activity of DT-diaphorase and superoxide dismutase against oxidative stress caused by certain quinones. The inhibition of H2O2 production resulting from the concerted mechanism of both enzymes (at variance with the sequence NADPH-cytochrome/)-450 reductasesuperoxide dismutase) is of main importance, for H202, if not intercepted by glutathione peroxidase or catalase and due to its facility to diffuse away from its site of formation, can enter a chain decomposition process upon its reduction to HO. by transition metals. It might be of interest to evaluate whether the cellular induction of DT-diaphorase exerted by Sudan III and other xenobiotics [51] is presently accompanied by a parallel induction of superoxide dismutase, thereby (according to the present studies) offering full protection against quinone-mediated oxidative stress. It could be hypothesized that such a parallel induction indeed would take place, considering that some of the xenobiotics that are known to induce DT-diaphorase undergo intracellular redox cycling with formation of O:9 as primary species. ACKNOWLEDGEMENTS This work was supported by Grant 2703-B89-01XA from the Swedish Cancer Foundation and Grants 7679 and 4481 from Swedish Medical Research Council.

74 REFERENCES 1

2

3 4

5

6

7 8 9 10 11 12

13 14 15 16

17 18 19

20 21 22

T.W. Gant, D.N. Ramakrishna Rao, R.P. Mason and G.M. Cohen, Redox cycling and sulphydryl arylation; their relative importance in the mechanism of quinone cytotoxicity to isolated hepatocytes, Chem.-Biol. Interact., 65 (1988) 157-173. H. Wefers and H. Sies, Hepatic low-level chemiluminescence during cycling of menadione and the menadione-glutathione conjugate. Relation to glutathione and NAD(P)H :quinone reduction (DT-diaphorase) activity, Arch. Biochem. Biophys., 224 (1983) 568-578. H. Wefers and H. Sies, Generation of photoemissive species during quinone redox cycling, Biochem. Pharmacol., 35 (1986) 22-24. G.D. Buffinton, K. Ollinger, A. Brunmark and E. Cadenas, DT-diaphorase-catalized reduction of 1,4-naphthoquinone derivatives and glutathionyl-quinone conjugates. Effect of substituents on autoxidation rates, Biochem. J., 257 (1989) 561-571. T.W. Gant, M. d'Arcy Doherty, D. Odowole, K.D. Sales and G.M. Cohen, Semiquinone anion radicals formed by the reaction of quinones with glutathione or amino acids, FEBS Lett., 201 (1986) 296-300. I. Wilson, P. Wardman, T.-S. Lin and A.C. Sartorelli, One-electron reduction of 2- and 6methyl-l,4-naphthoquinone bioreductive alkylating agents, J. Med. Chem., 29 (1986) 13811384. S. Orrenius, Oxidative stress studied in intact mammalian cells, Phil. Trans. R. Soc. Lond., B311 (1985) 673-677. M.G. Miller, A. Rodgers and G.M. Cohen, Mechanisms of toxicity of naphthoquinones to isolated hepatocytes, Biochem. Pharmacol., 35 (1986) 1177-1184. G. Bellomo, H. Thor, C. EklSw-L~stbom, P. Nicotera and S. Orrenius, Oxidative stress mechanisms of cytotoxicity. Chem. Scr., 27A (1987) 117-120. H.M. Hassan and I. Fridovich, Intracellular production of superoxide radicals and of hydrogen peroxide by redox active compounds, Arch. Biochem. Biophys., 196 (1979) 385-395. M.T. Smith and C.G. Evans, Inhibitory effect of superoxide-generating quinones on superoxide dismutase, Biochem. Pharmacol., 33 (1984) 3109--3110. W.F. Greenlee, J.D. Sun and J.S. Bus, A proposed mechanism of benzene toxicity: formation of reactive intermediates from polyphenol metabolites, Toxicol. Appl. Pharmacol., 59 (1981) 187-- 195. I. Fridovich, Oxygen radicals, hydrogen peroxide and oxygen toxicity, in: W.A. Pryor (Ed.), Free Radicals in Biology and Medicine, Academic Press, New York, 1976, pp. 239--275. J. Butler and B.M. Hoey, The apparent inhibition of superoxide dismutase activity by quinones, Free Radicals Biol. Med., 2 (1986) 77--81. C.C. Winterbourn, Cytochrome c reduction by semiquinone radicals can be directly inhibited by superoxide dismutase, Arch. Biochem. Biophys., 209 (1981) 159--167. B. HSjeberg, K. Blomberg, S. Stenberg and C. Lind, Biospecific adsorption of hepatic DTdiaphorase on immobilized dicoumarol. I. Purification of cytosolic DT-diaphorase from control and 3-methyl-cholanthrene-treated rats, Arch. Biochem. Biophys., 207 (1981) 205--216. L.L. Fan and B.S.S. Masters, Properties of purified kidney microsomal NADPH-cytochrome c reductase, Arch. Biochem. Biophys., 165 (1974) 665--671. J.M. McCord and I. Fridovich, Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein), J. Bol. Chem., 244 (1969) 6049-6055. D.J. Danner, P.J. Brignac, D. Arceneaux and V. Patel, The oxidation of phenol and its reaction product by horseradish peroxidase and hydrogen peroxidase, Arch. Biochem. Biophys., 156 (1973) 759--763. W.D. Hinsberg, K.H. Milby and R.N. Zare, Determination of insulin in serum by enzyme immunoassay with fluorimetric detection, Anal. Chem., 53 (1983) 1509-1512. T. Mukherjee, One-electron reduction of juglone (5-hydroxy-l,4-naphthoquinone): a pulse radiolysis study. Radiat. Phys. Chem., 29 (1987) 455-462. M. d'Arcy Doherty, A. Rodgers and G.M. Cohen, Mechanisms of toxicity of 2- and 5hydroxy-l,4-naphthoquinone; absence of a role for redox cycling in the toxicity of 2-hydroxy1,4-naphthoquinone to isolated hepatocytes, J. Appl. Toxicol., 7 (1987) 123--129.

75 23

24 25 26

27

28 29 30

31 32 33 34 35

36

37

38

E. Cadenas, D. Mira, A Brunmark, C. Lind, J. Segura-Aguilar and L. Ernster, Effect of superoxide dismutase on the autoxidation of various hydroquinones -- a possible role of superoxide dismutase as superoxide : semiquinone oxidoreductase, Free Radical Biol. Med., 5 (1988) 71--79. G. Powis and P.L. Appel, Relationship of the single-electron reduction potential of quinones to their reduction by flavoproteins, Biochem. Pharmacol., 29 (1980) 2567-2572. A.J. Swallow, Physical chemistry of semiquinones, in: B.L. Trumpower (Ed.), Function of Quinones in Energy Conserving Systems, Academic Press, London, 1982, pp. 59-72. S. Stenken, Oxidation of phenolates and phenylenediamines by 2-alkanonyl radicals produced from 1,2-dihydroxy- and 1-hydroxy-2-alkoxylalkyl radicals. J. Phys. Chem., 83 (1979) 595-- 599. C. Lind, P. Hochstein and L. Ernster, DT-Diaphorase as a quinone reductase: a cellular control device against semiquinone and superoxide radical formation, Arch. Biochem. Biophys., 216 (1982) 178--185. R.A. Morton, Introductory account of quinones, in: R.A. Morton (Ed.), Biochemistry of Quinones, Academic Press, London, 1965, pp. 1 - 22. T. Iyanagi, On the mechanisms of one- and two-electron transfer by flavin enzymes, Chem. Scr., 27A (1987) 3 1 - 3 6 . G. Powis, K.L. See, K.S. Santone, D.C. Melder and E.M. Hodnett, Quinoneimines as substrates for quinone reductase (NAD(P)H: (quinone-acceptor) oxidoreductase) and the effect of dicumarol on their cytotoxicity, Biocbem. Pharmacol., 36 (1987) 2473-2479. Y.A. Ilan, D. Meisel and G. Czapski, The redox potential of the OJO- system in aqueous solution, Israel J. Chem., 12 (1974) 891-895. K.B. Patel and R.L. Willson, Semiquinone free radicals and oxygen. Pulse radiolysis study of one-electron gransfer equilibria. J. Chem. Soc. Faraday 1 69 (1973) 8 1 4 - 825. D. Meisel, Free energy correlation of rate constants for electron transfer between organic systems in aqueous solutions, Chem.-Phys. Lett., 34 (1975) 263-266. D. Meisel and G. Czapski, One-electron transfer equilibria and redox potentials of radicals studied by pulse radiolysis, J. Phys. Chem., 79 (1975) 1503-1509. C. Auclair, Polycyclic phenols in alkaline solution: stable substrates for superoxide dismutase? in: J.V. Bannister and H.A.O. Hill (Eds.), Chemical and Biochemical Aspects of Superoxide and Superoxide Dismutase, Elsevier-North Holland, New York, 1980, Vol. l l A , pp. 318-327. D.T. Sawyer, T.S. Calderwood, C.L. Johlman and C.L. Wilkins, Oxidation by superoxide anion of catechols, ascorbic acid, dihydrophenazine, and reduced flavins to their respective anion radicals. A common mechanism with a sequential proton-hydrogen atom transfer, J. Org. Chem., 50 (1985) 1409-1412. D. Ballou, G. Palmer and V. Massey, Direct demonstration of superoxide anion production during the oxidation of reduced flavin and of its catalytic decomposition by erythrocuprein, Biochem. Biophys. Res. Commun., 36 (1969) 898-904. S. Marklund and G. Marklund, Involvement of superoxide anion radical in autoxidation of pyrogallol and a convenient assay for superoxide dismutase, Eur. J. Biochem., 47 (1974) 469 -

39 40

41

42 43

474.

R Munday, Dialuric acid autoxidation: effects of transition metals on the reaction rate and on the generation of 'active oxygen' species, Biochem. Pharmacol., 37 (1987) 409-413. C.C. Winterbourn, W.B. Cowden and H.C. Sutton, Auto-oxidation of dialuric acid, divicine and isouramil. Superoxide dependent and independent mechanisms, Biochem. Pharmacol., 38 (1989) 611 - 618. P. Gee and A.J. Davison, Intermediates in the aerobic autoxidation of 6-hydroxydopamine: relative importance under different reaction conditions, Free Radicals Biol. Med., 6 (1989) 271 -- 284. W.C. Barette, D.T. Sawyer, J.A. Fee and K. Asada, Potentiometric titrations and oxidationreduction potentials of several iron superoxide dismutase, Biochemistry, 22 (1983) 624--627. M. Rosi, A. Sgamellotti, F. Tarantelli, I. Bertini and C. Luchinat, A theoretical investigation of the copper-superoxide system. A model for the mechanism of copper-zinc superoxide dismutase, Inorgan. Chim. Acta, 107 (1985) L 2 1 - L22.

76 44 45

46 47

48

49

50 51

R. Osman, Effect of local environment and protein on the mechanism of action of superoxide dismutase, Enzyme, 36 (1986) 3 2 - 4 3 . L.M. Shubotz, M. Younes and U. Weser, Transient reaction of 2Cu,2Zn-superoxide dismutase with a superoxide generating system, in: J.V. Bannister and H.A.O. Hill (Eds.), Chemical and Biochemical Aspects of Superoxide and Superoxide Dismutase, Elsevier/NorthHolland, Amsterdam, 1980, Vol. 11A, pp. 328-335. L. Rossi, G.A. Moore, S. Orrenius and P.J. O'Brien, Quinone toxicity in hepatocytes without oxidative stress, Arch. Biochem. Biophys., 251 (1986) 25--35. P. Wardman and I. Wilson, Control of the generation and reactions of free radicals in biological systems by kinetic and thermodynamic factors, Free Radical Res. Commun., 2 (1987) 2 2 5 - 232. H. Wefers, T. Komai, P. Talalay and H. Sies, Protection against reactive oxygen species by NAD(P)H : quinone reductase induced by the dietary antioxidant butylated hydroxyanisole (BHA), FEBS Lett., 169 (1984) 63--66. H.J. Prochaska, P. Talalay and H. Sies, Direct protective effect of NAD(P)H : quinone reductase against menadione-induced chemiluminescence of postmitochondrial fractions of mouse liver, J. Biol. Chem., 262 (1987) 1391--1394. R.L. Willson, Organic peroxy radicals as ultimate agents in oxygen toxicity, in: H. Sies (Ed.), Oxidative Stress, Academic Press, London, 1985, pp. 41-72. A. Atallah, J.R. Randolph and P. Hochstin, P., DT-Diaphorase and quinone detoxification, Chem. Scr., 27A (1987) 141--144.