Singlet oxygen generation by the reaction of acrolein with peroxynitrite via a 2-hydroxyvinyl radical intermediate

Article abstract of DOI:10.1016/j.freeradbiomed.2020.03.003

Acrolein (2-propenal) is an environmental pollutant, food contaminant, and endogenous toxic by-product formed in the thermal decomposition and peroxidation of lipids, proteins, and carbohydrates. Like other α,β-unsaturated aldehydes, acrolein undergoes Michael addition of nucleophiles such as basic amino acids residues of proteins and nucleobases, triggering aging associated disorders. Here, we show that acrolein is also a potential target of the potent biological oxidant, nitrosating and nitrating agent peroxynitrite. In vitro studies revealed the occurrence of 1,4-addition of peroxynitrite (k₂ = 6 × 10³ M^?1 s^?1, pH 7.2, 25 °C) to acrolein in air-equilibrated phosphate buffer. This is attested by acrolein concentration-dependent oxygen uptake, peroxynitrite consumption, and generation of formaldehyde and glyoxal as final products. These products are predicted to be originated from the Russell termination of ^?OOCH=CH(OH) radical which also includes molecular oxygen at the singlet delta state (O₂ ¹Δ_g). Accordingly, EPR spin trapping studies with the 2,6-nitrosobenzene-4-sulfonate ion (DBNBS) revealed a 6-line spectrum attributable to the 2-hydroxyvinyl radical adduct. Singlet oxygen was identified by its characteristic monomolecular IR emission at 1,270 nm in deuterated buffer, which was expectedly quenched upon addition of water and sodium azide. These data represent the first report on singlet oxygen creation from a vinylperoxyl radical, previously reported for alkyl- and formylperoxyl radicals, and may contribute to better understand the adverse acrolein behavior in vivo.

Full text of DOI:10.1016/j.freeradbiomed.2020.03.003

Journal Pre-proof
Singlet oxygen generation by the reaction of acrolein with peroxynitrite via a 2-
hydroxyvinyl radical intermediate
Leticia C.P. Gonçalves, Júlio Massari, Saymon Licciardi, Fernanda M. Prado, Edlaine
Linares, Aline Klassen, Marina F.M. Tavares, Ohara Augusto, Paolo Di Mascio,
Etelvino J.H. Bechara
PII:
S0891-5849(20)30058-7
DOI:
https://doi.org/10.1016/j.freeradbiomed.2020.03.003
FRB 14628
Reference:
To appear in: Free Radical Biology and Medicine
Received Date: 9 January 2020
Revised Date: 26 February 2020
Accepted Date: 2 March 2020
Please cite this article as: L.C.P. Gonçalves, Jú. Massari, S. Licciardi, F.M. Prado, E. Linares, A.
Klassen, M.F.M. Tavares, O. Augusto, P. Di Mascio, E.J.H. Bechara, Singlet oxygen generation by the
reaction of acrolein with peroxynitrite via a 2-hydroxyvinyl radical intermediate, Free Radical Biology and
Medicine (2020), doi: https://doi.org/10.1016/j.freeradbiomed.2020.03.003.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Inc.

Singlet oxygen generation by the reaction of acrolein with
peroxynitrite via a 2-hydroxyvinyl radical intermediate
Leticia C. P. Gonçalves,1 Júlio Massari,1 Saymon Licciardi,1,2 Fernanda M. Prado,3
Edlaine Linares,3 Aline Klassen,1 Marina F. M. Tavares,1 Ohara Augusto,3 Paolo Di
Mascio,3 Etelvino J. H. Bechara*,1,2
1Departamento de Química Fundamental, Instituto de Química, Universidade de São
Paulo, São Paulo, SP, Brazil
2Departamento Ciências Exatas e da Terra, Instituto de Ciências Ambientais, Químicas
e Farmacêuticas, Universidade Federal de São Paulo, Diadema, SP, Brazil
3Departamento de Bioquímica, Instituto de Química, Universidade de São Paulo, São
Paulo, SP, Brazil
1
*Correspondence: Etelvino J. H. Bechara, Departamento de Química Fundamental,
Instituto de Química, Universidade de São Paulo, Av. Prof. Lineu Prestes, 748, 05508-
000, São Paulo, SP, Brazil, ejhbechara@gmail.com
1

Abstract
Acrolein (2-propenal) is an environmental pollutant, food contaminant, and endogenous
toxic by-product formed in the thermal decomposition and peroxidation of lipids,
proteins, and carbohydrates. Like other α,β-unsaturated aldehydes, acrolein undergoes
Michael addition of nucleophiles such as basic amino acids residues of proteins and
nucleobases, triggering aging associated disorders. Here, we show that acrolein is also a
potential target of the potent biological oxidant, nitrosating and nitrating agent
peroxynitrite. In vitro studies revealed the occurrence of 1,4-addition of peroxynitrite
3
-1 -1
⁰
(k2 = 6 × 10 M s , pH 7.2, 25 C) to acrolein in air-equilibrated phosphate buffer. This
is attested by acrolein concentration-dependent oxygen uptake, peroxynitrite
consumption, and generation of formaldehyde and glyoxal as final products. These
products are predicted to be originated from the Russell termination of
•OOCH=CH(OH) radical which also includes molecular oxygen at the singlet delta state
(O2 1∆g). Accordingly, EPR spin trapping studies with the 2,6-nitrosobenzene-4-
sulfonate ion (DBNBS) revealed a 6-line spectrum attributable to the 2-hydroxyvinyl
radical adduct. Singlet oxygen was identified by its characteristic monomolecular IR
emission at 1,270 nm in deuterated buffer, which was expectedly quenched upon
addition of water and sodium azide. These data represent the first report on singlet
oxygen creation from a vinylperoxyl radical, previously reported for alkyl- and
formylperoxyl radicals, and may contribute to better understand the adverse acrolein
behavior in vivo.
Keywords: acrolein, peroxynitrite, Russell reaction, singlet oxygen, carbonyl stress
2

Introduction
Short chain aldehydes (e.g., formaldehyde, acetaldehyde, malondialdehyde, n-hexanal),
2-alkenals (e.g., acrolein, crotonaldehyde, 4-hydroxy-2-nonenal), and α-dicarbonyls
(e.g., glyoxal, methylglyoxal, diacetyl, 4,5-dioxovaleric acid, 3-deoxyglucosone) are
highly electrophilic endogenous species generated by the oxidation and thermal
dehydration of lipids, proteins, and carbohydrates.[1-8] They can also be acquired by
environmental exposure and food ingestion. Once produced in tissues, these reactive
carbonyl catabolites undergo nucleophilic additions of the basic centers of protein
amino acid residues (Cys, His, Lys, Arg) and DNA bases, leading to covalent
modifications, aggregation, denaturation, and even activation of cell signals. When
deleterious to the organism, accumulation of these adducts triggers a condition often
named “carbonyl stress”.[9] Carbonyl adducts have been detected in a plethora of
acquired and innate diseases, among others, atherosclerosis, Alzheimer’s disease,
porphyria diseases, multiple sclerosis, autoimmune diseases, diabetes, food poisoning,
and cancer.[10-17]
The peroxynitrite anion formed by the diffusion-controlled reaction (k2 ~ 1010 M-
1 -1
•
•
¯
s ) between the superoxide anion radical (O2 ) and nitric oxide ( NO) upon oxidative
and nitrosative imbalance has been associated to different diseases.[18-22] Peroxynitrite
not only acts as a one- and two-electron oxidant, a nitrating and nitrosating agent but it
is also a strong nucleophile. Accordingly, the reaction of peroxynitrite with electrophilic
and biologically relevant dicarbonyls, such as glyoxal, methylglyoxal and diacetyl, was
well established. [23-25] In addition, singlet oxygen (O2 1∆g), a potent biological
oxidant, was detected in the reaction of glyoxal with peroxynitrite as a result from
formyl radical annihilation by ground state oxygen and subsequent bimolecular Russell
reaction of the formylperoxyl radical product (Scheme 1A).
3

The 2-alkenal acrolein (2-propenal) is a thermally processed food contaminant
and endogenous toxin (metabolism and lipid peroxidation) implicated in various
disorders including multiple sclerosis, Alzheimer’s disease, spinal cord injury,
cardiovascular disease, diabetes mellitus, and neuro-, hepato- and nephro-toxicity.[10,
26-28] Acrolein is as a powerful and harmful compound which can lead to protein-
acrolein adduct with Arg, Lys, His, and Cys residues of proteins as well as with
nucleobases, thereby disrupting the intracellular redox balance and driving mutagenesis
and carcinogenesis.[11, 29, 30] Recently, it has also been detected in the aerosols of
electronic cigarette devices.[31] The high reactivity of acrolein to nucleophiles lies on
the electronic delocalization along the double and the carbonyl bonds, favoring 1,4-
Michael addition.[30] Here we show that the nucleophilic addition of peroxynitrite to
acrolein in aerated phosphate buffer produces oxygen in the excited singlet state through
the formation of a carbon-centered radical bearing a germinal hydrogen atom which
reacts with molecular oxygen via Russell reaction mechanism (Scheme 1B). If
occurring in cell sites where both peroxynitrite and acrolein increase, singlet oxygen
could potentially exert noxious biological responses.[32]
4

Scheme 1. (A) Reaction mechanisms and final products of the reaction of peroxynitrite
with (A) diacetyl (R1 = R2 = CH3),[23] methylglyoxal (R1 = H, R2 = CH3)[25] and
glyoxal (R1 = R2 = H)[24] in aerated phosphate buffer, pH 7.2. (B) Proposed simplified
mechanism of oxidation of acrolein by peroxynitrite and expected products: (1) formic
1
acid, (2) glyoxal, (3) glycolaldehyde and singlet oxygen (O2 ∆g).
Experimental details
General information
All chemicals were purchased of the purest grade available from Sigma-Aldrich, Fluka
and Merck and used without further purification, unless otherwise stated. Stock
solutions and buffers were prepared using deionized water (18.2 MΩ cm at 25 ºC, TOC
≤ 4 ppb, Milli-Q, Millipore) and pretreated with Chelex-100 to remove metal
contaminants.
5

Preparation of peroxynitrite
Peroxynitrite (300.0 – 450.0 mmol L−1) was prepared from NaNO2 (0.60 mol L−1) and
H2O2 (0.70 mol L−1) in HCl (0.60 mol L−1) and NaOH (1.50 mol L−1) in a quench-flow
reactor as described previously.[33] Remaining H2O2 in the alkaline peroxynitrite
solution was eliminated with the addition of MnO2. Peroxynitrite and H2O2
concentrations were determined spectrophotometrically at 302 nm (ε = 1670 L mol−1
cm−1)[33, 34] and 240 nm (ε = 42 L mol−1 cm−1)[35], respectively. Peroxynitrite was
aliquoted and stored at −80°C. The concentration of an aliquot of peroxynitrite was re-
measured immediately before the experiment and kept on ice in the dark during the
experiments.
Oxygen consumption
The rate of oxygen consumption was monitored in a Hansatech oxygraph equipped with
a Clark type electrode. The measurement was recorded immediately after injection of
the peroxynitrite solution (0.25 − 3.0 mmol L−1) to the air-equilibrated solution of
acrolein (0.50 − 10.0 mmol L−1) in phosphate buffer (500.0 mmol L−1, pH 7.2) at 25°C.
Peroxynitrite depletion
The effect of acrolein (0 − 40 mmol L−1) in the decay rate of peroxynitrite (200 µmol
L−1) was monitored with a stopped-flow spectrophotometer (Applied Photophysics
model SX-18V) at 302 nm. The temperature was kept constant at 25.0 ± 0.2 °C, and the
pH values of the final reaction mixtures were confirmed by a simulated scaled up
experiment under the same experimental conditions. Pseudo-first-order rates, kobs (s−1)
were determined with a liner regression adjust of ln A vs time (0 – 300 ms) plots. The
results are the average of of 9−12 measurements. The apparent second-order rate
constants (k2,app) were determined from the slopes of the plots of kobs vs acrolein
concentration. The actual k2 value was calculated taking into account the degree of
6

acrolein hydration in the buffered solution (93%)[36]. The rate of peroxynitrite decay in
the buffer solution (k0), was subtracted from kobs before plotting the data against pH or
phosphate buffer concentration. All data were fitted using Origin 8 (OriginLab).
EPR spin trapping
The spin trap 3,5-dibromo-4-nitrosobenzenesulfonic acid (DBNBS) was synthesized as
previously described[37] and employed in the experiments using a Bruker EMX EPR
spectrometer. Spectra were recorded after 1.5 min and 5 min of reagents incubation in
phosphate buffer (500.0 mmol L−1, pH 7.2) pretreated with Chelex-100 at 22°C. The
final pH values were measured in order to check for subtle pH changes that might result
from the addition of the alkaline stock solutions of peroxynitrite to the buffered reaction
mixtures. No pH change was observed. The instrumental conditions were as follows:
microwave power, 20 mW; modulation amplitude, 0.1 mT; time constant, 81.92 ms; and
receiver gain, 1 × 10−5. Computer simulation of spectra was performed using the
program P.E.S.T written by Duling.[38]
Detection of singlet oxygen
Reaction mixtures of acrolein (0.05 – 10.0 mmol L−1) and peroxynitrite (1.0 mmol L−1)
in deuterated 500 mM phosphate buffer, pD approximately 7.4, at 25°C, were
monitored in an ultrasensitive photocounting equipment aiming to detect the
monomolecular photoemission of O2(1∆g) in the near-infrared (NIR) region (λ = 1,270
nm), as described previously.[39] All reactions were carried out in a quartz cuvette
under constant stirring at room temperature. The light emission was immediately
recorded by a FLSP 920 photon counter (Edinburgh Instruments, Edinburgh, UK)
consisting of two UV-Visible Hamamatsu detectors R9110, maintained at −20 °C by a
CO1 thermoelectric cooler (Edinburgh Instruments). The monomol light emission of O2
(1∆g) at 1,270 nm was monitored over time using a detector coupled to the device, a
7

Hamamatsu H10330A-45 apparatus (Hamamatsu city, Japan), preceded by a
monochromator. Peroxynitrite was infused to the reaction mixture in the cuvette after 50
s using a syringe with tubing for all experiments. Quenching of the emission intensity of
the reaction of acrolein (0.1 mmol L−1) with peroxynitrite (1 mmol L−1) by water (20%)
and sodium azide (NaN3, 0.125 mmol L−1) confirmed the singlet nature of the molecular
oxygen. The NIR spectrum of the reaction of acrolein (0.1 mmol L−1) with peroxynitrite
(1 mmol L−1) was filtered with a diffraction grate and the signal acquired digitally with
a computer. To determine the O2 (1∆g) yield, the N,N'-di(2,3-dihydroxypropyl)-1,4-
naphthalenedipropanamide endoperoxide (DHPNO2, 10 mmol L−1) was used as
standard. The NIR reference spectrum of O2 (1∆g) at 1,270 nm was produced by thermal
decomposition of DHPNO2 at 37 °C.[40] The presence of hydroquinone as stabilizer in
the Sigma acrolein did not affect the singlet oxygen emission at the concentration range
employed in our experiments.
Product Analysis
The expected aldehydes (formaldehyde and glyoxal) and carboxylic acids (formic and
acetic acids) products formed in the reaction of acrolein with peroxynitrite were
analyzed in a SCIEX PA800 capillary electrophoresis system equipped with a UV filter
wheel detector accommodating a 254 nm filter (10 nm bandwidth) for indirect
detection. A solution of 3,5-dinitrobenzoic acid (10 mmol L−1) containing
cetyltrimethylammonium bromide (0.20 mmol L−1) was employed as background
electrolyte (final pH = 4.5). Fused-silica capillaries with dimensions of 50 cm total
length (40 cm effective lengh) and 75 µm i.d. × 375 µm o.d. were used. Aliquots of the
reaction mixtures obtained by mixing peroxynitrite (1.45 mmol L−1) and acrolein (3
mmol L−1) adjusted to pH 7.2 − 7.5 were injected hydrodynamically at 3 kPa for 3 s and
the separation of products was observed at a constant applied voltage (−20 kV).
8

Aldehydes were previously derivatized with bisulfite to convert them into the respective
sulfonate adducts.[41] Standards of sodium nitrite, sodium nitrate, and carboxylic acids
at 10 mmol L-1 each as well as the derivatized aldehydes were injected separately and as
spiked final reaction mixtures to aid peak identification.
Results and Discussion
Peroxynitrite decay
The second order rate constant, pH profile and phosphate dependence of the
nucleophilic addition of peroxynitrite to acrolein were determined by tracing the fast
decay of peroxynitrite at 302 nm by stopped-flow absorption spectroscopy in air-
equilibrated phosphate buffer (500 mmol L−1, pH 7.2) at 25 °C (Figure 1, Figure S1).
The high phosphate concentration is needed to guarantee no variation in the pH of the
reaction mixture upon addition of the peroxynitrite stock solution containing 1.0 mmol
L−1 NaOH. The pseudo-first order rate constants (kobs) were found to increase linearly
with the initial concentration of acrolein, leading to the apparent second order rate
constant (k2,app) 4.0 × 102 M−1s−1 (Figure 1A). The fraction of hydrated acrolein in
aqueous medium is 93%[36]; hence, the k2 value for the reaction can be estimated as 6.0
× 103 M−1s−1. Therefore, acrolein is ~10-fold more reactive than monoalkanals like
ethanal, propanal, and isobutanal (~300 – 700 M−1s−1)[42, 43] and one order of
-
magnitude less reactive than HCO3 /CO2 (3 − 6 × 104 M−1s−1)[44]. Yet, its reactivity is
close to reactivity of diacetyl (k2 = 1.0 × 104 M−1s−1)[23]. These values are in
accordance with the expected relative electrophilicity of the carbonyl reactants: carbon
dioxide>glyoxals>acrolein>monoaldehydes.
The well-known general acid catalysis occurrence by the dihydrogen phosphate
anion in the peroxynitrite addition to carbonyls[23, 25, 45] was verified for the reaction
9

of peroxynitrite with acrolein (Figure 1B). The kobs-k0 values increase linearly approx.
8-fold with the increase of the phosphate concentration (50 − 1000 mM, pH 7.2) at 25
°C, being k0 related to the decomposition of peroxynitrite alone. The spontaneous
decomposition rate of peroxynitrite was not significantly affected by the phosphate
concentration. The bell-shaped pH profile of the peroxynitrite/acrolein reaction has a
maximum at pH ~ 6.8 in the range 6.2 to 8.2 (Figure 1C), and this result corroborates
the proposed phosphate catalysis. It roughly mirrors the overlap of the titration curve of
the peroxynitrous acid providing nucleophilic peroxynitrite anion (pKa
ONOOH/ONOO− = 6.8, increasing to over 7.0 at higher ionic strength)[23, 33] with
−
increased dissociation and loss of catalytic H2PO4 . All together these data are related to
those reported by Yang et al.[45] for the general-acid catalyzed addition of peroxynitrite
to diacetyl and other carbonyls. The bell-shaped pH profile and the linear phosphate
concentration dependence of the reaction rate points to a kinetic equation expressed as
rate = k2[acrolein][H2PO4 ][ONOO−]
−
10

Figure 1. Stopped-flow kinetics of the peroxynitrite decay in the presence of acrolein.
(A) Pseudo-first-order rate constants (kobs) for the decay of peroxynitrite (200 µmol L−1)
in the presence of acrolein (0 − 40 mmol L−1) in phosphate buffer (500 mmol L−1, pH
7.2). Calculation of the apparent second-order rate constant of peroxynitrite addition to
acrolein (k2,app). (B) Catalytic effect of phosphate anion on the observed rate constant of
peroxynitrite (200 µmol L−1) reaction with acrolein (20 mmol L−1) in phosphate buffer
pH 7.2. The ionic strength was corrected upon addition of NaCl. (C) pH profile of the
11

reaction of peroxynitrite (200 µmol L−1) with acrolein (20 mmol L−1) in phosphate
buffer (700 mmol L−1). All experiments were performed at 25 °C.
EPR spin trapping studies
EPR spin trapping studies with 10 mM DBNBS were performed in phosphate
buffer (500 mmol L−1, pH 7.2) treated with Chelex-100. The reaction was initiated by
the addition of peroxynitrite to the reaction mixture containing acrolein, DBNBS and
buffer. After 1.5 min of incubation at 22 °C, somewhat distorted six lines spectra were
recorded (Figure 2) and the values of the hyperfine coupling constants were estimated
from spectral analysis and simulation (aN = 13.33 G and the aH = 6.45 G). A six-line
⦁
spectrum is expected for a DBNBS radical adduct of the CH=CH(OH) hydroxyvinyl
radical, a possible intermediate in the mechanism of acrolein oxidation by peroxynitrite
(Scheme 1B). The formation of this radical was previously reported in gas phase studies
of the reaction of the hydroxyl radical with acrolein.[46]
The EPR signal was intensified with the increase in the concentration of acrolein
(Figure 2A). Conversely, addition of the bicarbonate/CO2 pair to the reaction mixture
decreased the EPR signal intensity (Figure 2B), indicating competition of carbon
dioxide with acrolein for the peroxynitrite anion. Control experiments in the absence of
either acrolein or peroxynitrite did not give significant response, as well as hydrogen
peroxide addition to acrolein (Figure S2). It should be noted that the lines of the EPR
spectra are distorted and broad, suggesting superposition of two radical adducts, most
likely the cis and trans forms of the 2-hydroxyvinyl radical. Indeed, attempts to split the
lines of the EPR spectra by changing the instrumental conditions (smaller modulation
amplitudes) failed (data not shown).
12

Figure 2. EPR spin trapping studies with DBNBS (10 mmol L−1) of the
acrolein/peroxynitrite reaction in phosphate buffer (500 mmol L−1, pH 7.2) after
incubation for 1.5 min at 22 °C. (A) Spectra of peroxynitrite (1 mmol L−1) with 5 (a), 10
(b), and 20 mmol L−1 acrolein (c). Computer simulation in (d). (B) Effect of carbonate
on the amplitude of the EPR signal of the adduct obtained from acrolein (10 mmol L−1)
and peroxynitrite (1 mmol L−1) without (a) and with 5 (b), 10 (c) and 25 mmol L−1
sodium bicarbonate (d).
Oxygen uptake
Addition of peroxynitrite (0 − 3 mmol L−1) to acrolein (0 − 10 mmol L−1) in phosphate
buffer (500 mmol L−1, pH 7.2) at 25 °C resulted in the consumption of a fraction of the
dissolved oxygen in less than one minute (Figure 3). Dissolved oxygen concentration
values were estimated assuming that the oxygen concentration in 500 mmol L−1
phosphate buffer is near that reported for 500 mmol L−1 NaCl at 25 °C, i.e., ~212 µmol
L−1.[47]. Increasing the acrolein concentration to 10 mmol L−1 in the presence of 1
mmol L−1 peroxynitrite, increased oxygen uptake to a saturation value of ~80 µmol L−1
13

oxygen (Figure 3A) whereas proportionally less oxygen was consumed with equal
concentrations of peroxynitrite and acrolein (3 mmol L−1) 170 µmol L−1. The non-linear
response of oxygen uptake upon increasing the concentration of acrolein may be due to
the well-known tendency of oxygen-exposed acrolein to undergo polymerization
(Figure 3A-3C). In addition, the Clark electrode used to measure has a low response
time, which may be response for distortions in the data. The effect of the concentration
of acrolein and peroxynitrite on the rate of O2 uptake was determined by the initial rate
fit of each curve shown in the Figure 3A-3B. The plot of rough initial rate vs.
concentration of acrolein and peroxynitrite (Figure 3C) may confirm a first order
dependence of oxygen consumption on the peroxynitrite concentration whereas do not
respond linearly to increasing acrolein concentration probably due to competing
acrolein polymerization.
14

Figure 3. Time courses of oxygen uptake during the reaction of peroxynitrite with
acrolein in air-equilibrated phosphate buffer (500 mmol L−1, pH 7.2) at 25 °C. (A)
Peroxynitrite (1 mmol L−1) in the absence of acrolein (a) and in the presence of 0.50 (b),
1.00 (c), 2.00 (d), 3.00 (e), 5.00 (f) and 10.00 mmol L−1 acrolein (g). (B) Acrolein (3
mmol L−1) in the absence of peroxynitrite (a) and in the presence of 0.25 (b), 0.50 (c),
15

1.00 (d) and 3.00 mmol L−1 peroxynitrite (e). (C) Effect of the concentration of acrolein
(black square) or peroxynitrite (blue circle) on the rate of O2 uptake (nmol/mL/s).
Assuming the 1,4-Michael addition of peroxyntrite to acrolein as previously
reported for alkanals and α-dicarbonyls, the carbon-centered radical originated by C1-
C2 acrolein cleavage, ⦁CH=CH(OH) (Scheme 1B), is expected to add molecular oxygen
yielding the 2-hydroxyvinylperoxyl radical, [CH(OO⦁)=C(H)OH]. The latter radical is
expected to undergo the bimolecular Russell reaction to form glycolaldehyde
[CH2(OH)-C(O)H], glyoxal [C(O)H-C(O)H] and excited molecular oxygen O2 (1∆g)
state ranging 3 to 14% yields [48]. Therefore, considering the stoichiometry of
peroxynitrite to consumed oxygen as 1.0 to 0.5, the fraction of peroxynitrite that reacted
with acrolein in Figure 3A is minimally 8% when calculated over initial 10 mM
acrolein at the experimental conditions stated in the figure legend. Parallel isomerization
of peroxynitrite to the nitrate, reformation of acrolein and dimerization of the
⦁CH=CH(OH) radical pari passu with oxygen depletion may contribute to the calculated
low yield of oxygen consumption.
Product analysis
Formaldehyde, glycolaldehyde, and glyoxal are expected to be formed upon the 1,4-
Michael addition of peroxynitrite to acrolein. Conversely, any unexpected contribution
of peroxynitrite addition to the aldehyde group would result in formate and vinyl
radical, whose products after oxygen addition would include acetate and acetaldehyde.
Electrophoresis separation and quantification of the final products generated by the
peroxynitrite/acrolein reaction were conducted to verify which final products were
formed. Aldehyde products are electrophoretically silent; thus, final reaction mixtures
were pre-derivatized with sodium bisulfite to the corresponding active sulfonate anions.
Authentic formaldehyde and glyoxal were used to attest the putative products (Figure
16

4A-B) whereas formate and acetate were used as reference compounds to rule out
peroxynitrite addition to the carbonyl group (Figure S3). In order to detect any expected
•
increase of the ratio nitrite/nitrate originated by lateral formation of the NO2 radical
during homolyzes of the peroxonitrous carbonyl intermediate, nitrite and nitrate peaks
were also monitored in the reference electropherogram. These anions are contaminants
of the peroxynitrite solutions and by-products formed upon hydrolysis of •NO2.
17

Figure 4. Analysis of the reaction products of the reaction of acrolein (3 mmol L−1)
with peroxynitrite (1.45 mmol L−1) in pH 7.2-7.5. (A) Electropherograms of
formaldehyde (a), glyoxal (b), and sodium bisulfite (c). (B) Electropherograms of the
final reaction mixtures doped glyoxal and formaldehyde (a), the final reaction mixtures
doped glyoxal (b) and the final reaction mixture without addition of reference
compounds (c). Reaction products were previously derivatized with sodium bisulfite.
Peak identification: (1) nitrite, (2) nitrate, (3) glyoxal, (4) bisulfite, (5) formaldehyde,
(6) phosphate, (*) system peak. Standard concentration: 10 mmol L−1. Experimental
conditions: see methods.
18

The electropherograms of the final reaction mixture in comparison to that
obtained from the standard mixture of analytes (Figure 4A-B) clearly show the
presence of formaldehyde and glyoxal, two of the final products of the 1,4-Michael
addition of peroxynitrite to acrolein. Accordingly, the electropherograms fingered the
absence of the formate and acetate ions, expected to be formed from peroxynitrite
addition to C1-acrolein (Figure S3).
Detection of singlet oxygen
Singlet oxygen exert deleterious effects to biomolecules and supramolecular cell
structures due to its electrophilicity and diversity of chemical targets (such as double
bonds, sulfur atoms, and heterocyclic rings).[32] To examine singlet oxygen O2 (1∆g)
generation during the oxidation of acrolein by peroxynitrite, we monitored its
characteristic monomolecular emission in the NIR region (λ= 1,270 nm)[49] running the
reaction in deuterated phosphate buffer, pD 7.4 (Figure 5). Addition of peroxynitrite to
acrolein showed a clear light emission that did not appear in the absence of any of the
reagents. Indeed, he light emission attributed to the minor conversion of peroxynitrite to
peroxynitrate, whose decomposition yields very low yields of singlet oxygen was not
observed under our experimental conditions (Figure 5A).[50] On the other hand,
addition of small amounts of either water or sodium azide to the reaction mixture in
D2O resulted in quenching of the singlet oxygen NIR emission as expected (Figure
5B).[51] Increments of acrolein concentration decreased the signal (Figure 5C), what
may be attributed to the increase in the water content of commercial acrolein (90%).
The emission spectrum from the acrolein/peroxynitrite system was recorded and
compared to that obtained from an already reported singlet oxygen generator, the
endoperoxide DHPNO2 [40] (Figure 5D). The singlet oxygen yield obtained from 0.1
mmol L−1 acrolein and 1 mmol L−1 peroxynitrite was calculated on the basis of the
19

emission area obtained from the decomposition of DHPNO2 (10 mmol L−1) at 1,270 nm
(Figure S4).[40] The calculated value , 6.1 %, is in range expected for reactions
involving the Russel reaction mechanism (3-14%).[48] Thus, the well-known Russell
reaction by annihilation of geminal hydrogen-lacking radical is here unveiled also to
occur with a vinylperoxyl radical, not only alkylperoxyl and acylperoxyl radicals. [24,
32]
Figure 5. Near infrared analysis at 1,270 nm of singlet oxygen O2 (1∆g) generated by
the acrolein/peroxynitrite system in D2O-prepared phosphate buffer. (A) Light emission
produced in the reaction of peroxynitrite (1.0 mmol L−1) with acrolein (0.1 mmol L−1)
(a), in acrolein alone (b) and peroxynitrite alone (c). (B) Chemical quenching of the
reaction (a) by water (b) and sodium azide (NaN3) (c). (C) Effect of the concentration of
acrolein in the excited oxygen signal. (D) Spectra of the reaction peroxynitrite/acrolein
20

and the standard endoperoxide DHPNO2 (Figure S4). Peroxynitrite was added to the
reaction mixture after 50 s in all experiments.
Conclusions
Current research on the toxic effects of acrolein to the human health focuses essentially
on its irreversible 1,4-Michael attachment to amino and thiolate groups of proteins and
to DNA, disregarding other biological nucleophiles. In this study, we show that acrolein
may act as a potential biological acceptor of peroxynitrite (k2 = 6.0 x 103 M-1s-1 at 25
°C, pH 7.4; optimum pH ~ 6.8) ultimately yielding formaldehyde, glyoxal, and singlet
oxygen. The aldehydes were analyzed by capillary electrophoresis after derivatization
with bisulfite. The absence of formate and acetate anions in the spent reaction mixtures
precluded the expected nucleophilic addition of peroxynitrite to the carbonyl group of
acrolein. The reaction was shown to be catalyzed by the dihydrogen phosphate anion
and to proceed via carbon- and oxygen-centered radicals. It terminates via the Russell
reaction of a secondary peroxyl radical [CH2(OO⦁)-CH(O] or a vinylperoxyl radical
[CH(OO⦁)=CH(OH) yielding molecular oxygen in the excited singlet state (Scheme 2).
Formation of the latter was confirmed in deuterated buffer by the characteristic NIR
monomolecular emission at 1,270 nm and by quenching of the ultra-weak red
chemiluminescence by H2O and NaN3. These data predict production of biologically
active singlet oxygen, albeit at very low yields, from the nucleophilic addition of
peroxynitrite to many putatively toxic alkanals and alkenals.
Although less reactive than CO2 and α-oxoaldehydes like methylglyoxal and
glyoxal,[24] acrolein is about ten-fold more reactive than alkanals and reported to be
more reactive than other α,β-unsaturated alkenals[12]. Despite being found in tissues at
much lower concentrations than the bicarbonate/carbon dioxide pair, one must keep in
mind that in cell compartments, the reactions are governed not only by thermodynamic
21

and kinetic factors but also by spatial and temporal factors.[52] This argument also
applies to the higher concentrations of phosphate in this study (500 mM) as beyond
redox potentials and rate constants, reagent compartmentalization and diffusion rates
also matters. In fact, various protein and DNA adducts of acrolein and of other
carbonyls have been isolated and identified in biological samples from experimental
animals and humans portraying redox unbalance.[53]
Scheme 2. Proposed mechanism for the reaction of acrolein with peroxynitrite yielding
1
singlet oxygen (O2 ∆g).
Acknowledgements
This work was supported by the São Paulo Research Foundation – FAPESP (EJHB,
2017/22501-2; LCPG, 2018/25842-8; OA and PDM, 2013/07937-8, PDM 2012/12663-
1, MFMT 2017/27059-6), the Brazilian National Council for Scientific and
Technological Development – CNPq (EJHB 444237/2014-3, EJHB 306460/2016-5).
FMP, EL, OA and PDM are members of NAP Redoxoma (PRPUSP), and the CEPID
Redoxoma (FAPESP).
References
[1] H. Esterbauer, R.J. Schaur, H. Zollner, Chemistry and biochemistry of 4-
hydroxynonenal, malonaldehyde and related aldehydes, Free Radic Biol Med 11(1)
(1991) 81-128.
[2] J. Cadet, T. Douki, J.-L. Ravanat, Oxidatively generated base damage to cellular
DNA, Free Radic Biol Med 49(1) (2010) 9-21.
22

[3] K.S. Fritz, D.R. Petersen, Exploring the Biology of Lipid Peroxidation-Derived
Protein Carbonylation, Chem Res Toxicol 24(9) (2011) 1411-1419.
[4] L.J. Marnett, J.N. Riggins, J.D. West, Endogenous generation of reactive oxidants
and electrophiles and their reactions with DNA and protein, J Clin Invest 111(5) (2003)
583-593.
[5] M.H.G. Medeiros, Exocyclic DNA Adducts as Biomarkers of Lipid Oxidation and
Predictors of Disease. Challenges in Developing Sensitive and Specific Methods for
Clinical Studies, Chem Res Toxicol 22(3) (2009) 419-425.
[6] A.N. Onyango, Small reactive carbonyl compounds as tissue lipid oxidation
products; and the mechanisms of their formation thereby, Chem Phys Lipids 165(7)
(2012) 777-786.
[7] E.R. Stadtman, R.L. Levine, Free radical-mediated oxidation of free amino acids and
amino acid residues in proteins, Amino Acids 25(3) (2003) 207-218.
[8] P.J. Thornalley, A. Langborg, H.S. Minhas, Formation of glyoxal, methylglyoxal
and 3-deoxyglucosone in the glycation of proteins by glucose, Biochem J 344 Pt 1(Pt 1)
(1999) 109-116.
[9] S. Hoon, M. Gebbia, M. Costanzo, R.W. Davis, G. Giaever, C. Nislow, A Global
Perspective of the Genetic Basis for Carbonyl Stress Resistance, G3 (Bethesda) 1(3)
(2011) 219.
[10] K. Abraham, S. Andres, R. Palavinskas, K. Berg, K.E. Appel, A. Lampen,
Toxicology and risk assessment of acrolein in food, Mol Nutr Food Res 55(9) (2011)
1277-1290.
[11] S. Pizzimenti, E. Ciamporcero, M. Daga, P. Pettazzoni, A. Arcaro, G. Cetrangolo,
R. Minelli, C. Dianzani, A. Lepore, F. Gentile, G. Barrera, Interaction of aldehydes
derived from lipid peroxidation and membrane proteins, Front Physiol 4 (2013) 242-
242.
[12] R. Sultana, M. Perluigi, D.A. Butterfield, Lipid peroxidation triggers
neurodegeneration: a redox proteomics view into the Alzheimer disease brain, Free
Radic Biol Med 62 (2013) 157-169.
[13] P.J. Thornalley, Protein and nucleotide damage by glyoxal and methylglyoxal in
physiological systems--role in ageing and disease, Drug Metabol Drug Interact 23(1-2)
(2008) 125-150.
[14] M. Tully, R. Shi, New insights in the pathogenesis of multiple sclerosis--role of
acrolein in neuronal and myelin damage, Int J Mol Sci 14(10) (2013) 20037-20047.
[15] A. Adhikari, C.A.A. Penatti, R.R. Resende, H. Ulrich, L.R.G. Britto, E.J.H.
Bechara, 5-Aminolevulinate and 4, 5-dioxovalerate ions decrease GABAA receptor
density in neuronal cells, synaptosomes and rat brain, Brain Res 1093(1) (2006) 95-104.
[16] T. Douki, J. Onuki, M.H.G. Medeiros, E.J.H. Bechara, J. Cadet, P. Di Mascio,
DNA Alkylation by 4,5-Dioxovaleric Acid, the Final Oxidation Product of 5-
Aminolevulinic Acid, Cheml Res Toxicol 11(2) (1998) 150-157.
[17] C.O. Soares, M. Boiani, L.J. Marnett, E.J.H. Bechara, Cytotoxicity of 1,4-diamino-
2-butanone, a putrescine analogue, to RKO cells: mechanism and redox imbalance, Free
Radic. Res. 47(9) (2013) 672-682.
[18] G. Ferrer-Sueta, R. Radi, Chemical Biology of Peroxynitrite: Kinetics, Diffusion,
and Radicals, ACS Chem Biol 4(3) (2009) 161-177.
[19] P. Pacher, J.S. Beckman, L. Liaudet, Nitric oxide and peroxynitrite in health and
disease, Physiol Rev 87(1) (2007) 315-424.
[20] C. Szabó, H. Ischiropoulos, R. Radi, Peroxynitrite: biochemistry, pathophysiology
and development of therapeutics, Nat Rev Drug Discov 6 (2007) 662.
23

[21] J.C. Toledo, O. Augusto, Connecting the Chemical and Biological Properties of
Nitric Oxide, Chem Res Toxicol 25(5) (2012) 975-989.
[22] Z.V. Varga, Z. Giricz, L. Liaudet, G. Haskó, P. Ferdinandy, P. Pacher, Interplay of
oxidative, nitrosative/nitrative stress, inflammation, cell death and autophagy in diabetic
cardiomyopathy, Biochim Biophys Acta 1852(2) (2015) 232-242.
[23] J. Massari, D.E. Fujiy, F. Dutra, S.M. Vaz, A.C.O. Costa, G.A. Micke, M.F.M.
Tavares, R. Tokikawa, N.A. Assunção, E.J.H. Bechara, Radical Acetylation of 2′-
Deoxyguanosine and l-Histidine Coupled to the Reaction of Diacetyl with Peroxynitrite
in Aerated Medium, Chem Res Toxicol 21(4) (2008) 879-887.
[24] J. Massari, R. Tokikawa, D.B. Medinas, J.P.F. Angeli, P. Di Mascio, N.A.
Assunção, E.J.H. Bechara, Generation of Singlet Oxygen by the Glyoxal–Peroxynitrite
System, J Am Chem Soc 133(51) (2011) 20761-20768.
[25] J. Massari, R. Tokikawa, L. Zanolli, M.F.M. Tavares, N.A. Assunção, E.J.H.
Bechara, Acetyl Radical Production by the Methylglyoxal−Peroxynitrite System: A
Possible Route for l-Lysine Acetylation, Chem Res Toxicol 23(11) (2010) 1762-1770.
[26] Q. Zhu, Z. Sun, Y. Jiang, F. Chen, M. Wang, Acrolein scavengers: Reactivity,
mechanism and impact on health, Mol Nutr Food Res 55(9) (2011) 1375-1390.
[27] A. Moghe, S. Ghare, B. Lamoreau, M. Mohammad, S. Barve, C. McClain, S.
Joshi-Barve, Molecular mechanisms of acrolein toxicity: relevance to human disease,
Toxicol Sci 143(2) (2015) 242-55.
[28] Y.J. Huang, M.H. Jin, R.B. Pi, J.J. Zhang, Y. Ouyang, X.J. Chao, M.H. Chen, P.Q.
Liu, J.C. Yu, C. Ramassamy, J. Dou, X.H. Chen, Y.M. Jiang, J. Qin, Acrolein induces
Alzheimer's disease-like pathologies in vitro and in vivo, Toxicol Lett 217(3) (2013)
184-91.
[29] H.-J.C. Chen, Analysis of DNA adducts in human samples: Acrolein-derived
exocyclic DNA adducts as an example, Mol Nutr Food Res 55(9) (2011) 1391-1400.
[30] R.M. LoPachin, T. Gavin, D.R. Petersen, D.S. Barber, Molecular Mechanisms of
4-Hydroxy-2-nonenal and Acrolein Toxicity: Nucleophilic Targets and Adduct
Formation, Chem Res Toxicol 22(9) (2009) 1499-1508.
[31] I.G. Gillman, K.A. Kistler, E.W. Stewart, A.R. Paolantonio, Effect of variable
power levels on the yield of total aerosol mass and formation of aldehydes in e-cigarette
aerosols, Regul Toxicol Pharmacol 75 (2016) 58-65.
[32] P. Di Mascio, G.R. Martinez, S. Miyamoto, G.E. Ronsein, M.H.G. Medeiros, J.
Cadet, Singlet Molecular Oxygen Reactions with Nucleic Acids, Lipids, and Proteins,
Chem Rev 119(3) (2019) 2043-2086.
[33] J.S. Beckman, T.W. Beckman, J. Chen, P.A. Marshall, B.A. Freeman, Apparent
hydroxyl radical production by peroxynitrite: implications for endothelial injury from
nitric oxide and superoxide, Proc Natl Acad Sci 87(4) (1990) 1620.
[34] M.N. Hughes, H.G. Nicklin, The chemistry of pernitrites. Part I. Kinetics of
decomposition of pernitrous acid, J Chem Soc A (0) (1968) 450-452.
[35] R.F. Beers, I.W. Sizer, A spectrophotometric method for measuring the breakdown
of hydrogen peroxide by catalase, J Biol Chem 195(1) (1952) 133-140.
[36] S.-Y. Oh, J. Lee, D.K. Cha, P.C. Chiu, Reduction of Acrolein by Elemental Iron:ꢀ
Kinetics, pH Effect, and Detoxification, Environ Sci Technol 40(8) (2006) 2765-2770.
[37] H. Kaur, A Water Soluble C-Nitroso-Aromatic Spin-Trap -3, 5-Dibromo-4-
Nitrosobenzenesulphonic Acid. The Perkins Spin-Trap, Free Rad. Res. 24(6) (1996)
409-420.
[38] D.R. Duling, Simulation of Multiple Isotropic Spin-Trap EPR Spectra, J Magn
Reson B 104(2) (1994) 105-110.
24

[39] C.M. Mano, F.M. Prado, J. Massari, G.E. Ronsein, G.R. Martinez, S. Miyamoto, J.
Cadet, H. Sies, M.H.G. Medeiros, E.J.H. Bechara, P. Di Mascio, Excited singlet
molecular O2 (1∆g) is generated enzymatically from excited carbonyls in the dark, Sci
Rep 4 (2014) 5938.
[40] C. Pierlot, J.-M. Aubry, K. Briviba, H. Sies, P.D. Mascio, [1] Naphthalene
endoperoxides as generators of singlet oxygen in biological media, Methods in
Enzymol., Academic Press2000, pp. 3-20.
[41] E.A. Pereira, M.F.M. Tavares, A.A. Cardoso, Alternative methodologies for the
determination of aldehydes by capillary electrophoresis, J J Aoac Int 82(6) (1999) 1562-
1570.
[42] L.S. Nakao, D. Ouchi, O. Augusto, Oxidation of Acetaldehyde by Peroxynitrite
and Hydrogen Peroxide/Iron(II). Production of Acetate, Formate, and Methyl Radicals,
Chem Res Toxicol 12(10) (1999) 1010-1018.
[43] R.M. Uppu, G.W. Winston, W.A. Pryor, Reactions of Peroxynitrite with
Aldehydes as Probes for the Reactive Intermediates Responsible for Biological
Nitration, Chem Res Toxicol 10(12) (1997) 1331-1337.
[44] A. Denicola, B.A. Freeman, M. Trujillo, R. Radi, Peroxynitrite Reaction with
Carbon Dioxide/Bicarbonate: Kinetics and Influence on Peroxynitrite-Mediated
Oxidations, Arch Biochem Biophys 333(1) (1996) 49-58.
[45] D. Yang, Y.-C. Tang, J. Chen, X.-C. Wang, M.D. Bartberger, K.N. Houk, L.
Olson, Ketone-Catalyzed Decomposition of Peroxynitrite via Dioxirane Intermediates, J
Am Chem Soc 121(51) (1999) 11976-11983.
[46] M.E. Davis, M.K. Gilles, A.R. Ravishankara, J.B. Burkholder, Rate coefficients for
the reaction of OH with (E)-2-pentenal, (E)-2-hexenal, and (E)-2-heptenal, Phys Chem
Chem Phys 9(18) (2007) 2240-2248.
[47] J. Robinson, J.M. Cooper, Method of determining oxygen concentrations in
biological media, suitable for calibration of the oxygen electrode, Anal Biochem 33(2)
(1970) 390-399.
[48] Q.J. Niu, G.D. Mendenhall, Yields of singlet molecular oxygen from peroxyl
radical termination, J Am Chem Soc 114(1) (1992) 165-172.
[49] A.U. Khan, M. Kasha, Chemiluminescence arising from simultaneous transitions in
pairs of singlet oxygen molecules, J Am Chem Soc 92(11) (1970) 3293-3300.
[50] S. Miyamoto, G.E. Ronsein, T.C. Corrêa, G.R. Martinez, M.H.G. Medeiros, P. Di
Mascio, Direct evidence of singlet molecular oxygen generation from peroxynitrate, a
decomposition product of peroxynitrite, Dalton Trans. (29) (2009) 5720-5729.
[51] C.S. Foote, R.W. Denny, L. Weaver, Y. Chang D, J. Peters B, Quenchin of Singlet
Oxygen Annals of the New York Academy of Sciences 171(1) (1970) 139-148.
[52] D.A. Wink, J.B. Mitchell, Chemical biology of nitric oxide: insights into
regulatory, cytotoxic, and cytoprotective mechanisms of nitric oxide, Free Radic Biol
Med 25(4) (1998) 434-456.
[53] S.E. Lee, Y.S. Park, Role of lipid peroxidation-derived α, β-unsaturated aldehydes
in vascular dysfunction, Oxid Med Cell Longev 2013 (2013) 629028-629028.
25

Graphical abstract
26

Highlights
• Acrolein is an endogenous toxicant, also found in thermoprocessed
food.
• In aerated medium acrolein adds peroxynitrite yielding
formaldehyde and glyoxal.
• Phosphate catalyses the reaction accompanied by singlet oxygen
emission.
• EPR spin trapping studies disclose a putative 2-hydroxyvinyl radical.
• Russell reaction of the vinyloxyl radical explains singlet oxygen
generation.