Comparative pulse radiolysis studies of alkyl- and methoxy-substituted semiquinones formed from quinones and hydroquinones

Article abstract of DOI:10.1039/a801082j

Absorption spectra and rate constants for the disproportionate of 12 alkyl- and methoxy-substituted semiquinone anion free radicals (Q^?-) produced by the one-electron reduction (using CO₂^?- as a reductant) of 1,4-benzoquinones and 1,4-naphthoquinone (Q) as well the oxidation (using N₃^? as an oxidant) of the corresponding hydroquinones (QH₂) were determined by pulse radiolysis in 50 mM sodium phosphate buffer, pH 7.40 at room temperature. Both spectral and kinetic characteristics of Q^?- only moderately depended on whether Q^?- was produced from Q or QH₂. Spectra of benzosemiquinones display two peaks with maximum at 310-320 nm and ca. 430 nm with the ratio of about 2-2.5. Molar absorption coefficients were determined. Rate constants for Q^?- disproportionation (2k₁) were correlated with the nature of substituents. While 2k₁ was scarcely affected by methyl substitution, Q^?- containing isopropyl, tert-butyl and methoxy substituents were visibly more stable than non-substituted and methyl-substituted Q^?-.

Full text of DOI:10.1039/a801082j

View Article Online / Journal Homepage / Table of Contents for this issue
Comparative pulse radiolysis studies of alkyl- and
methoxy-substituted semiquinones formed from quinones and
hydroquinones
Vitaly A. Roginsky,a Leonid M. Pisarenko,a Wolf Bors,b*¤ Christa Michelb and Manfred Saranb
a N. N. Semenov Institute of Chemical Physics, Russian Academy of Sciences, Kosygin St. 4,
117977 Moscow, Russia
b Institut f ur Strahlenbiologie, GSF Forschungszentrum f ur Umwelt und Gesundheit, 85764
Neuherberg, Germany
Absorption spectra and rate constants for the disproportionation of 12 alkyl- and methoxy-substituted semiquinone anion free
radicals (Q~~) produced by the one-electron reduction (using CO ~~ as a reductant) of 1,4-benzoquinones and 1,4-naphthoqui-
2
none (Q) as well the oxidation (using N ~ as an oxidant) of the corresponding hydroquinones (QH ) were determined by pulse
3
2
radiolysis in 50 mM sodium phosphate bu†er, pH 7.40 at room temperature. Both spectral and kinetic characteristics of Q~~ only
moderately depended on whether Q~~ was produced from Q or QH . Spectra of benzosemiquinones display two peaks with
2
maximum at 310È320 nm and ca. 430 nm with the ratio of about 2È2.5. Molar absorption coefficients were determined. Rate
constants for Q~~ disproportionation (2k ) were correlated with the nature of substituents. While 2k was scarcely a†ected by
1
1
methyl substitution, Q~~ containing isopropyl, tert-butyl and methoxy substituents were visibly more stable than non-substituted
and methyl-substituted Q~~.
Semiquinones (Q~~) hold a central position in redox trans-
so called “mixed solventÏ that is very popular in pulse
radiolysis studies, contains 5 M isopropanol and 1 M acetone6).
However, many characteristics of Q~~ are rather sensitive to
the solvent and thus some doubt is cast upon extending data
obtained with aqueousÈorganic solutions to that without
formations of quinones (Q) and hydroquinones (QH ). Kinetic
2
characteristics of Q~~ determine to a large degree various
properties of Q and QH , such as Q cytotoxicity,1 pharmaco-
2
logical activity of Q,2,3 QH oxidizability etc. Q~~ is known
2
as a reactive intermediate participating readily in various elec-
adding organic solvent.6,9 The use of QH as a source of Q~~
2
seems to be one solution to this problem as QH is more
2
tron transfer processes including reaction with O (for
2
example see ref. 4), ascorbate5 as well as with Q and other
soluble in water than the corresponding Q. However, there is
electron acceptors.6 When other reactions are hindered, dis-
proportionation of Q~~
no prior assurance that the properties of Q~~ produced from
Q and QH are in complete agreement. For this reason it
2
seems to be almost imperative to perform parallel studies of
2k1
Q~~ produced via pulse radiolysis of Q and QH in the same
Q~~ ] Q~~ ] 2H` ÈÈÈ Q ] QH
(1)
2
2
medium.
This was undertaken with the series of methyl- and
methoxy-substituted 1,4-benzoquinones and 1,4-hydro-
quinones (Fig. 1). The above Qs were widely applied as model
compounds in studying Q cytotoxicity;1 with two of them,
Q-910,11 and Q-11,12 reported as having outstanding anti-
is the basic pathway of Q~~ decay. Aside from the fact that the
rate constants for reaction (1), 2k , are of great methodologi-
1
cal signiÐcance, they are likely to be the only rate constants
which can be determined directly. This is one reason why 2k
is widely applied as a reference parameter in the determi-
1
tumor activity. For many Q and QH , which belong to these
2
nation of rate constants for other reactions with the partici-
series, detailed information on their thermodynamical proper-
pation of Q~~.
ties,13 reactivity (e.g. for reaction of Q with thiols14), and
cytotoxicity1 was reported. When these data are supplemented
with rate constants for Q~~ disproportionation, this will
provide insight into the nature of the relationship between
Much attention has been given to determining 2k basically
1
using pulse radiolysis combined with UVÈVIS spectrometry.
It is surprising that the available information on the dispro-
portionation of the most simple substituted 1,4-benzosemi-
quinones is very restricted, although the kinetics of this
process with Q~~ produced from substituted naphthoquinones
and anthraquinones and Q with a more complex structure has
been studied in much detail (ref. 7È9 and references therein).
One of the more serious problems which a researcher encoun-
ters when he studies Q~~ is a very poor solubility in aqueous
solution for the majority of Q, the most popular compounds
used as a source of Q~~. The problem is commonly solved by
adding organic solvents, sometimes in large quantities (e.g., a
chemical and biological activity of Q and QH and their
2
structure.
Experimental
Quinones and hydroquinones used to form the semiquinones
are presented in Fig. 1. Q-1, Q-5, Q-8, QH -7, QH -10 and
2
2
QH -12 were purchased from Aldrich (Steinheim); Q-2 and
2
QH -2 from Merck (Darmstadt); Q-6, QH -4 and QH -12
2
2
2
from Fluka (Neu-Ulm), Q-9 from Lancaster (Muhlheim/
Main), Q-11 from Sigma (Deisenhofen, Germany). Q-4 and
¤ E-mail: bors=gsf.de
QH -3 were a gift from H. B. Stegmann. Q-7 was prepared via
2
J. Chem. Soc., Faraday T rans., 1998, 94(13), 1835È1840
1835

View Article Online
System A
System B
e ~ ] N O(H O) ] N ] OH~ ] HO~
aq
2
2
2
HO~ ] HCOO~ ] CO ~~ ] H O
2
2
(2a) CO ~~ ] Q ] Q~~ ] CO
2
2
e ~ ] N O(H O) ] N ] OH~ ] HO~
aq
2
2
2
HO~ ] N ~ ] HO~ ] N ~
3
3
(2b) N ~ ] QH ] N ~ ] Q~~ ] 2H`
3
2
3
Depending on the Q (QH ) concentration and the dose in the
2
pulse (D), these reactions are accomplished within a few to
tens of microseconds, thereafter only Q~~ remained in the
irradiated solution. The decay of Q~~ occurred much more
slowly (Fig. 2) and thus the processes of build-up of Q~~ and
its decay are, as a rule, separated in time.
Examples of transient absorption spectra associated with
Q~~ are presented in Fig. 3. Parameters of Q~~ spectra are
listed in Table 1. Although spectra were recorded with 9 nm
intervals, and thus are not too precise in the UV region, they
seem to be representative for their basic features. When
spectra could be compared with those reported in the liter-
ature, they were found to be quite similar in appearance.
Spectra of Q~~ 3, 8, 9 and 11 were observed in this work for
the Ðrst time. The shape of the absorption spectra of 1,4-ben-
zosemiquinones changed only moderately with the nature and
position of substituents. In all cases two main peaks were
observed, with maxima at 310È320 nm (short wavelength
peak) and ca. 430 nm (long wavelength peak); the ratio of
intensities of the short wavelength peak to the long wave-
length peak was close to 2È2.5 (Fig. 3 and Table 1). In addi-
tion to these main peaks, spectra commonly displayed one or
more minor peaks whose position and relative intensity
depends on the substitution (Fig. 3). It is signiÐcant that the
Fig. 1 Structures of the quinones, hydroquinones and semiquinones
studied
the oxidation of QH -7 with K Fe(CN) in the mixture 1 : 1 of
2
3
6
benzene and diethyl ester. QH -1, QH -5, QH -6, QH -8,
2
2
2
2
QH -9 and QH -11 were reduced from the corresponding Q
by Zn powder in acetic acid. When the reaction mixture
became colorless, the solvent was removed with a rotary
2
2
evaporator, then QH was extracted with a suitable organic
2
solvent. Both purchased and synthesized Q and QH were
2
puriÐed by recrystallization, sublimation under vacuum or by
shape of the spectra of Q~~ produced from Q and QH ,
using silica gel (40È100 lm) column with CHCl as an eluent.
2
3
respectively, were similar (Fig. 3), although some di†erence in
Sodium phosphates, NaH PO and Na HPO , of highest
2
4
2
4
molar absorption coefficients (e) was observed (see below).
Values of e (in M~1 cm~1) as given in Table 1 were calcu-
lated by the formula
grade used to prepare bu†er solution, were purchased from
Merck. Other reagents were of the highest available grade.
The pulse radiolysis experiments were performed at room
temperature (ca. 22 ¡C) with 50 mM phosphate bu†er, pH
7.40 ^ 0.02. Solutions were prepared with distilled water
treated with Milli-Q system; the pH was adjusted by mixing
50 mM solutions of NaH PO and Na HPO without adding
e \ 106A /0.58D
(I)
'
where A
is the maximal absorbance (see Fig. 2) and D is the
'
dose in a pulse in Gy. Eqn. (I) is deduced under the assump-
2
4
2
4
tion that (i) the yield of the radicals CO ~~ or N ~ is about
any acid or base.
2
3
0.58 lM Gy~1 in both systems, based on the yield of the
primary ~OH radical; (ii) the radicals CO ~~ or N ~ are con-
The pulse radiolysis was performed using a Febetron accel-
erator (100 ns pulses of 2 MeV electrons) and computerized
process control and data handling as described earlier.5,15 To
convert the primary radicals into more selectively reacting
azide (formate) radicals, bu†er was added with 10 mM sodium
formate (when Q was used as a source of Q~~, system A) or 10
2
3
verted completely into Q~~ by reactions (2a) or (2b). Strictly
mM sodium azide (when Q~~ was prepared via QH , system
2
B). In both cases the solution was saturated with N O. Pulse
2
doses of 5È30 Gy were applied to generate 3È18 lM free rad-
icals. Every run was repeated at least three times and then
spectral and kinetic data were averaged to increase the signal-
to-noise ratio. Thiocyanate dosimetry16 with air-saturated
KSCN solution was used for measuring absorbed doses. The
absorption spectra and decay kinetics of the radicals were
evaluated by linear regression analysis of the digitized data.
Results
Absorption spectra and molar absorption coefficients of
semiquinones
Fig. 2 Initial part of the traces at 319 nm showing the build-up and
the beginning of decay of Q~~ produced by pulse radiolysis of various
concentrations of Q-7 in 50 mM phosphate bu†er, pH 7.40, containing
10 mM sodium formate. Dose in the pulse was 15.0 ^ 0.2 Gy in all
runs. (a) 40, (b) 100, (c) 200, (d) 400 lM.
Below are given the basic reactions resulting in the production
of Q~~.
1836
J. Chem. Soc., Faraday T rans., 1998, V ol. 94

View Article Online
Fig. 3 Transient spectra of semiquinones formed during pulse radiolysis of Q (open circles) and QH (Ðlled circles) in 50 mM phosphate bu†er,
2
pH 7.40. A, 500 lM Q-1: dose 15.2 Gy, and 800 lM QH -1: dose 9.2 Gy. B, 400 lM Q-6: dose 6.9 Gy, and 800 lM QH -6: dose 9.2 Gy, C, 400 lM
2
2
Q-9: dose 14.7 Gy, and 500 lM QH -9: dose 7.2 Gy. D, 300 lM Q-11: dose 6.5 Gy, and 500 lM QH -11: dose 8.1 Gy. Spectra were recorded
2
2
10È50 ls after completion of the pulse, when the absorbance peaked. The other details are given in the Experimental section.
speaking, the second assumption is not exactly correct. Along
with reactions (2a) and (2b), primary radicals decay by self-
termination
occurs rather slowly, a signiÐcant decay of Q~~ may start well
before the moment when the process of Q~~ build-up has been
completed (Fig. 2). All these circumstances may be a reason
why a maximal concentration of Q~~ is somewhat lower than
the concentration of a primary radical and the e value calcu-
lated by eqn. (I) is underestimated. It is evident that the higher
CO ~~ ] CO ~~ ] products
2
2
N ~ ] N ~ ] 3N
3
3
2
the concentration of Q or QH and the lower the dose, the
and probably by cross-termination
2
closer is the e value calculated by eqn. (I) to the true value.
CO ~~ ] Q~~ ] products
The values of e presented in Table 1 were determined within a
2
range of [Q] ([QH ]) and doses where the value of e calcu-
N ~ ] Q~~ ] products
2
3
lated by formula (I) no longer depended on the above param-
Besides, when [Q] ([QH ]) is low and the build-up of Q~~
eters.
2
Table 1 Absorption characteristics of semiquinones formed during pulse radiolysis of substituted 1,4-benzoquinones and 1,4-hydroquinones as
well as that of 1,4-naphthaquinone and 1,4-dihydroxynaphthalene in 50 mM phosphate bu†er, pH 7.40
e ] 10~3/M~1 cm~1b(j /nm)b
'
Q/QH a
Q
QH
literature values
ref.
2
2
1
2
19.8 ^ 0.8 (310)
7.8 ^ 0.3 (431)
15.4 ^ 0.3 (319)
7.6 ^ 0.1 (431)
ndf
15.9 ^ 0.3 (310)
6.0 ^ 0.1 (431)
16.0 ^ 0.3 (310)
6.0 ^ 0.2 (431)
15.2 ^ 0.4 (319)
6.7 ^ 0.1 (431)
15.7 ^ 0.3 (319)
6.5 ^ 0.2 (431)
12.5 ^ 0.1 (319)
6.0 ^ 0.1 (431)
14.7 ^ 0.6 (319)
6.8 ^ 0.1 (431)
13.9 ^ 0.4 (319)
6.8 ^ 0.2 (431)
13.1 ^ 0.5 (319)
6.1 ^ 0.2 (431)
12.6 ^ 0.4 (329)
5.3 ^ 0.1 (431)
12.9 ^ 0.2 (319)
6.7 ^ 0.1 (431)
12.2 ^ 0.3 (315)
6.7 ^ 0.1 (435)
11.5 ^ 0.3 (389)
6.1 (431)c
20 (315)d; 7.3 (430)e
6.2 (430)c
17
18
17
3
nrg
4
16.6 ^ 0.3 (319)
7.5 ^ 0.1 (431)
16.2 ^ 0.3 (319)
7.9 ^ 0.5 (431)
16.4 ^ 0.8 (319)
8.4 ^ 0.1 (431)
13.6 ^ 0.7 (319)
6.7 ^ 0.3 (431)
15.7 ^ 0.2 (319)
7.7 ^ 0.1 (431)
13.8 ^ 0.7 (329)
5.2 ^ 0.2 (431)
ndf
6.5 (432)h
6.1 (430) c
19
19
5
6
6.8 (435)c
7.2 (430)i
6.7 (430)c
17
20
17
7
8
nrg
nrg
9
10
11
12
6.7 (435)c
nrg
17
17
12.2 ^ 0.5 (315)
5.3 ^ 0.1 (435)
10.7 ^ 0.4 (389)
12.5 (390)c
a Numbers of Q, QH and Q~~ correspond to those in Fig. 1; b Values of e are given as the mean ^SD from three or more independent
2
experiments, values of e were corrected for the absorption of Q or QH as indicated in the text. Spectral intervals were monitored at 9 nm
2
intervals. c Pulse radiolysis of Q in aqueous solution, pH 7.0, containing 1 M acetone and 1 M isopropanol. d Estimated from the transient
absorbance plot presented in ref. 21 (pulse radiolysis of QH in phosphate bu†er, pH 7.40). e Pulse radiolysis of Q and QH in water at neutral
2
2
pH. f nd, not determined. g nr, not reported, h Pulse radiolysis of Q in phosphate bu†er, pH 7.0, containing 0.1 M acetone and 1 M isopropanol.
i Pulse radiolysis of Q in phosphate bu†er, pH 7.2.
J. Chem. Soc., Faraday T rans., 1998, V ol. 94
1837

View Article Online
Table 2 Rate constants for the disproportionation of semiquinones formed by pulse radiolysis of substituted quinones and hydroquinones in 50
mM phosphate bu†er, pH 7.40a
2k ] 10~8a/M~1 sec~1 (j/nm)
1
Q~~b
Q~~ produced from Q
Q~~ produced from QH
2
1
2
3
4
5
6
7
8
9
1.79 ^ 0.03 (310); 1.83 ^ 0.08 (431)
1.32 ^ 0.03 (319); 1.35 ^ 0.06 (431)
ndd
0.51 ^ 0.06 (319); 0.54 ^ 0.05 (431)
1.39 ^ 0.03 (319); 1.37 ^ 0.13 (431)
1.29 ^ 0.04 (319); 1.49 ^ 0.12 (431)
1.75 ^ 0.11 (319); 1.68 ^ 0.14 (431)
0.43 ^ 0.01 (319); 0.48 ^ 0.02 (431)
0.31 ^ 0.01 (329); 0.33 ^ 0.02 (431)
ndd
1.39 ^ 0.04 (310); 1.34 ^ 0.07 (431)
1.40 ^ 0.04 (310); 1.34 ^ 0.05 (431)
0.88 ^ 0.03 (319); 0.94 ^ 0.05 (431)
0.176 ^ 0.006 (319); 0.187 ^ 0.013 (431)
0.90 ^ 0.01 (319); 0.93 ^ 0.02 (431)
1.20 ^ 0.03 (319); 1.20 ^ 0.04 (431)
1.53 ^ 0.04 (319); 1.57 ^ 0.06 (431)
0.29 ^ 0.01 (319); 0.30 ^ 0.02 (431)
0.34 ^ 0.03 (329); 0.31 ^ 0.03 (431)
1.55 ^ 0.08 (319); 1.56 ^ 0.07 (431)
0.56 ^ 0.02 (315); 0.55 ^ 0.01 (435)
2.85 ^ 0.08 (389)
10
11
12c
0.55 ^ 0.03 (315); 0.51 ^ 0.04 (435)
2.66 ^ 0.06 (389)
a Values of 2k are given as the mean (SD from four or more independent experiments conducted at various concentrations of Q (QH ) and doses
1
2
in pulse. b Structures of Q~~ are given in Fig. 1. c Corrected for the absorption of Q or/and QH as indicated in text. d nd, not determined.
2
The tabulated values of e were corrected for the minor
Kinetics of semiquinone disproportionation
absorbances at the respective wavelengths of the original com-
With all Q~~ the decrease in transient absorbance occurred
nearly synchronously at all wavelengths (in Table 2 the values
are presented at the two major wavelengths for each Q~~). In
pounds, Q and QH , from which Q~~ was produced. Only in
2
three cases did these values exceed 5% (wavelengths are indi-
cated in parentheses): Q-9, 420 (431); Q-11, 420 (431); Q-12
most cases, Q~~ decays only by reaction (1), Q and QH being
830 (389). In most other cases e for Q and QH were 1% less
2
2
the only products. This is supported by the observation that
than that of Q~~ and the correction was not required.
in the mixture of Q and QH , where Q~~ is in equilibrium
When it was possible to compare the e values for Q~~ deter-
mined in this study with those reported in the literature, they
were in a reasonable agreement with each other (Table 1). Yet,
comparison was possible mostly with long wavelength peaks,
as the e values for the short wavelength peaks studied in this
work, with the exception of Q~~-1, were not reported earlier.
It may be noted that the absorption spectra of Q~~ recorded
2
with Q and QH , a steady-state concentration of Q~~ is kept
2
constant under anaerobic conditions for a long time (see ref.
36 for Q~~-9, for the other Q~~ the data will be published
elsewhere).
The decrease in transient absorbance with time that is rep-
resentative of the decay of Q~~ is best described by second
order kinetics to the point where [Q~~] is reduced by 80È90%
from the starting values as is exempliÐed in Fig. 4A. However,
in some cases the kinetics displayed a signiÐcant deviation
from the second order kinetics (Fig. 4B). The latter was evi-
dently caused by the fact that the decrease in the absorbance
due to Q~~ is superimposed on the increase in the absorbance
after pulse radiolysis of CH X substituted 1,4-benzoquinones
2
(X \ Cl, Br, OMe, and OPh)22 and that of tetraÑuoro-1,4-
benzoquinones23 are also in a reasonable agreement with
those obtained here. It should be borne in mind that the pre-
vious values of e are generally obtained in solutions with sig-
niÐcant addition of organic solvents in contrast to our study.
The strong e†ect of di†erent solvents on transient spectra and
e of simple semiquinones has been clearly shown by Rath and
Mukherjee.24
owing to the accumulation of Q and/or QH , the products of
2
Q~~ disproportionation. When this was properly taken into
account, the second order kinetics were observed up to the
very end of transformation (Fig. 4B).
As seen from Table 1, values of e determined for Q~~, pro-
duced from Q are as a rule somewhat higher than those of
Rate constants 2k for Q~~ disproportionation were calcu-
1
lated from experimental rate constants with regard to the e of
Q~~, produced from QH . The reasons for that are somewhat
2
Q~~ presented in Table 1. The determination of 2k was not
ambiguous, as three contributing factors may be involved. (i)
1
a†ected by the change in the concentration of Q (QH ) or the
2
The reported (applied) yields of primary radicals formed via
starting concentration of Q~~, as it is shown in Table 3. The
radiolysis of aqueous solutions saturated with N O or N
2
2
calculated values of 2k are presented in Table 2. Considering
vary in the range 0.56È0.72 lM Gy~1;7,17,20,25h30 this evi-
dently reÑects the increase of the initial yields if high solute
concentrations are used,31 but would be negligible in our
systems as both solutes had the same concentration. (ii) That
1
these data, the following should be noted.
(1) Rate constants determined for the short and long wave-
length peaks agreed with each other within the limits of
it may be due to a slightly higher yield of CO ~~ radicals
2
(system A) as compared to N ~ radicals (system B) is not
3
Table 3 E†ects of [QH ] and dose on the determination of the
reÑected in the various rate constants of H atoms with the
2
second order rate constants for the decay of Q~~ radical anions pro-
solutes. In fact, the opposite trend should prevail as azide
anions react with H atoms about ten times faster than formate
anions (1.9È2.4 ] 109 M~1 s~1, ref. 32 vs. 2.1 ] 108 M~1 s~1,
ref. 33, respectively). (iii) We therefore agree with a suggestion
by a referee, which is actually supported by data of our own,
that the most likely explanation of this di†erence lies in the
observation that certain semiquinones form complexes with
azide, observable by minor spectral and kinetic di†er-
ences.34,35 The fact that we never encountered this com-
plication with polyphenolic Ñavonoids,5 but only with phenols
with adjacent methyl or methoxy groups is in line with the
structures of the presently studied compounds, except that this
phenomenon now seems to include diphenolic structures.
duced by pulse radiolysis of various concentrations of QH -8 in 50
2
mM phosphate bu†er, pH 7.40, containing 10 mM sodium azide
2k ] 10~7/M~1 s~1
1
[QH ]/lM
dose Gy
319 nm
431 nm
2
50
100
150
6.2
7.7
7.2
3.03
3.08
3.22
3.08
2.87
3.25
250
500
15.0
7.2
2.95
2.76
2.75
3.01
500
500
16.4
22.1
2.82
2.63
2.75
3.01
average
2.93 ^ 0.16
2.96 ^ 0.15
1838
J. Chem. Soc., Faraday T rans., 1998, V ol. 94

View Article Online
OMe);22 2.3 ] 108 M~1 sec~1 for Q~~ formed from duroquin-
one.1 The above values are not far from those determined in
this work (Table 2).
Discussion
The number of studied Q~~ is insufficient to consider the
correlation of 2k with the substitution in Q~~ in detail; only
1
some tentative conclusions can be made. The number and the
position of methyl substituents in Q~~ has only moderate
e†ect on 2k ; the value varies within the narrow range from
1
0.9 ] 108 to 1.8 ] 108 M~1 s~1 (Table 2). Notice that the
reported 2k value of 2.3 ] 108 M~1 s~1 for tetramethyl-1,4-
1
benzosemiquinone1 is not far from the values given above. At
the same time, substitution with more bulky alkyl groups,
iso-Pr and tert-Bu (Q~~-4 and Q~~-8), as well as that with a
methoxy group (Q~~-9 and Q~~-11), results in an evident
decrease in 2k (Table 2).
1
The reduced reactivity of methoxy-substituted Q~~-9 and
Q~~-11 towards the disproportionation correlates with the
outstanding antitumor capability of the corresponding
Q.10h12 Notice that two more very potent antitumor benzo-
quinones, 2,5-bis(carboethoxyamino)-3,6-diaziridinyl-1,4-ben-
zoquinone (AZQ) and 2,5-bis(2-hydroxyethylamino)-3,6-
diaziridinyl-1,4-benzoquinone (BZQ) were reported to
produce very stable Q~~ (2k \ 8.9 ] 105 and 4.6 ] 105 M~1
1
Fig. 4 The fulÐllment of the second order kinetic law for the decay
of Q~~ produced by pulse radiolysis of QH -1 (A) and QH -11 (B). A:
s~1 for AZQ and BZQ, respectively).40 The correlation
between a low value of 2k and the pronounced antitumor
2
2
1
1, kinetic trace at 431 nm; 2, linear plot according to the second order
capability of Q may be attributable to the fact that reactions
equation; conditions: 800 lM QH -1, dose, 15.2 Gy. B: 1, kinetic trace
2
with the participation of Q~~ contribute signiÐcantly to prob-
at 435 nm; 2, linear display of the second order equation; trace 1@ and
able molecular mechanisms of Q cytotoxicity.1,3 All other
plot 2@ are the same as 1 and 2 but with subtraction of the absorbance
owing to Q formed via the disproportionation of Q~~; conditions: 500
things being equal, the lower the 2k value, the higher is the
1
latter value. In conclusion, a low value of 2k can be con-
1
lM QH -11, dose 23.5 Gy. The other details are given in the Experi-
2
mental section.
sidered as a likely contributing factor for Q antitumor activ-
ity.
experimental error. Among other things, the use of a short
wavelength peak for kinetic measurements thus provides the
opportunity to increase the sensitivity of the [Q~~] determi-
nation by a factor of ca. 2.
The work was Ðnancially supported by the Deutsche For-
schungsgemeinschaft (grant 436Rus 113/245/0) and the
Russian Foundation for Fundamental Researches (grant
96-03-00103).
(2) In most cases there was no signiÐcant di†erence in 2k
between the systems where Q~~ was produced from Q and
QH . With Q/QH 1, 5 and 8 the moderate di†erence might
1
References
2
2
be almost compensated if the same values of the e of Q~~
1
2
3
P. J. OÏBrien, Chem.-Biol. Interact., 1991, 80, 1.
G. Powis, Free Radical Biol. Med., 1989, 6, 63.
A. Denmark and E. Cadenas, Free Radical Biol. Med., 1989, 7,
435.
obtained with Q or QH were used in the calculation of 2k .
2
1
However, in the case of Q~~-4 the di†erence in 2k remained
1
substantial even after correction, i.e. when Q~~-4 was produc-
4
5
P. Wardman, Free Radical Res. Commun., 1990, 8, 219.
W. Bors, C. Michel and S. Schikora, Free Radical Biol. Med.,
1995, 19, 45.
ed from QH , the value of 2k was found to be less by a
2
1
factor of ca. 3. As we observed for Trolox34 and Etoposide,35
the kinetic changes due to complex formation of phenoxyl
radicals with azide are small, but not insigniÐcant and may
therefore explain these di†erences.
6
7
8
9
M. C. Rath, H. Pal and T. Mukherjee, J. Chem. Soc., Faraday
T rans., 1996, 92, 1891.
I. Wilson, P. Wardman, T-S. Lin and A. S. Sartorelli, J. Med.
Chem., 1986, 29, 1381.
(3) Only with a few Q~~ could the values of 2k determined
1
H. Pal, D. K. Palit, T. Mukherjee and J. P. Mittal, Radiat. Phys.
Chem., 1992, 40, 529.
in this study be compared with those reported in the liter-
ature. Values from 8 ] 107 to 1 ] 109 M~1 sec~1, reported as
H. Pal, T. Mukherjee and J. P. Mittal, Radiat. Phys. Chem., 1994,
44, 603.
2k for disproportionation of Q~~-1 at pH close to 7,1,37h39
1
and that for Q~~-12 (2 ] 108 M~1 sec~1)24 are in a reasonable
10 R. Pethig, P. R. C. Gascoyne, J. A. McLaughlin and A. Szent-
Gyorgyi, Proc. Natl. Acad. Sci., USA, 1983, 80, 129.
11 C. D. Morgan, S. H. Combs and J. Everse, Proc. Soc. Exp. Biol.
Med., 1995, 208, 294.
agreement with 2k determined in this study (Table 2).
1
However, reported values of 2k for Q~~-4 (7.5 ] 108
1
M~1 s~1)19 di†er greatly from those determined in this study
(note that in this case also the greatest discrepancy occurred
for the decay values at the two absorbance peaks). Further-
12 V. A. Roginsky, G. Bruchelt and O. Bartuli, Biochem. Pharmacol.,
1998, 55, 85.
13 P. Wardman, J. Phys. Chem. Ref. Data, 1989, 18, 1637.
14 J. Butler and B. M. Hoey, Free Radical Biol. Med., 1992, 12, 337.
15 M. Saran, G. Vetter, M. Erben-Russ, R. Winter, A. Kruse, C.
Michel and W. Bors, Rev. Sci. Instrum., 1987, 58, 363.
16 G. E. Adams, J. W. Boag, J. Currant and B. D. Michael, in Pulse
Radiolysis, ed. M. Ebert, J. P. Keene, J. Swallow and J. H. Baxen-
dale, Academic Press, London, 1965.
more it should be noted that the 2k determination reported
1
by Dohrmann and Bergmann19 was performed with solutions
containing signiÐcant amounts of organic solvents. Below are
given some more values of 2k reported in the literature for
1
disproportionation of Q~~ with structures similar to those
studied in this work: (0.7È1.8) ] 108 M~1 sec~1 for mono-
17 K. B. Patel and R. L. Willson, J. Chem. Soc., Faraday T rans. 1,
CH X-substituted 1,4-benzosemiquinones (X \ Cl, Br,
1973, 69, 814.
2
J. Chem. Soc., Faraday T rans., 1998, V ol. 94
1839

View Article Online
18 G. E. Adams and B. D. Michael, T rans. Faraday Soc., 1967, 63,
1171.
31 R. H. Schuler, A. L. Hartzell and B. Behar, J. Phys. Chem., 1981,
85, 192
19 J. K. Dohrmann and B. Bergmann, J. Phys. Chem., 1995, 99,
32 M. Ye, K. P. Madden, R. W. Fessenden and R.H. Schuler, J.
Phys. Chem., 1986, 90, 5397
1218.
20 Y. A. Ilan, G. Czapski and D. Meisel, Biochim. Biophys. Acta,
1976, 430, 209.
33 http://allen.rad.nd.edu
34 R. Goldman, W. Bors, C. Michel, B. W. Day and V. E. Kagan,
Environ. Nutr. Interact., 1997, 1, 97.
35 W. Bors and C. Michel, in L ipoic Acid in Health and Disease, ed.
L. Packer, J. Fuchs, G. Zimmer, Marcel Dekker, New York,
1997, p. 33.
21 E. J. Land, J Chem. Soc., Faraday T rans., 1993, 88, 803.
22 K. E. OÏShea and M. A. Fox, J. Am. Chem. Soc., 1991, 113, 611.
23 L. C. T. Shoute and J. P. Mittal, J. Phys. Chem., 1994, 98, 11 094.
24 M. C. Rath and T. Mukherjee, Radiat. Phys. Chem., 1997, 49, 29.
25 B. H. J. Bielski, D. A. Comstock and R. A. Bowen, J. Am. Chem.
Soc., 1971, 93, 5624.
36 V. A. Roginsky, G. Bruchelt and H. B. Stegmann, Biochemistry
(Moscow) 1988, 63, 240.
26 J. Butler, E. J. Land, A. J. Swallow and W. A. Pru
Ž
tz, Radiat.
37 D. Meisel, Chem. Phys. L ett., 1975, 34, 263.
38 I. Yamazaki and T. Ohnishi, Biochim. Biophys. Acta, 1966, 112,
469.
Phys. Chem., 1984, 23, 265.
27 Z. B. Alfassi and R. H. Schuler, J. Phys. Chem., 1985, 89, 3359.
28 I. Gyorgy, A. Blazovics, J. Feher and G. Foldiak, Radiat. Phys.
Chem., 1990, 36, 165.
39 H.R. Cooper, T rans. Faraday Soc., 1966, 62, 2865.
40 J. Butler, B. M. Hoey and J. S. Lea, Biochim. Biophys. Acta, 1987,
925, 144.
29 J. Lind, X. Shen, T. E. Eriksen and G. Merenyi, J. Am. Chem.
Soc., 1990, 112, 479.
30 J. A. LaVerne and S. M. Pimblott, J. Chem. Soc., Faraday T rans.,
1993, 89, 3527.
Paper 8/01082J; Received 6th February, 1998
1840
J. Chem. Soc., Faraday T rans., 1998, V ol. 94