BRAIN

RESEARCH
ELSEVIER Brain Research 725 (1996) 207-216

Research report

The superoxide production mediated by the redox cycling of xenobiotics in

rat brain microsomes is dependent on their reduction potential
Marie-Hel~ne Livertoux, Philippe Lagrange, Alain Minn *

CNRS URA No. 597, Centre du M£dicament, 30 rue Lionnois, F-54000 Nancy, France

Accepted 13 February 1996

Abstract

Several exogenous molecules undergo enzymatic one-electron reduction leading to radicals which can rapidly react with molecular
oxygen to form superoxide anions. We have previously shown that under aerobic conditions a significant superoxide anion production
occurred during the NADPH-dependent one-electron reduction of some drugs and xenobiotics by rat brain preparations. We report here
for several compounds a fairly good correlation between the reduction potentials (Epc vs. SCE) which ranged between - 230 and - 700
mV in aqueous medium (pH 7.4) or between - 700 mV and - 1100 mV in the aprotic solvent N,N-dimethylformamide, and the rate of
superoxide anion production during their metabolism by rat brain microsomes. The data obtained suggest that the redox potential of most
of the molecules assayed was related to their ability to undergo one-electron reduction mediated by flavoenzymes in the rat brain. The
main range of reduction potentials corresponding to a large superoxide anion production suggests that the redox cycling of these
chemicals was mediated by NADPH-cytochrome P-450 reductase. Therefore the measurement of reduction potentials of drugs and
xenobiotics able to reach the brain, and chemically related to quinones, nitroaromatics, nitroheterocyclics and iminiums, may provide
information both on their electron affinity and the possibility of one-electron transfer in vivo, and thus on their possible neurotoxicity due
to the production of oxygenated free radicals.

Keywords: Reduction potential; Electron affinity; Rat brain microsomes; Redox cycling of xenobiotics; Superoxide radical; NADPH-cytochrome P-450
reductase; Quinone; Nitroheterocycle;Bipyridinium;Brain ageing; Neurodegeneration

1. Introduction mental pollutants, resulting from the formation of reactive

It has been well established that the one-electron reduc-
tion of various aromatic nitro compounds and quinones to
free radicals is an essential step in their anti-infectious
[34], radiosensitizer [36] or antitumoral [4,35,39,41] mode
of action. The therapeutic effects of these drugs appear to
be mediated by toxic metabolites produced during their
enzymatic metabolism. As a consequence, several adverse
effects have been associated either with the intake of these
drugs or with an exposure to chemically related environ-
Abbreviations:DMSO, dimethylsulfoxide;DMF, N,N-dimethylform-
amide; MPP+/MPTP, N-methyl-4-phenyl-pyridinium/N-methyl-4-
phenyl-1,2,3,6-tetrahydropyridine;NHE, normalhydrogenelectrode; SCE,
saturated calomel electrode; Epc,7, cathodic peak reduction potentials in
phosphate (pH 7.0) buffer vs. SCE; Epc,DMF, cathodic peak reduction
potentials in anhydrous DMF vs. SCE; E~, single-electron reduction
potentials at pH 7.0 vs. NHE; Era,7, mid-pointpotentialof enzyme at pH
7.0 vs. NHE
Corresponding author. Fax: (33) (83) 32 13 22. E-maih
minn@ctrmd.u-nancy.fr
metabolites a n d / o r activated forms of oxygen. Despite
their toxicity, oxygen derived free radicals occur continu-
ously in aerobic cerebral metabolism. For instance, produc-
tion of hydrogen peroxide by dopamine desamination and
autooxidation has been proposed as one of the mechanisms
leading to the age-related depletion of dopaminergic neu-
rons in the substantia nigra [9,16]. Moreover, a deficiency
of the scavenger systems associated with an excess of
radical production should be responsible for an accelerated
destruction of these neurons, resulting in the appearance of
Parkinson's disease [7,53]. Accordingly, toxics that are
known to cause symptoms resembling Parkinson's disease
increase the production of free radicals. These include
6-hydroxydopamine [12], manganese [11], meperidine [7]
and paraquat [31], all responsible for a massive superoxide
production resulting in nigrostriatal damage [52].
Therefore, some of the neurotoxic effects resulting from
a chronic intake of compounds or drugs may be mediated,
at least in part, through the one-electron reduction to a
radical, a n d / o r through the subsequent formation of super-

0006-8993/96/$15.00 Copyright © 1996 Elsevier Science B.V. All rights reserved.

PI1 S0006-8993(96)00251-X
208 M.-H. Lit'ertoux et al. / Brain Research 725 (1996) 207-216
oxide anion radicals (O~-) during their reductive
metabolism. The reduction of some of these molecules to
radicals may be enzymatically catalyzed by brain
NADPH-cytochrome P-450 reductase (EC 1.6.2.4) and
possibly by some other reductases, the resulting radical
transferring in turn its unpaired electron to molecular
oxygen, thus forming superoxide anion [15]. The rates of
reduction by purified NADPH-cytochrome P-450 reduc-
tase have been correlated with the single-electron reduc-
tion potentials E ~ ( X / X - ) of quinones [37] and nitro
compounds [6] for molecules having single-reduction po-
tentials E~ ranging between - 4 0 0 mV and - 1 6 5 mV
[6,38]. In any case, O; production by redox cycling of
quinones and nitro compounds is maximal at an E~ value
of approx. - 7 0 mV [6,8,37].
Accordingly, various studies have shown that the rates
of reduction of a series of quinones [6,37], nitroheterocy-
cles [34] and iminium derivatives [31] by isolated cells or
mammalian liver microsomes, are related to their electron
affinity which can be evaluated from the reduction poten-
tials of these molecules. For the reduction of quinones and
nitro compounds by hepatic NADPH-cytochrome P-450
reductase, linear [6] or bell-shape dependences [8,38] of
logarithms of reaction rate o r Vmax/K m on the one-elec-
tron reduction potentials (E~) of compounds/drugs were
observed. Therefore, the ease of reduction is an important
factor for reactivity, and for some drugs chemically related
to the three chemical structures or precursors a correlation
has been described between the reduction potentials and
the pharmacological efficiency [22] or toxicity [35,36].
Thus the reduction potentials of these molecules should
provide information on the single-electron transfer possi-
bility in vivo, as the most cytotoxic molecules, benzo-
quinones and naphthoquinones, possess the highest redox
potentials, presumably because their higher electrophilicity
and reactivity towards thiol groups [39].
We previously showed that the NADPH-dependent pro-
duction of superoxide during the biotransformation of some
drugs was largely associated with the rat brain microsomal
fraction. Superoxide production was also observed with
microvessels and choroid plexus homogenates incubated
with the drugs under aerobic conditions [15,23]. Some of
the molecules we used in this study are benzodiazepines,
therefore drugs devoted to the brain, showing frequently
Parkinson's disease-like side effects when administered for
long periods of time (usually several years). All agents
used in this study may be transported by the blood flow
following an intoxication, therefore promote an oxidative
stress to cerebrovascular endothelial cells. This oxidative
stress results in an increase in passive permeability of the
blood-brain barrier ([33]; Lagrange, Romero, Minn and
Revest, submitted). Therefore, even if some of these
molecules initially poorly penetrate the brain, their influx
should increase as a consequence of an oxidative insult to
the blood-brain barrier, and, therefore, they can exert their
toxic effects on reaching neurons. The purpose of the
present investigation was to characterize the relationships
between the reduction potentials of some quinones, nitro-
and bipyridinium derivatives measured by cyclic voltam-
metry and the rate of superoxide anion radical formation
by microsomal preparations of rat brain.
2. Materials and methods
2.1. Drugs and chemicals
A series of compounds was selected from their reduc-
tion potentials and tested for their ability to produce
superoxide anion in microsomal fractions isolated from rat
brain. 2-Methyl-l,4-naphthoquinone (menadione), an-
thraquinone-2-sulfonate, potassium salt and 3-nitrobenzoic
acid were obtained from Aldrich Chemie (l'Isle D'Abeau,
France), and 2,5-dimethyl-l,4-benzoquinone and naphtho-
quinone-2-sulfonate, sodium salt, from Eastman Kodak
(Rochester, NY). 1,4-Benzoquinone and 2-methyl-l,4-
benzoquinone were purchased from Merck (Darmstadt,
Germany). MPP + was purchased from Research Biochem-
ical (RBI, Natick, MI), and diquat dibromide and indigo-
5,5-disulfonic acid, disodium salt (indigo carmine) were
obtained from Riedel-de-HaEn (Friedrichsthal, Germany).
1-Tetramethyl-l,4-benzoquinone (duroquinone), mito-
mycin C, 1,3-dinitrobenzene, nitrofurazone, metronidazole,
methyl and benzyl viologen were obtained from Sigma
Chemical (St. Louis, MO). Flunitrazepam and nitrazepam
were gifts from Roche Laboratories (Neuilly-sur-Seine,
France). Nimetazepam (1-methyl-nitrazepam) and nirida-
zole (1-(5-nitro-2-thiazolyl)-2-imidazolidinone) were gifts
from Sumitomo Laboratories (Osaka, Japan) and Ciba
Geigy (Basel, Switzerland), respectively. Cytochrome c,
NADPH and NADH were obtained from Boehringer
Mannheim (Meylan, France). Diphenyliodonium chloride
was obtained from Aldrich Chemie (l'isle D'Abeau,
France). All other reagents were of the highest purity
available commercially.
Acetylated ferricytochrome c was prepared as de-
scribed by Azzi et al. [3] using acetic anhydride in half-
saturated sodium acetate. The percentage of non-acetylated
cytochrome c in the final preparation was less than 10%.
2.2. Preparation o f rat brain subcellular fractions
Male Sprague-Dawley rats, weighing 200-250 g, ob-
tained from Iffa-Credo (St-Germain sur 1' Arbresle, France),
were killed by decapitation and the brains were rapidly
removed, dissected on ice and homogenized using a Dounce
glass hand homogenizer. The subcellular fractions were
obtained as described previously [14]. The purity of micro-
somal tractions was checked by measurement of activities
of NADPH-cytochrome P-450 reductase, a marker of en-
doplasmic reticulum [44] and NADH oxidase, a mitochon-
drial marker [32]. The contamination of microsomal frac-
 M.-H. Livertoux et aL/ Brain Research 725 (1996) 207-216 209

tions by mitochondrial particles was less than 8% in all
cases. Protein concentration was determined by the method
of Lowry et al. [28], using bovine serum albumin as a
standard.
2.3. Measurement of superoxide production
The rate of superoxide radical formation during micro-
somal NADPH-dependent redox cycling of the assayed
xenobiotics was followed by recording the reduction at
37°C of acetylated ferricytochrome c at 550 n m on a
Kontron Uvikon 820 spectrophotometer. The air-saturated
incubation medium contained 0.1 M Tris-HC1 (pH 7.40)
buffer (1 ml), 4 m M NADPH, 60 ~xM acetylated ferricy-
lochrome c and microsomal fraction ( 4 0 - 6 0 ~ g of pro-
tein). The cofactor was omitted in the reference cuvette.
Possible interferences such as basal endogenous superox-
ide anion production by other enzymatic systems, or a
direct reaction of the substrate with non-enzymatic compo-
nents, were prevented by the use of double-beam mode
spectrophotometry, and by a 3 min control absorbance
measurement which was subtracted from the assay mea-
surement. The reaction was started by the addition of the
drug (0.5 mM) in both cuvettes. The absorbance changes
allowed us to calculate the superoxide production, assum-
ing an extinction coefficient difference between the oxi-
dized and reduced forms of acetylated ferricytochrome c
of 19.6 m M - I cm -1 [3].
The xenobiotics (50 mM) were dissolved in N,N-di-
methylformamide (DMF), except for viologens, indigo
carmine and metronidazole, which were dissolved in 0.1 M
Tris-HC1 (pH 7.40) buffer. This concentration was higher
than the apparent K m for rat brain microsomal reductases
under aerobic conditions [23]. The amount of DMF (10 ~1)
added in each assay had no significant effect on NADPH
oxidation and acetylated ferricytochrome c reduction.
2.4. Involvement of NADPH-cytochrome P-450 reductase
in the formation of superoxide radicals
An inhibition of the activity of NADPH-cytochrome
P-450 reductase was obtained by addition of diphenyliodo-
nium chloride (10 m M stock solution in 10% DMSO) [47]
at a final concentration of 0.5 m M in samples prepared by
preincubating 60 p~M acetylated ferricytochrome c, micro-
somal fraction ( 4 0 - 6 0 ~ g of proteins) at 37°C in 0.1 M
Tris-HC1 (pH 7.40) buffer. The inhibitor (0.25 mM) was
added after 3 rain and the reaction was started by the
addition of N A D P H solution to give a final concentration
of 4 mM.

Table 1
Rates of reductionof acetylated ferricytochrome c (AcFe3+cyt c) reductionduring redox cycling of 0.5 mM quinones,nitroaromaticsand viologensby rat
brain microsomes, and apparent kinetic constants
Number Xenobiotics Rates of AcFe3+ cyt c reduction K m Vmax/K m
(nmol/min/mg protein) (mM)
Quinones
1 1,4-Benzoquinone ND
2 2-Methyl-1,4-benzoquinone ND
3 2,5-Dimethyl-1,4-benzoquinone 17.14 + 4.26 0.00812 4045.57
4 Naphthoquinone-2-sulfonate 49.66 + 4.92 a
5 Menadione 25.60 + 7.19 0.00415 5910.84
6 MitomycinC 13.51 0.252 11.15
7 Duroquinone 19.31 + 2.48 0.029 1115.02
8 Indigo carmine 12.73 + 1.42 0.105 108.48
9 Anthraquinone-2-sulfonate 10.47 + 0.91 0.983 26.55
Nitrocompounds
10 1,3-Dinitrobenzene 8.17 + 0.62 1.190 25.02
1l Nitrobenzoic acid ND
12 Nitrofurazone 22.01 + 5.09 0.296 91.32
13 Nitrazepam 0.36 _+0.25
14 Flunitrazepam 1.11 + 0.35 1.530 3.00
15 Nimetazepam ND
16 Niridazole 1.69 _+0.22 1.970 4.74
17 Metronidazole ND
Viologens
18 Benzyl viologen 23.55 + 0.70 0.025 1202.00
19 Diquat 19.81 _+ 1.70 0.028 1537.86
20 Methyl viologen 22.76 _+0.80 0.156 178.46
The resulting kinetic parameters correspond to the reciprocal intercepts and slopes of the Lineweaver-Burk plots. They are the means + S.D. of individual
experiments (n = 3 probes).
ND, not detectable.
Non-enzymaticreduction of acetylated ferricytochrome c.
210 M.-H. Livertoux et al. / Brain Research 725 (1996) 207-216

2.5. Determination of reduction potentials
The electron affinity of the xenobiotics used in this
study was established by cyclic voltammetric measure-
ments of their reduction potentials. This method allows the
calculation of the cathodic (Epc) reduction potentials.
Observed reduction potentials Epc were measured at
room temperature with a PAR model 174A polarographic
analyzer (EGG Instruments, Princeton, USA) piloted by a
VA scanner E612 (Metrohm, Herisau, Switzerland) and
associated with a Linseis LY 1810 model X-Y recorder.
molecules assayed at the concentration of 50 IxM, duro-
quinone, 2,5-dimethyl-1,4-benzoquinone, nitrofurazone and
the three viologens showed a relatively high rate of O~
production, whereas mitomycin C, indigo carmine, an-
thraquinone-2-sulfonate and 1,3-dinitrobenzene, were less
efficient. Rates of 0 2- production in the presence of
niridazole and flunitrazepam were approx. 10-fold lower
A
Menadione

Scan rates generally ranged from 50 to 200 m V / s . A

glassy carbon electrode (Tacussel, Villeurbanne, France)
was the working electrode, the reference electrode being a
saturated calomel electrode (SCE). The reduction poten-
tials were measured in both solvents, anhydrous DMF and 1°°° ~ n t i a l (mV)
aqueous 0.1 M phosphate (pH 7.0) buffered solution. The
electrolyte used in the unbuffered aprotic solvent was 0.1
M tetraethylammonium perchlorate (NEt4C104). Solutions
were deoxygenated by bubbling argon for 15 min. All
measurements were done at a final concentration of 0.1
mM immediately after the solubilization of the compounds
_ Methylviologen
under darkness conditions.

2.6. Data analysis

All results expressed as means _+ S.D. (n = 3 probes)

ooo/ =0o Potential (mV)
were analyzed using Student's t-test, and the non-paramet-
ric Spearman rank correlation analysis. P < 0.05 was con-
sidered to be significant.

3. Results

B
3.1. Production of superoxide by rat brain microsomes Menadione m
during the redox cycling of xenobiotics

Initial studies were performed to determine the rates of

superoxide formation by rat brain microsomes during the -15o0, s
redox cycling of xenobiotics related to quinoidal, nitroaro- Potential (mV)
matic and dipyridinium chemical families (Table 1). Under
aerobic conditions, most of the molecules tested were able
to produce superoxide anion radicals, but the rates of
production varied widely. In the absence of added xenobi-
otic, the basal superoxide production was very low (0.89
+ 0.21 n m o l / m i n / m g protein). The reactivity of nitroaro-
Methylvioiogen
matics was about an order of magnitude lower than those
of quinones and viologens, except for nitrofurazone and
1,3-dinitrobenzene. A very high superoxide production
-2000, -1 5 ~ . . . . . . . . . - - ~ ~ 0,_ ~ t'500
(49.7 _+ 4.9 n m o l / m i n / m g protein) was obtained in the -100o
presence of naphthoquinone-2-sulfonate with or without
brain microsomes, suggesting a non-enzymatic direct
~ Potential (mV)

transfer of one electron between the semi-quinone N Q

Fig. 1. Cyclic voltammograms of menadione (0.1 mM) and paraquat (0.1
and the acetylated ferricytochrome c [37]. The highest rate raM) in 0.1 M phosphate (pH 7.0) buffer (A) and in 0.1 M NEt4C104-
of superoxide production was obtained during the biotrans- anhydrous DMF solution (B). Potentials were reported vs. SCE. Scan rate
formation of menadione (Table 1). Among the other was 100 mV/s. See Section 2 for other experimental conditions.
 M.-H. Livertoux et a l . / Brain Research 725 (1996) 207-216 211

than with nitrofurazone. Nitrazepam rates approached the with classic or differential pulse polarography. This method
background values of acetylated ferricytochrome c reduc- consists in cycling the potential of the working electrode
tion by microsomes (Table 1). The other nitrobenzodi- over a given range and measuring the resulting current.
azepines assayed did not significantly produce superoxide The important parameters of the recorded cyclic voltam-
anions ( P > 0.05). mogram are the magnitudes of the peak currents and the
The apparent K r n and VmaX values have been calculated positions of the oxidation (Epa) and reduction (Epc)
from the rates of reduction of acetylated ferricytochrome c peaks. As we were unable to detect individual one-electron
during the redox cycling of xenobiotics by rat brain micro- systems by electrochemistry in an aqueous medium, the
somes with NADPH (4 mM) in aerobic conditions. Appar- reduction behavior of the assayed compounds was studied
ent Vmax values differed greatly and increased about 15 in both phosphate (pH 7.0) buffer and an unbuffered
times between mitomycin C (2.81 n m o l / m i n / m g protein) aprotic solvent, DMF. These assays were carried out be-
and diquat (43.06 n m o l / m i n / m g protein). The K m values cause the NADPH-cytochrome P-450 reductase membrane
obtained in the presence of quinones and nitro compounds environment is both polar and lipophilic, as the flavo-
ranged from approximately 2 mM (niridazole) to 4 txM protein is anchored to the phospholipid bilayer of the
(menadione). Kinetic parameters have not been determi- endoplasmic reticulum via a hydrophobic amino-terminal
nated, neither for nitrazepam, which did not produce peptide, and is exposed to the cytoplasmic face of this
enough O2- nor for naphthoquinone-2-sulfonate (No. 4), membrane system [5,29]. As a typical example, cyclic
which was able to reduce directly acetylated ferricy- voltammograms of menadione and paraquat recorded both
tochrome c [37]. V m a x / K r n values were calculated to char- in phosphate (pH 7.0) buffer and in anhydrous DMF with
acterize the relation between the ability of xenobiotics to 0.1 M NEt4C104 as electrolyte are shown in Fig. 1A and
produce superoxide anion and their electron affinity evalu- Fig. 1B, respectively. The observed reduction potential
ated from reduction potentials. peaks Epc were expressed in millivolts (mV) using the
SCE as reference electrode. A single reversible two-elec-
3.2. Reduction potentials of xenobiotics tron peak was obtained in phosphate (pH 7.0) buffer with
menadione (Fig. 1A) and most of the other quinones, in
The possibility of the in vivo one- or two-electron agreement with the postulated mechanism of quinone re-
transfer involvement was examined by means of electro- duction ( Q / Q H 2) in aqueous solvents, whereas two dis-
chemical data obtained by cyclic voltammetry, which gives tinct reduction peaks were observed in the aprotic solvent
the best resolution of the two redox steps in comparison (DMF). Similar observations were previously reported by

Table 2
R e d u c t i o n p o t e n t i a l s E p c referred to saturated c a l o m e l electrode ( S C E ) o f q u i n o n e s , n i t r o a r o m a t i c s a n d v i o l o g e n s o b s e r v e d in p h o s p h a t e ( p H 7.0) b u f f e r
( E p c , 7 , in m V ) and in a n h y d r o u s d i m e t h y l f o r m a m i d e ( E p c , D M F , in m V ) , a n d o n e - e l e c t r o n r e d u c t i o n p o t e n t i a l E~ f r o m the literature [51] referred to
standard hydrogen electrode (NHE)

Number Xenobiotics Epc,DMF Epc,7 E~ a

( m V vs. S C E ) ( m V vs. S C E ) ( m V vs. N H E )

l 1,4-Benzoquinone - 450 - 110 4- 9 9

2 Methyl-1,4-benzoquinone - 540 - 160 4- 23
3 2,5-Dimethyl- 1,4-benzoquinone - 610 - 220 - 67
4 Naphthoquinone-2-sulfonate - 635 - 230 - 60
5 Menadione - 708 - 330 - 162
6 Mitomycin C - 700 - 520 - 271
7 Duroquinone - 778 - 350 - 240
8 Indigo carmine - 782 - 380 - 247
9 9,10-Anthraquinone-2-sulfonate - 848 - 510 - 375
10 1,3-Dinitrobenzene - 835 - 580 - 345
11 N i t r o b e n z o i c acid - 880 - 775 - 425
12 Nitrofurazone - 890 - 560 - 257
13 Nitrazepam - 970 - 710 _ a
14 Flunitrazepam -980 -620 _ a
15 Nimetazepam - 990 - 670 _ a
16 Niridazole - 1010 - 620 _ a

17 Metronidazole - 1160 - 780 - 486

18 Benzyl viologen - 260 - 450 - 350
19 Diquat - 362 - 645 - 358
20 Methyl v i o l o g e n (paraquat) - 407 - 720 - 448

O n e - e l e c t r o n r e d u c t i o n p o t e n t i a l E~ v a l u e s are f r o m W a r d m a n [51].
-, not available.
212 M.-H. Livertoux et al. / Brain Research 725 (1996) 207-216

Klopman et Doddapaneni [21 ]. This reduction of quinones
was quasi-reversible. However, viologens exhibited two
reduction peaks in both solvents, corresponding to two
single-electron reductions of the two iminium groups [2].
Only methyl viologen (paraquat) underwent quasi-reversi-
ble reduction steps in both aqueous and aprotic solvents
(Fig. 1B). Under aqueous neutral conditions, all nitro
compounds assayed (nitroarenes, nitrobenzodiazepines, ni-
troimidazoles) showed only one reduction step of the nitro
group, without reoxidation peak. Also, only irreversible
voltammetric waves were observed. In the aprotic solvent
DMF, nitrobenzodiazepines and nitroimidazoles also
showed an irreversible nitro group redox system, whereas
nitroaryl derivatives exhibited two reduction peaks (results
not shown).
The reduction potentials obtained by cyclic voltamme-
try in phosphate (pH 7.0) buffer (Epc,7) and in DMF
(Epc,DMF) are listed in Table 2. The Epc of the first
reduction step of quinones in DMF and of viologens in
both solvents are also reported. For comparison, some data
on one-electron reduction potentials from the literature
30.
5 structural parameters of solvation on the compound's reac-
~20 ~19
Q 20
~E
0 6
<.- 10
• 10 ~9
14
17 13 15 16
0
-800 -600 -400
Reduction potential in phosphate buffer (mV)
30
(E~) usually determined by radiolysis have also been
included. These reference potentials measured at pH 7.0
were expressed throughout a normal or standard hydrogen
electrode (NHE). Our values (Epc,7 and Epc,DMF) re-
ferred to the saturated calomel electrode (SCE), which can
be converted to a normal hydrogen electrode (NHE) by
adding + 244 mV for measurements at T = 298°K [51]. E~
derived from redox equilibria measured using the pulse
radiolysis quoted in the literature have become available
for a limited number of quinones, bipyridinium and nitro
compounds [51]. When the E~ (vs. NHE) values of reduc-
tion of the quinones and nitro compounds in aqueous
media (pH 7.0) were plotted vs. those in anhydrous DMF
(Epc,DMF) a linear correlation was observed (R 2 = 0.86,
P = 0.02).
At pH 7.0, quinones were more easily reduced than
nitro compounds and iminium salts, which showed reduc-
tion potentials more negative than Epc,7 = - 560 mV (vs.
SCE), with the exception of benzyl viologen. It can be
seen that 1,4-benzoquinone (Epc,7 = - 2 0 0 mV) pos-
sesses the more positive reduction potential, therefore is
the most easily reducible molecule. The reduction poten-
tials of the other quinones decreased steadily with structure
complexity. Moreover, the Epc values of viologens varied
A strongly with the nature of the solvent used for the assay
(Table 2). This variation reflects in part the influence of
tivity. The particular electrochemical behavior of bipyri-
dinium ions in an aqueous neutral solution, as compared to
the situation in the aprotic solvent, was probably related to
the different solvation properties of the solvents toward
ionized molecules.
2 1 3.3. Relation between superoxide radical production and
-200
reduction potentials
A plot of rates of superoxide production vs. the reduc-
tion potential Epc,7 in phosphate (pH 7.0) buffer of all
compounds (0.5 mM) in Table 1 is presented in Fig. 2 and
B

t
shows that no apparent correlation does exist between the
5 two parameters. For quinones with reduction potentials
20~ ~ 19 ~ 18 greater than Epc,7 = - 1 6 0 mV (l,4-benzoquinone and
20
2-methyl-l,4-benzoquinone; E~ > 0 mV) (Fig. 2A) and
chemicals with redox potentials higher than - 7 5 0 mV
<.~ such as MPP ÷ (Epc,7 = - 1000 mV), no superoxide pro-
~E I0
duction was detected, indicating that the radicals released
9~ z0
during enzyme activity did not react with oxygen. Most
15 14 compounds having reduction potentials ranging from Epc
17 16_IL 13 11 2 I
0 ~. -~ .~. • . • . = - 2 3 0 mV to - 7 2 0 mV in phosphate (pH 7.0) buffer
- 1200 - 10t30 -800 -600 -400 -200 (and about - 7 0 0 mV to - 1 1 0 0 mV in non-aqueous
Reduction potential in DMF (mV)
DMF) (Fig. 2B) produced superoxide during their redox
Fig. 2. D e p e n d e n c e o f the rate constant (in n m o l / m i n / m g protein) for cycling. A maximal production was observed with mena-
reduction of acetylated f e r r i c y t o c h r o m e c by rat m i c r o s o m e s during redox
dione (No. 5), corresponding to an optimal reduction po-
c y c l i n g o f chemicals (0.5 m M ) at pH 7.4 a n d 37°C u p o n their reduction
potentials E p c , 7 (in p h o s p h a t e (pH 7.0) buffer) (A) a n d E p c , D M F (in
tential (Epc,7 = - 3 3 0 mV vs. SCE; E~ = - 2 0 0 mV vs.
a n h y d r o u s D M F ) (B) reported vs. SCE. Each point is the mean of three NHE). Nevertheless, a linear relation seems to exist for
measurements. See Table 2 for key to c o m p o u n d s . most of the quinones and nitro compounds which have
 M.-H. Livertoux et al. / Brain Research 725 (1996) 207-216
reduction potentials Epc,7 between - 3 3 0 mV and - 6 5 0
mV (vs. SCE) (Fig. 2A). It can be seen from Fig. 2B that
mitomycin C (No. 6) appears to react slower than would
be predicted. Mitomycin C which may undergo reactions
with microsomes other than a simple electron transfer
[4,6], and viologens (Nos. 18, 19, 20) which presented a
particular electrochemical behavior were excluded from
the study on linearity. The linearization of the rates of
reaction of acetylated ferricytochrome c with the radicals
formed during redox cycling for quinones and nitro com-
pounds, with Epc values in the favorable range of reduc-
tion potentials, as shown before, gave the following equa-
tions when mitomycin C was omitted:
potential measured in aqueous phosphate (pH 7.0)
buffer ( E p c , 7 ) :
rate ( n m o l / m i n / m g protein)
= 40.75 + 0.059 Epc,7 (mV)
( R 2 = 0.803, P : 0.009)
and:
potential measured in anhydrous DMF ( E p c , D M F ) :
rate ( n m o l / m i n / m g protein)
= 80.14 + 0.08 Epc,DMF (mV)
R 2 = 0.875, P = 0.008)
With respect to K m and Vmax value variations described
m Table 1, the influence of redox potentials on logarithmic
apparent kinetic parameters for quinones (except com-
pounds Nos. 4 and 6) and nitro compounds was investi-
gated. No significant correlation between Vmax values and
the reduction potentials (Epc,7 and Epc,DMF) was ob-
served. However, a good relation was observed with the
two terms, log K m ( R 2 = 0.845 in buffer and R 2 = 0.769
in DMF) and log V m ~ x / K m ( R 2 = 0.824 in buffer and
R 2 = 0.854 in DMF) for the compounds with reduction
potentials ranging from Epc,7 = - 3 3 0 mV (or E~ =
- 160 mV vs. NHE) to Epc,7 = - 650 mV (or E~ = - 400
mV vs. NHE). The data obtained suggest that the redox
potential of most of the molecules assayed was related to
their ability to undergo one-electron reduction mediated by
flavoenzymes in the rat brain.
4. Discussion
It is now well accepted that ageing is based on the
continuous production of free radicals during mitochon-
drial respiration, autooxidation of biological molecules and
chemicals or environmental pollutants and during irradia-
tion damage [27]. The brain is especially a target of
oxygen-derived free radicals as it shows high respiratory
activity and contains large amounts of substrates that are
susceptible to free radical attack [25]. The involvement of
213
these reactive species in several brain pathological states
such as trauma and stroke, in diseases of the central
nervous system such as Alzheimer's [49] and Parkinson's
diseases [1,10], and in some mechanisms of neurotoxicity
[25,30] are well documented. As a consequence of their
production in the human brain, modifications of proteins
by oxidation or glycation, and formation of lipid peroxides
or nucleic acid adducts may promote more or less specific
neuronal destruction, resulting either in accelerated cere-
bral ageing or in the pathogenesis of CNS diseases
[1,20,40,43,49]. Therefore it seemed to us very important
to assess the structure/risk factor relationships of
molecules possibly having access to the CNS by evaluat-
ing their electron affinity from their redox potentials.
The ability of exogenous compounds to undergo redox
cycling depends on the thermodynamics of electron trans-
fer reactions, and the equilibrium constants for electron
transfer reactions depend on the differences between the
reduction potentials of the electron donor, for instance
NADPH-cytochrome P-450 reductase, and xenobiotics
a n d / o r their metabolites, and oxygen. The reaction be-
tween oxygen and xenobiotics (e.g., quinones [45] and
nitro compounds [50]), occurs rapidely in the order of
106--107 M-1 S 1. The results of Butler and Hoey [6]
concerning the reaction of O 2 with acetylated ferricy-
tochrome c (rate constant in the order of 105 M-1 s 1)
essentially confirm that the enzymatic reduction of the
xenobiotics is the rate determining step.
The reactivity of quinones and nitro compounds and
their ability to be reduced by microsomal enzymes like
NADPH-cytochrome P-450 reductase seem to be related
to their single-electron reduction potential (E~) [6,8,34,37].
Usually, one-electron redox potentials were measured for
several systems using various methods ranging from con-
ventional potentiometric titration (mid-point potential
Em,7) to fast reaction methods as pulse radiolysis (E~)
[51]. This last method allows the determination of redox
potentials of short-lived radicals. Polarography was also
used for the estimation of redox potentials of single-elec-
tron steps for several systems, especially when the single
reduced intermediate was stable in non-aqueous media.
Cyclic voltammetry, which gave the best resolution of the
two redox one-electron steps in comparison with classic or
differential pulse polarography, was chosen to determine
the reduction potential of the first step of reduction in an
aprotic solvent (DMF). This method was generally similar
to the thermodynamic reversible E~ values, when ex-
pressed on the hydrogen scale (vs. NHE) [51]. However,
data obtained in these conditions cannot be related to the
component single-electron reduction potentials (E~) mea-
sured by pulse radiolysis. No data for nitrobenzodiazepines
and niridazole reduction potentials E~ have been found in
the literature. So, a complete measurement of the reduction
potentials of all xenobiotics used in this work was neces-
sary to relate them to superoxide anion production.
Our study on the relations between superoxide produc-
214 M.-H. Lit,ertoux et al. / Brain Research 725 (1996) 207-216

tion and the reduction potential of xenobiotics was in
accordance with results obtained by other authors both on
hepatic microsomes [34] and purified NADPH-cytochrome
P-450 reductase [6,8,37]. They also observed an important
superoxide production with quinones and most of nitro
compounds. Nevertheless, superoxide production during
redox cycling of xenobiotics occurs only for a limited
domain of potentials, which should be compatible with
0 2 / 0 2 - and enzymatic redox systems.
We observed that the larger superoxide productions
with quinones had the highest reduction potentials. Never-
theless, 1,4-benzoquinone and 2-methyl-l,4-benzoquinone
have a relatively high one-electron reduction potential
(E~ = + 99 mV and + 23 mV, respectively), but did not
reduce acetylated ferricytochrome c during their
metabolism by brain microsomes (Fig. 3). Thermodynami-
cally, N A D P H (mid-point potential Em,7
NADP+/NADPH = - 3 2 0 mV) [46] should be able to
reduce directly compounds with more positive reduction
potentials, including naphthoquinone-2-sulfonate (E~ =
- 6 0 mV) (Fig. 3); however, the rates would be extremely
slow [6]. Formed hydroquinones QH 2 slowly reduce acety-
lated ferricytochrome c [6]. Theoretically, if the one-elec-
tron reduction potential E~ of a quinone is below - 155
mV, which is the E~ for 0 2 / 0 2 on a molar basis, the
quinone radical will undergo redox cycling and form su-
peroxide anion radicals from molecular oxygen. Neverthe-
less, many quinones, nitroheterocyclic compounds and vio-
logens which produce superoxide anion have one-electron
reduction potentials E~ lower than that of the oxygen
couple O 2 / O 2- (Table 2, Fig. 3). In our experimental
conditions, using air-saturated buffer (02 concentration:
217 txM), E~ O 2 / O 2- was - 2 1 4 mV [38] and several
quinones appear more electropositive than this value.
Therefore, these compounds may form appreciable amounts
of superoxide. So, the redox properties of the one-electron
couple X / X provide only limited information on the
biological properties of these compounds. Nevertheless, an
absolute relationship between in vitro electrochemical be-
havior and physiological activity was not likely due to
many factors involved in vivo (e.g., stereochemistry, lipid
solubility, diffusion, enzymatic site binding, electron trans-
fer reactions described in terms of equilibrium and kinet-
ics). At potentials lower than E~ = - 4 0 0 mV, as was the
case for nitrobenzoic acid and metronidazole, the rate of
superoxide formation by the brain microsomal enzymes
was not significant. However, in the presence of viologens,
an important amount of superoxide was produced despite
their more negative potential ( - 3 5 0 mV < E~ < - 4 5 0
mV). The reduction capacity of these compounds should
be energetically favorable as a consequence of their elec-
trophilic properties.
The microsomal electron transport system contains two
flavin enzymes, NADPH-cytochrome P-450 reductase
(FAD-FMN) and NADH-cytochrome b 5 reductase (FAD)
[18]. The ability of rat brain tissue to catalyze the one-elec-
tron reduction of these xenobiotics was first suggested by
the relatively high cerebral NADPH-cytochrome P-450

Era,7 (mV) El,7 (mY)

+ 100 - - 1,4-benzoquinone (+99) + 100 --

2-methyl- 1,4-benzoquinone (+23)

0 -- 0 --

1,4-naphthoquinone-sulfonate (-60)
2,5-dimethylbenzoquinone (-67)
- 100 -- -100 --
FMN/FMNHo (-1 tO)

- 200 - -
\ menadione (- 1 6 2 ) - - ]~ o2/o2"
- 200 --
(-155)

FMN/FMNH2 • d u r o q u i n q n e (-240) IAE

/ i n a l g o c a r m i n e (-247)
FMNH°,'FMNH2 (-270) nitrofurazone (-257)
FAD/FADH° m i t o m y c i n C (-271) ~' X/X'-
- 300-- - 300 - -

NAD(P~" / NAD(P)H r FAD/FADH2

N 1 , 3 - d i n i t r o b e n z e n e (-345)
(-320) b e n z y l viqlogen (-35_0)
FADH°/FADH2 a i q u a t (-358)
9,10-anthraquinone-sulfonate (-375)
- 400 -- - 400 --
3 - n i t r o b e n z o i c a c i d (-425)
m e t h y l viologen (-448)
metronidazole (-486)
- 500 - - - 500 - -
I NADPH-CYTOCHROME P450 REDUCTASE I [ XENOBIOTICS X/X'' ]

Fig. 3. Mid-point oxidation-reduction potentials (pH 7.0) of m i c r o s o m a l N A D P H - c y t o c h r o m e P - 4 5 0 reductase ( F A D - F M N , E m , 7 ) and one-electron

reduction potentials of xenobiotics (E~). The values o f the e n z y m e are these determined by Iyanagi et al. [18] and those for chemicals are literature values
[51].
 M.-H. Livertoux et al. / B r a i n Research 725 (1996) 207-216
reductase activity [14]. In this study, we demonstrated that
in the brain there is only one molecule of microsomal
cytochrome P-450 for each molecule of N A D P H - c y t o -
chrome P-450 reductase, whereas in the liver one molecule
of N A D P H - c y t o c h r o m e P-450 reductase should supply
electrons to 20 molecules of microsomal cytochrome P-450
[14]. That means that in the CNS the relative concentration
of N A D P H - c y t o c h r o m e P-450 reductase, as compared to
cytochrome P-450, is 20-fold higher than in the liver,
allowing a high availability for this enzyme to catalyze
monoelectronic transfers to other acceptors than cy-
tochrome P-450. If the acceptors are molecules able to
undergo a monoelectronic redox cycling, this activity will
produce reactive oxygen species [15,37]. W e also demon-
strated recently that rat brain homogenates or microsomes,
microvessels or choroid plexus produced superoxide an-
ions when metabolizing drugs or xenobiotics related to
quinoidal, nitroheterocyclic or bipyridinium chemical fami-
lies [15,23]. The functional and immunological character-
istics of rat brain N A D P H - c y t o c h r o m e P-450 reductase
have been described [14]. This enzyme participates basi-
cally in the microsomal electron transport chain, using
both F A D and F M N as sites of entry and exit, respectively.
Studies on the enzymatic mechanism suggest that electron
transfer follows the pathway N A D P H --->F A D ~ F M N --->
e l e c t r o n a c c e p t o r X [48]. T h e redox couple
F M N H ' / F M N H 2 ( E m , 7 = - 270 mV) of N A D P H - c y t o -
chrome P-450 reductase may function as a one-electron
carrier in its catalytic c y c l e [17]. Inhibition by
diphenyliodonium cation DPI + (2.5 mM) [47] added to rat
brain microsomes incubated in the presence of menadione
showed a 60% decrease of the N A D P H - d e p e n d e n t super-
oxide production. So, the pattern of quinone-dependent
superoxide formation by microsomes (Table 2, Fig. 2)
exhibiting a m a x i m u m near an E~ = - 2 0 0 mV suggests
the implication of the N A D P H - c y t o c h r o m e P-450 reduc-
tase, which presents a close compatible redox potential.
W e demonstrated recently that incubation of rat brain
microsomes with N A D P H and several of the chemicals
studied in the present paper led to inactivation of two
isoforms of cytochrome P-450 [24]. It seems highly proba-
ble that this process is involved both in the intracellular
renewal of proteins and in cellular ageing [19,42,43]. This
may explain why a chronic intake of some drugs for the
treatment of neurologic disorders or an exposition to envi-
ronmental pollutants may promote chronic oxidative stress
[19]. This is especially important in the central nervous
system, where drug-induced parkinsonism accounts for an
important part of patients seen in neurology clinics for
parkinsonian symptoms [13]. Accordingly, it has been
suggested that among the drugs potentially capable of
producing drug-induced parkinsonism, several neuroleptics
promote free radical reaction in vivo, which contributes to
their side-effects [19] and possibly to neurodegenerative
disorders [26,42].
In conclusion, our results show that rat brain micro-
215
somes catalyze the production of superoxide anion during
the redox cycling of many chemicals, and suggest that the
rate of formation was dependent on the reduction poten-
tials between approx. Epc,7 = - 3 3 0 mV and - 6 5 0 mV
(vs. SCE). Through our use of the N A D P H - c y t o c h r o m e
P-450 reductase inhibitor DPI +, we showed that superox-
ide production involves this microsomal flavoprotein, as
suggested in some previous papers [15,23]. Therefore, the
present study should allow some prediction about the way
by which chemicals like quinones and nitroaromatics can
react with brain microsomes by measuring the reduction
potentials of drugs and xenobiotics able to cross the
b l o o d - b r a i n barrier. This may provide information on their
potential reactivity and their possible neurotoxicity result-
ing from the production of oxygen reactive species through
a redox cycling mechanism.
Acknowledgements
This study was supported by a grant from the Sandoz
Fondation for Gerontological Research. W e thank Prof. J.
Bessi~re, Department of Chemistry and Analytical Electro-
chemistry (UniversitE Henri PoincarE, Nancy 1, France)
for giving the opportunity to use the voltammetric appara-
tus and other facilities in his laboratory. W e also would
like to thank Dr. J. Magdalou (Centre du MEdicament,
Nancy) for fruitful discussions.
References
[1] Adams, J.D., Jr. and Odunze, I.N., Oxygen free radicals and Parkin-
son's disease, Free Rad. Biol. Med., 10 (1991) 161-169.
[2] Ames, J.R., Brand~inge, S., Rodriguez, B., Castagnoli, N., Ryan,
M.D. and Kovacic, P., Cyclic voltammetry with cyclic iminium
ions: implications for charge transfer with biomolecules (metabolites
of nicotine, phencyclidine, and spermine), Bioorg. Chem., 14 (1986)
228-241.
[3] Azzi, A., Montecucco, C. and Richter, C., The use of acetylated
ferricytochrome c for the detection of superoxide radicals produced
in biological membranes, Biochem. Biophys. Res. Commun., 65
(1975) 597-607.
[4] Bachur, N.R., Gordon, S.L. and Gee, M.V., A general mechanism
for microsomal activation of quinone anticancer agents to free
radicals, Cancer Res., 38 (1978) 1745-1750.
[5] Black, S. and Coon, M.J., Structural features of liver microsomal
NADPH-cytochrome P450 oxidoreductase: hydrophobic domain, hy-
drophilic domain, and connecting region, J. Biol. Chem., 257 (1982)
5929-5938.
[6] Butler, J. and Hoey, B.M., The one-electron reduction potential of
several substrates can be related to their reduction rates by cy-
tochrome P-450 reductase, Biochim. Biophys. Acta, 1161 (1993)
73-78.
[7] Calne, D.B., Duvoisin, R.C. and McGeer, E., Speculations on the
etiology of Parkinson's disease, Ann. Neurol., 40 (1984) 353-360.
[8] CEnas, N., Anusevicius, Z., BironaitE, D., Bachmanova, G.I., Ar-
chakov, A.I. and Ollinger, K., The electron transfer reactions of
NADPH:cytochrome P450 reductase with nonphysiological oxi-
dants, Arch. Biochem. Biophys., 315 (1994)400-406.
216 M.-H. Lil ertoux et al. / Brain Research 725 (1996) 207-216
[9] Cohen, G., Oxy-radical toxicity in catecholamine neurons, Neuro-
toxicology, 5 (1984) 77-82.
[10] Dexter, D.T., Carter, C.J., Wells, F.R., Javoy-Agid, F., Agid, Y.,
Lees, A., Jenner, P. and Marsden, C.D., Basal lipid peroxidation in
substantia nigra is increased in Parkinson's disease, J. Neurochem,
52 (1989) 381 389.
[l 1] Donaldson, J., The physiopathologic significance of manganese in
brain: its relation to schizophrenia and neurodegenerative disorders,
Neurotoxicology, 8 (1987) 451-462.
[12] Gee, P. and Davison, A.J., Intermediates in the aerobic autooxida-
tion of 6-hydroxydopamine: relative importance under different reac-
tion conditions, Free Rad. Biol. Med., 6 (1989) 271-284.
[13] Gershanik, O.S., Drug induced parkinsonism in the aged. Recogni-
tion and prevention, Drugs Aging, 5 (1994) 127 132.
[14] Ghersi-Egea, J.F., Minn, A., Daval, J.L., Jayyosi, Z., Arnould, V.,
Souhaili-E1 Amri, H. and Siest, G., NADPH-cytochrome P450(c)
reductase: biochemical characterization in rat brain and cultured
neurones and evolution of activity during development, Neurochem.
Res.. 14 (1989) 883-888.
[15] Ghersi-Egea, J.F., Livertoux, M.H., Minn, A., Perrin, R. and Siest,
G., Enzyme mediated superoxide radical formation initiated by
exogenous molecules in rat brain preparations, Toxicol. Appl. Phar-
macol.. 110 (1991) 107 117.
[16] Harman, D., Free radicals and aging, Mol. Cell. Biochem., 84 (1988)
155-161.
[17] Iyanagi, T., On the mechanisms of one- and two-electron transfer by
flavin enzymes, Chem. Scripm, 27(A)(1987) 31 36.
[18] [yanagi, T., Makino, N. and Mason, H.S., Redox properties of the
reduced nicotinamide adenine dinucleotide phosphate-cytocbrome
P-450 and reduced nicotinamide adenine dinucleotide-cytochrome
b5 reductases, Biochemisto,, 13 (1974) 1701-1710.
[19] Jeding, I., Evans, P.J., Akanmu, D., Dexter, D., Spencer, J.D.,
Aruoma, O.l., Jenner, P. and Halliwell, B., Characterization of the
potential antioxidant and pro-oxidant actions of some neuroleptic
drugs, Biochem. Pharmacol., 49 (1995) 359-365.
[20] Jenner, P., Oxidative damage in neurodegenerative disease, Lancet,
344 (1994) 796-798.
[21] Klopman, G. and Doddapaneni, N., Electrochemical behavior of
substituted polycyclic aromatic quinones. J. Phys. Chem.. 78 (1974)
1820-1825.
[22] Kovacic, P., Ames, J.R. and Ryan, M.D., Reduction potentials of
antimycobacterial agents; relationship to activity, Bioelectrochem.
Bioenerg., 21 (1989) 269-278.
[23] Lagrange, P., Livertoux, M.H., Grassiot, M.C. and Minn, A., Super-
oxide anion production during monoelectronic reduction of xenobi-
otics by subcellular fractions, Free Rad. Biol. Med., 17 (1994)
355-359.
[24] Lagrange, P., Livertoux, M.H., Chat, M. and Minn, A., Inhibition of
rat brain microsomal cytochromes P450 1A and 2B activities by
superoxide radicals. Submitted.
[25] Le Bel, B.E. and Bondy, S.C., Oxidative damage and cerebral aging,
Prog. Neurobiol., 38 (1992) 601 609.
[26] Lees, G.J., Contributory mechanisms in the causation of neurode-
generative disorders, Neuros~ience, 54 (1993) 287 322.
[27] Leibowitz, B.E. and Siegel, B.V., Aspects of free radical reactions in
biological systems: aging, J. Gerontol., 35 (1980) 45 56.
[28] Lowry, O.H., Rosenbrough, N.J., Farr, A.L. and Randall, R.J.,
Protein measurement with the Folin phenol reagent, J. Biol. Chem.,
193 (1951) 265-275.
[29] Magdalou, J., Thirion, C., Balland, M. and Siest, G., Conformation
studies of NADPH cytochrome P450 reductase by circular dichroism
interaction with phospholipids, Int. J. Biochem.. 17 (1985) 1103
1107.
[30] Maret, G., Tayar, N.E., Carrupt, P.A., Testa, B., Jenner, P. and
Baird, M., Toxication of MPTP (l-methyl-l,2,3,6-tetrabydropyri-
dine), Biochem. Pharmacol.. 40 (1990) 783 792.
[31] Mason, R.P. and Holtzman, J.L., The role of catalytic formation in
the 02 inhibition of nitroreductase, Biochem. Biophys. Res. Com-
mun., 67 (1975) 1267-1274.
[32] Mikani, S., Ringler, R.L. and Singer, T.P., Studies of the respira-
tory-chain-linked dihydrophosphopyridine nucleotide deshydroge-
nase, J. Biol. Chem., 237 (1976) 569-576.
[33] Olesen, S.P., Free radicals decrease electrical resistance of microvas-
cular endothelium in brain, Acta Physiol. Seand., 129 (1987) 181-
187.
[34] Olive, P.L., Correlation between metabolic reduction rates and elec
tron affinity of nitroheterocycles, Cancer Res., 39 (1979) 4512 4515.
[35] Olive, P.L., Inhibition of DNA synthesis by nitroheterocycles. 1.
Correlation with half-wave reduction potential, Br. J. Cancer, 40
(1979) 89-94.
[36] Olive, P.L. and Durand. R.E., Activation of radiosensitizers by
hypoxic cells, Br. J. Cancer, 37 (1978) 124-127.
[37] Powis, G. and Appel, P.L., Relationship of the single-electron
reduction potential of quinones to their reduction by flavoproteins,
Biochem. Pharmacol., 29 (1980) 2567-2572.
[38] Powis, G., Svingen, B.A. and Appel, P., Quinone-stimulated super-
oxide formation by subcellular fractions, isolated bepatocytes, and
other cells, Mol. Pharmacol., 20 (1981) 387 394.
[39] Powis, G., Free radical formation by antitumor quinones, Free Rad.
Biol. Med.. 6 (1989) 63 101.
[40] Richardson, J.S., Free radicals in the genesis of Alzheimer's disease,
Ann. NY Aead. Sci., 695 (1993)73-76.
[41] Sato, K., Akaike, T., Suga, M., Ando, M. and Merada, H., Genera-
tion of free radical from neocarsinostatin mediated by NADPH/cy-
tochrome P-450 reductase via activation of enediyne chromophore,
Biochem. Bioph3s. Res. Commun.. 205 (I 994) 1716 1723.
[42] Schapira. A.H.V.. Oxidative stress in Parkinson's disease - review,
Neuropathol. AppI. Neurobiol., 21 (1995) 3-9.
[43] Smith, M.A., Sayre, L.M., Monnier, V.M. and Perry, G., Radical
ageing in Alzbeimer' s disease, Trends Neurosei., 18 (1995) 172-176.
[44] Strobel, H.W. and Digman, J.D., Purification and properties of
NADPH cytochrome P-450 reductase, Methods Enz.ymol., 52 (1978)
89 96.
[45] Svingen, B.A. and Powis, G., Pulse radiolysis studies of antitumor
quinones: radical lifetimes, reactivity with oxygen and one-electron
reduction potentials, Arch. Biochem. Biophys., 209 (1981 ) 119-126.
[46] Swallow, A.J., In B.L. Trumpower (Ed.), Functions ~I~Quinones in
Energ) Comersing Systems~ Academic Press, London, 1982, pp.
59-72.
[47] Tew, D.G., Inhibition of cytochrome P450 reductase by the
diphenyliodonium cation. Kinetic analysis and covalent modifica-
tions, Biochemistry, 32 (1993) 10209 10215.
[48] Vermilion, J.L., Ballou, D.P., Massey, V. and Coon, M.J., Separate
roles for FMN and FAD in catalysis by liver microsomal NADPH-
cytochrome P-450 reductase, .L Biol. Chem., 256 ( 1981 ) 266-277.
[49] Volicer, L. and Crino, P.B., Involvement of free radicals in dementia
of the Alzheimer type: a hypothesis, Neurobiol. Aging, 11 (1990)
567-571.
[50] Wardman, P. and Clarke, E.D., Oxygen inhibition of nitroreductase:
electron transfer from nitro radical-anions to oxygen, Biochem.
Biophys. Res. Commun., 69 (1976) 942-949.
[51 ] Wardman, P., Reduction potentials of one-electron couples involving
free radicals in aqueous solution, ,L Phys. Chem. Re.[. Dam, 18
(1989) 1637 1755.
[52] Wooley, D.E., Gietzen, D.W., Gee, S.J.. Magdalou, J. and Ham-
mock, D.W., Does paraquat (PQ) mimic MPP + toxicity?, Proc.
West. Pharma~ol. Soc., 32 (1989) 191 193.
[53] Youdim. M.B.H., Ben-Shachar, D. and Riederer, P., Is Parkinson's
disease a progressive siderosis of substantia nigra resulting in iron
and melanin induced neurodegeneration?, Acre Neulvl. Scand., 126
(1989) 47 54.