Hydropropidine: A novel, cell-impermeant fluorogenic probe for detecting extracellular superoxide

Article abstract of DOI:10.1016/j.freeradbiomed.2012.09.018

Here we report the synthesis and characterization of a membrane-impermeant fluorogenic probe, hydropropidine (HPr⁺), the reduction product of propidium iodide, for detecting extracellular superoxide (O₂ ^?-). HPr⁺

Full text of DOI:10.1016/j.freeradbiomed.2012.09.018

Free Radical Biology and Medicine 54 (2013) 135–147
Contents lists available at SciVerse ScienceDirect
Free Radical Biology and Medicine
journal homepage: www.elsevier.com/locate/freeradbiomed
Original Contribution
Hydropropidine: A novel, cell-impermeant fluorogenic probe for detecting
extracellular superoxide
n
Radoslaw Michalski a,1, Jacek Zielonka a, Micael Hardy b, Joy Joseph a, Balaraman Kalyanaraman a,
a Department of Biophysics and Free Radical Research Center, Medical College of Wisconsin, 8701 Watertown Plank Road, Milwaukee, WI 53226, USA
b Institut de Chimie Radicalaire, Equipe SREP UMR 7273, Aix-Marseille Universite´, Campus de Saint Jerome, 13397 Marseille cedex 20, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Here we report the synthesis and characterization of a membrane-impermeant fluorogenic probe,
hydropropidine (HPrþ), the reduction product of propidium iodide, for detecting extracellular super-
Received 27 July 2012
Received in revised form
13 September 2012
ꢀ
oxide (O2ꢁ). HPrþ is a positively charged water-soluble analog of hydroethidine (HE), a fluorogenic
ꢀ
probe commonly used for monitoring intracellular O2ꢁ. We hypothesized that the presence of a highly
Accepted 14 September 2012
Available online 7 October 2012
localized positive charge on the nitrogen atom would impede cellular uptake of HPrþ and allow for
ꢀ
ꢁ
ꢀꢁ
exclusive detection of extracellular O2
.
Our results indicate that O2
reacts with HPrþ
(k¼1.2 ꢂ 104
M
ꢁ1 sꢁ1) to form exclusively 2-hydroxypropidium (2-OH-Pr2þ) in cell-free and cell-
Keywords:
ꢀ
ꢁ
Superoxide
based systems. This reaction is analogous to the reaction between HE and O2 (Zhao et al., Free Radic.
Biol. Med. 34:1359–1368; 2003). During the course of this investigation, we also reassessed the rate
Fluorescent probes
Hydroethidine
Hydropropidine
Propidium
ꢀ
ꢁ
constants for the reactions of O2 with HE and its mitochondria targeted analog (Mito-HE or MitoSOX
Red) and addressed the discrepancies between the present values and those reported previously by us.
Our results indicate that the rate constant between O2 and HPrþ is slightly higher than that of HE and
ꢀ
ꢁ
NADPH oxidase
ꢀ
ꢁ
ꢀ
O2 and is closer to that of Mito-HE and O2ꢁ. Similar to HE, HPrþ undergoes oxidation in the presence
of various oxidants (peroxynitrite-derived radicals, Fenton’s reagent, and ferricytochrome c) forming
the corresponding propidium dication (Pr2þ) and the dimeric products (e.g., Pr2þ-Pr2þ). In contrast to
HE, there was very little intracellular uptake of HPrþ. We conclude that HPrþ is a useful probe for
ꢀ
ꢁ
detecting O2 and other one-electron oxidizing species in an extracellular milieu.
& 2012 Published by Elsevier Inc.
ꢀ
ꢁ
Introduction
HE and O2 is 2-hydroxyethidium (2-OH-Eþ) [1,2,4,5]. Although
other oxidants (hydroxyl, nitrogen dioxide, and carbonate radicals
derived from peroxynitrꢁite) also react with HE to form different
Hydroethidine or dihydroethidium (HE or DHE) has become
the probe of choice for detecting intracellular superoxide radical
ꢀ
oxidation products, O2 remains as the only viable reactive
anion (O2ꢁ) [1–3]. The diagnostic product of the reaction between
oxygen species that reacts with HE to form 2-OH-Eþ [6–8]. In
ꢀ
an effort to selectively detect extracellular O2ꢁ, we synthesized
ꢀ
hydropropidine (HPrþ), a positively charged, water-soluble ana-
log of HE (Fig. 1). We reasoned that the presence of a highly
localized positive charge on the nitrogen atom of the alkyl group
will prevent the cellular uptake of HPrþ. Hydropropidine was
prepared from a two-electron reduction of propidium iodide
(Pr2þ ꢃ 2Iꢁ), a cell-impermeant fluorescent molecule (Fig. 1) that
is used frequently in tissue staining and in the identification of
dead cells [9]. Due to the structural similarity between HE and
HPrþ, we furꢁther reasoned that the reaction chemistry between
Abbreviations: HPrþ
,
hydropropidine; Pr2þ
,
propidium; 2-OH-Pr2þ
,
2-hydroxypropidium; HE, hydroethidine; NDS, nitrosodisulfonate radical dia-
nion; Mito-HE (or MitoSOX Red), mitochondria-targeted HE analog; DPBS, Dul-
becco’s PBS; DPBS-GP, Dulbecco’s PBS supplemented with pyruvic acid and
glucose; PMA, phorbol 12-myristate 13-acetate; MN, menadione bisulfite sodium
salt; HX, hypoxanthine; XO, xanthine oxidase; HRP, horseradish peroxidase; DTPA,
diethylenetriaminepentaacetic acid; BMPO, 5-tert-butoxycarbonyl 5-methyl-1-
pyrroline N-oxide; TFA, trifluoroacetic acid; EPR, electron paramagnetic reso-
nance; Pr2þ-Pr2þ, dipropidium (propidium-propidium homodimer); HPrþ-Pr2þ
hydropropidine-propidium heterodimer; HPrþ-HPrþ, hydropropidine-
hydropropidine homodimer.
,
ꢀ
HPrþ and O2 and other oxidants might be similar to that of HE.
In this study, we report that the reaction between HPrþ and
n Corresponding author. Fax: þ1 414 955 6512.
ꢀ
ꢁ
O2 yields 2-hydroxypropidium (2-OH-Prþ) (Fig. 1) as a diag-
E-mail addresses: rmichalski@mcw.edu,
radoslaw.michalski@p.lodz.pl (R. Michalski), jzielonk@mcw.edu (J. Zielonka),
micael.hardy@univ-provence.fr (M. Hardy), jjoseph@mcw.edu (J. Joseph),
nostic marker product in cell-free and cell-bꢁased systems. The
ꢀ
rate constant for the reaction between O2 and HPrþ was
balarama@mcw.edu (B. Kalyanaraman).
measured to be slightly higher than that of HE and O2ꢁ. In
ꢀ
1
On leave from the Institute of Applied Radiation Chemistry, Lodz University
of Technology, 90-924 Lodz, Poland.
addition, we reassessed the rate constant parameters previously
0891-5849/$ - see front matter & 2012 Published by Elsevier Inc.
http://dx.doi.org/10.1016/j.freeradbiomed.2012.09.018

136
R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
Fig. 1. Chemical structures of hydropropidine, 2-hydroxypropidium, other oxidation products of hydropropidine and other selected agents.
reported for HE and MitoSOX Red with O2ꢁ[1,10,11]. Results
indicate that these rate constants were overestimated by 30- to
100-fold in our previous studies [1,10]. Reasons for these dis-
crepancies are discussed. Similar to HE, HPrþ also undergoes
oxidation in the presence of other oxidants (derived from peroxy-
nitrite and peroxidases) forming propidium (Pr2þ) and character-
istic dimeric products such as dipropidium (Pr2þ-Pr2þ) (Fig. 1).
Biological implications for the use of this novel probe in monitoring
Syntheses and preparation of probes and reagents
ꢀ
Hydropropidine was prepared by reducing propidium in the
presence of sodium borohydride. Sodium borohydride (6 mg,
0.16 mmol) dissolved in 1 ml of MeOH was added dropwise to a
solution of propidium iodide (0.1 g, 0.15 mmol) in methanol
(5 ml) at 0 1C. After 30 min, the reaction product was extracted
with CH2Cl2. The extract was washed with water and brine and
dried over Na2SO4. The solvent was removed under vacuum
to give the hydropropidine cation (80 mg, 0.14 mmol, 98%).
The crude product was purified by HPLC, as described below for
2-hydroxypropidium. HPrþ stock solutions (20 mM) were pre-
pared in DMSO under anaerobic conditions, and small aliquots
ꢀ
ꢁ
extracellular generation of O2 and other one-electron oxidizing
species from NADPH oxidases are discussed.
Materials and methods
(20
m
l) were stored at -80 1C. All solutions containing HPrþ were
was
Chemicals
protected from light. 2-Hydroxypropidium (2-OH-Pr2þ
)
synthesized by reacting HPrþ with Fremy’s salt [4]. Due to the
instability of Fremy’s salt in acidic solutions, stock solutions were
prepared in phosphate buffer (100 mM, pH 7.5) containing
0.2 mM DTPA. Details of synthesis are identical to our previously
published procedure for 2-hydroxyethidium [4]. The final reaction
mixture contained 2-OH-Pr2þ as a major product, and propidium
as a minor product. 2-OH-Pr2þ was purified by HPLC on a
Xanthine oxidase (XO) from cow milk was purchased from Roche
Diagnostics GmbH. Hydroethidine (HE) and mito-hydroethidine
(Mito-HE or MitoSOX Red) were purchased from Invitrogen. Super-
oxide dismutase (bovine erythrocytes), ferricytochrome c (equine
heart), propidium iodide, hypoxanthine (HX), phorbol 12-myristate
13-acetate (PMA), potassium nitrosodisulfonate (Fremy’s salt), chlor-
anil, menadione bisulfite sodium salt (MN), hydrogen peroxide,
K3FeCN6, KH2PO4, K2PO4, and FeSO4 were purchased at the highest
available purity from Sigma Aldrich. Trifluoroacetic acid was pur-
chased from Thermo Scientific. Catalase (beef liver) was purchased
from Boehringer Mannheim. Peroxynitrite was prepared by reacting
nitrite with H2O2, according to the previously published procedure
[12]. 5-Tert-butoxycarbonyl-5-methyl-1-pyrroline N-oxide (BMPO)
was synthesized as described previously [13].
semipreparative C18 column (Beckmann Ultrasphere, 250
ꢂ
10 mm, 5 m) using a gradient of acetonitrile/water containing
m
0.1% trifluoroacetic acid (TFA). Dipropidium (Pr2þ-Pr2þ) was
prepared by oxidizing HPrþ with excess potassium ferricyanide
[14]. The reaction mixture contained dipropidium as a major
product, and other oxidation products as minor products. Dipro-
pidium was purified by HPLC under the same conditions as those
used in the purification of 2-OH-Pr2þ
.

R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
137
UV-Vis absorption and fluorescence measurements
and from 60 to 100% over the next 2 min, and kept at this level for
2.5 min. All analytes, in both separation methods, were eluted at a
flow rate of 1.5 ml/min. Fluorescence detection was carried out
using an excitation wavelength of 490 nm and an emission
wavelength of 567 and 596 nm. The absorption traces were
collected at 220, 242, 290, 370, and 500 nm.
The absorption spectra were recorded on an Agilent 8453
photodiode array spectrophotometer equipped with thermo-
stated cell holder. Fluorescence excitation/emission matrix
(FEEM) spectra were collected using a Perkin-Elmer LS-55 lumi-
nescence spectrometer.
Mass spectral analyses
HPLC measurements
The structural identity of products was confirmed by MS
analysis using two different mass spectrometers equipped with
electrospray ion sources. The MS spectra were recorded on an
Agilent 6460 and an IonSpec 7.0 T FT-ICR for higher resolution.
The compounds were dissolved in water/acetonitrile (1/1 v/v)
mixture containing 0.1% TFA by volume, and their concentration
HPrþ and its oxidation products were separated and moni-
tored by HPLC using an Agilent 1100 apparatus equipped with an
UV-Vis absorption and fluorescence detector. Typically, 50
sample was injected on C18 column (Phenomenex, Kinetex, 100 ꢂ
4.6 mm, 2.6 m) equilibrated with acetonitrile/water mobile
ml of a
m
phase (10/90 v/v) containing 0.1% TFA. Compounds were sepa-
rated by a linear increase of the acetonitrile concentration from
10 to 50% over 5 min. Next, the concentration of acetonitrile was
increased to 100% over 2 min and kept at this level for 2.5 min.
HE, Mito-HE, and their oxidation products were separated using a
higher initial concentration of acetonitrile phase, as described
previously [5,6,8]. The column was equilibrated with acetonitrile/
water mobile phase (20/80 v/v) containing 0.1% TFA. The concen-
tration of acetonitrile was increased from 20 to 60% over 5 min,
was in the range of 10–50 mM. The samples were directly injected
into the MS spectrometer without HPLC separation. The com-
pounds were detected in the positive-ion mode. The charges of
the molecular ions were calculated from the intervals between
isotopic peaks. In the case of Pr2þ-Pr2þ, ion cluster formation
with trifluoroacetate anion on MS analysis was observed. Results
showed an excellent agreement between experimental and cal-
culated MS spectral data. HRMS was also performed on a QStar
Elite (Applied Biosystems SCIEX) with API as a ionization source.
Fig. 2. Spectroscopic properties of hydropropidine and its oxidation products. (A) The UV-Vis absorption spectra of HPrþ (10
(10 M) (green) in phosphate buffer (50 mM, pH 7.4) containing 0.1 mM DTPA. (B) The fluorescence excitation (dashed line) and emission
M) (red), and Pr2þ-Pr2þ (10
spectra (solid line) of HPrþ (1 M), Pr2þ (10 M), 2-OH-Pr2þ (10 M), and Pr2þ-Pr2þ (10
M). (C) The fluorescence excitation (dashed line) and emission spectra (solid
line) of Pr2þ (1 M), 2-OH-Pr2þ (1 M), and Pr2þ-Pr2þ (10 M) in the presence of DNA (0.1 mg/ml). (D) The FEEM spectrum of 2-OH-Pr2þ (10
M) in the presence of DNA
(0.1 mg/ml). (E) The FEEM spectrum of Pr2þ (10
M) in the presence of DNA (0.1 mg/ml). The arrow marked in (C) and (D) indicates an additional excitation band for
2-OH-Pr2þ
lmax¼391 nm). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
m m
M) (black), Pr2þ (10 M) (blue), 2-OH-Pr2þ
m
m
m
m
m
m
m
m
m
m
m
(

138
R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
NMR analyses
self-dismutation of superoxide radical anion, even in the absence
of the competitor. In the competition kinetic approach, the
concentration of specific hydroxylation product of the probes
1H NMR and 13C NMR spectra were recorded at 400.13 and
75.54 MHz, respectively, using a Bruker DPX AVANCE 400 spec-
ꢀ
ꢁ
(2-OH-Xþ) depends on the rate constants of the reaction of O2
trometer equipped with a QNP probe. Chemical shifts (
d
) are
with both the competitor and the probes investigated (reactions
(1) and (2)).
reported in ppm and coupling constants (J) values in Hertz.
Assignments of 1H and 13C NMR signals of the compounds were
made with the help of the APT (attached proton test), HSQC
(heteronuclear single quantum correlation), and HMBC (hetero-
nuclear multiple bond correlation) sequences. The NMR spectral
analyses of compounds are given in the Supplemental Section
(Suppl. Figs. 1–4).
kProbe
Probe ðXÞþOꢀ2ꢁꢁꢁ!2-OH-X þ
ð1Þ
kComp
CompetitorþOꢀ2ꢁꢁꢁ!Product
ð2Þ
During the reaction there were no significant changes in the
concentration of the probes as well as the competitor. Thus, the
competition model assuming two pseudo-first-order reactions
can be applied, and the following equation can be written:
EPR measurements/kinetics
The electron paramagnetic resonance (EPR) spectra were
recorded on a Bruker EMX spectrometer at 9.85 GHz at room
temperature. Typical spectrometer parameters were scan range,
60 G; sweep time, 42 s; time constant, 1.28 s; modulation ampli-
tude, 1 G; modulation frequency, 100 kHz; microwave power,
5.0 mW. The competition kinetic approach was used to determine
½2-OH-X þ ꢄ0ꢁ½2-OH-Xþ ꢄ
½2-OH-Xþ ꢄ
kComp½Competitorꢄ
kProbe½Probeꢄ
¼
ð3Þ
where [2-OH-Xþ
]
is the concentration of 2-OH-Xþ in the
0
absence of a competitor (SOD). Two equations that fit a nonlinear
(Eq. (4)) and linear (Eq. (5)) relationship were used:
ꢀ
ꢁ
the rate constants of O2 reaction with fluorescent probes.
Superoxide dismutase (SOD) was used as a competitor in kinetic
experiments. The concentration of the probes was kept constant
throughout the experiment at the appropriate level to minimize
Â
Ã
kProbe½Probeꢄ
2-OH-X þ ¼ ½2-OH-X þ ꢄ0 ꢂ
ð4Þ
kProbe½ProbeꢄþkComp½Competitorꢄ
ꢀ
ꢁ
Fig. 3. HPLC and mass spectral analyses of products formed from O2 and Fremy’s salt-mediated oxidation of HPrþ. (A) HPLC traces of HPrþ (10
reaction mixtures containing Fremy’s salt and HPrþ or HPrþ and hypoxanthine/xanthine oxidase with and without SOD and catalase. Prior to HPLC analyses, HPrþ
(10 M), hypoxanthine (HX) (0.1 mM), and xanthine oxidase (XO) (10 mU/ml) were incubated in phosphate buffer (50 mM, pH 7.4) containing 0.1 mM DTPA for 30 min.
Where indicated, incubations contained SOD (0.1 mg/ml) or catalase (0.1 kU/ml). The incubation mixture containing HPrþ (10
M) and NDS (20 M) was analyzed
m mM),
M), Pr2þ (10
m
m
m
immediately after mixing the components. The HPLC traces were recorded at 290 nm. The HPLC traces were scaled by a factor 0.1 and 0.2, respectively, so as to fit the
absorbances of Pr2þ and HPrþ/Fremy’s salt product on the same scale. (B) The mass spectrum of HPrþ. (C) The mass spectrum of Pr2þ. (D) The mass spectrum of 2-OH-
Pr2þ. Arrows in (B), (C), and (D) indicate isotopic peaks.

R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
139
analyzed again to confirm that the level of 2-OH-Xþ was
unchanged over the time period of HPLC analysis. The second
control sample without XO was used to check for self-oxidation of
the probes during the course of the experiment. That sample was
analyzed at the end of experiment and appropriate corrections
were applied. Each rate constant was determined on the basis of
three independent experiments. All data points shown in Fig. 6
represent mean values, and error bars indicate standard devia-
tions. The molar concentration of SOD in the experiments was
determined by UV-Vis spectrophotometry using an extinction
coefficient of 1.03ꢁꢂ 104 Mꢁ1 cmꢁ1 at 258 nm.
ꢀ
The flux of O2 was determined before each experiment by
monitoring the ferricytochrome c reduction following an increase
in absorbance at 550 nm. For this purpose, a difference in the
values of the extinction coefficient between oxidized and reduced
forms of ferricytochrome c equal to 2.1 ꢂ 104 Mꢁ1 cmꢁ1 was used.
On this basis, the concenꢁtration of XO was adjusted to obtain the
ꢀ
flux of 0.2
m
M/min of O2 during the kinetic experiments [15].
Cell culture experiments
DMEM (Invitrogen) medium, supplemented with 10% heat-
inactivated FBS, 2 mM L-glutamine, 100 units/ml penicillin, and
100 mg/ml streptomycin, was used to grow the RAW 264.7 cells
(ATCC). Prior to experimentation, cells were washed three times
using DPBS supplemented with pyruvic acid and glucose (DPBS-
GP). Cells were then incubated with PMA (1
(100
M) in the presence of HPrþ in DPBS-GP. During the
incubation period, cells were stored at 37 1C in a CO2-free
incubator. After incubation, an aliquot of the medium (100 l)
mM) or menadione
m
m
was collected and immediately frozen in liquid nitrogen. The cells
were washed twice with ice-cold DPBS, scraped with 1 ml of
DPBS, transferred into 1.5 ml tube, and centrifuged (1 min,
1000g). After centrifugation, the remaining supernatant was
discarded and the cell pellets were frozen in liquid nitrogen.
The cell pellets and media were stored at -80 1C until the day
of HPLC analysis, and were processed according to the procedure
described previously [8,16]. Briefly, to 100
medium, a 0.2 M ice-cold solution of HClO4 in MeOH was added
(100 l) and the samples were vortexed for 10 s and stored on ice
for several minutes. In the next step, samples were centrifuged for
30 min (20,000g at 4 1C), and 100 l aliquots of the supernatant
ml of cell lysate or
m
m
were transferred to the tubes containing 1 M phosphate buffer
(pH 2.6). Again, the tubes were quickly vortexed and centrifuged
for 15 min (20,000g at 4 1C). Aliquots of 150 ml of supernatants
were used in HPLC analyses.
Results
Fig. 4. 1H NMR spectra of (A) hydropropidine (HPrþ), (B) 2-hydroxypropidium
(2-OH-Pr2þ), (C) propidium (Pr2þ), and (D) dipropidium (Pr2þ-Pr2þ), as measured
in DMSO-d6.
Spectroscopic properties of HPrþ and its oxidation products
Fig. 2 shows the UV-visible absorption and fluorescence
spectra of HPrþ and its oxidation products. HPrþ exhibits three
absorption maxima in the range below 400 nm (Fig. 2A) whereas
the products, 2-OH-Pr2þ, Pr2þ and Pr2þ-Pr2þ have a strong
absorption band in the UV range with an additional absorption
band in the visible region between 400 and 600 nm (Fig. 2A).
Fig. 2B shows the fluorescence excitation and emission spectra.
A comparison between the excitation and the emission spectra of
HPrþ and 2-OH-Pr2þ shows that HPrþ does not interfere with
kComp½Competitorꢄ
1
1
1
¼
þ
ꢂ
ð5Þ
½2-OH-Xþ ꢄ ½2-OH-Xþ ꢄ0 ½2-OH-Xþ ꢄ0
kProbe½Probeꢄ
ꢀ
ꢁ
The flux of O2 was generated by xanthine oxidase-catalyzed
oxidation of hypoxanthine. The samples containing hypoxanthine
(0.2 mM), XO (E1 mU/ml), the fluorogenic probe (40 or 50
and different concentrations of SOD were incubated in phosphate
buffer (pH 7.4, 50 mM) containing 100 M DTPA for 30 min at
mM),
the fluorescence of 2-OH-Pr2þ
. However, the spectrum of
m
Pr2þexhibits a significant overlap with 2-OH-Pr2þ. Thus, one
has to use the HPLC method in order to distinguish these oxidation
products. Both 2-OH-Pr2þ and Pr2þ display a strong fluorescence
intensity above 500 nm, but much like 2-hydroxyethidum and
room temperature. SOD (0.1 mg/ml final concentration) was then
added to stop the competition experiment, and the samples were
placed in an HPLC autosampler cooled to 6 1C and analyzed by
HPLC. After HPLC analysis of all the samples, the first sample was

140
R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
ethidum, the fluorescence intensity of 2-OH-Pr2þ in this region is
higher than that of Pr2þ. The fluorescence intensity of 2-OH-Pr2þ
and Pr2þ increases significantly on binding to DNA (Fig. 2C).
The FEEM spectra of 2-OH-Pr2þ and Pr2þ in the presence of DNA
are shown (Figs. 2D and 2E). Of interest is the observation that
2-OH-Pr2þ, but not Pr2þ, has an additional excitation band below
400 nm which is clearly seen in the presence of DNA (indicated by
an arrow in Figs. 2C and 2D). Table 1 lists the relevant absorption
and fluorescence parameters of HPrþ and its oxidation products.
The spectral properties of HPrþ and its oxidation products are very
similar to those described for HE and its oxidation products [5,6,8].
comparison with authentic standards of 2-OH-Pr2þand Pr2þ. As
shown in Fig. 3A, a new product eluting at 3.16 min was detected
under these conditions. The peak corresponding to the new
product was abolished by SOD and not by catalase, suggesting
ꢀ
that O2ꢁ, and not H2O2 or H2O2-derived oxidant, reacts with
HPrþ toꢁyield this new product (Fig. 3A). The product derived
from O2 reaction with HPrþ eluted at the same time as the
ꢀ
authentic standard, 2-OH-Pr2þ
, formed from Fremy’s salt-
mediated hydroxylation of HPrþ(Figs. 3A and D). The structures
of the oxidation products were determined by mass spectral and
NMR analyses, and by comparison with authentic standards
(Figs. 3C and D). Table 2 lists the mass spectral parameters for
HPrþ and its oxidation products.
Identification of the product of the reaction between superoxide
radical anion and HPrþ
The NMR spectral analyses of the hydroxylated product
ꢀ
ꢁ
formed from HPrþ and O2 reaction indicate that the hydroxyl
group is attached at position 2 and not at position 9. Figs. 4A–D
ꢀ
ꢁ
The reaction between HPrþ and O2 was investigated in
an incubation mixture containing hypoxanthine and xanthine
oxidase as a source of steady flux of superoxide at pH 7.4. The
identity of the product was confirmed by HPLC analysis by
show the NMR spectra of HPrþ, 2-OH-Pr2þ, Pr2þ, and Pr2þ-Pr2þ
.
Table 3 lists the chemical shifts and proton coupling constants of
HPrþ and its oxidation and hydroxylation products, Pr2þ, 2-OH-
Pr2þ, and Pr2þ-Pr2þ. Both NMR and MS data are consistent with
the proposed structures (Table 2 and supplementary Figs. 1–4).
Fremy’s salt-dependent oxidation of HPrþ: Reaction stoichiometry
Table 1
Absorption maxima max), extinction coefficients
(
l
(e), fluorescence excitation
(
lexc), and emission (l
em) maxima in the presence and absence of DNA for HPrþ
Fremy’s salt dissociates into nitrosodisulfonate radical dianion
(NDS) in water (Fig. 1) and has been used in the syntheses of
2-hydroxylated phenols and aromatic amines [17]. We have
previously used the Fremy’s salt to prepare 2-OH-Eþ, the product
and its oxidation products.
Compound UV-Vis absorption
Fluorescence
Without DNA
e
With DNA
ꢀꢁ
lmax
[nm]
of HE/O2 reaction [4]. We followed the same approach to
[Mꢁ1 cmꢁ1
]
prepare 2-OH-Pr2þ, the product of HPrþ and O2 reaction. As
NDS is a relatively stable nitroxide radical at pH 7.4, we moni-
tored the reaction between HPrþ and NDS by EPR [4]. Concomi-
tantly, the changes in HPrþ and 2-OH-Pr2þ concentration were
monitored by HPLC. Fig. 5A (top) shows the three-line EPR
ꢀ
ꢁ
lexc
[nm]
lem
lexc
lem
[nm]
[nm]
[nm]
HPrþ
264
350
258
350
479
403
1.15 ꢂ 104
1.1 ꢂ 104
2-OH-Pr2þ 269
483
Pr2þ
597
391
508
533
575
605
607a
spectrum of NDS (8
mM) in phosphate buffer. In the presence of
HPrþ (4
mM), the intensity of the three-line signal was greatly
288
495
485
614
5.2 ꢂ 103
diminished (Fig. 5A, bottom). Fig. 5B shows the changes in the
EPR signal intensity of NDS as a function of its initial concentra-
Pr2þ-Pr2þ 290
480a
618a
529a
ꢅ1.2 ꢂ 104
513
tion in the absence and presence of HPrþ (4
m
M). As shown, in the
M HPrþ the EPR signal intensity began to increase
mM NDS (Fig. 5B, solid squares). The EPR results indicate
The fluorescence intensity of Pr2þ-Pr2þ is at least 10 ꢂ lower than that of
a
presence of 4
m
Pr2þ and 2-OH-Pr2þ
.
above 8
Table 2
Mass spectral data for HPrþ and its oxidation products.
Compound
Molecular ion
HPrþ
Ionic charge
1
Distance between
Calc. masses
Calc. intensities
(%)
Exp. masses
Exp. intensities
(%)
isotopic peaks (m/z)
(m/z)
(m/z)
Hydropropidine (HPrþ
)
1
415.2862
416.2895
417.2929
100.0
30.0
4.3
415.2877
416.2907
417.2970
100.0
29.6
5.8
Propidium (Pr2þ
)
Pr2þ
2
0.5
207.1392
207.6409
208.1426
100.0
30.0
4.3
207.1392
207.6410
208.1425
100.0
28.5
5.0
2-Hydroxypropidium
(2-OH-Pr2þ
2-OH-Pr2þ
2
1
0.5
1
215.1367
215.6383
216.1400
429.2654
430.2688
431.2721
100.0
30.0
4.3
215.1361
215.6378
216.1393
429.2649
430.2676
431.2784
100.0
28.4
3.8
)
2-OH-Pr2þ(-Hþ
)
100.0
30.0
4.3
100.0
28.5
8.0
Dipropidium (Pr2þ-Pr2þ
)
Pr2þ-Pr2þ(-Hþ
)
3
2
0.33
0.5
275.1777
275.5122
275.8466
276.1811
526.2556
526.7573
527.2589
527.7606
100.0
60.1
17.7
3.4
275.1775
275.5125
275.8469
276.1814
526.2563
526.7580
527.2606
527.7593
100.0
55.7
13.0
2.4
Pr2þ-Pr2þ (þ2 ꢂ CF3COOꢁ
)
100.0
64.5
20.4
4.2
100.0
62.0
20.1
5.5

R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
141
Table 3
1H NMR chemical shifts ( , ppm) and coupling constants (J, Hz) of HPrþ and its oxidation products.a
d
Product
Chemical shifts (ppm) and coupling constants (Hz)
H1
H2
H4
H7
H9
H10
H6
HPrþ
7.29 (J¼8.8)
8.01
6.03 (J¼8.5)
6.01
6.34 (J¼2.3)
6.25 (J¼2.3)
6.26 (J¼2.5)
6.34 (J¼2.1)
6.47 (J¼8.3, 2.3)
7.52 (J¼9.0, 2.5)
7.54 (J¼9.3, 2.3)
7.54 (J¼9.0, 2.1)
7.32 (J¼8.3)
8.35 (J¼9.3)
8.64 (J¼9.3)
8.70 (J¼9.0)
5.34
2-OH-Pr2þ
Pr2þ
–
7.54
–
–
–
8.70 (J¼9.3)
8.69
7.36 (J¼9.3, 1.5)
7.56 (J¼1.5)
7.73 (J¼6.5)
Pr2þ-Pr2þ
–
a
Atom positions are indicated in structures shown in Fig. 4.
ꢀ
ꢁ
Kinetics of O2 reaction with HPrþ
ꢀ
ꢁ
The rate constant of the reaction between O2 and HPrþ was
determined by a competition kinetic technique in the presence
and absence of the enzyme bovine Cu, Zn-SOD, as discussed under
Materials and methods. Fig. 6A shows the HPLC traces of 2-OH-
Pr2þ formed from incubating HPrþ in the presence of hypox-
anthine and xanthine oxidase and different concentrations of SOD
for 30 min at room temperature. With increasing SOD concentra-
tion, the HPLC peak intensity due toꢁ 2-OH-Pr2þ decreased
ꢀ
(Fig. 6A). The rate constant between O2 and HPrþ was calcu-
lated from the nonlinear graph (Eq. (4)) as shown (Fig. 6D). The
curve fitted using a linear equation (Eq. (5)) is also shown (Fig. 6D,
inset). Fig. 6B shows the effect of SOD on 2-OH-Eþ formed from
incubating HE and HX/XO. The kinetic analysis of these data is
shown in Fig. 6E. Similar experiments were carried out for Mito-
HE, and the results are shown in Figs. 6C and F. Table 4 lisꢁts the
ꢀ
rate constants determined for the reaction between the O2 and
the fluorogenic probes (HE, HPrþ, and Mito-HE).
Oxidation of HPrþ by one-electron oxidants
Next, we determined the oxidation products formed from the
reaction between HPrþ and other biologically relevant oxidants
and compared with the products formed from reacting these
probes with ferricyanide (one-electron oxidant) and chloranil
(two-electron oxidant). Oxidants tested included peroxynitrite,
ferricytochrome c, horseradish peroxidase/H2O2 system, Fenton’s
reagent, and hydrogen peroxide. The two-electron oxidant (chlor-
anil) and one-electron oxidizing agent (ferricyanide anion) were
used in order to understand the reaction mechanisms of product
formation.
The HPLC traces obtained from the incubation mixtures con-
taining HPrþ and the various oxidants are shown in Fig. 7.
Incubation of HPrþ with chloranil leads primarily to Pr2þ. In
the presence of ferricyanide anion we detected several other
products in addition to Pr2þ. These products were ascribed to
various dimers of HPrþ (Fig. 7). A similar set of products was also
observed in the presence of peroxynitrite, ferricytochrome c,
Fenton’s reagent, and horseradish peroxidase/H2O2. Incubation
of HPrþ with hydrogen peroxide alone did not yield a significant
amount of the aforementioned reaction products. Lack of a
reaction between H2O2 and HPrþ is in agreement with results
obtained from hypoxanthine/xanthine oxidase system (Fig. 3B).
In the case of peroxynitrite, ferricytochrome c, and Fenton’s
reagent, small amounts of 2-OH-Pr2þ were formed during pro-
longed incubation. As explained in earlier publications, this small
increase in SOD-sensitive 2-OH-Prþ formation could be due to
DTPA-derived radicals [10].
Fig. 5. The stoichiometry of the reaction between HPrþ and Fremy’s salt. (A) The
EPR spectrum of NDS (8 M) in phosphate buffer (50 mM, pH 7.4) containing
DTPA (0.1 mM) in the absence (top) and presence (bottom) of HPrþ (4
M).
(B) The double-integrated EPR signal intensity of NDS as function of its
concentration in the absence (open circles) and presence (solid squares) of HPrþ
(4
M). (C) HPLC peak areas of HPrþ (open circles) and 2-OH-Pr2þ (solid squares)
m
m
a
m
as a function of NDS concentration. HPLC absorption traces were recorded at
290 nm. The peak areas of 2-OH-Pr2þ were scaled by a factor of 0.2 to fit on the
same scale as that of HPrþ peak area. The final concentration of HPrþ was 42
mM.
that two molecules of NDS react with one molecule of HPrþ. Next,
we investigated the reaction stoichiometry by monitoring the
HPLC peak areas of HPrþ and the product (which was determined
to be 2-OH-Pr2þ) in the presence of NDS (Fig. 5C). With increasing
NDS concentration, the peak area due to HPrþ decreased, and that
of 2-OH-Pr2þ increased, until the stoichiometry between HPrþ
and NDS reached approximately 1:2. The HPLC results also
suggest that one molecule of HPrþ reacts with two NDS radicals
to produce one molecule of 2-OH-Pr2þ. At a higher concentration
of NDS, HPrþ was completely consumed and NDS reacted with
2-OH-Pr2þ, leading to a decrease in the peak area corresponding
to this product.
The HPLC-based titration of HPrþ with ferricyanide anion
indicates formation of at least three dimeric oxidation products
of HPrþ (Fig. 8). At lower concentrations of ferricyanide, the
homodimer (HPrþ-HPrþ) was predominantly observed (Fig. 8A
and B). With increasing ferricyanide concentration, one of the

142
R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
ꢀ
ꢁ
Fig. 6. The competition kinetic data for the reaction between O2 and fluorogenic probes. (A) HPLC traces obtained from a mixture containing HPrþ (40
xanthine oxidase, DTPA, and SOD (where indicated), incubated in phosphate buffer (50 mM, pH 7.4) for 30 min. Prior to HPLC measurements, SOD (0.1 mg/ml) was added to
stop further oxidation of the probe. (B) Same as (A) except in the presence of HE (50 M). (C) Same as (A) except that Mito-HE (40 M) was used. (D) The dependence of the
peak area of 2-OH-Pr2þ on the concentration of SOD. The solid line trace represents the result of nonlinear fitting (Eq. (4)) to data points for 2-OH-Pr2þ using HPrþ(40
M) as a
mM), hypoxanthine/
m
m
m
probe. (Inset) Results of fitting the experimental data using the linear relationship (Eq. (5)). (E) Same as (D) except that HE was used. (F) Same as (E) except that Mito-HE
(MitoSOX Red) was included. HPLC traces were collected using a fluorescence detector with an emission wavelength at 595 nm and an excitation wavelength at 485 nm.
rings is oxidized to form a heterodimer (HPrþ-Pr2þ). The ESI-MS
analysis of product obtained from oxidation of HPrþ at a much
higher ferricyanide concentration revealed a homodimeric struc-
ture, i.e., dipropidium (Pr2þ-Pr2þ) (Fig. 8C). The structure of
dipropidium was confirmed by 1H and 13C NMR analyses
(Fig. 4D and Suppl. Fig. 4).
Table 4
Rate constants for the reactions among HPrþ, HE,
ꢀ
ꢁ
Mito-HE, and O2 as determined from the compe-
tition kinetics with SOD.
Probes
SODa
HPrþ
HE
(1.1970.05) ꢂ 104
M
ꢁ1 sꢁ1
ꢁ1 sꢁ1
ꢁ1 sꢁ1
Relative intracellular uptake of HE and HPrþ
(6.270.8) ꢂ 103
(1.670.1) ꢂ 104
M
M
Mito-HE
Because of the localized positive charge on the nitrogen atom
in HPrþ, we surmised that HPrþ will not penetrate into the cell.
k¼2 ꢂ 109
M .
ꢁ1 sꢁ1
a

R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
143
Fig. 7. HPLC profiles of the products of the reaction of HPrþ with various oxidants.
Incubations contained HPrþ (10
M) and 0.1 mM DTPA in phosphate buffer
(50 mM, pH 7.4) and various oxidants as indicated: chloranil (5 M), ferricyanide
anion (20 M), peroxynitrite (20 M), ferricytochrome (5 M), horseradish
peroxidase (HRP, 5 mU/ml)/H2O2 (5 M) and
M), Fenton’s reagent [Fe2þ(20
m
m
m
m
c
m
m
m
H2O2(1 mM)], or H2O2 alone (1 mM). With ferricyanide anion, peroxynitrite, and
HRP/H2O2 system, incubation mixtures were analyzed immediately after mixing
the components. With respect to chloranil, ferricytochrome, the Fenton’s reagent
or H2O2-dependent oxidation of HPrþ
, HPLC analysis was performed after
incubation at room temperature for 15–60 min. In the case of ferricytochrome,
the reaction was stopped by mixing (1:1) with an ice-cold solution of 0.2 M HClO4
in MeOH and the supernatant was diluted with phosphate buffer 1 M (pH 2.6).
HPLC traces were recorded at 290 nm. Because of differences in the intensities of
absorption peaks, the scale was set individually for each HPLC trace, as shown on
the figure.
This is in contrast to MitoSOX Red (or Mito-HE) where the
positive charge is delocalized over the aromatic rings and is
sequestered into mitochondria [11]. RAW macrophages were
treated with HE or HPrþ for 1 h, and their concentrations in cell
lysates and cell culture media were monitored by HPLC with
fluorescence detection. Results show that the cell lysate to cell
culture medium ratio for HE is significantly higher than that of
HPrþ (Fig. 9A). However, the reverse was true for HPrþ (in that a
significantly larger amount of HPrþ remained in the medium as
compared to cell lysate) (Fig. 9B).
Fig. 8. Oxidation of HPrþ by ferricyanide anion. (A) HPLC traces obtained after
incubating 40
the presence of 5, 45, 80, and 120
m
M HPrþ and 0.1 mM DTPA in phosphate buffer (50 mM, pH 7.4) in
mM potassium ferricyanide. HPLC traces were
recorded at 290 nm. (B) The profiles of different oxidation products with varying
concentrations of ferricyanide. (C) Mass spectrum of Pr2þ-Pr2þ dimer.
ꢀ
ꢁ
Measurement of extracellular O2 using hydropropidine
ꢀ
ꢁ
We used two sources of O2 generation in cells. RAW 264.7
macrophages were stimulated with PMA or treated with mena-
ꢀ
ꢁ
dione (Fig. 1), a redox-active quinone that generates O2 via
redox cycling of the semiquinone radical derived from mena-
dione. RAW macrophages were activated with PMA (1
mM) in
DPBS supplemented with pyruvic acid and glucose containing
HPrþ probe in the presence and absence of SOD and incubated for
1 h at 37 1C. The diagnostic marker product of HPrþ/O2 reaction
was subsequently monitored by HPLC in cell lysates and in the
media. As shown in Fig. 10A, 2-OH-Pr2þ formation was stimu-
lated by 3- to 4-fold in the extracellular compartment of macro-
phages stimulated with PMA. This increase in 2-OH-Pr2þ
formation was abrogated by SOD. A negligible amount of 2-OH-
ꢀ
ꢁ
Fig. 9. The cellular uptake of HE and HPrþ on incubation of the probes with
macrophages RAW 264.7. The cells were incubated with 10 m
M HE or HPrþ for 1 h
in cell culture media. Samples were then prepared according to the procedure
described under Materials and methods and analyzed by HPLC.

144
R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
Fig. 10. Measurement of extracellular superoxide radical anion production using the HPrþ probe. (A) RAW macrophages were stimulated with PMA (1
mM) in cell culture
media containing HPrþ (60 M) (as described under Materials and methods) in the presence and absence of SOD (0.1 mg/ml) for 1 h and 2-OH-Pr2þ levels were measured
m
in cell culture media (extracellular) and cell lysates (intracellular). To directly compare the amount of 2-OH-Pr2þ inside the cell or media, the amount of the product was
normalized to the total amount of protein in the cell lysates. (B) Time-course measurements of extracellular 2-OH-Pr2þ under different experimental conditions, as
indicated. Experimental conditions: PMA (1
m
M), HPrþ (60
m
M), and SOD (0.1 mg/ml). (C) Same as (A) except that superoxide production was stimulated by adding
M) instead of PMA.
menadione (MN, 100 M) and incubated for 30 min. (D) Same as (B) except that macrophages were treated with menadione (MN, 100
m
m
ꢀ
ꢁ
Pr2þ was detected in macrophage cell lysates. Fig. 10B shows the
time course of 2-OH-Pr2þ formation from the media isolated from
macrophages stimulated with and without PMA and SOD.
with O2 to ultimately form the corresponding hydroxylated
products (2-OH-Eþ, 2-OH-Pr2þ, and 2-OH-Mito-Eþ) (Fig. 11B).
Contrary to a previous assertion that ethidium is the major
product of HE oxidation [19], the hydroxylated products were
Addition of menadione to macrophages in the presence of
HPrþ induced 42-fold increase in 2-OH-Pr2þ formation in the
media (Fig. 10C). Time-course measurements also showed the
same trend (Fig. 10D). This increase in 2-OH-Pr2þ was abrogated
by SOD. A negligible amount of 2-OH-Pr2þ was detected in the
cell lysates as compared to the medium (Figs. 10A and C). These
results indicate that HPrþ ꢁis a suitable probe for detecting
detected as the only major product at various fluxes of O2ꢁwith
no significant change in the two-electron oxidation product, Eþ
ꢀ
.
Treatment with two-electron oxidation agent, chloranil, yielded
Eþ, Pr2þ, and Mito-Eþ from HE, HPrþ, and Mito-HE (Fig. 7) [5,6].
In the presence of other oxidants (peroxynitrite-derived radicals,
cytochrome c, and higher oxidants from peroxidase/H2O2 reac-
tion) that oxidize HE, HPrþ, and Mito-HE via a one-electron
transfer mechanism, various reduced homodimers (HE-HE,
ꢀ
extracellularly generated O2
.
HPrþ-HPrþ
, , -
Mito-HE-Mito-HE), heterodimers (HE-Eþ HPrþ
Pr2þ Mito-HE-Mito-Eþ), and oxidized homodimers (Eþ-Eþ
,
,
Discussion
Pr2þ-Pr2þ, Mito-Eþ-Mito-Eþ) are formed. The similarity in the
redox chemistry of these probes, combined with their different
cellular localization, potentially opens up new possibilities to
simultaneously monitor O2 generation in different cellular
locales.
We report here the development of a novel fluorogenic probe,
HPrþ, which is specific for measuring extracellular superoxide and
other one-electron oxidation species in cell-free and cell-based
ꢀ
ꢁ
systems. Reaction between O2 and HPrþ led to the identification
of a characteristic marker product, 2-hydroxypropidium. In the
presence of one-electron oxidizing species (peroxynitrite-derived
radicals, peroxidases), the corresponding propidium dication and
homo- and heterodimeric products were formed. The reaction
ꢀ
ꢁ
Discrepancies in the rate constants
ꢀ
ꢁ
The rate constant for O2 reaction with HE that is reported in
this work is more than an order of magnitude lower than the
previously reported value [1]. We attribute theꢁ higher rate
ꢀ
chemistry among HPrþ, O2ꢁ, and one-electron oxidants is analo-
gous to that of hydroethidine, a well-known probe for tracking
ꢀ
intracellular O2ꢁ. On the basis of similarity of HPrþ to HE and the
ꢀ
constant value previously reported for the HE/O2 reaction to
reaction stoichiometry of HPrþ with NDS (Fig. 5), we assume that
ꢀ
ꢁ
the higher literature rate constant value (for BMPO and O2
)
ꢀ
ꢁ
the reaction stoichiometry between HPrþ and O2 is 1:2. The
[20,21] that we used in the competition kinetic analysis [1].
ꢀ
ꢁ
hypothetical mechanism is shown in Fig. 11A.
We have reevaluated the BMPO/O2 rate constant using the
competition kinetic approach as described under Materials anꢁd
ꢀ
Similarities in redox chemistry among HE, Mito-HE, and HPrþ
methods (Fig. 12). The new rate constant for the BMPO and O2
reaction was calculated to be ca. 13 Mꢁ1 sꢁ1. In theꢁpulse
ꢀ
ꢁ
ꢀ
Results indicate that the reaction chemistry between O2 and
reduced phenanthridinium compounds investigated in this study
are identical [18]. HE, HPrþ, and Mito-HE (or MitoSOX Red) react
radiolysis work [10], the reaction between HE and O2 was
determined in a 1/1 ethanol/water mixture. In such a mixture, it
ꢀ
is likely that the pKa of the hydroperoxyl radical (HO2, the

R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
145
ꢀ
ꢀꢁ
Fig. 11. (A) Proposed mechanism of reaction of HPrþ with O2ꢁ. (B) Scheme showing the oxidation of HE, HPrþ, and Mito-HE probes by O2 to the hydroxylated
fluorescent product and by one-electron oxidants with the formation of fluorescent, nonhydroxylated cation and nonfluorescent dimeric products. For the sake of clarity,
the structures of other intermediate dimeric products are not shown.
protonated form of superoxide) is higher than in water. Assuming
H2O2 [28]. Nox5 is activaꢁted by elevated cytosolic calcium and
that HO2 reacts with HE much faster than O2ꢁ, this may explain
the higher reactivity of superoxide in such a solvent mixture, as
compared to water alone. Nevertheless, in this work (Table 2) we
have rectified the pꢁreviously reported incorrect rate constant
generates extracellular O2 [29]. The actual identity and site(s) of
ꢀ
ꢀ
ꢀ
ROS generation from Nox isoforms remain an active area of
investigation. Development of site-specific fluorogenic probes
will help in the determination of the identity, the source, and
the location of oxidants generated by Nox isoforms in cells
exposed to proinflammatory conditions.
ꢀ
values for the HE/O2 reaction [1,10].
ꢀ
ꢁ
Implications in the detection of cellular O2 in biological systems
Comparison of HPrþ with the existing probes
The NADPH oxidase (Nox) family (Nox 1–5, DUOX1 and 2) is
one of the major biological sources of oxidant generation [22,23].
Nox enzymes (Nox 1–3, and 5) catalyꢁze the NADPH-dependent
In addition to the artifact-prone assays employing probes
capable of self-generation of superoxide (e.g., lucigenin and
luminol), two other probes used for measurement of extracellular
superoxide radical anion include ferricytochrome c and spin traps
(e.g., BMPO, DEPMPO) [30]. The major advantage of both of those
probes is that the product is formed in a single step, which
minimizes the possibility of interference with assay.
However, the rate constant of the reaction of nitrone spin traps
with superoxide is very low and the resulting nitroxide spin adduct
is unstable. Also, there is a limited access to EPR instrumentation in
most biochemical labs. In the case of a ferricytochrome-based
method, the major disadvantage for cellular assays is the mode of
detection (spectrophotometric assay), which not only limits the
sensitivity of the assay but also dictates the need for time-
dependent probing the extracellular medium by transferring an
ꢀ
one-electron reduction of oxygen to O2 [22,23]. The only known
function of Nox enzymes is to generate ROS, and thꢁe function of
ꢀ
this enzyme is typically assessed by measuring O2 generation
[24]. Currently available assays for determining Nox activity use
nonspecific and artifact-prone probes for ROS [25,26]. The most
investigated Nox member is Nox2 that is expressed abundantly in
neutrophils and macrophages [27]. This enzyme complex consists
of different protein components that are located in the cytosolic
and membranous compartments [28]. On Rac1 or Rac2 activation,
the various components (e.g., gp91phox, p22phox, p67phox,
p47phox, and p40phox) are assembled at the plasma membrane
to form a functionally active NADPH oxidase which transfers
ꢀ
ꢁ
electrons from NADPH to oxygen, producing either O2 and/or

146
R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
stable product on reaction with superoxide and the rate constant is
at least three orders of magnitude higher than most nitrone spin
traps. The HPLC-based detection of the specific product can be
accompanied by fluorescence-based real time monitoring in cellular
systems to obtain both a reliable oxidant identification and the
dynamics of oxidant production.
Conclusions
We conclude that 2-OH-Pr2þ and the dimeric productsꢁ(e.g.,
ꢀ
Pr2þ-Pr2þ) derived from HPrþ are reliable indicators of O2 and
one-electron oxidizing species formed in an extracellular milieu.
Acknowledgments
This work was supported by grants R01 HL063119 and R01
NS039958 from the National Institutes of Health. The authors
thank Mrs. Monika Zielonka for her help in culturing of RAW
264.7 cells.
Appendix A. Supporting information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.freeradbiomed.2012.
09.018.
References
[1] Zhao, H.; Kalivendi, S.; Zhang, H.; Joseph, J.; Nithipatikom, K.; Va´squez-Vivar, J.;
Kalyanaraman, B. Superoxide reacts with hydroethidine but forms a fluorescent
product that is distinctly different from ethidium: potential implications in
intracellular fluorescence detection of superoxide. Free Radic. Biol. Med.
34:1359–1368; 2003.
[2] Zhao, H.; Joseph, J.; Fales, H.M.; Sokoloski, E.A.; Levine, R.L.; Va´squez-Vivar,
J.;Kalyanaraman, B. Detection and characterization of the product of hydro-
ethidine and intracellular superoxide by HPLC and limitations of fluores-
cence. Proc. Natl. Acad. Sci. USA 102:5727-5732; 2005. [Erratum: Proc. Natl.
Acad. Sci. USA 102:9086; 2005.].
[3] Dikalov, S.; Griendling, K. E.; Harrison, D. G. Measurement of reactive oxygen
species in cardiovascular studies. Hypertension 49:717–727; 2007.
[4] Zielonka, J.; Zhao, H.; Xu, Y.; Kalyanaraman, B. Mechanistic similarities
between oxidation at hydroethidine by Fremy’s salt and superoxide:
stopped-flow optical and EPR studies. Free Radic. Biol. Med. 39:853–863;
2005.
[5] Zielonka, J.; Hardy, M.; Kalyanaraman, B. HPLC study of oxidation products of
hydroethidine in chemical and biological systems: ramifications in super-
oxide measurements. Free Radic. Biol. Med. 46:329–338; 2009.
[6] Zielonka, J.; Va´squez-Vivar, J.; Kalyanaraman, B. Detection of 2-hydroxyethidium
in cellular systems: a unique marker product of superoxide and hydroethidine.
Nat. Protoc. 3:8–21; 2008.
[7] Palazzolo-Balance, A. M.; Suquet, C.; Hurst, J. K. Pathways for intracellular
generation of oxidants and tyrosine nitration by a macrophage cell line.
Biochemistry 46:3536–3548; 2007.
[8] Zielonka, J.; Zielonka, M.; Sikora, A.; Adamus, J.; Hardy, M.; Ouari, O.; Dranka, B. P.;
Kalyanaraman, B. Global profiling of reactive oxygen and nitrogen species in
biological systems: High-throughput real-time analyses. J. Biol. Chem.
287:2984–2995; 2011.
[9] Moore, A.; Donahue, C. J.; Bauer, K. D.; Mather, J. P. Simultaneous measure-
ment of cell cycle and apoptotic cell death. Methods Cell Biol. 57:265–278;
1998.
[10] Zielonka, J.; Sarna, T.; Roberts, J. E.; Wishart, J. F.; Kalyanaraman, B. Pulse
radiolysis and steady-state analyses of the reaction between hydroethidine
and superoxide and other oxidants. Arch. Biochem. Biophys. 456:39–47; 2006.
[11] Robinson, K. M.; Janes, M. S.; Pehar, M.; Monette, J. S.; Ross, M. F.; Hagen, T. M.;
Murphy, M. P.; Beckman, J. S. Selective fluorescent imaging of superoxide
in vivo using ethidium-based probes. Proc. Natl. Acad. Sci. USA 103:
15038–15043; 2006.
ꢀ
ꢁ
Fig. 12. Reevaluation of the kinetic data for the reaction between O2 and BMPO.
ꢀ
(A) The EPR spectra of BMPO- OOH adduct obtained from a mixture containing
ꢀ
ꢁ
BMPO (0.2 M), hypoxanthine/xanthine oxidase (O2 flux of 0.8 mM/min), 0.1 mM
DTPA, and SOD (where indicated) in phosphate buffer (pH, 7.4, 50 mM) after a 10-
min incubation at room temperature. The spectra shown are average of four
ꢀ
consecutive scans. (B) Changes in the signal height of BMPO- OOH spin adduct
with time (initial slopes of the fitted lines were used in the kinetic calculations).
(C) The dependence of a rate of spin adduct formation on the concentration
of SOD.
aliquot to spectrophotometer for the measurement in the case of
adherent cells. Also, the propensity of cytochrome c to undergo
superoxide-independent reduction (e.g., by flavoproteins, redox-
cycling intermediates, thiol compounds) limits the applicability of
the probe. On the other hand, hydropropidine gives unique and
[12] Kissner, R.; Beckman, J. S.; Koppenol, W. H. Peroxynitrite studied by stopped-
flow spectroscopy. Methods Enzymol. 301:342–352; 1999.
[13] Zhao, H.; Joseph, J.; Zhang, H.; Karoui, H.; Kalyanaraman, B. Synthesis and
biochemical applications of a solid cyclic nitrone spin trap: a relatively
superior trap for detecting superoxide anions and glutathiyl radicals. Free
Radic. Biol. Med. 31:599–606; 2001.

R. Michalski et al. / Free Radical Biology and Medicine 54 (2013) 135–147
147
[14] Zielonka, J.; Srinivasan, S.; Hardy, M.; Ouari, O.; Lopez, M.; Vasquez-Vivar, J.;
Avadhani, N. G.; Kalyanaraman, B. Cytochrome c-mediated oxidation of
hydroethidine and mito-hydroethidine in mitochondria: identification of
homo- and heterodimers. Free Radic. Biol. Med. 44:835–846; 2008.
[15] Sikora, A.; Zielonka, J.; Lopez, M.; Joseph, J.; Kalyanaraman, B. Direct
oxidation of boronates by peroxynitrite: Mechanism and implications in
fluorescence imaging of peroxynitrite. Free Radic. Biol. Med. 47:1401–1407;
2009.
[16] Dranka, B. P.; Zielonka, J.; Kathansamy, A. G.; Kalyanaraman, B. Alterations in
bioenergetics function induced by Parkinson’s disease mimetic compounds:
lack of correlation with superoxide generation. J. Neurochem. 122:941–951;
2012.
[17] Zimmer, H.; Lankin, D. C.; Horgan, S. W. Oxidations with potassium nitrosodi-
sulfonate (Fremy’s radical). The Teuber reaction. Chem. Rev. 71:229–246; 1971.
[18] Zielonka, J.; Kalyanaraman, B. Hydroethidine- and Mito-SOX-derived red
fluorescence is not a reliable indicator of intracellular superoxide formation:
Another inconvenient truth. Free Radic. Biol. Med. 48:983–1001; 2010.
[19] Kundo, K.; Knight, S. F.; Lee, S.; Taylor, W. R.; Murthy, N. A significant
improvement of the efficacy of radical oxidant probe by the kinetic isotope
effect. Angew Chem. Int. Ed. Engl. 49:6134–6138; 2010.
[20] Tsai, P.; Ichikawa, K.; Mailer, C.; Pou, S.; Halpern, H. J.; Robinson, B. H.;
Nielson, R.; Rosen, G. M. Esters of 5-carboxyl-5-methyl-1-pyrroline N-oxide:
a family of spin traps for superoxide. J. Org. Chem. 68:7811–7817; 2003.
[21] Rosen, G. M.; Tsai, P.; Weaver, J.; Porasuphatana, S.; Roman, L. J.; Starkov, A.
A.; Fiskum, G.; Pou, S. The role of tetrahydrobiopterin in the regulation of
neuronal nitric-oxide synthase-generated superoxide. J. Biol. Chem.
277:40275–40280; 2002.
[22] Lambeth, J. D. Nox enzymes and the biology of reactive oxygen. Nat. Rev.
Immunol. 4:181–189; 2004.
[23] Leto, T. L.; Morand, S.; Hurt, D.; Ueyama, T. Targeting and regulation of
reactive oxygen species generation by Nox family NADPH oxidases. Antioxid.
Redox Signal. 11:2607–2619; 2009.
[24] Bedard, K.; Krause, K. H. The NOX family of ROS-generating NADPH oxidases:
physiology and pathophysiology. Physiol. Rev. 87:245–313; 2007.
[25] Ambasta, R. K.; Schreiber, J. G.; Janiszewski, M.; Busse, R.; Brandes, R. P.
Noxa1 is a central component of the smooth muscle NADPH oxidase in mice.
Free Radic. Biol. Med. 41:193–201; 2006.
[26] Pietrowski, E.; Bender, B.; Huppert, J.; White, R.; Luhmann, H. J.; Kuhlmann, C.
R. Proinflammatory effects of interleukin-17A on vascular smooth muscle
cells involve NAD(PH)-oxidase derived reactive oxygen species. J. Vasc. Res.
48:52–58; 2011.
[27] Sareila, O.; Kelkka, T.; Pizzolla, A.; Hultqvist, M.; Holmdahl, R. Nox2 complex-
derived ROS as immune regulators. Antioxid. Redox Signal. 15:2197–2208;
2011.
[28] Streeter, J.; Thiel, W.; Brieger, K.; Miller, F.J. Jr. Opportunity Nox: the future of
NADPH oxidases as therapeutic targets in cardiovascular disease. Cardiovasc.
Ther.; 2012. 10.1111/j.1755-5922.2011.00310.x. [Epub ahead of print].
[29] Bedard, K.; Jaquet, V.; Krause, K. H. NOX5: from basic biology to signaling and
disease. Free Radic. Biol. Med. 52:725–734; 2012.
[30] Zielonka, J.; Kalyanaraman, B. Methods of investigation of selected radical
oxygen/nitrogen species in cell-free and cellular systems. in: Pantopoulos, K.,
Schipper, H. M., editors. Principles of free radical biomedicine, vol. 1. Nova
Science Publishers; 2012.