Myeloperoxidase-mediated oxidation of edaravone produces an apparent non-toxic free radical metabolite and modulates hydrogen peroxide-mediated cytotoxicity in HL-60?cells

Article abstract of DOI:10.1016/j.freeradbiomed.2019.08.021

Edaravone is considered to be a potent antioxidant drug known to scavenge free radical species and prevent free radical-induced lipid peroxidation. In this study, we investigated the effect of edaravone on the myeloperoxidase (MPO) activity, an enzyme res

Full text of DOI:10.1016/j.freeradbiomed.2019.08.021

FreeRadicalBiologyandMedicine143(2019)422–432
Contents lists available at ScienceDirect
Free Radical Biology and Medicine
journal homepage: www.elsevier.com/locate/freeradbiomed
Original article
Myeloperoxidase-mediated oxidation of edaravone produces an apparent
non-toxic free radical metabolite and modulates hydrogen peroxide-
mediated cytotoxicity in HL-60 cells
Lindsey Y.K. Suha, Dinesh Babua, Lusine Tonoyana, Béla Reizb, Randy Whittalb,
S. Amirhossein Tabatabaei-Dakhilia, Andrew G. Morgana, Carlos A. Velázquez-Martíneza,
Arno G. Sirakia,∗
a Faculty of Pharmacy & Pharmaceutical Sciences, Katz Group-Rexall Centre for Pharmacy and Health Research, University of Alberta, Edmonton, Alberta, T6G 2E1,
Canada
b Department of Chemistry, 11227 Saskatchewan Drive, University of Alberta, Edmonton, Alberta, T6G 2G2, Canada
A R T I C L E I N F O
A B S T R A C T
Keywords:
Edaravone
Myeloperoxidase
Free radicals
Electron spin resonance
HL-60 cells
Edaravone is considered to be a potent antioxidant drug known to scavenge free radical species and prevent free
radical-induced lipid peroxidation. In this study, we investigated the effect of edaravone on the myeloperoxidase
(MPO) activity, an enzyme responsible for the production of an array of neutrophil-derived oxidants that can
cause cellular damage. The addition of edaravone to the reaction of MPO and hydrogen peroxide (H2O2) sig-
nificantly enhanced the reduction of MPO Compound II back to native MPO. Interestingly, the MPO-mediated
production of toxic hypochlorous acid exhibited a concentration-dependent biphasic effect, with the apparent
optimal edaravone concentration at 10 μM. Oxidation of edaravone by MPO was examined by various analytical
methods. An MPO-catalyzed product(s) of edaravone was identified at 350 nm by kinetic analysis of UV–Vis
spectroscopy. Several MPO-catalyzed metabolites of edaravone were proposed from the LC-MS analyses, in-
cluding oxidized dimers from edaravone radicals. Electron spin resonance (ESR) spin trapping detected a carbon-
centred radical metabolite of edaravone. NMR studies revealed that there are two exchangeable hydrogens, one
of which is on the α-carbon, justifying the carbon-centred edaravone radical produced from MPO. Despite the
formation of an edaravone carbon-radical metabolite, it did not appear to effectively oxidize GSH (in comparison
with phenoxyl radicals). Viability (ATP) and cytotoxicity (LDH release) assays showed a concentration-depen-
dent effect of edaravone on HL-60 cells treated with either a bolus concentration of 30 μM H2O2 or a flux of H2O2
generated by 5 mM glucose and 10 mU/mL glucose oxidase. The H2O2-induced toxicity was ameliorated at high
edaravone concentrations (100–200 μM). In contrast, low concentrations of edaravone (1–10 μM) exacerbated
the H2O2-induced toxicity. However, the effect of edaravone at low concentration (0–10 μM) appeared more
prominent with the LDH assay only. The cellular findings correlated with the biochemical studies with respect to
hypochlorous acid formation. These findings provide interesting perspectives regarding the duality of edaravone
as an antioxidant drug.
Antioxidant
1. Introduction
It is believed that oxidative stress is involved in the pathobiology,
which occurs after an ischemic stroke [3]. Oxidative stress is likely
Edaravone (3-methyl-1-phenyl-2-pyrazolin-5-one), marketed under
the trade name Radicava® or Radicut®, is a small molecular weight drug
with antioxidant properties. In 2001, edaravone was approved in Japan
for the early treatment of acute cerebral infarction. Later, it was ap-
proved to slow the progression of amyotrophic lateral sclerosis (ALS) in
Japan and the United States in 2015 and 2017, respectively [1]. It was
just recently approved by Health Canada, in October 2018 [2].
manifested by multiple processes, which produces reactive oxygen
species (ROS) including hydroxyl radical (HO•), hydrogen peroxide
(H2O2), and superoxide anion radical (O2•−) [4]. These species partake
in vascular endothelial cell damage and can lead to membrane lipid
peroxidation [4,5]. Moreover, ROS are considered to contribute to the
pathogenesis of ALS, a disease characterized by progressive degenera-
tion of upper and lower motor neurons [6]. Free radical-mediated
∗ Corresponding author.
E-mail addresses: velazque@ualberta.ca (C.A. Velázquez-Martínez), siraki@ualberta.ca (A.G. Siraki).
https://doi.org/10.1016/j.freeradbiomed.2019.08.021
Received 3 March 2019; Received in revised form 13 August 2019; Accepted 20 August 2019
Availableonline21August2019
0891-5849/©2019ElsevierInc.Allrightsreserved.

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
oxidative stress affects presynaptic transmission, promotes inflamma-
tion, and causes lipid peroxidation [6]. Therefore, targeted antioxidant
treatments for the restoration of redox balance is an essential con-
sideration in the therapeutic approach.
2. Materials and methods
2.1. Reagents
The free radical scavenging (antioxidant) activity of edaravone is
believed to be its mechanism of action, i.e. neuroprotective activity [7].
As edaravone is amphiphilic, it readily crosses biological membranes
and distributes to both plasma and tissue, where it may scavenge a wide
range of water-soluble and lipid-soluble radicals [7]. Several studies
reported that edaravone has potent radical scavenging activity against
2,2-diphenyl-1-picrylhydrazyl (DPPH), HO•, tert-butyl peroxyl radical,
ascorbyl free radical, methyl radical and nitric oxide, although its re-
Human purified neutrophil MPO was purchased from Athens
Research & Technology (Athens, GA). 3-methyl-1-phenyl-2-pyrazolin-5-
one (edaravone) and 2-chloro-5,5-dimethyl-1,3-cyclohexanedione
(MCD) were purchased from Alfa Aesar (Tewksbury, MA).
Diethylenetriamine-pentaacetic acid (DTPA) was purchased from Fluka
Analytical (Buchs, Switzerland). Glucose oxidase (GOx), D-(+)-glucose,
hydrogen peroxide, L-glutathione reduced, protease inhibitor cocktail,
Ponceau S and guaiacol were purchased from Sigma (St. Louis, MO).
5,5-dimethyl-1-pyrroline-N-oxide (DMPO) and Cytotoxicity LDH Assay
Kit-WST were purchased from Dojindo Molecular Technologies, Inc.
(via Cedarlane Labs, Burlington, ON). Sodium chloride was from VWR
(West Chester, PA). CellTiter-Glo® Luminescent Cell Viability Assay kit
was purchased from Promega (Madison, WI). HL-60 human promye-
locytic leukemia cells (#CCL-240) were obtained from ATCC (via
Cedarlane Labs, Burlington, ON). Micro BCATM Protein Assay Kit was
purchased from Thermo Scientific (Rockford, IL). 4X Laemmli Sample
Buffer and Transfer-blot® semi-dry transfer cell were purchased from
Bio-Rad (Hercules, CA). BLUelf Prestained Protein Ladder was from
FroggaBio (North York, ON, Canada). Anti-human MPO (rabbit pAb)
and Immobilon Western Chemiluminescent HRP Substrate were pur-
chased from EMD Millipore (Billerica, MA). Anti-rabbit IgG, HRP-linked
antibody was purchased from Cell Signaling Technology (Whitby, ON,
Canada). A stock solution of edaravone (500 mM) was prepared in
DMSO and stored at −20 °C up to one month.
•−
activity with O2 is controversial [8,9]. Specifically, its therapeutic
activity has been attributed to suppressing lipid peroxidation and per-
oxynitrite-catalyzed cellular damage [7,10]. Under pathological con-
ditions such as (neuro)inflammation, production of peroxynitrite is
dramatically increased from the rapid reaction between nitric oxide and
•−
O2 [11]. Fujisawa and Yamamoto reported edaravone to be 30 times
more reactive with peroxynitrite compared to uric acid [12].
Although reducing oxidative stress is the postulated mechanism of
action of edaravone, the complete picture of its cellular redox regula-
tion is not entirely understood. Myeloperoxidase (MPO) is a heme-
containing enzyme abundantly expressed in neutrophils and found to a
lesser extent in monocytes [13,14]. Furthermore, some evidence sug-
gests MPO expression in Kupffer cells in the liver, microglia and neu-
ronal cells [15–17]. In response to injury or infection, these cells be-
come activated to release MPO, which can produce highly reactive
hypochlorous acid (HOCl) and other oxidants that contribute to mi-
crobial killing [18]. Such MPO-derived oxidants can also target pro-
teins, DNA, and lipids to carry out modifications like halogenation,
nitration and oxidative cross-linking [19]. Hence, persistent MPO ac-
tivation and excessive ROS production under pathological conditions
could lead to tissue damage and contribute to the pathogenesis of
various diseases [19]. A growing body of literature suggests that reg-
ulation of the MPO activity is particularly critical when it comes to
diseases involving acute or chronic inflammation [20]. Recent clinical
and experimental evidence supports the association of MPO in a ple-
thora of pathological conditions such as atherosclerosis, myocardial
infarction, cancer, neurodegenerative diseases, glomerulonephritis,
2.2. UV–Vis spectroscopic analysis of the MPO active site
The effect of edaravone on the catalytic activity of MPO was ex-
amined by measuring the kinetic changes in the UV–Vis absorption
spectra of the heme active site of the MPO. Reactions of 500 nM MPO
with or without 100 μM edaravone was prepared in 0.1 M sodium
phosphate buffer (pH 7.4) containing 100 μM DTPA (PB). After in-
itiating the reaction with 100 μM H2O2, the reaction was allowed to
proceed for 1 h. The region of the UV–Vis spectrum spanning the ab-
sorption maxima (350–550 nm) of the resting MPO heme
(λmax = 429 nm) and for MPO Compound II (λmax = 456 nm) were
measured at t = 0, 2, 5, 10, 20 and 60 min using Thermo Scientific
NanoDrop 2000c dual-mode UV–vis spectrophotometer (Wilmington,
DE). The absorption spectrum of 500 nM resting MPO in the buffer was
recorded as a reference.
ischemia/reperfusion injury and cystic fibrosis, to name
a few
[16,17,21–30]. Hence modulation of MPO activity to ameliorate the
adverse effects has been targeted as a viable therapeutic approach [20].
Directing the MPO redox pathway is essential in modulating the
MPO-mediated HOCl production. MPO becomes activated when it re-
acts with H2O2 to form an active redox intermediate (Compound I).
Once produced, Compound I can either enter the chlorination cycle, in
which toxic HOCl is produced from chloride ion (Cl−), or enter the
peroxidase cycle, oxidizing other substrates (electron donors) [19].
Despite the critical contribution of MPO in neuropathology and other
diseases, currently, there is no report on the effect of edaravone on MPO
activity and MPO-mediated oxidative stress. According to the hy-
pothetical scheme described by Yamamoto et al., edaravone scavenges
free radicals by donating one electron to the radical species [7]. As
edaravone apparently acts as an efficient electron donor, we postulated
that edaravone would serve as a substrate of MPO. In this report, we
provide evidence that edaravone acts as an MPO substrate. In addition,
we characterized the edaravone free radical as well as edaravone me-
tabolites produced as a result of the MPO-mediated oxidation, followed
by evaluation of the cytotoxic potential of the edaravone free radical.
Lastly, we describe intriguing findings regarding the concentration-
dependent effects of edaravone in both exacerbating and suppressing
H2O2-mediated HL-60 cell toxicity, which we propose to occur via
modulation of MPO-mediated HOCl generation.
2.3. UV–Vis spectroscopic analysis of MPO-mediated MCD chlorination
The rate of MPO-catalyzed MCD chlorination into dichlorodimedon
(DCD) is an indirect measure of the rate of HOCl production by MPO
from H2O2 and Cl−. The effect of edaravone on the rate of MPO cata-
lyzed halogenation was investigated using MCD as a chlorination target
of HOCl. Reactions were prepared by mixing 50 nM MPO, 200 mM
NaCl, 40 μM MCD and different concentrations of edaravone (0, 0.3, 3,
10, 30° and 100 μM), and then initiating the reaction with
300 μM H2O2. Kinetic changes in the maximum absorption wavelength
(λmax = 291 nm) of MCD were recorded every 20 s for 100 s. All re-
actions were carried out in PB. Data were expressed as a change in
absorbance at 291 nm from t = 0 (or ΔAbs at 291 nm).
2.4. UV–Vis spectroscopic analysis of MPO-catalyzed oxidation of
edaravone
Formation of the MPO-catalyzed product of edaravone was detected
by measuring kinetic changes in the UV–Vis spectra. 500 nM of MPO
and 100 μM of edaravone was mixed, and the reaction was triggered by
100 μM H2O2, and the UV–Vis spectrum was measured at t = 0, 2, 5, 10,
423

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
20 and 60 min to identify the wavelength(s) subject to change. All re-
actions were carried out in PB. UV–Vis spectra were acquired as de-
scribed above.
d6; Cambridge Isotopes Laboratories, Tewksbury, MA) for 1H-NMR
spectra acquisition using a Bruker FT-600 MHz NMR spectrometer. 1H-
NMR and 13C-NMR spectra were determined using tetramethylsilane as
a reference. Chemical shifts (δ) and coupling constants (J) are expressed
in parts per million and Hertz, respectively.
2.5. Liquid-chromatography mass-spectroscopy (LC-MS) analysis of
edaravone metabolites
2.8. Metabolic activity (ATP) assay of the HL-60 cells
Edaravone (500 μM) was mixed with 500 nM MPO and
100 μM H2O2 in a 200 μL reaction mixture using PB. The reaction was
carried out for 1 h on a shaker and then stopped by addition of 200 μL
ice-cold acetonitrile. Reaction components were separated by adding
ice-cold acetonitrile to the mixture, followed by vigorous mixing on a
vortex for 5 min and then centrifugation at 3000 rpm for 5 min at 4 °C.
The organic phase was collected and analyzed by high-resolution LC-
MS. RP-HPLC-UV-MS was performed using an Agilent 1200 SL HPLC
System with a Phenomenex Kinetex 1.7 μm, 100 Å, 50 × 2.1 mm, C8
reverse-phase analytical column with guard (Phenomenex, Torrance,
CA, USA), thermostated at 40 °C followed by UV and mass spectro-
metric detection. An aliquot was loaded onto the column and flushed
for 1 min to remove salts at a flow rate of 0.50 mL/min and an initial
buffer composition of 99% of 0.1% formic acid as mobile phase A and
1% of 0.1% acetonitrile as mobile phase B. Elution of the analytes was
achieved by using a linear gradient from 1% to 99% mobile phase B
over a period of 9 min, held at 99% mobile phase B for 1 min to remove
all analytes from the column and 99%–1% mobile phase B over a period
of 1 min. Mass spectra were acquired in positive mode ionization using
an Agilent 6220 Accurate-Mass TOF HPLC/MS system (Santa Clara, CA,
USA) equipped with a dual sprayer electrospray ionization source with
the second sprayer providing a reference mass solution. Mass spectro-
metric conditions were drying gas 9 L/min at 325 °C, nebulizer 25 psi,
mass range 100–1000 Da, acquisition rate of ~1.03 spectra/sec, frag-
mentor 150 V, skimmer 65 V, capillary 3800 V, instrument state 4 GHz
High Resolution. Mass correction was performed for every individual
spectrum using peaks at m/z 121.05087 and 922.00979 from the re-
ference solution. Data acquisition was performed using the Mass Hunter
software package (ver. B.04.00). Analysis of the HPLC-UV-MS data was
done using the Agilent Mass Hunter Qualitative Analysis software (ver.
B.07.00).
HL-60 human promyelocytic cells were seeded in a 96-well micro-
plate at 1 × 105 cells/mL in 100 μL phosphate-buffered saline (PBS).
Cells were treated with increasing concentration of edaravone (0, 1, 5,
10, 25, 50, 100 and 200 μM), then co-stimulated with either a bolus
concentration of 30 μM H2O2 or 5 mM glucose and 10 mU/mL GOx to
induce toxicity. Cells receiving no treatment and 0.04% DMSO treat-
ment were included as a negative control (NC) and vehicle control,
respectively. Following 3 h incubation, CellTiter-Glo® Luminescent Cell
Viability Assay was performed following the manufacturer's instruction.
Luminescence values were presented as a percentage of the vehicle
control and expressed as mean
SD. All treatments were carried out
in triplicates, and the assay was repeated three independent times.
Statistical analysis was carried out by one-way ANOVA with Bonferroni
multiple comparisons (n = 3) using GraphPad Prism 6 software.
2.9. Cytotoxicity (LDH) assay of the HL-60 cells
HL-60 cells were treated with edaravone and H2O2 as described
above. The LDH activity assay was carried out according to the man-
ufacturer's instruction. Briefly, 10 μL of lysis buffer was added to the
untreated cells for the last 30 min of incubation to establish 100% cy-
totoxicity (positive control, PC). Similarly, cells in other wells were
treated with PBS (negative control, NC), and vehicle control (DMSO).
The dye mixture in assay buffer (100 μL) was added to each well and
incubated in the dark for 30 min. Stop solution (50 μL) was added, and
the absorbance was recorded at 490 nm. The percent cytotoxicity was
calculated using the following formula:
[Cytotoxicity (%) = (ATest substance-APC) / (APC-ANC)]
The percent cytotoxicity was expressed as mean
SD. Statistical
analysis was carried out by one-way ANOVA with Bonferroni multiple
comparisons (n = 3) using GraphPad Prism 6 software.
Data were normalized as the percentage of negative control and
positive control for CellTiter-Glo and LDH assay, respectively.
2.6. Electron spin resonance (ESR) spectroscopy spin-trapping of edaravone
and glutathionyl radical
Free radical metabolites of edaravone were detected by ESR spin
trapping, where short-lived radical species form covalent adducts with
the nitrone spin trap agent DMPO to produce a relatively stable para-
magnetic species. The reaction mixture consisted of 6.4 mM edaravone,
100 mM DMPO, and 0.5 μM MPO, all prepared in PB. The reaction was
initiated by a sustained generation of H2O2, produced from the reaction
between 5 mM glucose and 25 mU/mL GOx. After briefly vortexing, the
reaction mixture was transferred into three separate 50 μL micro-ca-
pillary tubes, inserted simultaneously into a 3 mm ESR tube which was
then placed in the resonator. ESR spectra were obtained with a Bruker
Elexsys E−500 spectrometer (Billerica, MA) with the following para-
meters: Center field = 3505 G, sweep width = 100 G, Field
modulation = 1 G, Microwave frequency = 9.8 GHz, Microwave
power = 20 mW, and sweep time = 120 s. Spectra were recorded after
two scans. Oxidation of glutathione by edaravone radical into glu-
tathionyl radical was also assessed by ESR spin trapping with DMPO.
Reaction mixtures consisted of 100 mM DMPO, 1 mM GSH, and 50 nM
MPO in PB were mixed with either 10 μM phenol, 10 μM edaravone or
DMSO (0.02%). After triggering with 100 μM H2O2, ESR spectroscopic
analyses were carried out as described above.
2.10. Western blot analysis of MPO amount in HL-60 cells
HL-60 total cell protein lysates were prepared using RIPA buffer
(1% Triton X-100, 0.5% Sodium deoxycholate, and 0.1% SDS in PBS pH
7.4) containing protease inhibitor cocktail (10 μL per 107 cells), fol-
lowed by three freeze-thaw cycles. Mixtures were centrifuged at
15,000 rpm for 15 min at 4 °C, and the supernatant was collected.
Human purified MPO (0.3, 1, 3 and 9 μg) and the HL-60 total cell
protein lysate (containing 0.1, 0.3 and 0.9 × 106 cells) were recon-
stituted in 4X Laemmli buffer and subjected onto the same gel for
electrophoretic separation, followed by semi-dry transfer onto ni-
trocellulose membranes. Membranes were blocked with blocking buffer
consisting of 5% non-fat skim milk in Tris-buffered saline with 0.1%
Tween-20 (TBST) for 1 h at room temperature and incubated overnight
at 4 °C with primary anti-MPO Rabbit pAb (1:4000) in blocking buffer.
The secondary goat anti-rabbit IgG, HRP-linked antibody (1:5000) was
incubated for 1 h at room temperature. Membranes were washed three
times with TBST after each step of immunoblotting. Immunoblots were
visualized with Immobilon Western Chemiluminescent HRP Substrate
using ImageQuant LAS 4000mini Luminescent image analyzer (GE
Healthcare, Pittsburgh, PA). Band intensities were quantified using
2.7. NMR studies for exchangeable proton analysis in edaravone
Edaravone (30 mg) was dissolved in 2 mL deuterated DMSO (DMSO-
ImageJ software (NIH, United States) and expressed as mean
SD
424

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
Fig. 1. Effect of edaravone on reducing MPO to its resting state. The reaction of 500 nM MPO in the absence (A) or presence (B) of 100 μM edaravone was
prepared in 0.1 M sodium phosphate buffer (pH 7.4) containing 100 μM DTPA (PB). After triggering the reaction with 100 μM H2O2, UV–Vis spectra were acquired at
t = 0, 2, 5, 10, 20 and 60 min. Absorption peaks in the Soret region show native MPO (λmax = 429 nm) and MPO Compound II (λmax = 456 nm).
(n = 3).
MCD (Fig. 2A). Interestingly, 30 μM of edaravone resulted in dimin-
ished MCD chlorination, which was further pronounced at 100 μM
edaravone (Fig. 2B). Therefore, our results indicate that under the given
conditions, 0.3–10 μM edaravone increased the rate of MCD chlorina-
tion (HOCl production), whereas 30 and 100 μM attenuated the rate of
HOCl production.
3. Results
3.1. The effect of edaravone on the rate of MPO regeneration
Kinetic UV–Vis spectroscopy studies in
500 nM MPO and 100 μM H2O2 revealed the rapid formation of
Compound II at max = 456 nm. Regeneration of native MPO
a reaction containing
3.3. MPO peroxidase activity catalyzes UV–Vis spectral changes of
edaravone
λ
(λmax = 429 nm) from MPO Compound II occurred slowly over 60 min,
which is typical in the absence of a donor substrate (Fig. 1A) [31].
When 500 μM of edaravone was added to the full reaction, the initial
formation of the Compound II was observed but rapidly regenerated to
native MPO within 2 min, significantly increasing the rate of MPO
peroxidase cycle (Fig. 1B). These data demonstrate that edaravone acts
as an efficient electron donor substrate for the MPO peroxidase activity.
The data presented thus far revealed a role for edaravone in the
MPO activity. Kinetic UV–Vis analysis was performed to confirm that
the peroxidase activity of MPO catalyzed edaravone into its corre-
sponding metabolite(s) (Fig. 3). The UV–Vis spectra of the reaction
between 100 μM edaravone and 500 nM MPO/100 μM H2O2 were
monitored for 60 min. Examination of the entire UV–Vis spectrum de-
tected a significant increase in absorbance at 350 nm by 10.3-fold after
60 min compared to t = 0 (Fig. 3). The increased absorption at this
wavelength is expected to correspond to the MPO-oxidized metabolite
(s) of edaravone. No other significant metabolite was detected by the
UV–Vis spectroscopy.
3.2. The effect of edaravone on MPO-mediated HOCl formation
(chlorination of MCD)
Chloride ion is readily oxidized to HOCl by the MPO Compound I.
Chlorination of MCD into DCD by HOCl was used as a surrogate marker
of HOCl production from MPO/H2O2 reaction [32]. The absorbance
maximum of MCD at 291 nm decreases upon chlorination into DCD.
The change in absorbance of MCD at each time point from the t = 0
reflects the extent of MCD chlorination, and with time, the relative rate
of HOCl produced. In a reaction with 50 nM MPO, 200 mM NaCl and
300 μM H2O2, MCD was chlorinated at a constant rate over 100 s. In-
troduction of 0.3, 3 and 10 μM of edaravone to the reaction con-
centration-dependently augmented the MPO-mediated chlorination of
3.4. Metabolite formation by MPO-mediated oxidation of edaravone
To further characterize the MPO-mediated oxidation of edaravone,
the reaction mixture of MPO/H2O2/edaravone was analyzed by LC-MS.
The total ion current (TIC) scan was used to acquire the retention time
for metabolites and provide m/z by high-resolution MS. We were not
able to propose structures for all metabolites, but certain major ones
previously reported to occur through oxidation of edaravone were
Fig. 2. Concentration-dependent effect of edaravone on the MPO-mediated chlorination of MCD. Each 300 μL of reaction in PB contained 50 nM MPO, 200 mM
NaCl, 40 μM MCD with 0, 0.3, 3, 10 (A), 0, 10, 30 or 100 μM (B) of edaravone. Following the addition of 300 μM H2O2, absorbance at 291 nm was measured every 20 s
for 100 s. The change in the absorbance was obtained by subtracting the absorbance at t = 0 from each measured time point.
425

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
Fig. 3. Accumulation of MPO-catalyzed edaravone product(s) detected by UV–Vis spectroscopy. A reaction of 500 nM and 100 μM edaravone was prepared in
PB. After triggering the reaction with 100 μM H2O2, UV–Vis spectra were recorded at t = 0, 2, 5, 10, 20 and 60 min. The figure inset shows the formation of the
edaravone product over 60 min at 350 nm. An absorption spectrum of 500 nM resting MPO in the PB was recorded as a reference.
referred to for guidance [7,10].
where θ is the dihedral angle, ρN is the spin density on the nitrogen
atom of the nitroxyl function, B is the proportionality constant for each
radical adduct, and B0 is the same constant when θ = 90°. For nitr-
oxides, B0 ≈ 0, ρ ≅ 0.5. For the spectrum observed with edaravone,
The m/z for (2) appeared to elute multiple times (3.97, 5.40, 5.56);
it is possible that this is due to an isomer having the same elemental
composition. The latter is possible as structures in (4) and (5) are
proposed to be a trimer and tetramer of edaravone that necessitates
bond formation between the phenyl rings. The retention times, for-
mulas, m/z and calculated m/z are shown in Table 1, and the extracted
ion chromatograms along with proposed structures as determined using
high-resolution LC-MS are shown in Fig. 4.
αβH = 28.1 G, andNB = 62 G, therefore
θ ≈ 18°
. If the same calculation
was applied to a much smaller free radical trapped by DMPO (i.e.,
hydroxyl radical), the result is θ ≈ 36°. These calculations and findings
are in agreement with observations reported by Janzen et al., 198533
,
where large free radical molecules would produce a small dihedral
angle.
3.5. MPO-catalyzed edaravone radical is detected by ESR spin trapping with
DMPO
3.6. Proton exchange to determine the site of ionization on edaravone
Using d6-DMSO, studies were carried out to determine if the hy-
drogens on the α-carbon (adjacent to the carbonyl) were exchangeable
with deuterium oxide (Fig. 6). Edaravone in DMSO-d6 was present in
the enol form (Fig. 6A), with the following proton splittings: δ 11.47 (s,
1H), 7.75–7.70 (m, 2H), 7.44–7.38 (m, 2H), 7.22–7.17 (m, 1H), 5.35 (s,
1H), 2.12 (s, 3H). Upon addition of D2O, the hydrogens on the α-carbon
and the enol both exchanged with 2D (Fig. 6B).
In Fig. 5, we have shown a free radical trapped during MPO-cata-
lyzed oxidation of edaravone using DMPO spin trapping with ESR
spectroscopy. The DMPO/●edaravone adduct produced
a six-line
spectrum indicative of a carbon-centred edaravone radical (Fig. 5A).
The signal intensity significantly diminished when GOx (the H2O2
source) was removed from the system (Fig. 5B). The
DMPO/●edaravone signal was not detected if MPO was omitted from
the complete system (Fig. 5C). A simulated spectrum of the experi-
mental spectrum in Fig. 5A is shown in Fig. 5D. The simulation revealed
an unprecedentedly large hyperfine splitting constant. In order to va-
lidate the simulated and experimental findings, we calculated the di-
hedral angle between the C–H bond and nitrogen π-orbital electrons by
the following equation [33]:
3.7. Edaravone radical-induced glutathionyl radical formation
The potency of edaravone radical to oxidize glutathione was as-
sessed by ESR spin trapping using DMPO. The edaravone radical pro-
duced from the reaction of 50 nM MPO/100 μM H2O2 oxidized glu-
tathione into glutathionyl radical, which was trapped using DMPO to
produce
a characteristic four-line spectrum corresponding to the
αβH = (B0 + Bcos2θ)ρN
DMPO/●SG (Fig. 7a). However, the intensity of the spectrum was minor
Table 1
Chromatographic details for edaravone metabolites.
Name
RT
Formula
m/z
Calc. m/z (exact mass + H+
)
Δ(ppm)
Edaravone
3.41
3.97, 5.40, 5.56
4.92
5.48
5.96
C10H10N2O
175.0868
347.1503
363.1443
519.2125
691.2754
175.0866
347.1503
363.1452
519.2144
691.2781
−1.2
−0.14
2.39
2.73
3.15
(2)
(3)
(4)
(5)
C
C
C
C
20H18N4O2
20H18N4O3
30H26N6O3
40H34N8O4
426

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
Fig. 5. Formation of the edaravone free radical metabolite through per-
oxidase activity of MPO analyzed by ESR. (A) The experimental spectrum run
with PB consisted of 6.4 mM edaravone, 100 mM DMPO, 500 nM MPO, 5 mM
glucose and 25 mU/mL glucose oxidase (GOx). (B) GOx and (C) MPO were
removed from the complete system, respectively in the control reactions. ESR
spectra were acquired by transferring the reaction into 3 separate 50 μL micro-
capillary tubes, inserted simultaneously into a 3 mm ESR tube which was then
placed in the resonator. ESR spectrometer settings: Center field = 3505 G,
sweep
width = 100 G,
Field
modulation = 1 G,
Microwave
frequency = 9.8 GHz, Microwave power = 20 mW, sweep time = 120 s,
number of scans = 2. (D) The simulated spectrum was well correlated to the
experimental spectrum (r = 0.99, produced using Winsim 2002), where the
a
N = 16.5 G and aH = 28.1 G.
Fig. 4. High-resolution LC-MS analysis of MPO-catalyzed oxidation pro-
ducts of edaravone. The reaction mixture consisted of 500 μM edaravone,
500 nM MPO and 100 μM H2O2 was incubated for 1 h. Edaravone (1) and its
MPO-catalyzed metabolites (2, 3, 4 and 5) were analyzed after liquid-liquid
extraction. For clarity, we have shown the extracted ion chromatograms for
each specific metabolite due to the overlapping retention times for some of the
metabolites.
decrease in ATP and the increase in cytotoxicity; conversely, the ad-
dition of 1–10 μM of edaravone exacerbated the H2O2-induced cyto-
toxicity in G/GOx-treated cells, and appeared to do the same in H2O2-
treated cells, but the latter was not statistically significant. The con-
centration-dependent biphasic toxic-protective effect of edaravone on
H2O2-induced toxicity was reproduced in the LDH assay (Fig. 8C). A
bolus concentration of 30 μM H2O2 induced significant cytotoxicity,
which was significantly augmented by the addition of 1 and 5 μM of
edaravone, whereas 25–100 μM of edaravone provided reversal of the
H2O2-induced toxicity. The LDH assay was not conducted for G/GOx-
treated cells due to colorimetric interference (data not shown).
compared to the vehicle control (Fig. 7c) and was significantly less
intense compared to the DMPO/●SG produce by phenoxyl radicals
(Fig. 7b).
3.8. Edaravone exhibits a concentration-dependent biphasic effect on H2O2-
induced toxicity of HL-60 cells
3.9. MPO estimation in HL-60
Human acute promyelocytic leukemia (HL-60) cells constitutively
express a high level of MPO. Cell viability (metabolic activity using
CellTiter-Glo™) and cytotoxicity (LDH assay) were performed to ex-
amine the effect of edaravone on H2O2-induced toxicity of HL-60 cells
(Fig. 8). A bolus concentration of 30 μM H2O2 (Fig. 8A) and a flux of
H2O2 produced by glucose/GOx (G/GOx) (Fig. 8B) induced a significant
decrease in cell viability (drop in ATP) over 3 h. The 100 and 200 μM of
edaravone dose-dependently protected against the H2O2-induced
Cellular MPO protein content was approximated by immunoblotting
the purified human MPO and the HL-60 cell total protein lysate (Fig. 9).
The band intensities for the purified MPO increased linearly with the
concentration (0.33, 1, 3 and 9 μM). Based on a calibration curve, the
amount of MPO in 106 HL-60 cells was estimated to be 11.5 μg. This
estimation was performed based on the intensities of the MPO heavy
subunit (55 kDa).
427

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
Figure 6. 600 MHz 1H-NMR spectrum of edaravone in DMSO-d6 with the subsequent addition of D2O. Edaravone in DMSO-d6 appeared to be in the keto form
(A), but after addition of D2O, the spectrum demonstrated that the hydrogens (on the enol and the α-carbon) exchanged with deuterium (B).
4. Discussion
the U.S. Food and Drug Administration, in addition to the existing oral
drug riluzole [1,34]. Since 2015, Mitsubishi Tanabe Pharma Corpora-
tion has also gained marketing approval for edaravone in Japan, South
Korea, the United States and Canada [1,2,35].
Edaravone has received much attention in the last few years as the
first ALS treatment option in over two decades to receive approval from
428

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
Fig. 7. Glutathionyl radical detection produced by edaravone or phenol
oxidation by MPO/H2O2. The characteristic four-line spectra correspond to
the DMPO/●SG spin adduct. Reactions were carried out in PB, which contained
100 mM DMPO, 1 mM GSH, 50 nM MPO, and 10 μM edaravone (a), 10 μM
phenol (b), or DMSO (vehicle control, c). Reactions were initiated by the ad-
dition of 100 μM H2O2, vortexed briefly, then transferred to 3 × 50 μL micro-
capillary tubes which were placed in an ESR tube for acquiring spectra. ESR
parameters were the same as described previously, except one scan was per-
formed.
Although a considerable amount of work has been conducted to
establish the antioxidant properties of edaravone, it remains con-
troversial whether its pharmacological effects can be ascribed entirely
to free radical scavenging. Pathobiology of a wide range of diseases
involve influx of neutrophils to the site of injury, which release ROS and
exacerbate local oxidative damage, inflammation and consequent tissue
injury [36]. As a potent antioxidant drug, numerous studies provide
evidence that edaravone is a good candidate for treating diseases in-
volving oxidative stress for its protective effects, including the reduced
neutrophil infiltration to the site of injury. Yang et al. have shown that
edaravone provided significant protective effects against neutrophil
infiltration and tissue injury in acute pancreatitis and associated lung
injury in rats [37]. In an acute liver injury model induced by co-ad-
ministration of D-galactosamine and lipopolysaccharide, edaravone re-
duced neutrophil infiltration to the rat liver tissues, and attenuated the
oxidative stress and inflammation [38]. However, the direct effect of
edaravone on MPO activity in neutrophils has not yet been eluciated to
our knowledge. The present study illustrates a new activity of edar-
avone by demonstrating its role in modulating the MPO-mediated
oxidative stress.
Mounting evidence points toward MPO as a critical contributing
player in the progression and pathogenesis of various diseases, espe-
cially those characterized by acute or chronic inflammation [19]. Ex-
cessive generation of oxidants by MPO has been correlated with tissue
damage in neurogenerative, cardiovascular, renal, lung, and other
chronic inflammatory diseases [19]. Much of the MPO-mediated oxi-
dative stress is caused by HOCl (or other hypohalous acids) [39]. The
short-lived and highly reactive HOCl targets a wide range of cellular
biomolecules exerting cytotoxic effect [40]. Oxidation of cysteine re-
sidues by HOCl can affect the functions of critical transcriptional factors
and regulatory proteins that contain cysteine in their active site [40].
HOCl can also chlorinate amines and amides in phospholipids, glyco-
saminoglycan, DNA and RNA [40]. These reactions result in the for-
mation of long-lived chloramines, some of which retain oxidizing ac-
tivities and contribute to prolonged tissue damage [41,42]. Oxidization
of glutathione by HOCl further disrupts the cellular redox balance,
exacerbating the oxidative damage and kindling inflammatory pro-
cesses [40]. Moreover, HOCl acts as a precursor to other reactive spe-
cies such as HO•, singlet oxygen (1O2) and ozone (O3) [43–45].
Therefore, an MPO inhibitor that interferes with pathologically persis-
tent activation of MPO has been considered a viable target for ther-
apeutic intervention. For example, AZD5904 is a new compound being
developed (Phase 1 studies have been completed) as a clinical inhibitor
of MPO based on 2-thioxanthines [46,47].
(caption on next page)
depending on its oxidation status. The resting enzyme is in its ferric
form [MPO-Fe(III)] and is converted to Compound I containing oxy-
ferryl [MPO-Fe(IV)]O] heme center upon reacting with H2O2. In the
halogenation cycle, halides and pseudo-halides directly reduce
Compound I back to its resting state. This cycle produces reactive hy-
pohalous acids (HOCl, HOBr) or hypothiocyanate. Alternatively,
Production of HOCl depends on the competition between the two
competing MPO redox pathways (i.e., chlorination vs. peroxidation).
The active heme center of MPO exists in multiple redox intermediates
429

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
Fig. 8. Concentration-dependent effect of edaravone on H2O2-induced
cytotoxicity of HL-60 cells. HL-60 cells were treated with a bolus concentra-
tion of 30 μM H2O2 (A and C) or 5 mM glucose (G) and 10 mU/mL GOx (B) for
3 h in PBS with or without 1, 5, 10, 25, 50, 100 and 200 μM of edaravone (Eda).
Cell viability (ATP) and cytotoxicity (LDH) were assessed by CellTiter-Glo (A
and B) and LDH (C), respectively. Data were analyzed by One-way ANOVA with
Bonferroni multiple comparisons (n = 3). The results represent the mean
halogenation cycle or the peroxidase cycle depending on the apparent
bimolecular rate constant and donor concentrations [40,48]. Kinetic
UV–Vis analyses revealed edaravone plays a role in the cycling of MPO
redox intermediates by serving as a peroxidase substrate of MPO, as
evident by its effect on Compound II reduction back to the resting state.
This efficient enzymatic turnover, however, may play a role in the
enhancement of HOCl formation, as evidenced by MCD chlorination.
These data collectively suggest that edaravone at low concentration
(0.3–10 μM) enhanced the resting MPO regeneration through the per-
oxidase cycle, increasing the rate of Compound I formation that reacts
with Cl− to produce HOCl. It is likely that a relatively higher con-
centration of MCD (40 μM) continuously consumed HOCl from the
system, shifting the reaction equilibrium towards the HOCl production.
In contrast, the rate of HOCl production concentration-dependently
decreased when edaravone concentrations were increased to 30 μM and
above. This observation can be explained by the HOCl-producing ha-
logenation cycle being outcompeted by the introduction of a high
concentration of edaravone as the MPO substrate and target for HOCl.
Moreover, as the peroxidase pathway becomes dominant under high
donor (edaravone) concentrations, the amount of Compound I available
to undergo the halogenation pathway decreases. Therefore, a low
concentration of edaravone can enhance the halogenation activity and
the HOCl production in the MPO/H2O2/Cl− system, whereas a high
concentration of edaravone divert the enzymatic activity towards the
peroxidase cycle and reduce HOCl production.
values
to the negative control (NC); ##p < 0.01 and ###p < 0.001 compared to
H2O2; **p 0.01 and ***p 0.001 compared to G/GOx. NC, negative
SD. Statistically significant comparisons: +++p < 0.001 compared
<
<
control (PBS); PC, positive control (lysis buffer); DMSO, 0.04% DMSO solvent
control; Eda, edaravone.
MPO-catalyzed metabolites of edaravone were analyzed and com-
pared to previously reported findings from edaravone oxidation by free
radicals. Yamamoto et al. described the time-course formation of the
oxidation products of edaravone [7]. Initially, edaravone anion is
thought to donate an electron to a radical, converting the radical to the
corresponding anion [7]. The resulting edaravone radical is converted
to edaravone peroxyl radical by reacting with oxygen [7]. Subsequent
formation of 4-oxoedaravone was observed as the main product, along
with small amounts of a hydroxylated edaravone dimer (BPOH) [7].
The 4-oxoedarvone was hydrolyzed into 2-oxo-3-(phenylhydrazono)
butanoic acid [7].
Fig. 9. Intracellular MPO content in HL-60 cells compared to isolated
MPO. Human purified MPO (0.33, 1, 3 and 9 μg) and the HL-60 cell total
protein lysate (0.1, 0.3 and 0.9 × 106 cells) were subjected to western blot
analysis. Band intensities for the MPO heavy chain subunit (55 kDa) were
quantified using ImageJ. Data is presented as mean
SD (n = 3).
In our experiment, the time-dependent formation of the MPO-cat-
alyzed metabolites was confirmed by kinetic UV–Vis spectroscopy. The
product formed at λmax of 350 nm could be the same product reported
at λmax of 345 nm formed by peroxyl and azidyl radicals [49]. Several
different metabolite structures of edaravone were detected by LC-MS. In
contrast to earlier findings, we did not find species that corresponded to
4-oxoedaravone or its hydrolyzed product. However, we detected a
structure corresponding to BPOH, and a molecule structurally similar to
the BPOH only without the 4-hydroxyl group. Interestingly, dimers,
trimers, and tetramers of edaravone were also detected. We detected a
trace amount of dimer present in the starting material (data not shown),
though this may not be sufficient to partake in radical dimerization.
As we demonstrated in this study that edaravone serves as a per-
oxidase substrate of MPO, a free radical intermediate was hypothesized
to be produced during the peroxidase cycle of MPO. ESR spin-trapping
with DMPO revealed the formation of a primary edaravone radical
through MPO-mediated oxidation. The stability of the edaravone ra-
dical was previously investigated by Ono et al. [50]. Based on the
electron density function calculation, they concluded the unpaired
electron on the pyrazoline radical was highly delocalized, and thus its
reactivity should be much less than that of reactive oxygen radicals
such as HO●, alkyl peroxyl radicals, and alkoxyl radicals [50]. How-
ever, our findings conclusively demonstrated that a carbon-centred
radical was formed and trapped. To further validate the formation of a
carbon-centred free radical, we performed NMR studies to determine
proton exchange using D2O. Our findings indicated that the α-carbon
could be ionized, and its proximity to the carbonyl group likely facil-
itates its ionization (pKa = 6.9) [51]. An extreme example of this is
phenylbutazone, which has an acidic carbon between two ketone
groups (pKa = 4.5) [52]. The antioxidant activity of edaravone (and
Fig. 10. The role of edaravone in modulating the MPO catalytic cycles. At
physiological concentration, the chloride ion is a potential substrate for oxi-
dation by MPO Compound I and hypochlorous acid is the major product.
Edaravone serves as a peroxidase substrate, enhancing the rate of Compound II
reduction and native MPO regeneration. Edaravone is oxidized by Compound I/
II to a free radical metabolite. We propose that the halogenation cycle is pro-
nounced when edaravone concentrations are low (up to 10 μM in this study).
On the other hand, the non-HOCl producing peroxidase cycle is favoured when
the edaravone concentration is high (≥ 30 μM in this study).
Compound I may also oxidize substrates in two sequential one-electron
oxidation steps through the formation of another redox intermediate
Compound II (MPO-Fe(IV)–OH]. Compound I can either follow the
430

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
Fig. 11. Proposed mechanisms of MPO-
mediated oxidation of edaravone based
on observations in this study. Edaravone
undergoes keto-enol tautomerization as
well as ionization. The carbanion-keto form
of edaravone is shown for simplicity, which
is apparently detected as dimer structures
using LC-MS (i.e., the enol is no longer de-
tected after MPO oxidation). The carbon-
centred radical of edaravone was verified by
ESR spin trapping with DMPO, and some of
the dimer products previously reported
were verified using LC-MS in this study, in
addition to new proposed metabolites [10].
thus its ionization potential) has been ascribed in part to its ionization
which facilitates electron transfer [53]. To investigate the relative
oxidizing potential of the edaravone radical, we tested its ability to
oxidize glutathione in comparison to the highly reactive phenoxyl ra-
dical using ESR spin-trapping of the glutathionyl radical with DMPO. In
agreement with the finding of Ono et al., 1997, edaravone radical
showed only a minor reactivity with glutathione compared to control
and significantly less than that of phenoxyl radical [41].
In order to evaluate the intracellular effects of edaravone, H2O2-
treated HL-60 cells (abundant in MPO) were assessed for cell viability.
Interestingly, edaravone exacerbated the H2O2-induced toxicity at a
low concentration range while attenuating it at a high concentration
range. This observation correlates with the concentration-dependent
biphasic effect of edaravone on the rate of HOCl production, as this
phenomenon cannot be explained by the free radical scavenging ac-
tivity of edaravone alone.
stroke and ALS. Research of its pharmacological effects has mainly been
focused on its antioxidant activity. In this report, we have provided
evidence that edaravone serves as a peroxidase substrate of MPO, and
that this is likely facilitated by acting as an efficient donor substrate for
the peroxidase cycle of MPO (Fig. 10). The consequence of donating
electrons results in edaravone carbon-centred radicals; though the
latter, or its dimerized products, do not appear to adversely affect cells
or GSH oxidation significantly (Fig. 11). Most importantly, our study
elucidated that edaravone exerts a concentration-dependent biphasic
effect on the MPO/H2O2/Cl− system to modulate the rate of toxic HOCl
production and consequent H2O2-induced HL-60 cell death. Further
studies should be carried out to expand our knowledge about the in-
tracellular effect of edaravone in HOCl-mediated oxidative stress, and
to determine if MPO modulation by edaravone is a functional me-
chanism of action in vivo.
The cell viability assay assumes the cell count based on the ATP
quantitation [54], and the LDH assay result verifies that H2O2 indeed
induced cell death via membrane lysis. Many studies have demon-
strated that H2O2-induced apoptotic and necrotic death in HL-60 cells is
dependent on the HOCl generation with subsequent formation of
chloramines by reaction with amines and amides [55,56]. Although
there is no strict consensus on the exact concentration, there is a
threshold concentration of HOCl, above which the majority of cell
death is manifested by necrosis rather than apoptosis [57]. For ex-
ample, Wagner et al. has demonstrated that 10–15 μM of H2O2 induces
apoptosis in HL-60 cells, and necrosis at concentrations > 20 μM based
on the DNA fragmentation assay. Under our experimental conditions,
30 μM H2O2 induced significant cell death in HL-60 cells, which was
increased or decreased depending on the edaravone concentration, al-
though the mechanism of cell death was not clarified as it is beyond the
scope of the current study.
Lastly, MPO content in HL-60 total cell protein lysate was estimated.
Surprisingly, the MPO amount we found by immunoblotting was higher
than what was previously reported [58]. Firstly, small-scale purification
poses significant variation in protein content during the processing. In
addition, MPO is a multimeric protein composed of two heavy subunits
(55 kDa) and two light subunits (15 kDa) which are denatured into
several different combinations during the SDS-PAGE, which makes it
challenging for quantification [59]. Also, the measurement of band
intensities is only semi-quantitative. The MPO concentration used in
cell viability experiments cannot be compared directly to that in the ex-
vitro experiments, as the interaction of edaravone with MPO in the
cellular system involves cellular uptake and interaction with other
proteins.
Acknowledgements
The authors acknowledge funding support from the Natural
Sciences and Engineering Research Council (NSERC) grant number
RGPIN-2014-04878.
References
[1] J.D. Rothstein, Edaravone: a new drug approved for ALS, Cell 171 (4) (2017) 725
doi: S0092-8674(17)31198-4 [pii].
[2] K. Dangerfiekld, ALS drug approved by health Canada could slow down progression
of fatal disease, Available from: https://globalnews.ca/news/4519791/health-
canada-approves-als-drug-radicava/, (Oct 5th, 2019) , Accessed date: 1 March
2019.
[3] C.L. Allen, U. Bayraktutan, Oxidative stress and its role in the pathogenesis of
ischaemic stroke, Int. J. Stroke 4 (6) (2009) 461–470, https://doi.org/10.1111/j.
1747-4949.2009.00387.x [doi].
[4] I. Olmez, H. Ozyurt, Reactive oxygen species and ischemic cerebrovascular disease,
Neurochem. Int. 60 (2) (2012) 208–212, https://doi.org/10.1016/j.neuint.2011.11.
009 [doi].
[5] M.M. Gaschler, B.R. Stockwell, Lipid peroxidation in cell death, Biochem. Biophys.
Res. Commun. 482 (3) (2017) 419–425 doi: S0006-291X(16)31771-5 [pii].
[6] S.C. Barber, R.J. Mead, P.J. Shaw, Oxidative stress in ALS: a mechanism of neu-
rodegeneration and a therapeutic target, Biochim. Biophys. Acta 1762 (11–12)
(2006) 1051–1067 doi: S0925-4439(06)00052-4 [pii].
[7] Y. Yamamoto, T. Kuwahara, K. Watanabe, K. Watanabe, Antioxidant activity of 3-
methyl-1-phenyl-2-pyrazolin-5-one, Redox Rep. 2 (5) (1996) 333–338, https://doi.
org/10.1080/13510002.1996.11747069 [doi].
[8] O. Tokumaru, Y. Shuto, K. Ogata, et al., Dose-dependency of multiple free radical-
scavenging activity of edaravone, J. Surg. Res. 228 (2018) 147–153 doi: S0022-
4804(18)30185-9 [pii].
[9] E. Kamogawa, Y. Sueishi, A multiple free-radical scavenging (MULTIS) study on the
antioxidant capacity of a neuroprotective drug, edaravone as compared with uric
acid, glutathione, and trolox, Bioorg. Med. Chem. Lett 24 (5) (2014) 1376–1379,
https://doi.org/10.1016/j.bmcl.2014.01.045 [doi].
[10] K. Watanabe, M. Tanaka, S. Yuki, M. Hirai, Y. Yamamoto, How is edaravone ef-
fective against acute ischemic stroke and amyotrophic lateral sclerosis? J. Clin.
Biochem. Nutr. 62 (1) (2018) 20–38, https://doi.org/10.3164/jcbn.17-62 [doi].
[11] R. Radi, Peroxynitrite, a stealthy biological oxidant, J. Biol. Chem. 288 (37) (2013)
26464–26472, https://doi.org/10.1074/jbc.R113.472936 [doi].
5. Conclusion
Edaravone is a novel antioxidant drug approved for the treatment of
431

L.Y.K. Suh, et al.
FreeRadicalBiologyandMedicine143(2019)422–432
[12] A. Fujisawa, Y. Yamamoto, Edaravone, a potent free radical scavenger, reacts with
peroxynitrite to produce predominantly 4-NO-edaravone, Redox Rep. 21 (3) (2016)
98–103, https://doi.org/10.1179/1351000215Y.0000000025 [doi].
[13] S.J. Klebanoff, Myeloperoxidase: friend and foe, J. Leukoc. Biol. 77 (5) (2005)
598–625.
[14] J. Schultz, K. Kaminker, Myeloperoxidase of the leucocyte of normal human blood.
I. content and localization, Arch. Biochem. Biophys. 96 (3) (1962) 465–467,
https://doi.org/10.1016/0003-9861(62)90321-1 https://www.sciencedirect.com/
science/article/pii/0003986162903211.
[15] K.E. Brown, E.M. Brunt, J.W. Heinecke, Immunohistochemical detection of mye-
loperoxidase and its oxidation products in kupffer cells of human liver, Am. J.
Pathol. 159 (6) (2001) 2081–2088, https://doi.org/10.1016/S0002-9440(10)
63059-3 https://www.sciencedirect.com/science/article/pii/
approval-in-europe, (2018) , Accessed date: 1 March 2019Updated.
[36] S. Fujishima, N. Aikawa, Neutrophil-mediated tissue injury and its modulation,
Intensive Care Med. 21 (3) (1995) 277–285, https://doi.org/10.1007/BF01701489
https://www.ncbi.nlm.nih.gov/pubmed/7790621.
[37] T. Yang, Y.F. Mao, S.Q. Liu, et al., Protective effects of the free radical scavenger
edaravone on acute pancreatitis-associated lung injury, Eur. J. Pharmacol. 630
(1–3) (2010) 152–157, https://doi.org/10.1016/j.ejphar.2009.12.025 [doi].
[38] K. Ito, H. Ozasa, Y. Noda, S. Arii, S. Horikawa, Effects of free radical scavenger on
acute liver injury induced by d-galactosamine and lipopolysaccharide in rats,
Hepatol. Res. : Off. J. Jpn. Soc. Hepatol. 38 (2) (2008), https://doi.org/10.1111/j.
1872-034X.2007.00252.x https://www.ncbi.nlm.nih.gov/pubmed/17727650.
[39] J.M. Pullar, M.C. Vissers, C.C. Winterbourn, Living with a killer: the effects of hy-
pochlorous acid on mammalian cells, IUBMB Life 50 (4–5) (2000) 259–266,
https://doi.org/10.1080/713803731 [doi].
S0002944010630593.
[16] P.S. Green, A.J. Mendez, J.S. Jacob, et al., Neuronal expression of myeloperoxidase
is increased in alzheimer's disease, J. Neurochem. 90 (3) (2004) 724–733, https://
doi.org/10.1111/j.1471-4159.2004.02527.x [doi].
[17] R.M. Nagra, B. Becher, W.W. Tourtellotte, et al., Immunohistochemical and genetic
evidence of myeloperoxidase involvement in multiple sclerosis, J. Neuroimmunol.
78 (1–2) (1997) 97–107, https://doi.org/10.1016/S0165-5728(97)00089-1
https://www.ncbi.nlm.nih.gov/pubmed/9307233.
[40] M.J. Davies, C.L. Hawkins, D.I. Pattison, M.D. Rees, Mammalian heme peroxidases:
from molecular mechanisms to health implications, Antioxidants Redox Signal. 10
(7) (2008) 1199–1234, https://doi.org/10.1089/ars.2007.1927 [doi].
[41] D.B. Learn, V.A. Fried, E.L. Thomas, Taurine and hypotaurine content of human
leukocytes, J. Leukoc. Biol. 48 (2) (1990) 174–182.
[42] E.L. Thomas, M.B. Grisham, M.M. Jefferson, Cytotoxicity of chloramines, Methods
Enzymol. 132 (1986) 585–593.
[18] S.J. Klebanoff, Myeloperoxidase: friend and foe, J. Leukoc. Biol. 77 (5) (2005)
598–625 doi: jlb.1204697 [pii].
[19] B.S. van der Veen, M.P. de Winther, P. Heeringa, Myeloperoxidase: molecular
mechanisms of action and their relevance to human health and disease,
Antioxidants Redox Signal. 11 (11) (2009) 2899–2937, https://doi.org/10.1089/
ARS.2009.2538 [doi].
[20] R.S. Ray, A. Katyal, Myeloperoxidase: bridging the gap in neurodegeneration,
Neurosci. Biobehav. Rev. 68 (2016) 611–620 doi: S0149-7634(16)30048-3 [pii].
[21] A. Daugherty, J.L. Dunn, D.L. Rateri, J.W. Heinecke, Myeloperoxidase, a catalyst for
lipoprotein oxidation, is expressed in human atherosclerotic lesions, J. Clin.
Investig. 94 (1) (1994) 437–444, https://doi.org/10.1172/JCI117342 [doi].
[22] C.C. Winterbourn, R. Senthilmohan, V.A. Cameron, et al., Plasma concentrations of
myeloperoxidase predict mortality after myocardial infarction, J. Am. Coll. Cardiol.
49 (20) (2007) 1993–2000, https://doi.org/10.1016/j.jacc.2007.02.040 https://
www.clinicalkey.es/playcontent/1-s2.0-S0735109707008340.
[23] I. Cascorbi, S. Henning, J. Brockmoller, et al., Substantially reduced risk of cancer of
the aerodigestive tract in subjects with variant -463A of the myeloperoxidase gene,
Cancer Res. 60 (3) (2000) 644–649 http://cancerres.aacrjournals.org/cgi/content/
abstract/60/3/644.
[43] L.P. Candeias, K.B. Patel, M.R. Stratford, P. Wardman, Free hydroxyl radicals are
formed on reaction between the neutrophil-derived species superoxide anion and
hypochlorous acid, FEBS Lett. 333 (1–2) (1993) 151–153 doi: 0014-5793(93)
80394-A [pii].
[44] J.R. Kanofsky, J. Wright, G.E. Miles-Richardson, A.I. Tauber, Biochemical require-
ments for singlet oxygen production by purified human myeloperoxidase, J. Clin.
Investig. 74 (4) (1984) 1489–1495, https://doi.org/10.1172/JCI111562 [doi].
[45] A.J. Kettle, B.M. Clark, C.C. Winterbourn, Superoxide converts indigo carmine to
isatin sulfonic acid: implications for the hypothesis that neutrophils produce ozone,
J. Biol. Chem. 279 (18) (2004) 18521–18525, https://doi.org/10.1074/jbc.
M400334200 [doi].
[46] AZD5904 mechanism of action: myeloperoxidase (MPO) inhibitor, https://
openinnovation.astrazeneca.com/azd5904.html , Accessed date: 1 March 2019.
[47] A.K. Tiden, T. Sjogren, M. Svensson, et al., 2-thioxanthines are mechanism-based
inactivators of myeloperoxidase that block oxidative stress during inflammation, J.
Biol. Chem. 286 (43) (2011) 37578–37589, https://doi.org/10.1074/jbc.M111.
266981 [doi].
[48] P.G. Furtmuller, M. Zederbauer, W. Jantschko, et al., Active site structure and
catalytic mechanisms of human peroxidases, Arch. Biochem. Biophys. 445 (2)
(2006) 199–213 doi: S0003-9861(05)00403-0 [pii].
[49] M. Lin, Y. Katsumura, K. Hata, Y. Muroya, K. Nakagawa, Pulse radiolysis study on
free radical scavenger edaravone (3-methyl-1-phenyl-2-pyrazolin-5-one), J.
Photochem. Photobiol., B 89 (1) (2007) 36–43 doi: S1011-1344(07)00110-8 [pii].
[50] S. Ono, K. Okazaki, M. Sakurai, Y. Inoue, Density functional study of the radical
reactions of 3-methyl-1-phenyl-2-pyrazolin-5-one (MCI-186): Implication for the
biological function of MCI-186 as a highly potent antioxidative radical scavenger, J.
Phys. Chem. A 101 (20) (1997) 3769–3775.
[51] K. Chegaev, C. Cena, M. Giorgis, et al., Edaravone derivatives containing NO-donor
functions, J. Med. Chem. 52 (2) (2009) 574–578, https://doi.org/10.1021/
jm8007008 [doi].
[52] A.B. Gutman, P.G. Dayton, T.F. Yu, et al., A study of the inverse relationship be-
tween pKa and rate of renal excretion of phenylbutazone analogs in man and dog,
Am. J. Med. 29 (1960) 1017–1033 doi: 0002-9343(60)90082-6 [pii].
[53] A. Perez-Gonzalez, A. Galano, Ionization energies, proton affinities, and pKa values
of a large series of edaravone derivatives: implication for their free radical
scavenging activity, J. Phys. Chem. B 115 (34) (2011) 10375–10384, https://doi.
org/10.1021/jp2047163 [doi].
[24] J. Ahn, M.D. Gammon, R.M. Santella, et al., Myeloperoxidase genotype, fruit and
vegetable consumption, and breast cancer risk, Cancer Res. 64 (20) (2004)
7634–7639, https://doi.org/10.1158/0008-5472.CAN-04-1843 http://cancerres.
aacrjournals.org/cgi/content/abstract/64/20/7634.
[25] W.F. Reynolds, E. Chang, D. Douer, E.D. Ball, V. Kanda, An allelic association im-
plicates myeloperoxidase in the etiology of acute promyelocytic leukemia, Blood 90
(7) (1997) 2730–2737 https://www.ncbi.nlm.nih.gov/pubmed/9326240.
[26] P. Wheatley‐Price, K. Asomaning, A. Reid, et al., Myeloperoxidase and superoxide
dismutase polymorphisms are associated with an increased risk of developing
pancreatic adenocarcinoma, Cancer 112 (5) (2008) 1037–1042, https://doi.org/10.
1002/cncr.23267 https://onlinelibrary.wiley.com/doi/abs/10.1002/cncr.23267.
[27] D. Choi, S. Pennathur, C. Perier, et al., Ablation of the inflammatory enzyme
myeloperoxidase mitigates features of Parkinson's disease in mice, J. Neurosci. 25
(28) (2005) 6594–6600, https://doi.org/10.1523/JNEUROSCI.0970-05.2005
http://www.jneurosci.org/cgi/content/abstract/25/28/6594.
[28] H.J. Grone, E.F. Grone, E. Malle, Immunohistochemical detection of hypochlorite-
modified proteins in glomeruli of human membranous glomerulonephritis, Lab.
Investig. 82 (1) (2002) 5–14.
[29] P. Heeringa, R.A. Matthijsen, C.J. Peutz-Kootstra, et al., Myeloperoxidase is criti-
cally involved in the induction of organ damage after renal ischemia reperfusion,
Am. J. Pathol. 171 (6) (2007) 1743–1752, https://doi.org/10.2353/ajpath.2007.
070184 https://www.clinicalkey.es/playcontent/1-s2.0-S0002944010624376.
[30] A.J. Kettle, T. Chan, I. Osberg, et al., Myeloperoxidase and protein oxidation in the
airways of young children with cystic fibrosis, Am. J. Respir. Crit. Care Med. 170
(12) (2004) 1317–1323, https://doi.org/10.1164/rccm.200311-1516OC http://
ajrccm.atsjournals.org/cgi/content/abstract/170/12/1317.
[31] P.G. Furtmuller, W. Jantschko, G. Regelsberger, C. Jakopitsch, N. Moguilevsky,
C. Obinger, A transient kinetic study on the reactivity of recombinant unprocessed
monomeric myeloperoxidase, FEBS Lett. 503 (2–3) (2001) 147–150 S0014-
5793(01)02725-9 [pii].
[54] Grace Ka Yan Chan, Tracy L. Kleinheinz, David Peterson, John G. Moffat, A simple
high-content cell cycle assay reveals frequent discrepancies between cell number
and ATP and MTS proliferation assays, PLoS One 8 (5) (2013) e63583, , https://doi.
org/10.1371/journal.pone.0063583 https://www.ncbi.nlm.nih.gov/pubmed/
23691072.
[55] S.R. Khan, A. Baghdasarian, P.H. Nagar, et al., Proteomic profile of aminoglu-
tethimide-induced apoptosis in HL-60 cells: role of myeloperoxidase and arylamine
free radicals, Chem. Biol. Interact. 239 (2015) 129–138, https://doi.org/10.1016/j.
cbi.2015.06.020 [doi].
[56] T. Matsura, M. Kai, Y. Fujii, H. Ito, K. Yamada, Hydrogen peroxide-induced apop-
tosis in HL-60 cells requires caspase-3 activation, Free Radic. Res. 30 (1) (1999)
73–83.
[32] L.K. Folkes, L.P. Candeias, P. Wardman, Kinetics and mechanisms of hypochlorous
acid reactions, Arch. Biochem. Biophys. 323 (1) (1995) 120–126 doi: S0003-
9861(85)70017-3 [pii].
[33] E.G. Janzen, H.J. Stronks, C.M. Dubose, J.L. Poyer, P.B. McCay, Chemistry and
biology of spin-trapping radicals associated with halocarbon metabolism in vitro
and in vivo, Environ. Health Perspect. 64 (1985) 151–170, https://doi.org/10.
1289/ehp.8564151 [doi].
[34] R.P. Dash, R.J. Babu, N.R. Srinivas, Two decades-long journey from riluzole to
edaravone: revisiting the clinical pharmacokinetics of the only two amyotrophic
lateral sclerosis therapeutics, Clin. Pharmacokinet. 57 (11) (2018) 1385–1398,
https://doi.org/10.1007/s40262-018-0655-4 [doi].
[35] Novel ALS therapy edaravone considered for approval in europe, https://www.
thepharmaletter.com/article/novel-als-therapy-edaravone-to-be-considered-for-
[57] Brett A. Wagner, Garry R. Buettner, Larry W. Oberley, Christine J. Darby, C. Patrick
Burns, Myeloperoxidase is involved in H2O2-induced apoptosis of HL-60 human
leukemia cells, J. Biol. Chem. 275 (29) (2000) 22461–22469, https://doi.org/10.
1074/jbc.M001434200 http://www.jbc.org/content/275/29/22461.abstract.
[58] H.R. Hope, E.E. Remsen, C. Lewis, et al., Large-scale purification of myeloperox-
idase from HL60 promyelocytic cells: characterization and comparison to human
neutrophil myeloperoxidase, Protein Expr. Purif. 18 (3) (2000) 269–276, https://
doi.org/10.1006/prep.1999.1197 https://www.sciencedirect.com/science/article/
pii/S104659289991197X.
[59] K.L. Taylor, J. Pohl, J.M. Kinkade J, Unique autolytic cleavage of human myelo-
peroxidase. implications for the involvement of active site MET409, J. Biol. Chem.
267 (35) (1992) 25282 http://www.jbc.org/content/267/35/25282.abstract.
432