Free Radical Biology & Medicine, Vol. 27, Nos. 7/8, pp.

864 – 872, 1999

Copyright © 1999 Elsevier Science Inc.
Printed in the USA. All rights reserved
0891-5849/99/$–see front matter

PII S0891-5849(99)00134-3

Original Contribution
REASSIGNMENT OF ORGANIC PEROXYL RADICAL ADDUCTS

SERGEY I. DIKALOV1 and RONALD P. MASON

Laboratory of Pharmacology and Chemistry, National Institute of Environmental Health Sciences, National Institutes of Health,
Research Triangle Park, NC, USA

(Received 28 April 1999; Revised 14 June 1999; Accepted 14 June 1999)

Abstract—The study of the important role of peroxyl radicals in biological systems is limited by their difficult detection
with direct electron spin resonance (ESR). Many ESR spectra were assigned to 5,5-dimethyl-1-pyrroline N-oxide
(DMPO)/peroxyl radical adducts based only on the close similarity of their ESR spectra to that of DMPO/superoxide
radical adduct in conjunction with their insensitivity to superoxide dismutase, which distinguishes the radical adduct
from DMPO/superoxide radical adduct. Later, the spin-trapping literature reported that DMPO/peroxyl radical adducts
have virtually the same hyperfine coupling constants as synthesized alkoxyl radical adducts, raising the issue of the
correct assignment of peroxyl radical adducts. However, using 17O-isotope labelling, the methylperoxyl and methoxyl
radical adducts should be distinguishable. We have reinvestigated the spin trapping of the methylperoxyl radical. The
methylperoxyl radical was generated in aerobic solution with 17O-molecular oxygen either in a Fenton system with
dimethylsulfoxide or in a chloroperoxidase system with tert-butyl hydroperoxide. Two different spin traps, DMPO and
2,2,4-trimethyl-2H-imidazole-1-oxide (TMIO), were used to trap methylperoxyl radical. 17O-labelled methanol was
used to synthesize methoxyl radical adducts by nucleophylic addition. It was shown that the 17O hyperfine coupling
constants of radical adducts formed in methylperoxyl radical– generating systems are identical to that of the methoxyl
radical adduct. Therefore, methylperoxyl radical–producing systems form detectable methoxyl radical adduct, but not
detectable methylperoxyl radical adducts at room temperature. One of the possible mechanisms is the decomposition of
peroxyl radical adduct with the formation of secondary alkoxyl radical adduct. These results allow us to reinterpret
previously published data reporting detection of peroxyl radical adducts. We suggest that detection of 17O-alkoxyl
radical adduct from 17O-labelled molecular oxygen can be used as indirect evidence for peroxyl radical generation.
© 1999 Elsevier Science Inc.

Keywords—Peroxyl radical, Alkoxyl radical, Spin trap, Radical adduct, Peroxidase, oxygen-17, DMPO, TMIO, Free
radicals

INTRODUCTION lifetimes and can not be observed directly in biological

samples.
Peroxyl radicals play an important role in chemistry and
The spin trapping technique has been successfully
biology [1]. They are involved in many radical chain
used for detection of various oxygen-centered radicals
reactions, for example lipid peroxidation and protein
[6]. It was proposed that spin traps could be used for
damage [1,2]. Tertiary peroxyl radicals can be detected
detection of peroxyl radicals [7–9]. However, it was
directly by electron spin resonance (ESR) [3]. However,
found that peroxyl radical adducts of the spin trap N-tert-
primary and secondary alkylperoxyl radicals dispropor-
butyl-␣-phenylnitrone (PBN) are very unstable [7,8].
tionate with a nearly diffusion-controlled, bimolecular
Janzen et al. reported that peroxyl radical adducts of
rate constant [4,5] and, therefore, have extremely short
PBN are not persistent at temperatures above 230°K [8].
Moreover, it was found that at a temperature higher than
Address correspondence to: Sergey Dikalov, Laboratory of Pharma-
cology and Chemistry, National Institute of Environmental Health 250°K only the alkoxyl radical adduct was detected [8].
Sciences/National Institutes of Health, 111 Alexander Drive, PO Box Beginning in 1980 [9], many ESR spectra were as-
12233, Research Triangle Park, NC 27709, USA; Tel: (919) 541-2687; signed to 5,5-dimethyl-1-pyrroline N-oxide (DMPO)/
Fax: (919) 541-1043; E-Mail: dikalov@niehs.nih.gov. .
1
Permanent Address: Institute of Chemical Kinetics & Combustion, peroxyl radical adducts based only on the close similarity
Novosibirsk, 630090, Russia of their ESR spectra to DMPO/superoxide radical adduct

864
 Reassignment of organic peroxyl radical adducts
in conjunction with their insensitivity to superoxide dis-
mutase [3], which distinguishes the radical adduct from
DMPO/superoxide radical adduct. Moreover, according
to the current literature, the hyperfine coupling constants
for some primary peroxyl and alkoxyl radical adducts are
reported to be virtually identical [9 –11]. However, this is
very unlikely because of differences in the electronic
structure of peroxyl and alkoxyl radical adducts [7,8]. It
was reported that the chemical structure of these radical
adducts could be identified using 17O-centered radicals
[7,12]. In this case, hyperfine coupling constants from
17
O for alkylperoxyl and alkoxyl radical adducts are
expected to be different.
In this article we used two different spin traps, DMPO
[13] and 2,2,4-trimethyl-2H-imidazole-1-oxide (TMIO)
[14,15], to trap methylperoxyl radical. Now we report
that methylperoxyl radicals produce ESR-detectable meth-
oxyl radical adducts, but not methylperoxyl radical ad-
ducts. This allows us to reinterpret previously published
data reporting detection of DMPO/methylperoxyl radical
[11,16,17] adduct and, presumably, many other DMPO/
peroxyl radical adducts. According to our data, previ-
ously reported formation of DMPO/cumylperoxyl and
DMPO/tert-butylperoxyl radical adducts [9,18 –23],
which are known to be methyl radical-derived, oxygen-
dependent, and assigned as DMPO/methylperoxyl radi-
cal adduct [11,16,17], are most probably the DMPO/
methoxyl radical adduct. We suggest that formation of
the alkoxyl radical adduct from the originally trapped
peroxyl radical is a general phenomenon that is not
dependent on the nitrone spin trap or peroxyl radical
structures.
MATERIALS AND METHODS
Chemicals
Spin trap DMPO was obtained from Aldrich (Milwau-
kee, WI, USA). DMPO was vacuum-distilled twice at
room temperature and stored under nitrogen at ⫺70°C.
Oxygen-17O2 (not less than 85 atom % 17O) and meth-
anol-17O (20 atom % 17O) were from ISOTEC, Inc.
(Miamisburg, OH, USA). Spin trap TMIO was obtained
from Alexis Corporation (Läufelfingen, Switzerland)
[14]. Hydrogen peroxide was from Fisher Scientific (Fair
Lawn, NJ, USA). Deferoxamine (Desferal), manganese
superoxide dismutase (EC 1.15.1.1), dimethylsulfoxide
(DMSO), tert-butyl hydroperoxide (tBuOOH), FeCl3,
FeSO4, and chloroperoxidase were purchased from
Sigma Chemical Co. (St. Louis, MO, USA). Analytical
grade Chelex-100 (Bio-Rad Laboratories, Hercules, CA,
USA) was used.
865
ESR spin trapping experiments
The ESR spectra were recorded using a Bruker EMX
spectrometer operating at 9.78 GHz with a modulation
frequency of 100 kHz and a super-high Q microwave
cavity. All ESR samples were placed in a 10-mm flat
cell. In order to perform experiments with a lower con-
centration of transition metals, phosphate buffer was
treated with 5 g of Chelex-100 per 100 ml of solution for
2 h followed by filtration using Millex-HA 0.45-␮m
filters (Millipore Corp., Bedford, MA, USA). The ESR
instrumental settings for experiments with methylper-
oxyl radical generation were as follows: field sweep, 80
G; microwave frequency, 9.78 GHz; microwave power,
40 mW; modulation amplitude, 0.5 G; conversion time,
656 ms; time constant, 1310 ms; and receiver gain, 1 ⫻
105. ESR spin-trapping experiments were done at least
three times. The ESR instrumental settings for experi-
ments with methanol plus ferric ion system were the
following: field sweep, 80 G; microwave frequency, 9.78
GHz; microwave power, 20 mW; modulation amplitude,
0.5 G; conversion time, 328 ms; time constant, 656 ms;
and receiver gain, 1 ⫻ 105.
Synthesis of methoxyl radical adducts
DMPO-methoxyl (DMPO/•OCH3) radical adduct was
synthesized in an aqueous solution of 0.5 mM FeCl3 plus
5% methanol [10,24]. TMIO/methoxyl radical adduct
was synthesized in an aqueous solution of 0.5 mM FeCl3
plus 20% methanol. To prepare DMPO/•17OCH3 radical
adduct, 17O-methanol was used (it contained 20% of
CH3-17OH). The reaction was initiated by the addition of
50 mM DMPO or TMIO to a water solution of FeCl3
plus methanol.
Spin trapping in a Fenton system with DMSO
A Fenton system with DMSO was used to generate
methylperoxyl radical: 0.5 mM FeSO4 was added to 1
mM H2O2 with 7% DMSO in 150 mM sodium-phos-
phate buffer, pH 7.40. DMPO (5 mM) or TMIO (10 mM)
was used as a spin trap. Some spin-trapping experiments
were done in the presence of 500 U/ml manganese-
containing superoxide dismutase.
Spin trapping in a chloroperoxidase system with
tert-butyl hydroperoxide
Chloroperoxidase with tert-butyl hydroperoxide was
used to generate methylperoxyl radical adduct. For this
purpose 5 ␮M chloroperoxidase was added to a solution
of spin trap with 50 mM tert-butyl hydroperoxide in 150
mM sodium-phosphate buffer, pH 7.40. The spin trap-
866 S. I. DIKALOV and R. P. MASON

ping experiments were done in the presence of 500 U/ml

manganese superoxide dismutase.

Computer simulation
Computer simulations and spin trap database searches
were performed using programs that are available to the
public through the Internet (http://epr.niehs.nih.gov/).
The details of this computer simulation program have
been described elsewhere [25]. Hyperfine coupling con-
stants are expressed as an average of ESR parameters
obtained from computer simulations using at least three
experimental spectra, which provided accuracy not less
than 0.05 G.

RESULTS

Synthesis of DMPO-methoxyl radical adduct

Previously, it was shown that the DMPO/•OCH3 radical
adduct can be synthesized by mixing the spin trap DMPO
with a water solution of ferric salt plus methanol [10,24].
Scheme 1 shows the mechanism of this reaction. We used
CH3OH (Fig. 1, spectrum A) and CH3-17OH (Fig. 1, spec-
tra B and D) to synthesize the DMPO/methoxyl radical
adducts. Computer simulations of the latter spectrum (Fig.
1 spectra C and E) demonstrate the presence of two nitrox-
ides, namely, DMPO/•OCH3 and DMPO/•17OCH3 (Fig. 1,
spectra F and G). Both radical adducts have hyperfine
coupling constants from nitrogens, ␤-hydrogen, and ␥-hy-
drogen (Table 1). The hyperfine coupling constant from
17
O, with spin 5/2, is 6.51 G for DMPO/•17OCH3, which is
very different from either the hydroxyl or the superoxide
radical adduct (Table 1). The hyperfine coupling constants
for DMPO/•16OCH3 are very close to the literature values
[10] and very similar to those reported as DMPO/•OOCH3
[11, 17].
Spin trapping in a Fenton system with
dimethylsulfoxide using DMPO
It is known that the Fenton system (Fe2⫹ ⫹ H2O2)
reacts with DMSO to produce the methyl radical [26]
which, in the presence of oxygen, gives the methylper-
oxyl radical (Scheme 2). We used this Fenton system
with DMSO to generate methylperoxyl radical.
In the Fenton system in the presence of DMPO we
found the four-line ESR spectrum of the hydroxyl radical
adduct (Fig. 2, spectrum A, a N ⫽ 14.98 G and a H␤ ⫽
14.75 G). Addition of DMPO to a nitrogen-bubbled
Fenton system with DMSO produced the six-line ESR
spectrum of the methyl radical adduct (Fig. 2, spectrum
B, a N ⫽ 16.36 G and a H␤ ⫽ 23.40 G). Spin trapping
under aerobic conditions revealed significant formation
Fig. 1. ESR spectra of DMPO/methoxyl radical adducts. (Spectrum A)
ESR spectrum of an aqueous solution of DMPO (50 mM) with 5%
methanol plus FeCl3 (0.5 mM). (Spectrum B) ESR spectrum of an
aqueous solution of DMPO (50 mM) with 5% methanol-17O (20 atom
% 17O) plus FeCl3 (0.5 mM). (Spectrum C) Composite computer
simulation of spectrum B. (Spectrum D) ESR spectrum B with 10-fold
higher gain. (Spectrum E) Composite computer simulation of spectrum
D. (Spectrum F) Computer simulation of the DMPO/•16OCH3 compo-
nent (molar ratio 0.75) of spectrum E. (Spectrum G) Computer simu-
lation of the DMPO/•17OCH3 component (molar ratio 0.20) of spectrum
E. A minor contribution of DMPO/•OH (molar ratio 0.05) is present in
the composite simulation E.
of an O-centered radical adduct (Fig. 2, spectrum C, a N
⫽ 14.53 G, a H␤ ⫽ 10.75 G, and a H␥ ⫽ 1.33 G) with
traces of the methyl radical adduct. The addition of 500
U/ml of manganese containing superoxide dismutase did
not affect the spin trapping results (data not shown). In
order to determine the chemical structure of the O-
centered radical adduct detected (Fig. 2, spectrum D), we
performed spin trapping experiments with a 17O2-bub-
bled solution (Fig. 2, spectrum E). The ESR spectrum of
the 17O2-bubbled solution contains additional lines (Fig.
2, spectrum E) for the hyperfine coupling constant of 17O
that, with spin 5/2, splits each line of the 16O-radical
adduct into six lines. Computer simulations of the 16O-
radical adduct (Fig. 2, spectra D and G) and 17O-radical
adduct (Fig. 2, spectrum H) were nearly identical to that
of the DMPO/•OCH3 radical adduct (Table 1). In partic-
ular, the hyperfine coupling constant for 17O was 6.50 G,
 Reassignment of organic peroxyl radical adducts 867

Table 1. Hyperfine Coupling Constants of Radical Adducts of Various Oxygen-Centered Radicals

Radical
17
adduct N H␤ H␥ O Experiment condition Ref.
•
PBN/ O2R1 13.42 0.95 — 2.9 Photolysis of tBu2CO ⫹ O2, ⫺60°C
17
7
PBN/•OR1 13.62 1.72 — 5.05 Photolysis of tBu2CO ⫹ 17O2, O°C 7
PBN/•O2R2 12.84 1.23 — — a13C ⫽ 5.05, photolysis of AIBN, 205°K 8
PBN/•OR2 13.87 2.06 — — a13C ⫽ 4.7, photolysis of AIBN, 250°K 8
DMPO/•OH 14.9 14.8 — 4.68 DMPO ⫹ Fe3⫹ in H2O17 10
DMPO/•OH 15.01 15.01 — 4.66 Xanthine oxidase ⫹ Fe2⫹ ⫹ 17O2 12
DMPO/•OH 15.04 14.80 — 4.68 DMPO ⫹ Fe2⫹ ⫹ H217O2 32
DMPO/•O2H 14.2 11.34 1.25 5.9 Microsomes ⫹ paraquat ⫹ 17O2 12
DMPO/•OCH3 14.5 10.7 1.34 — DMPO ⫹ Fe3⫹ ⫹ 5% CH3OH 10
DMPO/•OCH3 14.45 10.61 1.35 6.51 DMPO ⫹ Fe3⫹ ⫹ 5% CH3⫺17OH this work
DMPO/•OCH3 14.53 10.72 1.33 6.50 DMPO ⫹ Fe2⫹ ⫹ 5% DMSO ⫹ H2O2 ⫹ 17
O2 this work
DMPO/•OCH3 14.43 10.70 1.32 6.50 Chloroperoxidase ⫹ tBuOOH ⫹ 17O2 this work
TMIO/•OCH3 13.62 15.42 — — TMIO ⫹ Fe3⫹ ⫹ 20% CH3OH this work
TMIO/•OCH3 13.65 15.40 — 5.71 TMIO ⫹ Fe3⫹ ⫹ 20% CH3 ⫹ 17OH this work
TMIO/•OCH3 13.60 15.40 — 5.62 DMPO ⫹ Fe2⫹ ⫹ 5% DMSO ⫹ H2O2 ⫹ 17
O2 this work
TMIO/•OCH3 13.67 15.49 — 5.70 Chloroperoxidase ⫹ tBuOOH ⫹ 17O2 this work

R1 ⫽ tert-butyl
R2 ⫽ 2-cyano-2-propyl

which was the same as for the DMPO/• 17OCH3 radical
adduct obtained in the methanol system (Table 1).
Spin trapping system with tert-butyl hydroperoxide
using DMPO
Previously, it was found that chloroperoxidase with
tert-butyl hydroperoxide produces the tert-butoxyl radi-
cal [20], which is decomposed through ␤-scission into
acetone and the methyl radical (Scheme 3). The methyl
radical reacts with oxygen to give the methylperoxyl
radical. Therefore, at a high spin trap concentration (⬎
160 mM), the primary tert-butoxyl radical will be
trapped whereas at a low spin trap concentration (⬍ 40
mM), DMPO will mainly react with its decomposition
products, the methyl or methylperoxyl radicals [11].
In the sample of chloroperoxidase with tert-butyl hy-
droperoxide plus 0.4 M DMPO, we found a four-line
ESR spectrum of the tBuO• radical adduct as expected
(Fig. 3, spectrum A). Addition of 10 mM DMPO to
nitrogen-bubbled chloroperoxidase with tert-butyl hy-
droperoxide produced an ESR spectrum (Fig. 3, spec-
trum B) that resulted from superposition of the previ-
ously reported [13,22] four-line ESR spectrum of
DMPO/•OtBu radical adduct (aN ⫽ 14.91 G and aH␤ ⫽
16.04 G) and the six-line ESR spectrum of the methyl
radical adduct (Fig. 2, spectrum B, aN ⫽ 16.36 G and
aH␤ ⫽ 23.40 G). At this low DMPO concentration, spin
trapping under aerobic conditions revealed significant
formation of an O-centered radical adduct (aN ⫽ 14.40
G, aH␤ ⫽ 10.70 G, and aH␥ ⫽ 1.32 G) and a trace amount
of the tBuO• radical adduct (Fig. 3, spectrum C). The
addition of 500 U/ml of manganese-containing superox-
ide dismutase did not affect the spin trapping results
(data not shown). In order to clarify the chemical struc-
ture of the O-centered radical adduct observed (Fig. 3,
spectrum C), we performed spin-trapping experiments in
a 17O2-bubbled solution (Fig. 3, spectrum D). The ESR
spectrum of the 17O2-bubbled solution contains addi-
tional lines (Fig. 3, spectrum D) from the hyperfine
coupling of 17O. Computer simulation of this spectrum
(Fig. 3, spectrum E) required three radical adducts. The
first was DMPO/•OtBu radical adduct (Fig. 3, spectrum
F); the second one was DMPO/•16OCH3 radical adduct
(Fig. 3, spectrum G); and the third one was the DMPO/
•17
OCH3 radical adduct (Fig. 3, spectrum H). The hyper-
fine coupling constant for 17O was 6.50 G, which was the
same as that found for the DMPO/•17OCH3 radical ad-
duct obtained in the methanol system (Table 1). The ESR
spectra of DMPO/•OtBu and DMPO/•OCH3 are very
distinct (Fig. 3, spectra A and C) because of the differ-
ences in the conformation of the alkoxyl radical adducts
due to the steric effect of the bulky tert-butyl group,
which strongly affects the hyperfine coupling constants
of DMPO radical adducts [6,29].

Synthesis of TMIO-methoxyl radical adduct

Previously, it was shown that the mixing of the ni-
Scheme 1. Synthesis of methoxyl radical adduct. trone compound DMPO with ferric salt plus methanol
868 S. I. DIKALOV and R. P. MASON

Scheme 2. Formation of radical adducts in hydroxyl radical– generating

system with DMSO and DMPO.

5 (Fig. 5, spectrum B) consisted of three radical adducts,

TMIO/•OCH3, TMIO/•OH, and TMIO/•17OCH3 (Fig. 5
Fig. 2. Formation of radical adducts in the Fenton system with DMSO spectra C, D, and E). Hyperfine coupling constants of
and spin trap DMPO. ESR spectrum of a phosphate buffer solution these radical adducts are described in Table 1. This is the
(0.15 M, pH 7.4) of DMPO (5 mM) and H2O2 (1 mM) initiated with first report to our knowledge of the TMIO/•OCH3 radical
FeSO4 (0.5 mM). (Spectrum B) Same as in (spectrum A) with 7%
DMSO, bubbled with nitrogen for 20 min and then initiated with FeSO4 adduct with coupling constants aN ⫽ 13.65 G, aH␤ ⫽
(0.5 mM). (Spectrum C) Same as in (spectrum A) with 7% DMSO, 15.40 G, and a17O ⫽ 5.71 G.
bubbled with oxygen for 2 min, and then initiated with FeSO4 (0.5
mM). (Spectrum D) Computer simulation of the major species in
(spectrum C). (Spectrum E) Same as in (spectrum B) bubbled with 17O2
(85 atom % 17O) for 2 min, and then initiated with FeSO4 (0.5 mM).
(Spectrum F) Composite computer simulation of spectrum E. (G)
Computer simulation of the DMPO/•16OCH3 component (molar ratio
0.16) of spectrum E. (Spectrum H) Computer simulation of the DMPO/
•17
OCH3 component (molar ratio 0.79) of spectrum E. A minor con-
tribution of DMPO/•CH3 (molar ratio 0.05) is present in the composite
simulation E.

produces the methoxyl radical adduct [10,24]. We used

the same reaction (Scheme 1) for the nitrone-type spin
trap TMIO. The ESR spectrum of a water solution of
TMIO with 20% methanol plus FeCl3 is shown in Fig.
4A. Computer simulation (Fig. 4, spectrum B) of spec-
trum A in Fig. 4 revealed two radical adducts, TMIO/
•
OCH3 and TMIO/•OH (Fig. 4, spectra C and D). The
presence of TMIO/•OH is due to parallel nucleophilic
addition of both water and methanol to the spin trap
molecule [10]. The ESR spectrum obtained with CH3-
17
OH revealed a hyperfine splitting from 17O (Fig. 5,
spectrum A). Computer simulation of spectrum A in Fig. Scheme 3. Radical formation by peroxidase with tBuOOH and DMPO.
 Reassignment of organic peroxyl radical adducts
Fig. 3. Formation of radical adducts in the chloroperoxidase system
with tert-butyl hydroperoxide and spin trap DMPO. (Spectrum A) ESR
spectrum of phosphate buffer solution (0.15 M, pH 7.4) of DMPO (0.4
M) with t-BuOOH (50 mM) plus CPO (5 ␮M). (Spectrum B) ESR
spectrum of nitrogen-bubbled (20 min) phosphate buffer solution (0.15
M, pH 7.4) of DMPO (10 mM) with tBuOOH (50 mM) plus CPO (5
␮M). (Spectrum C) ESR spectrum of aerobic phosphate buffer solution
(0.15 M, pH 7.4) of DMPO (10 mM) with tBuOOH (50 mM) plus CPO
(5 ␮M). (Spectrum D) ESR spectrum of 17O2 (85 atom % 17O)-bubbled
(2 min) phosphate buffer solution (0.15 M, pH 7.4) of DMPO (10 mM)
with tBuOOH (50 mM) plus CPO (5 ␮M). (Spectrum E) Composite
computer simulation of spectrum D. (Spectrum F) Computer simula-
tion of the DMPO/•OC(CH3)3 component (molar ratio 0.20) of spec-
trum D. (Spectrum G) Computer simulation of the DMPO/•16OCH3
component (molar ratio 0.15) of spectrum D. (H) Computer simulation
of the DMPO/•17OCH3 component (molar ratio 0.65) of spectrum D.
Spin trapping in a chloroperoxidase system with
tert-butyl hydroperoxide using TMIO
As discussed above, chloroperoxidase with tert-butyl
hydroperoxide produces tert-butoxyl radical [20], which
decomposes to acetone and methyl radical, which then
reacts with oxygen to give methylperoxyl radical
(Scheme 3). Thus, at high spin trap concentration, tert-
butoxyl radical will be trapped whereas at lower spin trap
concentrations, the spin trap will react with methyl or
methylperoxyl radicals [11].
Chloroperoxidase with tert-butyl hydroperoxide in the
presence of 0.2 M TMIO gave a six-line ESR spectrum
of the tBuO• radical adduct as expected (Fig. 6, spectrum
A). Addition of 10 mM TMIO to nitrogen-bubbled chlo-
roperoxidase with tert-butyl hydroperoxide produced an
869
Fig. 4. ESR spectra of TMIO radical adducts formed in 20% methanol
with ferric ions. (Spectrum A) ESR spectrum of an aqueous solution of
TMIO (50 mM) with 20% methanol plus FeCl3 (0.5 mM). (Spectrum
B) Composite computer simulation of spectrum A. (Spectrum C) Com-
puter simulation of the TMIO/•OCH3 component (molar ratio 0.75) of
spectrum A. (D) Computer simulation of the TMIO/•OH component
(molar ratio 0.25) of spectrum A.
ESR spectrum (Fig. 6, spectrum B) that consisted of the
previously reported [14,15] tBuO• radical adduct (aN ⫽
14.13 G and aH␤ ⫽ 16.36 G) and the methyl radical
adduct (aN ⫽ 15.69 G and a H␤ ⫽ 22.33 G). At this low
TMIO concentration, spin trapping under aerobic condi-
tions revealed significant formation of the O-centered
radical adduct (aN ⫽ 13.65 G and aH␤ ⫽ 15.43 G), a
small amount of the tBuO• radical adduct (Fig. 5, spec-
trum C), and traces of the methyl radical adduct. In order
to determine the chemical structure of the O-centered
radical adduct (Fig. 6, spectrum C), we performed the
spin trapping experiments in a 17O2-bubbled solution.
The ESR spectrum from the 17O2-bubbled solution con-
tains additional lines from the 17O-hyperfine coupling
constants (Fig. 6, spectrum D). Computer simulation of
spectrum D (Fig. 6, spectrum E) revealed three major
radical adducts. The first was the TMIO adduct of the
tBuO• radical (Fig. 6, spectrum F). The second was the
870 S. I. DIKALOV and R. P. MASON
Fig. 5. ESR spectra of TMIO radical adducts formed in 20% 17O-
methanol with ferric ions. (Spectrum A) ESR spectrum of an aqueous
solution of TMIO (50 mM) with 20% methanol-17 (20 atom % 17O)
plus FeCl3 (0.5mM). (Spectrum B) Composite computer simulation of
spectrum A. (Spectrum C) Computer simulation of the TMIO/•16OCH3
component (molar ratio 0.59) of spectrum A. (Spectrum D) Computer
simulation of the TMIO/•OH component (molar ratio 0.26) of the
spectrum A. (Spectrum E) Computer simulation of the TMIO/•17OCH3
component (molar ratio 0.15) of spectrum A.
TMIO/•16OCH3 radical adduct as obtained in the meth-
anol system (Fig. 6, spectrum G). The hyperfine coupling
constants of the third radical adduct (Fig. 6, spectrum
H) were those of the TMIO/•17OCH3 radical adduct
(Table 1).
Spin trapping with TMIO in a Fenton system plus
DMSO in a 17O2-bubbled solution revealed formation of
the TMIO/•17OCH3 radical adduct (data not shown). The
hyperfine coupling constant for 17O was 5.62 G, which
was very similar to that for the TMIO/•17OCH3 radical
adduct obtained in the methanol/Fe3⫹ system (Table 1).
DISCUSSION
It is known that the spin trap PBN forms radical
adducts with peroxyl radicals distinct from the alkoxyl
radical adduct (Table 1). However, it was found that the
Fig. 6. Formation of radical adduct in the chloroperoxidase system with
tert-butyl hydroperoxide and spin trap TMIO. (Spectrum A) ESR
spectrum of a phosphate buffer solution (0.15 M, pH 7.4) of TMIO (0.2
M) with tBuOOH (50 mM) plus CPO (5 ␮M). (Spectrum B) ESR
spectrum of nitrogen-bubbled (20 min) phosphate buffer solution (0.15
M, pH 7.4) of TMIO (10 mM) with tBuOOH (50 mM) plus CPO (5
␮M). (Spectrum C) ESR spectrum of aerobic phosphate buffer solution
(0.15 M, pH 7.4) of TMIO (10 mM) with tBuOOH (50 mM) plus CPO
(5 ␮M). (Spectrum D) ESR spectrum of 17O2 (85 atom % 17O)-bubbled
(2 min) phosphate buffer solution (0.15 M, pH 7.4) of TMIO (10 mM)
with tBuOOH (50 mM) plus CPO (5 ␮M). (Spectrum E) Composite
computer simulation of spectrum D. (Spectrum F) Computer simula-
tion of the TMIO/•OC(CH3)3 component (molar ratio 0.37) of spectrum
D. (Spectrum G) Computer simulation of the TMIO/•16OCH3 compo-
nent (molar ratio 0.34) of spectrum D. (Spectrum H) Computer simu-
lation of the TMIO/•17OCH3 component (molar ratio 0.25) of spectrum
D. A minor contribution of DMPO/•CH3 (molar ratio 0.04) was present
in the composite simulation E.
PBN-peroxyl radical adducts are very unstable [7,8].
Janzen et al. reported that peroxyl radical adducts of
PBN are not persistent above 230°K [8]. Moreover, it
was found that at temperatures higher then 250°K, only
the PBN-alkoxyl radical adduct was detected [8]. The
mechanism proposed for the formation of the alkoxyl
radical adduct in this peroxyl radical-generating system
is shown in Scheme 4.
In this work we show that methylperoxyl radical
forms methoxyl radical adducts of DMPO and TMIO at
 Reassignment of organic peroxyl radical adducts
Scheme 4. Formation of radical adducts with azo initiator and PBN.
room temperature. Based on the terminal instability of
the peroxyl radical adduct described for PBN, we suggest
the first of two possible mechanisms of methoxyl radical
formation in methylperoxyl-generating systems
(Schemes 2 and 3). According to the first mechanism,
DMPO reacts with methylperoxyl radical to form
DMPO/•OOCH3, which rapidly decomposes to the me-
thoxyl radical. Then methoxyl radical reacts with another
molecule of DMPO to form DMPO/•OCH3. This mech-
anism could explain why methylperoxyl radical produc-
tion forms methoxyl radical adducts, but not any detect-
able methylperoxyl radical adducts.
Alternatively, the decomposition of methylperoxyl
radicals can give two methoxyl radicals and molecular
oxygen (Scheme 3) [27]. Formation of methoxyl radicals
was estimated as at least 6% from disproportionated
methylperoxyl radicals [28] whereas most of the meth-
ylperoxyl radicals are decaying via the Russell mecha-
nism, giving methanol, formaldehyde, and singlet oxy-
gen [30,31]. Despite the minor role of methoxyl radical
formation in methylperoxyl radical decomposition, it
may be responsible for the formation of the methoxyl
radical adduct (Schemes 2 and 3).
In either case, it is clear that formation of methoxyl
radical adduct takes place via the reaction of a spin trap
with methoxyl radical. It is known that methoxyl radical
intramolecularly converts to hydroxymethyl radical [33].
Indeed, in the chloroperoxidase system, using the
2-methyl-2-nitrosopropane spin trap, hydroxymethyl
radical was detected [20], confirming methoxyl radical
formation.
Both DMPO and TMIO form methoxyl radical ad-
ducts in the methylperoxyl radical-generating systems at
room temperature. Previously, it was described that
above 250°K, reaction of spin trap PBN with tert-butylp-
eroxyl and 2-cyano-2-propylperoxyl radicals also results
in alkoxyl radical adducts [8]. Therefore, the formation
of alkoxyl radical adducts is not specific to a particular
spin trap or peroxyl radical; it appears to be a general
phenomenon.
Our data allow us to reinterpret previously published
data reporting detection of peroxyl radical adducts [9,11,
15–23]. Previously, formation of DMPO/•OOCumene
and DMPO/•OOtBu radical adducts from hydroperoxides
871
has been reported [9,18 –23]. However, it was recently
found that species thought to be DMPO/•OOCumene or
DMPO/•OOtBu were dependent on the presence of oxy-
gen, whereas under anaerobic conditions, the methyl
radical adduct was detected [11]. We have confirmed
these experiments [11,16,17], which are inconsistent
with the previously reported interpretation of these ESR
spectra as arising from DMPO/•OOCumene or DMPO/
•
OOtBu [9,18 –23]. Formation of these “peroxyl”-like
radical adducts were assigned to the DMPO/methylper-
oxyl radical [11,16,17]. Nevertheless, according to the
current literature, the nitrogen, ␤-hydrogen, and ␥-hy-
drogen hyperfine coupling constants for DMPO/•OOC-
umene, DMPO/•OOtBu, and DMPO/•OOCH3 were all
the same as for the DMPO/methoxyl radical adduct
[9 –11, 16 –23]. The data obtained show that the DMPO
radical adducts with hyperfine coupling constants aN ⫽
14.4 ⫾ 0.1 G, aH␤ ⫽ 10.7 ⫾ 0.1 G, and aH␥ ⫽ 1.3 ⫾ 0.1
G are likely to be DMPO/•OCH3 (a17O␤ ⫽ 6.5 G), but not
DMPO/•OOCumene, DMPO/•OOtBu, or DMPO/•OOCH3
as has been suggested in many publications [9,11,16 –23].
Recently, formation of the TMIO/•OOtBu radical ad-
duct has been reported in the chloroperoxidase system
[15]. However, reported hyperfine coupling constants
(aN ⫽ 13.49 G and aH␤ ⫽ 15.52 G) [15] are the same as
we found for TMIO/•OCH3 (Table 1). Moreover, accord-
ing to our oxygen-17 experiments in the chloroperoxi-
dase system, only TMIO/•OCH3 and the very different
TMIO/•OtBu (aN ⫽ 14.13 G and aH␤ ⫽ 16.36 G) were
observed (Fig. 6D, E). Therefore, the previously reported
TMIO/•OOtBu radical adduct actually was the TMIO/
•
OCH3 radical adduct.
Formation of DMPO/•OOC2H5 has been reported in a
chloroperoxidase system with ethyl hydroperoxide
[3,20]. The reported hyperfine coupling constants are
close to those for DMPO/•OC2H5 [10]. Methyl peroxyl
and ethyl peroxyl are very similar radicals; therefore, one
can conclude that the previously assigned DMPO/
•
OOC2H5 radical adduct was, in fact, DMPO/•OC2H5.
There are many articles reporting formation of
DMPO/lipid peroxyl radical adducts. One could suggest
a reassignment of these radical adducts to DMPO/lipid
alkoxyl. However, an unambiguous proof of structure of
a secondary alkoxyl radical adduct has not been pub-
lished to our knowledge. The correct assignment of
DMPO/lipid peroxyl radical adducts is a separate issue
that has yet to be addressed.
Although the detection and correct assignment of
PBN/peroxyl radical adducts is possible at low temper-
ature using 17O2 or 13C-labelled spin traps [7,8], our data
show that at room temperature in aqueous solution only
alkoxyl radical adducts of DMPO and TMIO are de-
tected. The detection of oxygen-dependent alkoxyl rad-
ical adduct formation may possibly be of use as indirect
872 S. I. DIKALOV and R. P. MASON
evidence for the formation of some peroxyl radicals, but
the methoxyl radical adduct is, in fact, not the methyl
peroxyl adduct that it was thought to be. Specifically, we
suggest that detection of 17O-alkoxyl radical adduct from
17
O-labelled molecular oxygen can be used as good
indirect evidence for peroxyl radical generation.
REFERENCES
[1] Frei, B.; Stocker, R.; Ames, B. N. Antioxidant defenses and lipid
peroxidation in human blood plasma. Proc. Natl. Acad. Sci. USA
85:9748 –9752; 1988.
[2] Marnett, L. J. Peroxyl free radicals: potential mediators of tumor
initiation and promotion. Carcinogenesis 8:1365–1373; 1987.
[3] Kalyanaraman, B.; Mottley, C.; Mason, R. P. A direct electron
spin resonance and spin-trapping investigation of peroxyl free
radical formation by hematin/hydroperoxide systems. J. Biol.
Chem. 258:3855–3858; 1983.
[4] Nikolaev, A. I.; Safiullin, R. L.; Enikeeva, L. R.; Komissarov,
V. D. Kinetics of the loss of methylperoxyl radicals in a liquid-
phase. Khimicheskaya Fizika 11:69 –72; 1992.
[5] Bennett, J. E. J. A kinetic study of the self-reaction of prop-2-
ylperoxyl radicals in solution using ultraviolet absorption spec-
troscopy. Chem. Soc., Faraday Trans. I 83:1805–13; 1987.
[6] Janzen, E. G.; Haire D. L. Two decades of spin trapping. Adv.
Free Radic. Chem. 1:253–295; 1990.
[7] Howard, J. A.; Tait, J. C. Electron paramagnetic resonance spec-
tra of the tert-butylperoxy and tert-butoxy adducts to phenyl
tert-butyl nitrone and 2-methyl-2-nitrosopropane. Oxygen-17 hy-
perfine coupling constants. Can. J. Chem. 56:176 –178; 1978.
[8] Janzen, E. G.; Krygsman, P. H.; Lindsay, D. A.; Haire, D. L.
Detection of alkyl, alkoxyl, and alkyperoxyl radicals from the
thermolysis of azobis(isobutyronitrile) by ESR/spin trapping. Ev-
idence for double spin adducts from liquid-phase chromatography
and mass spectroscopy. J. Am. Chem. Soc. 112:8279 – 8284; 1990.
[9] Rosen, G. M.; Rauckman, E. J. Spin trapping of the primary
radical involved in the activation of the carcinogen N-hydroxy-
2-acetylaminofluorene by cumene hydroperoxide-hematin. Mol.
Pharmacol. 17:233–238; 1980.
[10] Hanna, P. M.; Chamulitrat, W.; Mason, R. P. When are metal
ion-dependent hydroxyl radical and alkoxyl radical adducts of
5,5-dimethyl-1-pyrroline N-oxide artifacts? Arch. Biochem. Bio-
phys. 296:640 – 644; 1992.
[11] Barr, D. P.; Martin, M. V.; Guengerich, F. P.; Mason, R. P.
Reaction of cytochrome P450 with cumene hydroperoxide: ESR
spin-trapping evidence for the homolytic scission of the peroxide
O-O bond by ferric cytochrome P450 1A2. Chem. Res. Toxicol.
9:318 –325; 1996.
[12] Mottley, C.; Connor, H. D.; Mason, R. P. [17O] Oxygen hyperfine
structure for the hydroxyl and superoxide radical adducts of the
spin traps DMPO, PBN and 4-POBN. Biochem. Biophys. Res.
Comms. 141:622– 628; 1986.
[13] Buettner G. R. Spin trapping: ESR parameters of spin adducts.
Free Radic. Biol. Med. 3:259 –303; 1987.
[14] Dikalov, S.; Kirilyuk, I.; Grigor’ev, I. Spin trapping of O-, C-, and
S-centered radicals and peroxynitrite by 2H-imidazole-1-oxides.
Biochem. Biophys. Res. Comms. 218:616 – 622; 1996.
[15] Krainev, A. G.; Williams, T. D.; Bigelow, D. J. Oxygen-centered
spin adducts of 5,5-dimethyl-1-pyrroline N-oxide (DMPO) and
2H-imidazole 1-oxides. J. Mag. Reson. Series B 111:272–280;
1996.
[16] Barr, D. P.; Mason, R. P. Mechanism of radical production from
the reaction of cytochrome c with organic hydroperoxides. J. Biol.
Chem. 270:12709 –12716; 1995.
[17] Van der Zee, J.; Barr, D. P.; Mason, R. P. ESR spin trapping
investigation of radical formation from the reaction between
hematin and tert-butyl hydroperoxide. Free Radic. Biol. Med.
20:199 –206; 1996.
[18] Thornalley, P. J; Trotta, R. J.; Stern, A. Free radical involvement
in the oxidative phenomena induced by tert-butyl hydroperoxide
in erythrocytes. Biochem. Biophys. Acta 759:16 –22; 1983.
[19] Davies, M.J. Detection of peroxyl and alkoxyl radicals produced
by reaction of hydroperoxides with rat liver microsomal fractions.
Biochem. J. 257:603– 606; 1989.
[20] Chamulitrat, W.; Takahashi, N.; Mason, R. P. Peroxyl, alkoxyl,
and carbon-centered radical formation from organic hydroperox-
ides by chloroperoxidase. J. Biol. Chem. 264:7889 –7899; 1989.
[21] Kennedy, C. H.; Dyer, J. M.; Church, D. F.; Winston, G. W.;
Pryor, W. A. Radical production in liver-mitochondria by perox-
idic tumor promoters. Biochem. Biophys. Res. Comms. 160:1067–
1072; 1989.
[22] Kohno, M.; Yamada, M.; Mitsuta, K.; Mizuta, Y.; Yoshikawa, T.
Spin-trapping studies on the reaction of iron complexes with
peroxides and effects of water-soluble antioxidants. Bull. Chem.
Soc. Jpn. 64:1447–1453; 1991.
[23] Kennedy, C. H.; Church, D. F.; Winston, G. W.; Pryor, W. A.
Tert-butyl hydroperoxide-induced radical production in rat liver
mitochondria. Free Radic. Biol. Med. 12:381–387; 1992.
[24] Makino, K.; Hagiwara, T.; Hagi, A.; Nishi, M.; Murakami, A.
Cautionary note for DMPO spin trapping in the presence of iron
ion. Biochem. Biophys. Res. Commun. 172:1073–1080; 1990.
[25] Duling, D. R. Simulation of multiple isotropic spin-trap EPR
spectra. J. Magn. Res., Series B 104:105–110; 1994.
[26] Veltwisch, D.; Janata, E.; Asmus, K.-D. Primary processes in the
reaction of OH-radicals with sulphoxides. J. Chem. Soc. Perkin
Trans. 2:146 –153; 1980.
[27] von Sonntag, C.; Schuchman, H. P. The elucidation of peroxyl
radical reactions in aqueous solution with the help of radiation-
chemical methods. Angew. Chem. Int. Ed. Engl. 30:1229 –1253;
1991.
[28] Schuchman, H. P.; von Sonntag C. Methylperoxyl radicals: a
study of the ␥-radiolysis of methane in oxygenated aqueous
solutions. Z. Naturforsch. B39:217–221; 1984.
[29] Janzen, E.G. Spin trapping. Acc. Chem. Res. 4:31– 40; 1971.
[30] Russell, G. A. Deuterium-isotope effects in the autoxidation of
aralkyl hydrocarbons. Mechanism of the interaction of peroxy
radicals. J. Am. Chem. Soc. 79:3871–3877; 1957.
[31] Ingold, K. U. Peroxy radicals. Accts. Chem. Res. 2:1–9; 1969.
[32] Lloyd, R. V.; Hanna, P. M.; Mason, R. P. The origin of the
hydroxyl radical in the Fenton reaction. Free Radic. Biol. Med.
22:885– 888; 1997.
[33] Ledwith, A.; Russell, P. J.; Sutcliffe, L. H. Alkoxy radical inter-
mediates in the thermal and photochemical oxidation of alcohols.
Proc. R. Soc. Lond. A. 332:151–166; 1973.