Revisiting the peroxidase oxidation of 2,4,6-trihalophenols: ESR detection of radical intermediates

Article abstract of DOI:10.1021/tx200215r

The peroxidase oxidation of 2,4,6-trichlorophenol (TCP) has been clearly shown to result in 2,6-dichloro-1,4-benzoquinone (DCQ). DCQ is a 2-electron oxidation product of TCP that has undergone para dechlorination. Many peroxidases show similar oxidation of the substrate, TCP, to yield the quinone, DCQ. Depending on the substrate, peroxidases are thought to carry out both 1- and 2-electron oxidations; the mechanism can be confirmed by the detection of both enzyme and substrate intermediates. This article presents ESR evidence for the transient 2,4,6-trichlorophenoxyl radical intermediate (TCP?), which exists free in solution, i.e., is not enzyme associated. These data are best explained as a 1-electron peroxidase oxidation of TCP to form TCP?, followed by enzyme-independent radical reactions leading to the 2-electron oxidized product. Also presented are data for the peroxidase oxidation of 2,4,6-trifluorophenol and 2,6-dichloro-4-fluorophenol.

Full text of DOI:10.1021/tx200215r

ARTICLE
pubs.acs.org/crt
Revisiting the Peroxidase Oxidation of 2,4,6-Trihalophenols: ESR
Detection of Radical Intermediates
Bradley E. Sturgeon,*,† Benjamin J. Battenburg,† Blake J. Lyon,† and Stefan Franzen‡
†Department of Chemistry, Monmouth College, Monmouth, Illinois 61462, United States
‡Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695, United States
S Supporting Information
b
ABSTRACT: The peroxidase oxidation of 2,4,6-trichlorophenol (TCP) has been clearly
shown to result in 2,6-dichloro-1,4-benzoquinone (DCQ). DCQ is a 2-electron
oxidation product of TCP that has undergone para dechlorination. Many peroxidases
show similar oxidation of the substrate, TCP, to yield the quinone, DCQ. Depending on
the substrate, peroxidases are thought to carry out both 1- and 2-electron oxidations; the
mechanism can be confirmed by the detection of both enzyme and substrate inter-
mediates. This article presents ESR evidence for the transient 2,4,6-trichlorophenoxyl
radical intermediate (TCP•), which exists free in solution, i.e., is not enzyme associated.
These data are best explained as a 1-electron peroxidase oxidation of TCP to form TCP•,
followed by enzyme-independent radical reactions leading to the 2-electron oxidized
product. Also presented are data for the peroxidase oxidation of 2,4,6-trifluorophenol
and 2,6-dichloro-4-fluorophenol.
’ INTRODUCTION
can act as a 2-electron reducing substrate for compound I, hence
forgoing the formation of compound II, and resulting in XAO+,
then the quinone (Figure 1, 2-electron mechanism). Third, the
radical generated via 1-electron oxidation of TCP can undergo an
apparent radicalꢀradical dismutation reaction resulting in the
original reducing substrate and the quinone (reaction 1).
A peroxidase in the presence of H2O2 will carry out the
oxidization of phenolic compounds such as 2,4,6-trichlorophenol
(TCP). In the case of ligin peroxidase,1 horseradish peroxidase
(HRP),2,3 chloroperoxidase,4,5 dehaloperoxidase-hemoglobin,6
and myoglobin,7 the main product is reported to be 2,6-di-
chloro-1,4-benzoquinone (DCQ). DCQ is the 2-electron oxidized
form of TCP that has undergone para dechlorination (Scheme 1).
The typical peroxidase mechanism is outlined in Figure 1 and
is initiated by the 2-electron oxidation of the enzyme resting state
(RS) by an oxidizing substrate (eg., H2O2) to form compound I.
Depending on the origin of the peroxidase, the structural form of
compound I consists of Fe(IV)dO and either a heme radical
cation or an amino acid radical.8 Compound I then undergoes a
1-electron reduction by a reducing substrate (eg., a halophenol,
XAOH) to form compound II and the corresponding radical,
XAO•. Compound II then undergoes a second 1-electron reduc-
tion by a second molecule of the reducing substrate to regenerate
RS and an additional radical.
This peroxidase mechanism is commonly referred to as an
irreversible ping-pong mechanism.8 The name makes reference
to a very short-lived enzymeꢀsubstrate complex; the product
departs as fast as the substrate contacts the enzyme, like a ping-
pong ball when it strikes a paddle. To account for the formation
of the 2-electron oxidized product, DCQ, three different path-
ways/reactions are considered. First, the radical, XAO•, can act
as a reducing substrate for compound I and/or compound II
forming a transient cyclohexadienone cation (XAO+). XAO+
would rapidly react with water (or OHꢀ), dehalogenate, and
result in the quinone product (OdAdO)1ꢀ4,7(Figure 1, 1-
electron mechanism). Second, the reducing substrate (eg., XAOH)
2XAO• þ H O f XAOH þ O A O þ Hþ þ Clꢀ
d d
2
ð1Þ
For a given enzyme/substrate system, it is important to know
which mechanism is in operation. The difference between the
1-electron and 2-electron mechanism involves two key points:
(1) compound II is formed during the 1-electron mechanism
and not during the 2-electron mechanism, and (2) a radical (eg.,
XAO•) is formed during the 1-electron mechanism and not
during the 2-electron mechanism. The significance of which
mechanism is in operation can be briefly summarized by the fol-
lowing comments: (1) the formation of radical intermediates
provides opportunities for other reactions/products to occur, and
(2) the more involved enzyme kinetics of the 1-electron mech-
anism, the more intermediates that can be targeted to modulate
enzyme function.
The 1-electron mechanism has been most frequently used to
explain the formation of DCQ.1ꢀ4,7 Compound II oxidation of
the reducing substrate is usually the rate limiting step in the
enzymatic cycle and hence can be detected under steady-state
conditions. In cases where the form of compound II involves the
Received: May 21, 2011
Published: September 27, 2011
r
2011 American Chemical Society
1862
dx.doi.org/10.1021/tx200215r Chem. Res. Toxicol. 2011, 24, 1862–1868
|

Chemical Research in Toxicology
ARTICLE
Scheme 1. Peroxidase oxidation of 2,4,6-trichlorophenol to
form the quinone product
Table 1. Molar Absorptivity Values Used to Generate Con-
centration Data in Figure 3
TCP
DCQ
a
a
a
a
pH
ε274
ε312
ε274
ε312
6.2
7.4
9.0
1.18
0.96
1.03
3.39
4.17
4.58
12.64
12.17
8.37
0.43
0.41
1.89
a In units of mMꢀ1 cmꢀ1
.
Figure 1. Outline of peroxidase mechanism with H2O2 as the oxidizing
substrate and a halophenol (XAOH) as the reducing substrate. Enzyme
intermediates are labeled as resting state (RS), compound I, and com-
pound II. The dotted borders indicate the reactionsspecifically involvedin
the 1-electron and 2-electron mechanisms. Reactions presented in gray
lead to the formation of a quinone (a two-electron oxidized product).
Figure 2. ESR spectrum observed 13 min after a solution of HRP
(4.47 μM), H2O2 (0.5 mM), and TCP (0.5 mM) at pH 7.4 was mixed
and transferred to the mounted Aqua-X ESR sample holder. The inset
monitors the ESR intensity at 3510 G vs time for (a) 4.47 μM HRP and
(b) 0.447 μM HRP. ESR parameters: 9.85 GHz microwave frequency,
1 mW microwave power, 100 kHz modulation frequency, 0.2 Gpp modulation
amplitude, and 82 s scan time with a 164 ms time constant.
heme radical cation, UVꢀvis detection is fairly straightforward,
although when the form of compound II involves a protein
radical, UVꢀvis detection is more challenging.9 This 1-electron
mechanism is also supported by the observation of compound II
under single-turnover conditions; if the 2-electron mechanism
were involved, compound II would not be detected as is the case
for the oxidation of thioanisole by CCPO,5 iodide by HRP,10 and
sulfite by HRP.11
In this article, we present ESR data related to the reaction of
HRP/H2O2 with TCP and other TCP-analogues and present
evidence to support the 1-electron oxidation mechanism media-
ted by both enzyme and radical chemistry.
ESR Spectroscopy. Detection of all radicals was done using a
Bruker EMX spectrometer (Billerica, MA) equipped with a High Sensi-
tivity cavity. X-band (∼9.8 GHz) ESR conditions are listed in the figure
captions. When detecting radicals from a standard mixing experiment
(Figure 2), the sample was introduced into the cavity using the Bruker
Aqua-X sample holder. When conducting IE-ESR experiments, enzymes
were loaded into a WG-812-Q flatcell from Wilmad Glass (Vineland, NJ)
as described.14 ESR simulations were performed using the WINSIM
program;15 all presented simulations use best-fit parameters as deter-
mined by the WINSIM program.
’ EXPERIMENTAL PROCEDURES
’ RESULTS
Materials. HRP (Type-VIa), TCP, TFP, DCFP, and horse heart
myoglobin were obtained from Sigma Chemical Company (St. Louis,
MO). Dehaloperoxidase was purified as described previously.12 All other
chemicals were of the highest purity available.
UVꢀVisible Spectroscopy. The oxidation of TCP by HRP/
H2O2 was followed using an HP8453 photodiode array UVꢀvisible
spectrometer. A cuvette containing TCP and H2O2 was scanned (230ꢀ
400 nm) prior to the addition of enzyme (t = 0 s). The reaction was
initiated by the addition of a negligible volume of HRP, mixed, and scan-
ned at 30 s; additional scans were taken every 30 s. Since the absorption
of TCP was well separated from DCQ, data were analyzed using the
method presented by Harris13 using the data in Table 1.
Shown in Figure 2 is the ESR spectrum detected when a solu-
tion of TCP/H2O2 (pH 7.4) was mixed with HRP and quickly
(∼20 s) transferred into the ESR spectrometer. The resulting
spectrum (aH = 2.4 G) is equivalent to the ESR spectrum pre-
viously detected by Wiese et al. and is unlikely to be TCP•.2
The time course for the observed radical is shown in the inset
of Figure 2. The radical signal was not detected immediately
and was dependent on HRP concentration. At 4.47 μM HRP
(Figure 2, inset a), the radical signal appeared at ∼350 s after
mixing, whereas at 0.447 μM HRP (Figure 2, inset b), the
radical signal appeared at ∼970 s after mixing. Previous work
1863
dx.doi.org/10.1021/tx200215r |Chem. Res. Toxicol. 2011, 24, 1862–1868

Chemical Research in Toxicology
ARTICLE
Figure 4. IE-ESR spectra collected using HRPi/H2O2 and halophenols
at pH 7.4. (A) ESR spectrum of TCP• (g = 2.008) collected using HRPi
(3 mg/mL), H2O2 (1 mM), TCP (0.5 mM), and a 2.0 mL/min flow rate.
ESR parameters: 9.798 GHz microwave frequency, 20 mW microwave
power, 1 Gpp modulation amplitude, and an average of 10ꢀ42 s scans
with a time constant of 82 ms; (A0) high-resolution TCP• collected using
the same conditions as A but with 0.5 Gpp moduation amplitude and
an average of 20ꢀ167 s scans with a time constant of 167 ms; (B) TFP•
(g = 2.006) collected using HRPi (3.0 mg/mL), H2O2 (1 mM), TFP
(1.0 mM), and a 2.0 mL/min flow rate. ESR parameters: 9.802 GHz
microwave frequency, 20 mW microwave power, 1 Gpp modulation
amplitude, and an average of 10ꢀ84 s scans with a time constant of
164 ms; (B0) overlay of simulation (dotted) ESR spectrum, see text for
simulation parameters; (C), DCFP• (g = 2.005) collected using HRPi
(0.7 mg/mL), H2O2 (1 mM), DCFP (0.5 mM), and a 2.0 mL/min flow
rate. ESR parameters: 9.775 GHz microwave frequency, 20 mW micro-
wave power, 2 Gpp modulation amplitude, and an average of 10ꢀ42 s
scans with a time constant of 82 ms; and (C0) high-resolution DCFP•
collected using the same conditions as those in C, except for 0.5 Gpp
modulation amplitude and an average of 110ꢀ42 s scans with a time
constant of 82 ms.
Figure 3. UVꢀvisible monitoring of TCP oxidation by HRP and H2O2.
Solutions contained HRP (0.2 μM), H2O2 (0.2 mM), and TCP
(0.2 mM) at A, pH 6.2, B, pH 7.4, and C, pH 9.0. Inset data are the
calculated concentration of TCP and DCQ as described in the Experi-
mental Procedures.
combined with the peculiar radical persistence lead us to inves-
tigate further the early events in the enzymatic cycle.
by Wiese et al.2 used 23 μM HRP, which apparently resulted in an
almost immediate detection of this radical signal. The effect of
TCP or H2O2 concentration on the time course of the radical was
not investigated. Similar attempts at ESR data collected with
HRP (0.447 μM), TCP, and H2O2 at pH 6.2 and 9.0 observed no
radical within a 40 min period. The time course of this reaction
was followed by UVꢀvisible spectroscopy at pH 6.2, 7.4, and 9.0
(Figure 3) indicating that oxidation of TCP was rapid at pH 6.2
and pH 7.4, although quite slow at pH 9.0.
In order to observe highly reactive radicals formed during the
early events of the enzymatic cycle, the ESR techniques of fast-
flow ESR or immobilized enzyme ESR (IE-ESR) are required.14
Shown in Figure 4A is the IE-ESR spectrum observed when a
TCP/H2O2 solution (pH 7.4) was flowed over immobilized HRP
(HRPi). Under these experimental conditions, radicals with life-
times of milliseconds have been detected.14,16,17 The very broad,
featureless spectrum (Figure 4A) provides no clear information
with respect to the identification of TCP•, although when ESR con-
ditions wereset to be sensitive to spectral features and not signal-to-
noise (i.e., modulation amplitude changed from 1.0 Gpp to 0.5 Gpp
with signal averaging), the ESR spectrum shown in Figure 4A0 was
observed. This spectrum showed clear hyperfine interactions that
can be simulated (Figure S1, Supporting Information) using two
ortho-chlorine hyperfine coupling constants (hfcc) (aoCl = 0.90 G),
a para-chlorine hfcc (apCl = 1.82 G), and two meta-proton hfcc
(amH = 1.62 G); this spectrum and simulation are consistent with
the TCP• formed in organic solvents by photolysis.18
Although the published TCP• ESR hyperfine couplings2 are
consistent with a diortho substituted phenoxyl radical, in the case
of TCP• two questions arise: (1) the time course for the radical
formation/persistence is unexpected for a phenoxyl radical, and
(2) the spectrum is missing chlorine hyperfine couplings. Chlor-
ine hyperfine couplings often go undetected since the gyromag-
netic ratios for 35Cl and 37Cl are quite small (and different for
each isotope), which typically result in inhomogeneous line
broadening. As for the time course, a phenoxyl radical like TCP•
is expected to form during the early events of the enzymatic cycle
and is expected to be considerably less stable (more reactive)
than the observed radical signal. The lack of chlorine hyperfine
In order to overcome the complexity associated with chlorine
nuclei, the same IE-ESR experiments were performed with the
1864
dx.doi.org/10.1021/tx200215r |Chem. Res. Toxicol. 2011, 24, 1862–1868

Chemical Research in Toxicology
ARTICLE
Figure 5. IE-ESR spectra collected at varying substrate flow rates for
HRPi/H2O2/TCP at pH 7.4. (A) ESR spectrum of TCP• at 2.0 mL/min,
same data as that presented in Figure 4A; (B) same as A, except for
0.25 mL/min; (B0) simulation of B with relative contribution of TCP•
and the three-line spectrum indicated; (C) same as A, except for the flow
is stopped, and only 5 scans are averaged.
fluoro-analogue of TCP, 2,4,6-trifluorophenol (TFP). Shown in
Figure 4B is the IE-ESR spectrum observed when a TFP/H2O2
solution (pH 7.4) was flowed over HRPi. Similar spectra were
observed when enzymatic oxidation was carried out by dehalo-
peroxidase or myoglobin (Figure S2, Supporting Information).
Unlike TCP, this ESR spectrum was rich in hyperfine couplings.
On the basis of previous ESR data on the tyrosyl radical19 and the
phenoxyl radical,20 the observed ESR spectra can be simulated
(Figure 4B0) using two large ortho-fluorine hfcc (aoF = 13.5 G),
an even larger para-fluorine hfcc (apF = 25.0 G), and two meta-
Figure 6. IE-ESR data collected for HRPi (8 mg/mL)/H2O2
(1.0 mM)/TFP (1.0 mM) at varying substrate flow rates and the
simulated ESR data. All data were collected using the same ESR data
acquisition parameters. The hyperfine coupling constants used in the
simulations are presented in the text; the % contribution to the over-
all spectrum is stated on the Figure. (A) 2.0 mL/min with ESR para-
meters similar to those in Figure 4B; (A0) simulation of A; (B) 1.0 mL/
min; (B0) simulation of B; (C) 0.50 mL/min; (C0) simulation of C; (D)
0.25 mL/min; (D0) simulation of D; (E) 0.125 mL/min; (E0) simulation
of E; (F) 0 mL/min; and (F0) simulation of F.
H
proton hfcc (am = 1.5 G). Additionally, IE-ESR data were
collected for the peroxidase oxidation of 2,6-dichloro-4-fluoro-
phenol (DCFP). When a DCFP/H2O2 solution (pH 7.4) was
flowed over HRPi, a doublet ESR spectrum was observed with a
the ESR signal detected under standard mixing conditions
(Figure 2). This result suggests that the radical seen under
standard mixing conditions is not the primary phenoxyl radical
but a secondary radical species. At an intermediate flow rate of
0.25 mL/min (Figure 5B), both the primary phenoxyl radical and
the secondary radical are visible. Note the differences in radical
g-values.
F
single hfcc (ap = 25.2 G) assigned to the para-fluorine nuclei
(Figure 4C). When ESR conditions were set to be sensitive to
spectral features and not signal-to-noise (ie., modulation ampli-
tude changed from 2 Gpp to 0.5 Gpp with signal averaging), as was
done with TCP•, the ESR spectrum revealed clear hyperfine
interactions (aoCl = 0.84 G and amH = 1.44 G) consistent with the
values observed in TCP• (Figure 4A0). The simulation (low field
only) of the DCFP• radical spectrum is presented in Figure S1 of
the Supporting Information. IE-ESR experiments were per-
formed under acidic conditions (pH 5) for TCP, TFP, and
DCFP; the ESR spectra were nearly identical to the pH 7.4 data.
No high resolution ESR data were collected at pH 5. No IE-ESR
data were collected under basic (pH∼9) conditions since UVꢀ
vis data (Figure 3) showed that the enzymatic activity was low.
A unique aspect of the IE-ESR experiment is that when the
substrate (XAOH/H2O2) flow rate is high (2 mL/min), early
time (msec) events are observed. When flow rates are decreased,
additional chemistry is allowed to occur in the active region of the
ESR detection system. Shown in Figure 5 are the ESR spectra
collected under a set of different flow conditions for HRPi/
H2O2/TCP. The most obvious result was that the ESR signal
(aH=2.4 G) seen under slow-flow (0.25 mL/min) (Figure 5B)
and stopped-flow conditions (Figure 5C) is nearly identical to
When variable flow rate IE-ESR data were collected for the
HRPi/H2O2/TFP system (Figure 6), the resolved fluorine
hyperfine allowed for the observation of additional radical
species. All of the data in Figure 6 can be simulated using three
different radical species; species 1 (TFP•) [amH(2) = 1.42 G,
F
F
ao (2) = 13.90 G, and ap (1) = 25.08 G], species 2 [a(1) =
1.99 G, a(1) = 2.86 G, and a(1) = 10.7 G], and species 3 [a(1) =
1.86 G and a(1) = 2.92 G].
’ DISCUSSION
The observation of the chlorine and fluorine hyperfine on the
primary radicals generated from the HRPi/H2O2/TCP, HRPi/
H2O2/TFP, and HRPi/H2O2/DCFP systems is clear evidence of
a 1-electron peroxidase mechanism in HRP; we believe this to be
the case with dehaloperoxidase and myoglobin. This conclusion
was correctly presented by Wiese et al.2 but based on incorrectly
assigned ESR data. It is important to stress that the detection of
1865
dx.doi.org/10.1021/tx200215r |Chem. Res. Toxicol. 2011, 24, 1862–1868

Chemical Research in Toxicology
ARTICLE
Scheme 2. Enzyme Independent, RadicalꢀRadical Coupling Reaction Leading to the Quinone Product
TCP•, TFP•, and DCFP• is support for the 1-electron peroxidase
mechanism, although without correlating the radical concentra-
tion with H2O2 consumed, which is difficult due to high radical
reactivity, these results can only say that the 1-electron mechan-
ism is clearly in operation; they do not rule out the 2-electron
mechanism. In addition, the detection of such an isotropic (sym-
metric) ESR spectrum is clear evidence that the radical product is
not associated with the reaction center, hence in solution. Our
ESR data suggest that a most appropriate mechanism for the
formation of DCQ is a combination of 1-electron peroxidase
oxidation of the halophenol, followed by the radicalꢀradical coupl-
ing chemistry (reaction 1) originally put forth, as one option, by
Wiese et al.2
The presented ESR data show a large hfcc at the para position,
indicating large unpaired spin density; hence, the ether inter-
mediate would be a favorable radicalꢀradical coupling product
(Scheme 2). This enzyme/radical mechanism is much more
likely than the halophenoxyl radical as a peroxidase substrate
since the steady-state concentration of the halophenoxyl radical
is quite low compared to the halophenol. This enzyme-indepen-
dent radical chemistry of the peroxidase-generated radicals com-
monly dictates product formation as with the case of the tyrosyl
radical forming dimers and trimers21 and ascorbate radical
dismutation.22
So, what is the origin of the radical shown in Figure 2 and
presented by Wiese et al.?2 We believe that the radical shown in
Figure 2 is 2,6-dichloro-1,4-benzosemiquinone for the following
reasons: (1) the radical’s persistence is consistent with those
observed for chlorinated benzosemiquinones;23 (2) the observed
1:2:1 structure of the ESR spectrum is consistent with the expec-
ted hfcc for the two equivalent ring protons; (3) the absence of
the chlorine hyperfine is an indication of small couplings that
will contribute to only line width;23,24 and (4) the radical shown
in Figure 2 was not observed when the reaction was carried out
at pH < 6.2; benzosemiquinones are known to not persist at or
below pH values <pKa of the benzosemiquinone.25
As presented in Figures 5 and 6, the IE-ESR method can vary
the flow rate to detect secondary radicals. We were hopeful that
the detection of secondary radicals for the fluoro-analogue of
TCP (TFP•) would assist in the analysis of the TCP• secondary
radical, although this was not the case. If the TFP• under went the
same radical evolution as TCP•, then the final ESR spectrum
seen under 0 mL/min flow conditions (Figure 6F) would have
been from the unique 2,6-difluoro-1,4-benzosemiquinone. Ap-
parently, the oxidation of TFP does not only result in para
dehalogenation, as is the case with the oxidation of TCP,1ꢀ3 but
also in ortho dehalogenation as well. It is important to note that
since the 0 mL/min flow ESR spectrum for TCP oxidation does
not show any chlorine hyperfine splitting, these data cannot
be used to prove that dechlorination does not occur at the ortho
Scheme 3. Proposed Radical Evolution Based on ESR Data
in Figure 6
positions. Hammel and Tardone1 have reported that only one
inorganic chloride per TCP is detected during the peroxidase
oxidation of TCP; hence, this result alone has led to the belief
that there is no ortho dechlorination occurring in TCP. The
ortho defluorination of TFP can be seen from the data presented
in Figure 6. Under slower flow(1.0 mL/min> flow > 0.125 mL/min;
Figure 6BꢀE), additional radicals are observed. For the data
presented in Figure 6, species 2 can be simulated as a mono-
fluorinated radical (aoF = 10.7 G) with two nonidentical protons
(a1H = 1.99 G and a2H = 2.96 G), and species 3 can be simulated
as an asymmetric radical (a proposed semiquinone) with two
nonidentical protons (a1H = 1.86, a2H = 2.92) (Scheme 3, where
Ra and Rb are ESR silent nuclei). Further investigation into the
complete oxidation of TFP is required to fully understand the
mechanism, although as a preliminary thought, it appears as
though the radicalꢀradical coupling reaction with TFP• involves
the ether intermediate resulting from coupling at not only the
para position but also the ortho position.
The ESR data requires additional discussion. Presented in
Figure 4 are the experimental (4B) and simulated (4B0) spectra
of TFP•. These IE-ESR data are extremely reproducible; IE-ESR
data have no scan time-dependent features.14 The simulation of
these data fit very well with the only exception being the intensity
of the outer ESR transitions (3460ꢀ3470 G and 3515ꢀ3525 G);
the experimental intensity of the outer transitions is lower in
amplitude than that of the simulated. The broadening of the
higher field outer transitions (>3500 G) is a result of small hyper-
fine anisotropy commonly seen in ESR spectra with large hfcc,
although the lower than expected intensity of the lowest and
highest field lines (3460ꢀ3470 G and 3515ꢀ3525 G) remains
unclear. Similar ESR observations have been seen by others for
the fluoranil radical anion26 and 9,10-perfluoroanthraquinone
radical anion.27
In regard to the ESR simulation of the TCP•, the simulations
did not require the use of both 35Cl and 37Cl isotopes. Reported
simulations of TCP•18 did use both isotopes, but this early publi-
cation did not have access to sophisticated computer methods.
To further interpret the TCP• ESR data, the well-determined
F
F
fluorine hfcc from TFP• (ao = 13.5 G, ap = 25.0 G)
1866
dx.doi.org/10.1021/tx200215r |Chem. Res. Toxicol. 2011, 24, 1862–1868

Chemical Research in Toxicology
ARTICLE
were scaled using the gyromagnetic ratios to predict chlorine
6-trifluorophenol; TFP•, 2,4,6-trifluorophenoxyl radical; DCFP,
2,6-dichloro-4-fluorophenol; DCFP•, 2,6-dichloro-4-fluorophe-
noxyl radical; XAOH, unspecified halophenol; XAO•, unspeci-
fied halophenoxyl radical ; XAO+, unspecified halocyclo-
hexadienone cation; DCQ, 2,6-dichloro-1,4-benzoquinone; HRP,
horseradish peroxidase;RS, enzyme resting state; HRPi, immo-
bilized HRP; hfcc, hyperfine coupling constants;IE-ESR, immo-
bilized enzyme ESR
37Cl
hfcc (ao35Cl = 1.4 G and ap35Cl = 2.6 G; ao37Cl = 1.2 G and ap
=
2.2 G). These values, along with the meta-protons (amH = 1.5 G)
and natural isotopic abundance, were used to generate a pre-
dicted TCP• ESR spectrum. Two observations were made:
(1) the predicted aoCl and apCl from the scaled TFP• hfcc were
larger than the observed, indicating that the spin density on the
fluorine nuclei is slightly larger than that for the chlorine nuclei,
and (2) an adequate simulation could be achieved using only
the 35Cl hfcc. When isotopic abundance of 35Cl and 37Cl was
included in the simulation, the outermost lines were only slightly
broadened. In our case, the low signal-to-noise in the high resolu-
tion ESR spectra (Figure 4A0,C0; Figure S1, Supporting Informa-
tion) does not allow for clear observation of these outermost
lines.
’ REFERENCES
(1) Hammel, K. E., and Tardone, P. J. (1988) The oxidative
4-dechlorination of polychlorinated phenols is catalyzed by extracellular
fungal lignin peroxidases. Biochemistry 27, 6563–6568.
(2) Wiese, F. W., Chang, H. C., Lloyd, R. V., Freeman, J. P., and
Samokyszyn, V. M. (1998) Peroxidase-catalyzed oxidation of 2,4,6-
trichlorophenol. Arch. Environ. Contam. Toxicol. 34, 217–222.
(3) Ferrari, R. P., Laurenti, E., and Trotta, F. (1999) Oxidative
4-dechlorination of 2,4,6-trichlorophenol catalyzed by horseradish per-
oxidase. J. Biol. Inorg. Chem. 4, 232–237.
(4) Osborne, R. L., Raner, G. M., Hager, L. P., and Dawson, J. H.
(2006) C. fumago chloroperoxidase is also a dehaloperoxidase: oxidative
dehalogenation of halophenols. J. Am. Chem. Soc. 128, 1036–1037.
(5) Osborne, R. L., Coggins, M. K., Terner, J., and Dawson, J. H.
(2007) Caldariomyces fumago chloroperoxidase catalyzes the oxidative
dehalogenation of chlorophenols by a mechanism involving two one-
electron steps. J. Am. Chem. Soc. 129, 14838–14839.
(6) Chen, Y. P., Woodin, S. A., Lincoln, D. E., and Lovell, C. R.
(1996) An unusual dehalogenating peroxidase from the marine tere-
bellid polychaete Amphitrite ornata. J. Biol. Chem. 271, 4609–4612.
(7) Osborne, R. L., Coggins, M. K., Walla, M., and Dawson, J. H.
(2007) Horse heart myoglobin catalyzes the H2O2-dependent oxidative
dehalogenation of chlorophenols to DNA-binding radicals and qui-
nones. Biochemistry 46, 9823–9829.
’ SUMMARY
We have detected the primary 2,4,6-trihalophenoxyl radical
intermediates resulting from the 1-electron peroxidase oxidation
of TCP, TFP, and DCFP. Since the observed IE-ESR spectra are
isotropic, we conclude that the primary radicals are not asso-
ciated with the reaction center and are free to react further in
solution. The ESR data provide support for the hypothesis that
the formation of DCQ involves a 1-electron peroxidase oxidation
of the TCP, followed by radical coupling chemistry resulting in
TCP+, which is attacked by OHꢀ in solution to form DCQ and
Clꢀ. Secondary radicals observed under slower-flow IE-ESR
experiments in both TCP (Figure 5) and TFP (Figure 6) are
tentatively assigned as halogenated/dehalogenated 1,4-semiqui-
nones; the mechanism of semiquinone formation will be pre-
sented in a subsequent paper.
(8) Dunford, H. B. (1999) Heme Peroxidases, Wiley-VCH, New York.
(9) Feducia, J., Dumarieh, R., Gilvey, L. B., Smirnova, T., Franzen, S.,
and Ghiladi, R. A. (2009) Characterization of dehaloperoxidase com-
pound ES and its reactivity with trihalophenols. Biochemistry 48, 995–
1005.
(10) Roman, R., and Dunford, H. B. (1972) pH Dependence of the
oxidation of iodide by compound I of horseradish peroxidase. Biochem-
istry 11, 2076–2082.
’ ASSOCIATED CONTENT
S
Supporting Information. Details of the ESR simulation
b
of TCP•/DCFP•; TFP• spectrum from dehaloperoxidase and
myoglobin with kinetics. This material is available free of charge
via the Internet at http://pubs.acs.org.
(11) Roman, R., and Dunford, H. B. (1973) Studies on horseradish
peroxidase. XII. A kinetic study of the oxidation of sulfite and nitrite by
compounds I and II. Can. J. Chem. 51, 588–596.
(12) Belyea, J., Gilvey, L. B., Davis, M. F., Godek, M., Sit, T. L.,
Lommel, S. A., and Franzen, S. (2005) Enzyme function of the globin
dehaloperoxidase from Amphitrite ornata is activated by substrate
binding. Biochemistry 44, 15637–15644.
’ AUTHOR INFORMATION
Corresponding Author
*Tel: 309-457-2368. E-mail: besturgeon@monm.edu.
Funding Sources
This project was in part supported by a grant from the Research
Corporation for Science Advancement: Cottrell College Science
Award #7943 (B.E.S.) and U.S. Army Research Office grant
#52278 (S.F.).
(13) Harris, D. C. (2006) Quantitative Chemical Analysis, 7th ed., W.
H. Freeman and Company, New York.
(14) Sturgeon, B. E., Chen, Y. R., and Mason, R. P. (2003)
Immobilized enzyme electron spin resonance: a method for detecting
enzymatically generated transient radicals. Anal. Chem. 75, 5006–5011.
(15) Duling, D. R. (1994) Simulation of multiple isotropic spin-trap
EPR spectra. J. Magn. Reson. 104, 105–110.
(16) Chen, Y. R., Deterding, L. J., Sturgeon, B. E., Tomer, K. B., and
Mason, R. P. (2002) Protein oxidation of cytochrome c by reactive
halogen species enhances its peroxidase activity. J. Biol. Chem. 277,
29781–29791.
(17) Sturgeon, B. E., Glover, R. E., Chen, Y. R., Burka, L. T., and
Mason, R. P. (2001) Tyrosine iminoxyl radical formation from tyrosyl
radical/nitric oxide and nitrosotyrosine. J. Biol. Chem. 276, 45516–45521.
(18) Graf, F., Loth, K., and Guenthard, H. H. (1977) Chlorine
hyperfine splittings and spin density distributions of phenoxy radi-
cals. An ESR and quantum chemical study. Helv. Chim. Acta 60,
710–721.
’ ACKNOWLEDGMENT
We wish to thank Garry R. Buettner and Brett A. Wagner of
the University of Iowa ESR Facility and Ronald P. Mason and
Jean Corbett of the National Institute of Environmental Sciences
(NIEHS/NIH) for the use of ESR instrumentation and con-
sultation. We wish to acknowledge Eric M. Todd of Monmouth
College for review of the manuscript.
’ ABBREVIATIONS
TCP, 2,4,6-trichlorophenol; TCP•, 2,4,6-trichlorophenoxyl radi-
cal; TCP+, 2,4,6-trichlorocyclohexadienone cation; TFP, 2,4,
1867
dx.doi.org/10.1021/tx200215r |Chem. Res. Toxicol. 2011, 24, 1862–1868

Chemical Research in Toxicology
ARTICLE
(19) Sealy, R. C., Harman, L., West, P. R., and Mason, R. P. (1985)
The electron-spin resonance-spectrum of the tyrosyl radical. J. Am.
Chem. Soc. 107, 3401–3406.
(20) Mottley, C., and Mason, R. P. (2001) Sulfur-centered radical
formation from the antioxidant dihydrolipoic acid. J. Biol. Chem. 276,
42677–42683.
(21) Jacob, J. S., Cistola, D. P., Hsu, F. F., Muzaffar, S., Mueller,
D. M., Hazen, S. L., and Heinecke, J. W. (1996) Human phagocytes
employ the myeloperoxidase-hydrogen peroxide system to synthesize
dityrosine, trityrosine, pulcherosine, and isodityrosine by a tyrosyl
radical-dependent pathway. J. Biol. Chem. 271, 19950–19956.
(22) Bors, W., and Buettner, G. R. (1997) The vitamin C radical and
its reactions, in Vitamin C in Health and Disease (Packer, L., and Fuchs, J.,
Eds.) pp 75ꢀ94, Marcel Dekker, Inc., New York.
(23) Song, Y., Buettner, G. R., Parkin, S., Wagner, B. A., Robertson,
L. W., and Lehmler, H. J. (2008) Chlorination increases the persistence
of semiquinone free radicals derived from polychlorinated biphenyl
hydroquinones and quinones. J. Org. Chem. 73, 8296–8304.
(24) Song, Y., Wagner, B. A., Lehmler, H. J., and Buettner, G. R.
(2008) Semiquinone radicals from oxygenated polychlorinated biphe-
nyls: electron paramagnetic resonance studies. Chem. Res. Toxicol. 21,
1359–1367.
(25) Wong, S. K., Sytnyk, W., and Wan, J. K. S. (1972) Electron spin
resonance study of the self-disproportionation of some semiquinone
radicals in solution. Can. J. Chem. 50, 3052–3057.
(26) Lubitz, W., Dinse, K. P., Moebius, K., and Biehl, R. (1975)
Fluorine and proton ENDOR of aromatic radicals in solution. Chem.
Phys. 8, 371–383.
(27) Rakitin, A. R., Yff, D., and Trapp, C. (2003) Fluorine hyperfine
splittings in the electron spin resonance (ESR) spectra of aromatic
radicals. An experimental and theoretical investigation. J. Phys. Chem. A
107, 6281–6292.
1868
dx.doi.org/10.1021/tx200215r |Chem. Res. Toxicol. 2011, 24, 1862–1868