|
|
|
|||||||
|
|||||||||
(Received for publication, August 1,
1995; and in revised form, September 28, 1995) From the
The iron chelator deferoxamine (Desferal; DSFL) reacts with
peroxidases and H
Deferoxamine (Desferal; DSFL) ( We recently demonstrated
that human neutrophils, monocytes, and eosinophils generate
HO
All EPR experiments were performed in 20 mM
phosphate buffer, pH 7.4, pretreated with Chelex 100 analytical grade
chelating resin (Bio-Rad) to remove adventitious iron(13) . In
addition, to hinder any remaining redox-active metals, 100 µM DTPA was added to all samples during EPR experiments.
Figure 1:
DSFL
DSFL
Figure 2:
Enhanced DSFL
Figure 3:
DSFL
Figure 4:
Chemical structures of 4-POBN, PBN, MNP,
PNO, 4-methyl-PNO, DMPO, and pyridine.
One of the possible explanations
for the partial activity of PNO was its lack of a side chain at the
4-position of the aromatic ring. We thus tested 4-picoline N-oxide (4-methyl-PNO) (Fig. 4) in the
peroxidase/H We were also interested to see if this enhancement
was seen with other nitrone spin traps. The spin trap PBN is
structurally similar to 4-POBN, differing only at the 4-position of the
aromatic ring (Fig. 4). This feature makes PBN far more
lipophilic than 4-POBN. At a concentration of 10 mM, the
highest concentration we were able to effectively utilize due to
solubility constraints, PBN demonstrated no enhancement of
DSFL Another commonly used nitrone
spin trap is DMPO. Aside from the shared nitrone group, DMPO is
structurally dissimilar to 4-POBN (Fig. 4). Addition of up to
100 mM DMPO to the peroxidase/H
Addition of 4-POBN to either the HRP or LPO
system failed to demonstrate any significant increase in peroxidase
activity (Table 1). DMPO, PBN, and 4-methyl-PNO, all of which
failed to demonstrate an increase in DSFL
We can thus conclude that the enhanced
DSFL Soriani et
al.(5) have proposed that peroxidase-mediated
DSFL
When 10 mM 4-POBN was added to either the
HRP/H
Figure 5:
EPR spectra representing the
DMPO/
Further evidence for the formation of a
4-POBN radical during the reaction of the peroxidases,
H
Figure 6:
EPR spectra representing the
4-POBN/
The detection
of the above spin adducts strengthens the possibility that the
enhancement of DSFL In summary, we have shown that under certain conditions, the nitrone
spin trap 4-POBN, the nitroso compound MNP, and the nitrogen oxide
compound PNO can enhance DSFL radical formation by several peroxidases.
It is likely, however, that these compounds work in different ways to
achieve the same end. The biological implications of these observations
remain unclear. Nevertheless, nitrones and other related compounds have
been investigated as pharmacological agents for the treatment of
various pathological
states(8, 9, 10, 11) . Our data
suggest the need for caution in the use of such agents in conditions in
which peroxidase-derived oxidants may be involved. Peroxidase-mediated
oxidation of compounds other than DSFL could potentially be enhanced in
the presence of compounds such as 4-POBN. 4-POBN has been observed to
have toxic effects. Normally perfused rat hearts demonstrated a 40%
loss of cardiac function following a 15-min exposure to 4-POBN. (
Volume 270,
Number 49,
Issue of December 8, 1995 pp. 29265-29269
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
-(4-Pyridyl-1-oxide)-N-tert-butylnitrone
Stimulates Peroxidase-mediated Oxidation of Deferoxamine
IMPLICATIONS FOR PHARMACOLOGICAL USE OF SPIN-TRAPPING AGENTS (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
O to form the DSFL radical
(DSFL), which can be detected by EPR spectroscopy. We
have found that DSFL
formation resulting from exposure
to H
O and any of a number of different
peroxidases is greatly enhanced in the presence of the nitrone spin
trap 
-(4-pyridyl-1-oxide)-N-tert-butylnitrone
(4-POBN). This enhancement was seen at 4-POBN concentrations as low as
200 µM. We observed a modest enhancement of
DSFL formation with 2-methyl-2-nitrosopropane.
However, no enhancement was seen with 5,5-dimethyl-1-pyrroline 1-oxide
(DMPO) or phenyl-tert-butylnitrone. A modest enhancement was
also seen with the nitrone compound pyridine N-oxide.
2-Methyl-2-nitrosopropane and pyridine N-oxide were
additionally capable of increasing enzymatic peroxidase activity as
measured by o-dianisidine and/or tetramethylbenzidine
oxidation. Furthermore, at high concentrations of 4-POBN (50
mM) in the absence of DSFL, we detected a
peroxidase/H
O-dependent 12-line EPR spectrum
that likely represents a 4-POBN/4-POBN
nitrogen-centered spin adduct. In the presence of both 4-POBN (10
mM) and DMPO (100 mM), an 18-line EPR spectrum was
observed consistent with formation of a DMPO/
4-POBN
nitrogen-centered spin adduct. Thus, the nitrone spin trap 4-POBN can
enhance the peroxidase-mediated formation of DSFL
,
possibly via the formation of a transient 4-POBN radical species. These
data suggest the importance of assessing the potential for nitrone spin
traps to both inhibit and enhance biological oxidation prior to their
use as potential pharmacological agents.
)is a potent iron
chelator used clinically to treat iron overload(1) . By binding
iron in such a way that it greatly hinders redox cycling, DSFL can
reduce the potential for damage to normal tissue from iron-catalyzed
HO formation(2) . Reaction of various
peroxidases (e.g. HRP, LPO, MPO, and eosinophil peroxidase)
with H
O in the presence of DSFL leads to the
formation of the DSFL radical (DSFL), a nitroxide
radical that can be detected by EPR spectroscopy(3) .
DSFL
is capable of damaging vitamins and
enzymes(4) , and its formation is thought to be dependent on
formation of Compound II of the peroxidase(5) , although the
exact reaction process remains ill defined.
via a peroxidase-dependent mechanism involving MPO
or eosinophil peroxidase(6, 7) . In the course of work
examining the ability of DSFL to modulate this HO
formation in its role as an iron chelator, we observed that
MPO-dependent formation of DSFL
increased dramatically
in the presence of the nitrone spin trap 4-POBN. Due to their
oxidant-scavenging properties, nitrone spin traps have been studied as
potential therapeutic modalities for a variety of pathological
processes believed to be mediated by reactive oxidant
species(8, 9, 10, 11) . Our
observations with 4-POBN and DSFL raised the possibility that some
nitrone spin traps could enhance rather than inhibit
peroxidase-mediated oxidant injury under some circumstances.
Accordingly, we examined 1) whether 4-POBN enhances the ability of
other peroxidases to generate DSFL
, 2) the extent to
which spin traps other than 4-POBN enhance peroxidase-mediated
DSFL
formation, and 3) the mechanism(s) involved.
Materials
HRP and bovine LPO were obtained from Sigma. Human MPO was
provided by Dr. William Nauseef (University of Iowa, Iowa City, IA).
Human eosinophils were isolated as described previously (7, 12) and lysed by repeated freeze-thaw cycles in
order to obtain crude lysates with eosinophil peroxidase activity.
DSFL, L-tryptophan, 3-hydroxybenzaldehyde, 4-POBN,
5,5-dimethyl-1-pyrroline 1-oxide (DMPO), 2-methyl-2-nitrosopropane
(MNP) dimer, and phenyl-tert-butylnitrone (PBN) were also
obtained from Sigma. Salicylic acid was obtained from Fisher. Pyridine N-oxide (PNO), 4-methyl-PNO, pyridine, and
diethylenetriaminepentaacetic acid (DTPA) were obtained from Aldrich.EPR Spectroscopy
For the generation of DSFL, 100
µM H
O was added to a solution
containing 1-100 µM DSFL and one of the following:
HRP (2-20 µg/ml), LPO (2-20 µg/ml), MPO
(40-400 units/ml), or lysed eosinophils. After 10 min, samples
were transferred to an EPR flat cell, and the resulting EPR spectra
were obtained at room temperature using a Bruker Model ESP 300
spectrometer. Unless otherwise stated, instrument settings were as
follows: microwave power, 20 milliwatts; modulation frequency, 100 kHz;
modulation amplitude, 0.9 G; sweep rate, 0.4 G/min; and response time,
0.6 s.Peroxidase Assays
Peroxidase activity was determined by two methods, each of
which is based on its ability to oxidize a specific substrate.Tetramethylbenzidine
Peroxidase activity was initially
determined by the method of Bozeman et al.(14) . HRP
(5-20 ng/ml) or LPO (0.4-2 µg/ml) was diluted in 50
mM acetate buffer, pH 5.4, in the presence of the various
compounds. Tetramethylbenzidine (Sigma) and HO were added to yield final concentrations of 1.4 and 0.3
mM, respectively (final volume of 0.7 ml). After 3 min, the
reaction was terminated by addition of 0.02 ml of bovine liver catalase
(0.3 mg/ml stock solution; Sigma) and 0.68 ml of cold acetic acid (0.2 M stock solution). The absorbance of the resulting solution
was then determined at 635 nm. Effects of spin traps on peroxidase
activity are expressed as -fold increase over control (peroxidase only)
values.o-Dianisidine
Peroxidase activity was also
determined as the ability to oxidize o-dianisidine as
described previously (15) . In this assay, HRP (5-20
ng/ml) or LPO (0.4-2 µg/ml) was added to a 10 mM sodium phosphate buffer solution, pH 6.0, containing 0.88
mM HO, 260 µMo-dianisidine (Sigma), and the various compounds. Change
in absorbance at 460 nm was determined for the first 30 s of the
reaction. Effects of spin traps on peroxidase activity are expressed as
-fold increase over control (peroxidase only) values.
Formation of the DSFL Radical by the
Peroxidase/H
DSFL, an iron
chelator used clinically in cases of iron overload, is susceptible to
enzymatic oxidation by various peroxidases(3, 5) .
Consistent with the results of others(3) , addition of DSFL to
a HRP/HO SystemO system yielded the characteristic
DSFL as detected by EPR (Fig. 1). The
steady-state concentration of DSFL
increased as the
level of DSFL added to the system increased, but it was not linear over
the 100-fold range of DSFL concentrations used. Under identical
experimental conditions, we observed DSFL
formation by
MPO, LPO, and lysed eosinophils (i.e. eosinophil peroxidase)
in the presence of H
O (data not shown). In each
case, both HO and peroxidase were required for
the formation of DSFL (data not shown). Formation of
this radical by the HRP/H
O system was clearly
observed at DSFL concentrations as low as 10 µM. When
administered to humans, stable plasma concentrations of DSFL as high as
20 µM have been reported(16) . Thus, at in
vivo sites of inflammation in such treated individuals,
neutrophil-derived MPO and eosinophil-derived eosinophil peroxidase
could lead to the formation of DSFL.
formation by a
peroxidase/H
O system. A, EPR spectrum
seen following addition of 1 µM DSFL to a system
containing 100 µM HO and 2 mg/ml
HRP; B, same as A, but in the presence of 10
µM DSFL; C, same as A, but in the
presence of 100 µM DSFL. Results are representative of
three separate experiments. Note that for the first spectrum shown, the
instrument gain was increased 4-fold. EPR parameters for
DSFL are a
= 7.84 G and a
(2) = 6.23 G.
seen in the above experiments is likely a
direct product of the action of the peroxidase/H
O system on DSFL(3) . It has been reported that
DSFL can be formed directly by a scavenging reaction
between DSFL and HO
(3) . This suggested that
the DSFL
we observed could be a result of DSFL
reacting with HO
generated by the
peroxidase/H
O system, forming DSFL as a consequence. However, using the highly sensitive
4-POBN/ethanol spin-trapping
system(6, 17, 18) , we were unable to detect
significant production of HO
by the combination of
H
O and HRP, LPO, or MPO (data not shown). These
data suggest that peroxidase enzymes form DSFL predominantly via their oxidizing intermediates, i.e. Compounds I and/or II, and not via formation of a HO
intermediate.
Enhanced DSFL Radical Formation in the Presence of
4-POBN
In the course of the above experiments, we surprisingly
observed a 2-5-fold enhancement of DSFL formation in the presence of the nitrone spin trap 4-POBN at the
concentration (10 mM) typically employed in our spin-trapping
studies (Fig. 2). Results were qualitatively similar regardless
of the peroxidase employed, implying a common mechanism of action in
the enhancement of DSFL
formation. However, the
magnitude of increase varied with the peroxidase employed. In the
presence of 20 µg/ml HRP, this enhancement was typically 4-fold,
whereas with 20 µg/ml LPO, this enhancement was as high as 20-fold.
Enhanced DSFL
formation could be observed at 4-POBN
concentrations as low as 200 µM and increased linearly
over the range of 200 µM to 10 mM 4-POBN.
Furthermore, by reducing the levels of peroxidase 10-fold, the effects
of 4-POBN became more dramatic. Thus, in the presence of 2 µg/ml
LPO and 200 µM 4-POBN, DSFL
was easily
detected (Fig. 3B), whereas in the absence of 4-POBN (Fig. 3A), it was not seen.
formation
in the presence of the nitrone spin trap 4-POBN. A, EPR
spectrum obtained with a horseradish peroxidase/H
O system (20 µg/ml HRP, 100 µM
HO) containing 10 mM 4-POBN; B, same as A, but containing 10 µM DSFL
(no 4-POBN); C, same as A, but containing both 10
mM 4-POBN and 10 µM DSFL. Results are
representative of three separate experiments. Note that for the first
spectrum shown, the instrument gain was increased
4-fold.
formation in the
presence of various spin traps. A, EPR spectrum obtained with
a lactoperoxidase/H
O system (2 µg/ml LPO,
100 µM HO) containing 10
µM DSFL; B, same as A, but in the
presence of 200 µM 4-POBN; C, same as A,
but in the presence of 100 mM DMPO; D, same as A, but in the presence of 10 mM PBN; E, same
as A, but in the presence of 10 mM MNP; F,
same as A, but in the presence of 10 mM PNO. Results
are representative of three separate experiments. Note the
concentration of peroxidase used is 10-fold less than in Fig. 1and Fig. 2.
Effect of Other Compounds on DSFL Radical
Formation
In an effort to better understand the mechanism(s)
responsible for the enhancement of DSFL formation, we
tested the ability of various compounds representative of the
structural moieties of 4-POBN to enhance the formation of
DSFL
by the various peroxidase/H
O systems. Initially, we examined the two main structural
components of 4-POBN. The nitroso spin trap MNP is contained in the
structure of 4-POBN, but has no aromatic ring structure (Fig. 4). In contrast, the compound PNO, while not a spin trap,
has a pyridine ring structure that is identical to that found in 4-POBN (Fig. 4), i.e. it includes a nitrogen oxide group, but
lacks the MNP-like side chain. Surprisingly, at a concentration of 10
mM, both MNP and PNO demonstrated a significant enhancement of
DSFL formation, but to a lesser extent than 4-POBN (Fig. 3, compare A with E and A with F). Qualitatively similar increases in DSFL
formation in the presence of 4-POBN and MNP, but not PNO, have
been observed with MPO and isolated eosinophils (data not shown). This
raised the possibility that the nitrone moiety and the aromatic ring
structure are at least partially responsible for the enhancement of
DSFL
formation.
O system. Surprisingly, no
enhancement of DSFL formation was seen in the presence
of this compound (data not shown). Another related compound of interest
was pyridine, representing the aromatic ring of PNO without the oxide
group (Fig. 4). Addition of 10 mM pyridine to the
peroxidase/H
O system also demonstrated no
increase in DSFL formation (data not shown). Thus, two
compounds, MNP and PNO, which share structural features with 4-POBN,
were able to enhance DSFL
formation, implying a role
for the nitrone moiety as well as the nitrogen oxide moiety in the
aromatic ring.
formation (Fig. 3, A and D). This implies that the spin-trapping nitrone moiety is not
involved, but rather it is the nitrogen oxide group in the aromatic
ring (Fig. 4) that is important for the enhanced radical
formation in the presence of 4-POBN.
O systems did not lead to enhancement of peroxidase-mediated
DSFL formation under our experimental conditions (Fig. 3, A and C). The lack of effect with
DMPO suggests that the enhancement of peroxidase-mediated
DSFL
formation is not a general property of nitrone
spin traps, but that features unique to the 4-POBN molecule are
responsible for the effect.
Effect of Spin Traps and Related Compounds on Peroxidase
Activity
We next wanted to determine whether the enhancement of
DSFL formation was the result of a direct interaction
between the nitrone compound and DSFL or simply the result of a
generalized increase in peroxidase activity. We chose to measure
enzymatic peroxidase activity using two distinct assay systems, the
oxidation of tetramethylbenzidine and the oxidation of o-dianisidine.
formation,
also failed to demonstrate increased peroxidase activity (Table 1). MNP and PNO, however, did demonstrate an enhancement
of peroxidase activity (Table 1). The increases in enzymatic
peroxidase activity seen with MNP and PNO were of the same magnitude as
those seen in the spin-trapping experiments ( Table 1and Fig. 3). Pyridine also demonstrated a large increase in
enzymatic HRP activity, which was not seen with LPO (Table 1).
The large increase in enzymatic HRP activity observed in the presence
of pyridine was surprising given the lack of enhancement of
DSFL
formation seen with this compound in the
spin-trapping studies.
formation seen with the
peroxidase/H
O system in the presence of 4-POBN
is not attributable to its ability to increase peroxidase activity. It
is possible that the ability of both MNP and PNO to enhance
DSFL formation is explainable on the basis of an
increase in enzymatic activity, but this is in contrast to the pyridine
data showing an increase in enzymatic activity with no concomitant
increase in DSFL
formation. These data suggest that it
may be the ring structure, which 4-POBN and PNO share, that is of
critical importance to the ability of 4-POBN to enhance
peroxidase-dependent DSFL
formation.
formation requires the formation of Compound II
of the peroxidase. Inclusion of the Compound II-promoting agents L-tryptophan, 3-hydroxybenzaldehyde, or salicylate (all at 10
mM) in the DSFL/H
O/peroxidase system
yielded no observable increase in DSFL formation under
our experimental conditions (data not shown). Thus, the enhancement of
DSFL
formation seen with 4-POBN is likely not due to
an increase in peroxidase Compound II formation.
Spin Trapping a Radical from 4-POBN in the
Peroxidase/H
Rather than
directly increasing the enzymatic activity of the peroxidase, it is
possible that 4-POBN is an oxidizable substrate for the peroxidase,
forming a radical species that in turn abstracts an electron (hydrogen
atom) from DSFL to form DSFLO System. This would have the net
effect of increasing the steady-state level of DSFL
.
If this transient 4-POBN species were a free radical
(4-POBN
), we hypothesized that it might be possible to
detect its presence directly using EPR. However, previous experiments
described above showed that the reaction of either HRP or LPO with
H
O (without DSFL) in the presence of 10 mM 4-POBN yields no detectable spin-active species. Since this could
be related to a very short half-life of such a 4-POBN radical, we
explored the possibility that spin trapping could be used to detect
such a species.O or LPO/HO system
in the presence of 100 mM DMPO, the EPR spectrum shown in Fig. 5A was generated. No spectrum above background was
observed in the absence of 4-POBN. We propose that this spectrum
represents a DMPO/4-POBN spin adduct (a
= 15.9 G, a
= 13.85 G, and a
= 4.64 G). A
computer-simulated spectrum with these parameters is shown in Fig. 5B.
4-POBN spin adduct. A, EPR spectrum seen
following addition of 10 mM 4-POBN and 100 mM DMPO to
a system containing 100 µM H
O and
20 µg/ml HRP; B, computer simulation of a
DMPO/4-POBN spin adduct, where a
= 15.9 G, a =
13.85 G, and a = 4.635 G.
Results are representative of three separate
experiments.
O, and 4-POBN was obtained by increasing the
concentration of 4-POBN to 50 mM. Under such conditions, we
were able to routinely observe a 12-line EPR spectrum consistent with
the formation of a 4-POBN/4-POBN spin adduct, where a
= 1.8 G, a = 14.9 G, and a =
1.85 G (Fig. 6A). A computer-simulated spectrum with
these parameters is shown in Fig. 6B.
4-POBN spin adduct. A, EPR spectrum
seen following addition of 50 mM 4-POBN to a system containing
100 µM H
O and 20 µg/ml HRP; B, computer simulation of a 4-POBN/4-POBN
spin adduct, where a
= 1.8 G, a = 14.9 G, and a = 1.85 G. Results are
representative of three separate
experiments.
formation by 4-POBN is due to the
generation of an oxidizing 4-POBN radical species. This radical in turn
enhances the effectiveness of peroxidase-mediated oxidation of DSFL. We
assume that the 4-POBN radical species being trapped is derived from
the pyridyl-associated nitrone moiety, but cannot rule out completely
an association with the tert-butyl-associated nitrone moiety.
)In addition, rat hearts reperfused following 30 min of
ischemia, which typically recover 40-55% of their preischemic
function, demonstrated no recovery if they received prior exposure to
4-POBN. These observations may relate to 4-POBN's
susceptibility to oxidation by peroxidase-like processes. In light of
evidence demonstrating a role for peroxidase-mediated oxidation
products in human pathology and given the increasing use of nitrone
spin traps as pharmacological agents, it is important to evaluate the
potential for these nitrone spin traps to both abrogate and enhance
free radical-mediated oxidation.
), deferoxamine radical; HRP, horseradish
peroxidase; LPO, lactoperoxidase (bovine); MPO, myeloperoxidase
(human); 4-POBN,
-(4-pyridyl-1-oxide)-N-tert-butylnitrone; DMPO,
5,5-dimethyl-1-pyrroline 1-oxide; MNP, 2-methyl-2-nitrosopropane; PBN,
phenyl-tert-butylnitrone; PNO, pyridine N-oxide;
DTPA, diethylenetriaminepentaacetic acid.)
We thank Dr. William Nauseef for kindly providing
human MPO.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  B. A. Wagner, G. R. Buettner, L. W. Oberley, C. J. Darby, and C. P. Burns Myeloperoxidase Is Involved in H2O2-induced Apoptosis of HL-60 Human Leukemia Cells J. Biol. Chem., July 14, 2000; 275(29): 22461 - 22469. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |