Redox mediation in the peroxidase-catalyzed oxidation of aminopyrine: Possible implications for drug-drug interactions

Article abstract of DOI:10.1021/tx950186t

Many drugs, industrial pollutants, and other xenobiotics are known to be oxidized by peroxidases to potentially harmful free-radical intermediates. We have examined the possibility that certain compounds, acting as efficient peroxidase substrates, may stimulate the formation of reactive free radicals by acting as mediators of electron transfer reactions (redox mediators). To explore this hypothesis, we have investigated the interaction of two well- known peroxidase substrates, chlorpromazine and aminopyrine. As shown by ESR and UV-visible spectroscopy, chlorpromazine radical was able to oxidize aminopyrine to aminopyrine cation radical. The rate constant for this rapid, pH-dependent, reaction was estimated to be 1 x 10⁷ M^-1 s^-1 at pH 4.5. Transient-state and steady-state kinetic studies both showed that rate constants for chlorpromazine oxidation to its cation radical by horseradish peroxidase (HRP) were about 100-fold greater than for the corresponding HRP- catalyzed oxidation of aminopyrine to its cation radical. When both aminopyrine and chlorpromazine were present with HRP and H₂O₂, aminopyrine cation radical formation was stimulated 100-fold. Concomitantly, the accumulation of chlorpromazine cation radical was completely inhibited in the presence of aminopyrine. Similar results were obtained when lactoperoxidase, myeloperoxidase, or the myeloperoxidase mimic HOCl were substituted for HRP. These data suggest that chlorpromazine can act as a redox mediator for peroxidase-catalyzed oxidation of aminopyrine and other chemicals. We suggest that some peroxidase substrates, acting as redox mediators, may stimulate the production of toxic free-radical intermediates from various drugs and other xenobiotics. As such, this may have implications for a number of adverse effects caused by these xenobiotic chemicals.

Full text of DOI:10.1021/tx950186t

4
76
Chem. Res. Toxicol. 1996, 9, 476-483
Red ox Med ia tion in th e P er oxid a se-Ca ta lyzed Oxid a tion
of Am in op yr in e: P ossible Im p lica tion s for Dr u g-Dr u g
In ter a ction s
Douglas C. Goodwin, Thomas A. Grover, and Steven D. Aust*
Biotechnology Center, Utah State University, Logan, Utah 84322-4705
Received November 6, 1995X
Many drugs, industrial pollutants, and other xenobiotics are known to be oxidized by
peroxidases to potentially harmful free-radical intermediates. We have examined the possibility
that certain compounds, acting as efficient peroxidase substrates, may stimulate the formation
of reactive free radicals by acting as mediators of electron transfer reactions (redox mediators).
To explore this hypothesis, we have investigated the interaction of two well-known peroxidase
substrates, chlorpromazine and aminopyrine. As shown by ESR and UV-visible spectroscopy,
chlorpromazine radical was able to oxidize aminopyrine to aminopyrine cation radical. The
7
-1 -1
rate constant for this rapid, pH-dependent, reaction was estimated to be 1 × 10 M
s
at
pH 4.5. Transient-state and steady-state kinetic studies both showed that rate constants for
chlorpromazine oxidation to its cation radical by horseradish peroxidase (HRP) were about
1
00-fold greater than for the corresponding HRP-catalyzed oxidation of aminopyrine to its cation
radical. When both aminopyrine and chlorpromazine were present with HRP and H O ,
2
2
aminopyrine cation radical formation was stimulated 100-fold. Concomitantly, the accumula-
tion of chlorpromazine cation radical was completely inhibited in the presence of aminopyrine.
Similar results were obtained when lactoperoxidase, myeloperoxidase, or the myeloperoxidase
mimic HOCl were substituted for HRP. These data suggest that chlorpromazine can act as a
redox mediator for peroxidase-catalyzed oxidation of aminopyrine and other chemicals. We
suggest that some peroxidase substrates, acting as redox mediators, may stimulate the
production of toxic free-radical intermediates from various drugs and other xenobiotics. As
such, this may have implications for a number of adverse effects caused by these xenobiotic
chemicals.
In tr od u ction
The heme-containing peroxidases are known to cata-
lyze the one-electron oxidation of a wide range of struc-
turally diverse aromatic compounds (1-3). This is
usually accomplished through the catalytic cycle first
This catalytic cycle differs slightly for substrates that
donate electrons rather than hydrogen atoms when
oxidized by peroxidases (7). In this case, the protons for
liberation of the ferryl oxygen atom as water do not come
from the substrate and II, but instead are likely to be
obtained from bulk solution.
described by Chance (4) for horseradish peroxidase
1
A unique feature of peroxidase-catalyzed reactions is
the oxidation of substrates to diffusible free-radical
intermediates (8). These substrate radicals are known
to participate in a variety of nonenzymatic reactions
including disproportionation, polymerization, and elec-
tron transfer (oxidative or reductive) (9, 10). One par-
ticularly important reaction of radicals (especially cation
radicals) derived from peroxidases is that of redox
mediation or mediation by electron transfer (11-13). This
(
HRP) and later refined by Dunford and other investiga-
tors (1, 5, 6). Hydrogen peroxide oxidizes ferric peroxi-
dase by two electrons to the enzyme intermediate com-
pound I, described as an oxyferryl porphyrin π cation
•
+
radical (reaction 1). The porphyrin radical (P ) accepts
k1
9
III
IV
•+
Fe + H O
8 Fe dO[P ] + H O
(1)
(2)
(3)
2
2
2
k2
9
IV
•+
IV
•
•
+
Fe dO[P ] + RH
8 Fe dO + RH
mechanism proposes that radicals (A ) generated by a
peroxidase may act as diffusible oxidants to oxidize
secondary molecules (B) (reaction 4). This type of reaction
2
k3
IV
III
•
Fe dO + RH
98 Fe + RH + H O
2
2
k4
9
A•+ + B
8 A + B
•+
(4)
one electron from an organic substrate yielding the
corresponding substrate free radical and an oxyferryl
heme intermediate known as compound II (reaction 2).
A subsequent one-electron reduction of II by a second
substrate molecule yields ferric peroxidase (reaction 3).
can have several consequences for the kinetics of peroxi-
dase-catalyzed oxidation reactions. Foremost is that the
rate of generation of secondary radicals (B ) can be
•+
dramatically stimulated in the presence of a redox
mediator (A). This type of mechanism may have toxico-
*
To whom correspondence should be addressed: Biotechnology
Center, Utah State University, Logan, UT 84322-4705. Phone, (801)
•
+
7
97-2730; FAX, (801) 797-2755; E-mail, sdaust@cc.usu.edu.
logical implications if B or a subsequent metabolite is
toxic. Under these circumstances, the presence of A may
stimulate the production of toxic metabolites produced
X
Abstract published in Advance ACS Abstracts, February 1, 1996.
1
•+
Abbreviations: AP, aminopyrine; AP , AP cation radical; CPZ,
chlorpromazine; CPZ•+, CPZ cation radical; HRP, horseradish peroxi-
dase.
•
+
by peroxidases. Furthermore, the accumulation of A
0
893-228x/96/2709-0476$12.00/0 © 1996 American Chemical Society
+
+

Peroxidase-Catalyzed Aminopyrine Oxidation
Chem. Res. Toxicol., Vol. 9, No. 2, 1996 477
adding 50 µM H to 50 nM HRP, 500 µM CPZ and 50 mM
can be completely inhibited in the presence of B. This
may have a range of consequences desirable or undesir-
able depending on the destructive and/or beneficial
2 2
O
acetate (pH 4.5). The spectrum of this reaction was then
recorded. AP (500 nmol) was then added and another spectrum
recorded. To observe spectral changes for the same reaction at
pH 7.0, a pH jump method was employed. A 500 µL solution of
•
+
•+
properties of A . If, for example, A had some thera-
peutic role, the presence of B would interfere with that
function.
•
+
2
00 µM CPZ was generated as described above except that
the concentration of acetate buffer (pH 5.0) was 2 mM. Upon
One reaction catalyzed by peroxidases that may be
affected by redox mediation is the oxidation of aminopy-
•+
•+
completion of CPZ formation, this 500 µL solution of CPZ
was added to a 500 µL solution of 2 mM AP in 100 mM
phosphate buffer (pH 7.0) and a spectrum recorded.
CP Z•+ Red u ction by AP . Solutions of CPZ•+ were prepared
•
+
rine (AP) to its radical (AP ). AP has been implicated
as a cause of agranulocytosis (14). While the exact
mechanism of toxicity of this compound is not known, it
has been proposed that it must first be oxidized before it
can exert its toxic effects (15, 16). Cytochromes P450,
prostaglandin synthase, and a variety of peroxidases
have been shown to oxidize AP to a number of potentially
in pH 4.5 acetate buffer (20 mM) using 50 nM HRP, 1 mM H
2 2
O ,
and 2.2 mM CPZ. This reaction was allowed to proceed to
•+
completion. In order to stabilize the CPZ , this solution (1 mL)
was then transferred to 9 mL of 0.1 M NaCl (pH 1.0). Rate
•
+
constants for CPZ reduction by AP were determined using a
three-syringe stopped-flow spectrophotometer from KinTek
•
+
toxic metabolites, including AP (15-18). Oxidation of
AP by cytochromes P450, however, is a relatively slow
reaction, and it has been shown that the rates for HRP
oxidation of this chemical are 2-3 orders of magnitude
faster (18). Myeloperoxidase has also been shown to
oxidize AP through the production of HOCl (15, 16, 19).
This is particularly interesting given the immunological
side effects of AP. For these reasons, the rate of oxidation
of AP by peroxidases becomes important in examining
the toxicity of AP and other similar chemicals with toxic
side effects.
•
+
Instruments (State College, PA). Solutions of CPZ (<30 µM)
were loaded into the first syringe with 0.1 N HCl in 0.1 M NaCl
(HCl/NaCl). For these experiments double mixing was not
required, so the second syringe was loaded only with HCl/NaCl.
The third syringe was loaded with varying concentrations of AP
•
+
in HCl/NaCl. This arrangement allowed for mixing of CPZ
with AP just prior to entering the observation cell. Concentra-
•
+
tions of AP were at least 10-fold higher than CPZ to ensure
pseudo-first-order conditions. Rate constants were obtained in
this manner over a pH range of 0.0-2.0. Determination of rate
constants for this reaction at higher pH were complicated by
•
+
the increasing instability of CPZ and the increasing rate of
the reaction. A rate constant at pH 4.5 was estimated from
results obtained at the lower pH values.
We propose that the rate of oxidiation of AP catalyzed
by HRP, myeloperoxidase, and other peroxidases may be
stimulated in the presence of other peroxidase substrates
acting as redox mediators. To examine this hypothesis,
we have used chlorpromazine (CPZ), a common peroxi-
dase substrate, as a redox mediator. Using HRP, AP,
ESR Sp ectr a of Ra d ica l In ter m ed ia tes. All ESR spectra
were recorded at room temperature using a ECS-106 spectrom-
eter (Bruker Instruments, Billerica, MA). The instrument was
operated at 9.75 GHz with a 50 kHz modulation frequency, and
•
+
4
•+
and CPZ in a model system, we demonstrate that CPZ ,
a gain of 2 × 10 . The reaction between CPZ and AP was
•
+
examined using an aqueous flat cell flow system. A solution of
0 µM CPZ at pH 1.0 (prepared as described above) flowed
generated by HRP, can oxidize AP to AP and that CPZ
stimulates the rate of AP oxidation by 2 orders of
magnitude. Furthermore, we show that a similar effect
is observed with the myeloperoxidase mimic HOCl and
the mammalian peroxidases lactoperoxidase and myelop-
eroxidase. The implications of the kinetics of this mech-
anism in the toxicity of AP and other similar compounds
is discussed.
•
+
5
through one line into the flat cell. The other line contained
either 800 µM AP in 100 mM sodium acetate (pH 7.5) or buffer
alone. The pH of the final reaction was 4.9. The spectrometer
was set to 20 mW microwave power and 4.74 G modulation
amplitude. The time constant was 40.96 ms with a sweep rate
of 166 G/s. All other spectrometer settings are indicated in the
•
+
•+
figure legend. Formation of CPZ and AP by HRP under
various conditions was determined using the same spectrometer
equipped with a dielectric mixing resonator flow system (Bruker
Instruments). For these experiments a microwave power of 4.0
mW was used, and the modulation amplitude was set to 0.376
G. The time constant was set to 655.36 ms, and the sweep rate
was 35.8 G/min. All other settings were as indicated in the
figure legend. The three-syringe flow system (see Figure 3)
Ma ter ia ls a n d Meth od s
Ca u tion : The following chemicals are hazardous and should
be handled carefully: AP, CPZ, and the phenothiazines listed
in Table 2. Proper protection should be used while these
compounds are handled.
Ch em ica ls. Hydrogen peroxide (30%), CPZ, and the other
phenothiazines were purchased from Sigma (St. Louis, MO) and
used without further purification. AP was purchased from
•
+
was set up such that 100 µM CPZ was generated by mixing
0 nM HRP (syringe A) with 500 µM CPZ and 50 µM H
syringe B) in a delay line. The residence time of the reactants
5
2 2
O
(
•
+
in the delay line was about 40 s. The CPZ was then mixed
with H O from syringe C and the resulting spectrum recorded.
Aldrich (Milwaukee, WI). Concentrations of H
2 2
O were deter-
-1
-1
2
mined at 240 nm using an extinction coefficient of 39.4 M cm
•
+
The contents of syringe B were then changed to contain either
500 µM AP alone or 500 µM AP and 500 µM CPZ. The resulting
ESR spectra were then recorded. The flow rate was regulated
using a Harvard Apparatus Model 22 syringe pump (South
Natick, MA).
(
20). Concentrations of CPZ were determined at 525 nm using
4
-1
-1
an extinction coefficient of 1.21 × 10 M cm (7). Concentra-
tions of AP• were measured at 600 nm using an extinction
+
3
-1
-1
coefficient of 2.25 × 10 M cm derived from the extinction
3
-1
-1
coefficient of 2.43 × 10 M cm at 580 nm obtained by Sayo
and Masui (21).
Sin gle-Tu r n over Exp er im en ts. Single-turnover experi-
ments were performed using the stopped-flow spectrophotometer
described above. Compound I of HRP or lactoperoxidase was
formed as described previously for HRP (22). I was then reacted
with CPZ or AP, and the return of the enzyme to ferric
peroxidase was monitored at 398 (HRP) or 412 nm (lactoper-
oxidase).
En zym es. The HRP (Type VI), bovine milk lactoperoxidase,
and human leukocyte myeloperoxidase used in these experi-
ments were purchased from Sigma. Concentrations of HRP
were determined at 402 nm using an extinction coefficient of
5
-1
-1
1
.02 × 10 M cm (1), and concentrations of lactoperoxidase
were determined at 412 nm using an extinction coefficient of
5
-1
-1
1
.14 × 10 M cm (1).
Tr a n sien t-Sta te Kin etics. Rate constants for reduction of
HRP compound I to HRP compound II and reduction of HRP
compound II to ferric peroxidase by CPZ and AP were also
determined using the stopped-flow spectrophotometer. I was
UV-Visible Sp ectr a of Ra d ica l In ter m ed ia tes. All spec-
tra were recorded using a Shimadzu UV2101-PC spectropho-
tometer. A 995 µL solution of 100 µM CPZ was generated by
•
+
+
+

4
78 Chem. Res. Toxicol., Vol. 9, No. 2, 1996
Goodwin et al.
F igu r e 1. Electron spin resonance spectra obtained during the
•
+
reaction between CPZ and AP. Spectrum A was recorded
•
+
while 50 µM CPZ flowed in one line together with buffer in a
second line to the ESR flow cell. Spectrum B was recorded
exactly as in (A) except that the second line also contained 800
µM AP. The sweep width for both spectra was 200 G. No ESR
•
+
spectra were observed when CPZ was absent.
formed as previously described (22, 23). Reduction of I and II
was monitored at 428 nm, the isosbetic wavelength for ferric
HRP and HRP compound I (24). The rate constant for reduction
of HRP compound I by CPZ was determined using a method
recently developed by this laboratory (22).
HRP Stea d y-Sta te Kin etics. To ensure that true initial
•
+
•+
rates were obtained, oxidation of CPZ to CPZ and AP to AP
by HRP was monitored using the stopped-flow spectrophotom-
eter. The instrument was in the single-mixing mode such that
5
00 µM H
before entering the observation cell. Initial rates of CPZ
formation were monitored at 525 nm. Initial rates of AP
2 2
O was mixed with substrate and 50 nM HRP just
•
+
•
+
oxidation to AP in the presence and absence of CPZ were
monitored at 600 nm.
F igu r e 2. Absorption spectra of CPZ•+ and AP•+ and change
in absorbance at 525 nm obtained during the reaction between
CPZ and AP. The absorption spectrum of CPZ (panel A, solid
Rea ction s of Myelop er oxid a se a n d HOCl. Oxidation of
•+
•+
CPZ to CPZ• and AP to AP by HOCl were monitored using
the KinTek stopped-flow apparatus set at 525 and 580 nm,
respectively. The instrument was placed in the single-mixing
mode such that HOCl and AP or CPZ were mixed just before
entering the observation cell. Experiments with myeloperoxi-
dase were performed using a Shimadzu UV-2101PC spectro-
photometer. Myeloperoxidase activity was measured using the
method described by Desser and co-workers (25).
+
•+
line) was obtained by reaction of 500 µM CPZ with 50 nM HRP
2 2
and 50 µM H O in 50 mM acetate buffer, pH 4.5. The
•+
absorption spectrum of AP (panel A, dashed line) was obtained
•
+
by adding 500 µM AP to the CPZ generated as above. Panel
•
+
B shows the change in absorbances at 525 nm during CPZ
reduction by AP at pH 1.0. Concentrations after mixing were
•
+
3
.3 µM CPZ and 500 µM AP.
rate constant for this reaction increased with increasing
•+
pH. The increasing instability of CPZ , and the increas-
ing rates of reaction at higher pH, made it difficult to
obtain a rate constant at pH 4.5. However, on the basis
of the change in rate constant from pH 0 to pH 2, we
Resu lts
•
+
•+
AP Oxid a tion by CP Z . The oxidation of AP to AP
•+
by CPZ was shown (Figure 1) using an ESR spectrom-
eter equipped with an aqueous flat cell flow system.
7
-1 -1
have made a conservative estimate of 1 × 10 M
s
•
+
•
+
for the rate constant of AP oxidation by CPZ at pH 4.5.
When a solution of CPZ flowed into the flat cell and
mixed with buffer, the resulting spectrum was that of
While we were unable to obtain rate constants for AP
•+
•+
•+
•+
CPZ . However, when CPZ was mixed with buffer and
AP in the flat cell, the resulting spectrum was that of
AP . The spectra shown in Figure 1 were obtained
under overmodulating conditions (4.74 G modulation
amplitude) because of the need for increased sensitivity
oxidation to AP by CPZ at higher pH values, spectral
changes similar to those shown in Figure 2A were also
observed at pH 7.0. This indicated that the rapid
•+
•
+
oxidation of AP by CPZ also occurred at physiological
pH.
•
+
(
especially for AP ). Similar results were also obtained
Oxid a tion of CP Z a n d AP by HRP : Stea d y-Sta te
•
+
using a lower modulation amplitude (0.946 G).
The oxidation of AP by CPZ•+ was confirmed using
visible spectroscopy (Figure 2). A reaction mixture
a n d Tr a n sien t-Sta te Kin etics. Generation of CPZ
•+
and AP by HRP under multiple-turnover conditions was
carried out using the ESR three-syringe flow system
shown in Figure 3. The formation of CPZ was ac-
complished in the delay line by mixing the contents of
•
+
containing CPZ and HRP with limited H
2
O
2
(50 µM) was
•
+
used to generate 100 µM CPZ (Figure 2A, solid line).
The visible absorption spectrum of this radical was
consistent with a previous report (26). Upon addition of
AP, the observed spectrum immediately changed to one
syringe A (HRP) and syringe B (CPZ and H
2
O
2
). When
•+
H
2
O (syringe C) was mixed with CPZ in the ESR cavity,
•
+
the resulting spectrum was that of CPZ (Figure 4A)
•
+
characteristic of AP (Figure 2A, dashed line) (21). A
(27). Conversely, when AP was placed in syringe B with
•
+
typical reaction curve monitored at 525 nm is shown in
CPZ and H
spectrum typical of the AP was observed (Figure 4B)
(18, 21). When AP and H were placed in syringe B
2 2
O , CPZ was no longer detected, and a
•
+
•+
Figure 2B. The kobs for CPZ reaction with AP was
linearly dependent on AP concentration, and a second-
2 2
O
4
-1 -1
•+
order rate constant of 3.2 × 10 M
s
was obtained for
without CPZ, a barely detectable AP spectrum was
observed (Figure 4C).
this reaction at pH 1.0. Our results indicated that the
+
+

Peroxidase-Catalyzed Aminopyrine Oxidation
Chem. Res. Toxicol., Vol. 9, No. 2, 1996 479
F igu r e 5. Change in absorbance at 398 nm during a single
turnover of HRP in the presence of CPZ or AP. Concentrations
of reactants after mixing were 1.0 µM HRP compound I (from
reaction between 1.1 µM HRP and 1.0 µM H O ) and 10 µM
2 2
CPZ (a) or 10 µM AP (b). Both reactions were carried out in 30
mM acetate buffer, pH 4.5.
F igu r e 3. Schematic diagram representing the three-syringe
continuous-flow system used for obtaining ESR spectra. The
dielectric mixing resonator cavity is represented by the X.
F igu r e 4. Electron spin resonance spectra of CPZ•+ and AP•+
obtained during reactions between HRP, H O , and CPZ and/
2 2
or AP. All radicals were generated using the three-syringe flow
system diagrammed in Figure 3. Spectrum A was generated
using 150 nM HRP in syringe A, 150 µM H
2 2
O and 1.5 mM CPZ
in syringe B, and H O in syringe C. Spectrum B was obtained
2
under the same conditions as spectrum A, except 1.5 mM AP
was placed in syringe B with CPZ. Spectrum C was obtained
under the same conditions as spectrum B except that CPZ was
excluded from syringe B. All reactions were carried out using
F igu r e 6. Effect of CPZ on AP oxidation by HRP. All reactions
were carried out at pH 4.5 using 50 mM acetate buffer with 50
2 2
nM HRP, 500 µM H O , and varying concentrations of AP (b)
or 1 mM AP and varying concentrations of CPZ (O). The points
represent the mean of triplicate measurements. The standard
deviation for each point was less than 1% of the Y-axis scale.
3
0 mM acetate buffer, pH 4.5. The sweep width for all spectra
collected was 200 G.
The results of this experiment suggested that HRP
directly oxidized AP much more slowly than CPZ. This
was confirmed by the fact that the rate of a single
turnover of HRP is much faster in the presence of 10 µM
CPZ than with 10 µM AP (Figure 5). Moreover, the
apparent second-order rate constants for reduction of
Ta ble 1. Tr a n sien t-Sta te a n d Stea d y-Sta te Ra te
Con sta n ts for Oxid a tion of Am in op yr in e a n d
Ch lor p r om a zin e by Hor ser a d ish P er oxid a se
transient state
k2 (M-1 k3 (M-1
steady state
kapp (M-1
Km (mM)
-1
-1
-1
substr
AP
s
)
s
)
s )
5
3
3
(1.0 ( 0.1) × 10 (4.1 ( 0.3) × 10 (7.1 ( 0.1) × 10 >10
both HRP compound I (k
2
) and HRP compound II (k
3
)
CPZ (1.7 ( 0.2) × 10
7
(4.5 ( 0.2) × 10
5
(7.1 ( 0.4) × 10
5
2.1 ( 0.4
were about 2 orders of magnitude greater with CPZ than
with AP (Table 1).
•+
oxidize AP to AP , suggest that the oxidation of AP by
HRP would be greatly stimulated in the presence of CPZ.
Oxid a t ion of AP b y HR P in t h e P r esen ce of
P h en oth ia zin es. In the absence of CPZ, increasing
The apparent second-order rate constants for steady-
state oxidation of AP and CPZ, monitored by the forma-
tion of their respective cation radicals, are also shown in
Table 1. These data show that all of the transient-state
rate constants can account for the observed steady-state
behavior. Furthermore, comparison of steady-state and
transient-state rate constants indicates that compound
II reduction is the rate-limiting step in the catalytic cycle
for oxidation of CPZ and AP by HRP. These rate
•+
concentrations of AP increased the rate of AP formation
consistent with the observed rate constants in Table 1
(Figure 6, closed circles). When oxidation of 1 mM AP
was examined in the presence of increasing concentra-
•
+
tions of CPZ, the rate of AP formation was stimulated
dramatically (Figure 6, open circles). The apparent
second-order rate constant obtained from this curve was
•
+
constants, coupled with the observation that CPZ can
+
+

4
80 Chem. Res. Toxicol., Vol. 9, No. 2, 1996
Goodwin et al.
F igu r e 8. Effect of CPZ on the rate of AP oxidation to AP•+ by
HOCl. The reactions contained 20 µM HOCl, 200 µM AP, and
F igu r e 7. Change in absorbance at 412 nm during a single
turnover of lactoperoxidase in the presence of CPZ or AP.
Concentrations of reactants after mixing were 1.0 µM lacto-
peroxidase compound I (from reaction between 1.1 µM lacto-
2
0 µM CPZ (a) or no CPZ (b). The visible absorption spectra
(
inset) were obtained using 100 µM HOCl, 500 µM AP, and 500
µM CPZ (solid line) or no CPZ (dashed line). All reactions were
carried out in 30 mM acetate buffer, pH 5.0.
peroxidase and 1.0 µM H
b). Both reactions were carried out in 30 mM phosphate buffer,
pH 6.0.
2 2
O ) and 4 mM CPZ (a) or 4 mM AP
(
over of lactoperoxidase was much slower than HRP with
both substrates.
Ta ble 2. Ap p a r en t Secon d -Or d er Ra te Con sta n ts for
Stim u la tion of AP •+ F or m a tion by Va r iou s
The presence of CPZ had a similar effect on the rate
of AP oxidation by myeloperoxidase and its chemical
analog HOCl. It has been demonstrated that myelo-
a
P h en oth ia zin es
1
compound
structure
kapp (M- s-1
)
6
6
6
6
6
5
4
thioridazine
promazine
R ) SCH3; R′ ) 4
R ) H; R′ ) 1
(5.0 ( 0.3) × 10
(3.3 ( 0.3) × 10
(2.3 ( 0.2) × 10
(1.7 ( 0.1) × 10
(1.3 ( 0.1) × 10
(1.8 ( 0.2) × 10
(8.3 ( 0.4) × 10
-
peroxidase/H
2
O
2
/Cl or HOCl alone can oxidize AP to
•
+
AP (15, 16). Using stopped flow, we determined a rate
trifluoperazine
trimeprazine
fluphenazine
promethazine
trifluopromazine
R ) CF3; R′ ) 6
R ) H; R′ ) 2
6
-1 -1
•+
constant of (1.7 ( 0.1) × 10 M
s
for CPZ formation
R ) CF3; R′ ) 5
R ) H; R′ ) 3
R ) CF3; R′ ) 1
by HOCl. Similarly, a rate constant of (2.07 ( 0.05) ×
3 -1 -1 •+
1
0 M
s
was previously determined for AP formation
by HOCl (16). According to these rate constants, the
S
•+
presence of CPZ should stimulate AP oxidation to AP
by HOCl by 3 orders of magnitude. The effect of CPZ on
N
R
•+
AP formation by HOCl (monitored at 580 nm) is shown
in Figure 8. In the absence of CPZ, there was little
R′
–(CH2)3N(CH3)2
–(CH2)2CH(CH3)N(CH3)2
–CH2CH(CH3)N(CH3)2
1
2
3
•+
accumulation of AP . However, when 20 µM CPZ was
•
+
–(CH2)2
present with HOCl and AP, AP rapidly accumulated.
–(CH2)3N
N(CH2)2OH
–(CH2)3N
NCH3
H3CN
A linear relationship was observed between the kobs for
AP formation and CPZ concentration. A rate constant
5
6
4
•+
a
6
-1 -1
Experiments performed exactly as described for Figure 6
except that CPZ was replaced by each listed phenothiazine.
of (1.8 ( 0.1) × 10 M
s
was obtained from the slope
of this line. Within error, this is identical to the rate
constant for CPZ oxidation by HOCl. To ensure that the
change in absorbance observed at 580 nm was not due
to CPZ , spectra were obtained for AP oxidation by HOCl
in the presence and absence of CPZ (Figure 8, inset).
5
-1
(
7.8 ( 0.3) × 10 M s-1 and the K
m
was 2.1 ( 0.2 mM.
These values are nearly identical to those obtained for
direct oxidation of CPZ by HRP (see Table 1).
•+
A number of other phenothiazine compounds showed
results similar to those obtained with chlorpromazine.
The apparent second-order rate constants for AP oxida-
tion in the presence of increasing concentrations of each
phenothiazine compound are shown in Table 2. It is
important to point out that each rate constant exceeds
that obtained for direct oxidation of AP by HRP.
Effect of CP Z on AP Oxid a tion by Oth er P er oxi-
d a ses/P er oxid a se Mim ics. Similar to the kinetics ob-
served for HRP, lactoperoxidase turnover in the presence
of CPZ was much faster than that observed with the
same concentration of AP (Figure 7). Consistent with
these results, the rate of CPZ oxidation by lactoperoxi-
dase was about 30-fold greater than AP oxidation by the
same enzyme (data not shown). As with HRP, the
presence of CPZ greatly stimulated AP oxidation by
lactoperoxidase. It should be noted, however, that turn-
•
+
These spectra clearly show that no CPZ accumulated
during the reaction.
When myeloperoxidase was used, CPZ had a similar
•+
stimulatory effect on AP oxidation to AP (Figure 9). This
-
effect was observed in the presence and absence of Cl .
-
When Cl was absent, CPZ had a more dramatic effect,
•
+
but the maximal rate of AP formation was observed in
-
the presence of both CPZ and Cl . It is important to point
out that in each reaction progressive inactivation of
myeloperoxidase was observed. Furthermore, the rate
of inactivation of myeloperoxidase was correlated with
•+
the rate of AP generation. The most rapid inactivation
-
occurred in the presence of Cl , CPZ, and AP. In the
absence of AP (Figure 9B, dotted line), much less my-
eloperoxidase inactivation was observed.
+
+

Peroxidase-Catalyzed Aminopyrine Oxidation
Chem. Res. Toxicol., Vol. 9, No. 2, 1996 481
lated during the reaction until all of the AP had been
oxidized. This is consistent with the determined rate
constants for each of the reactions involved. In each case,
•+
the reaction between CPZ and AP was not rate limiting.
•
+
Thus, CPZ was reduced by AP as rapidly as it was
•
+
formed by the peroxidases, and no CPZ accumulated
during these reactions. This is a typical kinetic feature
of redox mediation in peroxidase catalysis (11, 12, 22).
•
+
Often the mediation reaction (i.e., A + B) exceeds the
rate of turnover of a given peroxidase. Thus, the ac-
•
+
cumulation of oxidized mediator (A ) is completely
inhibited.
The myeloperoxidase/Cl /H
-
2
O
2
and HOCl reactions are
more complicated than the corresponding HRP and
lactoperoxidase reactions. However, the same stimula-
•
+
tion of AP formation was observed in the presence of
•+
CPZ. It has been proposed that formation of AP by the
reaction of HOCl with AP proceeds through a multistep
mechanism. This mechanism is believed to involve the
formation of a chlorinated AP intermediate followed by
-
the loss of Cl to form a reactive AP dication (16).
Reaction of the AP dication with AP is believed to yield
•
+
two equivalents of AP (16). The rate-limiting reaction
would appear to be the last step, which occurs with a
3
-1 -1
rate constant of 2.07 × 10 M
s
(16). While we have
•
+
not explored such a mechanism for CPZ formation by
HOCl, it is possible that a similar mechanism may apply.
Most importantly, we have determined that the reaction
between CPZ and HOCl to form CPZ occurs with a rate
constant that is nearly 3 orders of magnitude higher than
•+
F igu r e 9. Effect of CPZ on AP oxidation to AP•+ by myelop-
•
+
-
AP formation by HOCl. Given the rapid reaction
between CPZ and AP, it is not surprising that AP
2 2
eroxidase and H O in the presence and absence of Cl . Panel
•
+
•+
•+
A shows the change in AP concentration with time in the
-
absence of Cl . Reactions depicted in panel B contained 100
formation by HOCl is stimulated by 3 orders of magni-
tude in the presence of CPZ.
-
mM Cl . All reactions contained 0.35 unit of myeloperoxidase,
50 µM H O , 1 mM AP, and 1 mM CPZ (a) or no CPZ (b). The
2 2
dotted line in panel B indicates CPZ formation in the absence
of AP. All reactions were carried out in 50 mM acetate buffer,
pH 5.0.
2
•
+
Similar to the results obtained with HOCl and the
other peroxidases, CPZ had a stimulating effect on the
oxidation of AP by myeloperoxidase. Interestingly, Cl-
was not required to observe this effect of CPZ. This
indicates that CPZ is directly oxidized by myeloperoxi-
dase at a significant rate. Indeed, CPZ showed the most
Discu ssion
Oxidation of AP to AP•+ by peroxidases is well docu-
mented (15-19). We demonstrate here that AP forma-
tion by peroxidases can be stimulated in the presence of
another peroxidase substrate. In this study, the presence
of CPZ stimulated AP formation by each peroxidase or
peroxidase mimic examined by 30-1000-fold. Kinetic
analysis of this stimulation suggests that CPZ acts as a
redox mediator in peroxidase-catalyzed AP oxidation.
•+
dramatic enhancement of AP formation in the absence
•+
-
•+
of Cl . Maximal rates of AP formation, however, were
-
observed when both Cl and CPZ were included in the
reaction mixture.
•
+
Of particular interest is the apparent inactivation of
•+
myeloperoxidase in the presence of AP . Our results
•+
indicate that increased rates of AP accumulation lead
to increased rates of myeloperoxidase inactivation. One
proposed mechanism of AP-induced agranulocytosis sug-
gests that covalent modification of some neutrophilic
protein (possibly myeloperoxidase) leads to an autoim-
mune response against neutrophils and other granulo-
This conclusion is based on three observations. First,
CPZ can oxidize AP to AP . This is consistent with
the observed one-electron reduction potentials for AP /
AP (about 0.58 V vs NHE) (21) and CPZ /CPZ (0.78 V
vs NHE) (28). Our results suggest that, typical of many
free-radical reactions, AP oxidation by CPZ occurs at
high rates. Furthermore, the rate of this reaction in-
creases dramatically with increasing pH. Second, redox
mediation is evident from the comparative rates of direct
oxidation of CPZ and AP by each peroxidase or peroxidase
mimic. In each case, CPZ was oxidized by 1-3 orders of
magnitude faster than AP. Thus, when both substrates
•+
•+
•
+
•
+
•
+
cytes (29). Inactivation of myeloperoxidase by AP
suggests the possibility that this protein may be co-
•
+
•
+
valently modified by AP . As such, this could represent
a mechanism by which AP causes agranulocytosis. An-
other group has recently suggested that the toxic me-
•
+
tabolite of AP may be an AP dication as opposed to AP
(16). Further study will be required to determine the
mechanism of AP-induced agranulocytosis.
•
+
•+
were present in the same reaction, the rate of AP
The relative rates of AP formation by peroxidases in
formation could not be explained by direct oxidation of
AP by each peroxidase. Conversely, the rate of CPZ
oxidation by each peroxidase was sufficient to explain the
observed rate of AP oxidation when both substrates were
present. Finally, when both CPZ and AP were present
the presence and absence of CPZ suggests that AP
oxidation by these enzymes can be dramatically stimu-
lated in the presence of other chemicals. The use of AP
as an analgesic medication led to numerous reports of
agranulocytosis (14). For this reason, AP is no longer
licensed for use in many countries (16). It should be
•
+
together in the peroxidase reactions, no CPZ accumu-
+
+

4
82 Chem. Res. Toxicol., Vol. 9, No. 2, 1996
Goodwin et al.
noted, however, that AP and other toxic drugs have been
found in some imported herbal pain remedies and that
use of these remedies has also caused agranulocytosis
Ack n ow led gm en t. This work was supported by
NIEHS Grant ES05056. D.C.G. is generously supported
by a Willard Eccles Family Foundation Fellowship. The
authors thank Isao Yamazaki for helpful discussions,
Curtis Takemoto and J oseph Benson for excellent techni-
cal assistance, and Terri Maughan for excellent secre-
tarial assistance.
(30). While the exact mechanism of toxicity of AP has
not been elucidated, some have suggested that it may
•
+
arise from the formation of AP (15, 19, 31). Thus,
stimulation of the production of this intermediate by
other chemicals acting as peroxidase substrates may
dramatically increase the toxicity of AP. Furthermore,
the toxicity and/or the carcinogenicity of many other
aromatic compounds is also thought to occur through
free-radical intermediates generated by peroxidases (31-
Refer en ces
(
(
(
(
1) Dunford, H. B., and Stillman, J . S. (1976) On the function and
mechanism of action of peroxidases. Coord. Chem. Rev. 19, 187-
251.
2) Yamazaki, I., and Piette, L. H. (1963) The mechanism of aerobic
oxidase reactions catalyzed by peroxidase. Biochim. Biophys. Acta
3
3). The rate of formation of these radicals by peroxi-
dases may be similarly affected by efficient peroxidase
substrates such as CPZ.
It is known that CPZ•+ is a potent oxidant capable of
7
7, 47-64.
3) Marklund, S., Ohlsson, P.-I., Opara, A., and Paul, K.-G. (1974)
The substrate profiles of the acidic and slightly basic horseradish
peroxidases. Biochim. Biophys. Acta 350, 304-313.
4) Chance, B. (1952) The kinetics and stoichiometry of the transition
from the primary to the secondary peroxidase-peroxide complexes.
Arch. Biochem. Biophys. 41, 416-424.
oxidizing a wide range of organic compounds including
NADPH, ascorbate, and dopamine (31). Indeed, we have
•
+
observed that CPZ can oxidize N,N-dimethylanaline,
(
5) Dunford, H. B. (1982) Peroxidases. Adv. Inorg. Biochem. 41, 41-
dimethyl-p-toluidine, isoniazid, iproniazid, and other
chemicals.2 Previous publications on oxidation of these
68.
(
6) Dunford, H. B. (1992) Horseradish peroxidase: Structure and
kinetic properties. In Peroxidases in Chemistry and Biology
(Everse, J ., Everse, K. E., and Grishman, M. B., Eds.) Vol. II, pp
and similar chemicals by HRP (34, 35) suggest that the
presence of CPZ would also stimulate oxidation of these
chemicals. Thus, the production of many free-radical
intermediates could be accomplished through the oxida-
tion of CPZ by peroxidases.
1
-24, CRC Press, Boca Raton, FL.
(
7) Kelder, P. P., de Mol, N. J ., Fischer, M. J . E., and J anssen, L. H.
M. (1994) Kinetic evaluation of the oxidation of phenothiazine
derivatives by methemoglobin and horseradish peroxidase in the
presence of hydrogen peroxide: implications for the reaction
mechanism. Biochim. Biophys. Acta 1205, 230-238.
The phenothiazine structure of CPZ is a common
component of many medications used as antipsychotics,
antihistaminics, antiemetics, antipruritics, antihelmin-
tics, and tranquilizers (36). Phenothiazines are also
widely known to act as good peroxidase substrates (7,
(
8) Yamazaki, I., Mason, H. S., and Piette, L. (1960) Identification,
by electron paramagnetic resonance spectroscopy, of free radicals
generated by peroxidase. J . Biol. Chem. 235, 2444-2449.
(9) Yamazaki, I. (1959) The oxidoreductive feature of intermediates
formed in the reaction of peroxidase. Proc. Int. Symp. Enzyme
Chem. (Tokyo and Kyoto, 1958) pp 224-229, Marazen, Tokyo.
10) Pryor, W. A. (1976) The role of free radical reactions in biological
systems. In Free Radicals in Biology (Pryor, W. A., Ed.) Vol. I,
pp 1-49, Academic Press, Inc., New York.
11) Goodwin, D. C., Aust, S. D., and Grover, T. A. (1995) Evidence
for veratryl alcohol as a redox mediator in lignin peroxidase-
catalyzed oxidation. Biochemistry 34, 5060-5065.
2
7, 37). Moreover, the one-electron reduction potentials
of many of these compounds are sufficient to carry out
oxidation of many drugs, xenobiotics, and other chemi-
cals. Indeed, we have shown that many phenothiazine
compounds can replace CPZ as redox mediators. We
suggest that redox mediation, similar to that observed
between CPZ and AP, may play an important role in the
toxicity of some medications and xenobiotics. This would
suggest that peroxidases may also be involved in the
activation of these foreign compounds to toxic (or possibly
carcinogenic) intermediates. Indeed, peroxidases (par-
ticularly myeloperoxidase) have been implicated as a
causal factor for the toxic side effects of many medica-
tions, including a number of idiosyncratic drug reactions
(
(
(
(
12) Chung, N., and Aust, S. D. (1995) Veratryl alcohol mediated
indirect oxidation of phenol by lignin peroxidase. Arch. Biochem.
Biophys. 316, 733-737.
13) Yamazaki, I. (1977) Free radicals in enzyme-substrate reactions.
In Free Radicals in Biology (Pryor, W. A., Ed.) Vol. III, pp 183-
2
18, Academic Press, Inc., New York.
(14) Madison, F. W., and Squier, T. L. (1934) The etiology of primary
granulocytopenia (agranulocytic angina) J . Am. Med. Assoc. 102,
7
55-759.
(
15) Kalyanaraman, B., and Solne, P. G. (1985) Generation of free
radical intermediates from foreign compounds by neutrophil-
derived oxidants. J . Clin. Invest. 75, 1618-1622.
(29, 38). It is possible that these or similar side effects
may be amplified by other compounds that can act as
efficient peroxidase substrates and redox mediators. The
mechanism proposed here also suggests that compounds
with no known toxic side effects may become toxic in the
presence of other simultaneously administered chemicals.
As such, this may represent a new mechanism for
adverse drug-drug interactions.
(16) Uetrecht, J . P., Ma, H. M., MacKnight, E., and McClelland, R.
(
1995) Oxidation of aminopyrine by hypochlorite to a reactive
dication: possible implications for aminopyrine-induced agranu-
locytosis. Chem. Res. Toxicol. 8, 226-233.
(17) Eling, T. E., Mason, R. P., and Sivarajah, K. (1985) The formation
of aminopyrine cation radical by the peroxidase activity of
prostaglandin H synthase and subsequent reactions of the radical.
J . Biol. Chem. 260, 1601-1607.
In conclusion, redox mediation in peroxidase catalysis
represents a possible mechanism by which the production
of harmful free-radical intermediates may be dramati-
cally stimulated. We have shown that CPZ, acting as a
redox mediator, can stimulate peroxidase-catalyzed oxi-
dation of AP by up to 1000-fold. This type of interaction
may have implications for the toxicity and/or carcinoge-
nicity of many xenobiotics or medications whose toxicity
is thought to occur due to the production of reactive free-
radical species.
(18) Griffin, B. W. (1977) Free radical intermediate in the N-de-
methylation of aminopyrine by horseradish peroxidase-hydrogen
peroxide. FEBS Lett. 74, 139-143.
(
19) Sayo, H., and Saito, M. (1990) The mechanism of myeloperoxidase-
catalyzed oxidation of aminopyrine. Xenobiotica 20, 957-965.
20) Nelson, D. P., and Kiesow, L. A. (1972) Enthalpy of decomposition
of hydrogen peroxide by catalase at 25 °C (with molar extinction
(
coefficients of H
2 2
O in the UV). Anal. Biochem. 49, 474-478.
(21) Sayo, H., and Masui, M. (1973) Anodic oxidation of amines. Part
III. cyclic voltammetry and controlled potential electrolysis of
4
-dimethylaminoantipyrine (4-dimethylamino-2,3-dimethyl-1-
3
phenyl-∆ -pyrolizin-5-one) in acetonitrile. J . Chem. Soc., Perkin
Trans. 2, 1640-1645.
(
22) Goodwin, D. C., Aust, S. D., and Grover, T. A. (1995) Determi-
nation of rate constants for rapid peroxidase reactions. Anal.
Biochem. 231, 333-338.
2
Goodwin, D. C., Aust, S. D., and Grover, T. A., unpublished ob-
servations.
+
+

Peroxidase-Catalyzed Aminopyrine Oxidation
Chem. Res. Toxicol., Vol. 9, No. 2, 1996 483
(
23) Kuan, I.-C., J ohnson, K. A., and Tien, M. (1993) Kinetic analysis
of manganese peroxidase: the reaction with manganese com-
plexes. J . Biol. Chem. 268, 20064-20070.
(31) Mason, R. P. (1982) Free-radical intermediates in the metabolism
of toxic chemicals. In Free Radicals in Biology (Pryor, W. A., Ed.)
Vol. V, pp 161-222, Academic Press, Inc., New York.
(
24) Huang, J ., and Dunford, H. B. (1991) One-electron oxidative
activation of 2-aminofluorene by horseradish peroxidase com-
pounds I and II: spectral and kinetic studies. Arch. Biochem.
Biophys. 287, 257-262.
(32) T’so, P. O. P., Caspary, W. J ., and Lorentzen, R. J . (1977) The
involvement of free radicals in chemical carcinogenesis. In Free
Radicals in Biology (Pryor, W. A., Ed.) Vol. III, pp 251-303,
Academic Press, Inc., New York.
(
33) Aust, S. D., Chignell, C. F., Bray, T. M., Kalyanaraman, B., and
Mason, R. P. (1993) Free radicals in toxicology. Toxicol. Appl.
Pharmacol. 120, 168-178.
(25) Desser, R. K., Himmelhoch, S. R., Evans, W. H., J anuska, M.,
Mage, M., and Shelton, E. (1972) Guinea pig heterophil and
eosinophil peroxidase. Arch. Biochem. Biophys. 148, 452-465.
(26) Borg, D. C., and Cotzias, G. C. (1962) Interaction of trace metals
with phenothiazine drug derivatives, I: structure reactivity
correlations. Proc. Natl. Acad. Sci. U.S.A. 48, 617-623.
(27) Piette, L. H., Bulow, G., and Yamazaki, I. (1964) Electron-
paramagnetic-resonance studies of the chlorpromazine free radi-
cal formed during enzymatic oxidation by peroxidase-hydrogen
peroxide. Biochim. Biophys. Acta 88, 120-129.
(
34) Kedderis, G. L., and Hollenberg, P. F. (1983) Characterization of
the N-demethylation reactions catalyzed by horseradish peroxi-
dase. J . Biol. Chem. 258, 8129-8138.
(
35) Sinha, B. K. (1983) Enzymatic activation of hydrazine deriva-
tives: a spin-trapping study. J . Biol. Chem. 258, 796-801.
36) Merck index: an encyclopedia of chemicals, drugs and biologicals
(
(1989) (Budavari, S., O’Neil, M. J ., Smith, A., and Heckelman, P.
E., Eds.) 11th ed., Merck & Co., Inc., Rathway, NJ .
37) Vasquez, A., Tudela, J ., Varon, R., and Garcia-Canovas, F. (1992)
A kinetic study of the generation and decomposition of some
phenothiazine free radicals formed during enzymatic oxidation
of phenothiazines by peroxidase-hydrogen peroxide. Biochem.
Pharmacol. 44, 889-894.
38) Uetrecht, J . P. (1990) Drug metabolism by leukocytes and its role
in drug-induced lupus and other idiosyncratic drug reactions.
Crit. Rev. Toxicol. 20, 213-235.
(
(
(
28) Pelizzetti, E., Meisel, D., Mulac, W. A., and Neta, P. (1979) On
the electron transfer from ascorbic acid to various phenothiazine
radicals. J . Am. Chem. Soc. 101, 6954-6959.
29) Uetrecht, J . P. (1992) The role of leukocyte-generated reactive
metabolites in the pathogenesis of idiosyncratic drug reactions.
Drug. Metab. Rev. 24, 299-366.
30) Ries, C. A., and Sahud, M. A. (1975) Agranulocystosis caused by
chinese herbal medicines: dangers of medications containing
aminopyrine and phenylbutazone. J . Am. Med. Assoc. 231, 352-
(
(
3
55.
TX950186T
+
+