Hypobromous acid, a powerful endogenous electrophile: Experimental and theoretical studies

Article abstract of DOI:10.1016/j.jinorgbio.2015.02.014

Abstract Hypobromous acid (HOBr) is an inorganic acid produced by the oxidation of the bromide anion (Br^-). The blood plasma level of Br^- is more than 1,000-fold lower than that of chloride anion (Cl^-). Consequently, the endogenous production of HOBr is also lower compared to hypochlorous acid (HOCl). Nevertheless, there is much evidence of the deleterious effects of HOBr. From these data, we hypothesized that the reactivity of HOBr could be better associated with its electrophilic strength. Our hypothesis was confirmed, since HOBr was significantly more reactive than HOCl when the oxidability of the studied compounds was not relevant. For instance: anisole (HOBr, k₂ = 2.3 × 10² M^{- 1} s^{- 1}, HOCl non-reactive); dansylglycine (HOBr, k₂ = 7.3 × 10⁶ M^{- 1} s^{- 1}, HOCl, 5.2 × 10² M^{- 1} s^{- 1}); salicylic acid (HOBr, k₂ = 4.0 × 10⁴ M^{- 1} s^{- 1}, non-reactive); 3-hydroxybenzoic acid (HOBr, k₂ = 5.9 × 10⁴ M^{- 1} s^{- 1}, HOCl, k₂ = 1.1 × 10¹ M^{- 1} s^{- 1}); uridine (HOBr, k₂ = 1.3 × 10³ M^{- 1} s^{- 1}, HOCl non-reactive). The compounds 4-bromoanisole and 5-bromouridine were identified as the products of the reactions between HOBr and anisole or uridine, respectively, i.e. typical products of electrophilic substitutions. Together, these results show that, rather than an oxidant, HOBr is a powerful electrophilic reactant. This chemical property was theoretically confirmed by measuring the positive Mulliken and ChelpG charges upon bromine and chlorine. In conclusion, the high electrophilicity of HOBr could be behind its well-established deleterious effects. We propose that HOBr is the most powerful endogenous electrophile.

Full text of DOI:10.1016/j.jinorgbio.2015.02.014

Journal of Inorganic Biochemistry 146 (2015) 61–68
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Hypobromous acid, a powerful endogenous electrophile: Experimental
and theoretical studies
a,
b
a
Valdecir Farias Ximenes ⁎, Nelson Henrique Morgon , Aguinaldo Robinson de Souza
a
Department of Chemistry, Faculty of Sciences, São Paulo State University (UNESP), 17033-360, Bauru, São Paulo, Brazil
Department of Chemistry, Institute of Chemistry, Campinas State University (UNICAMP), 13083-861, Campinas, São Paulo, Brazil
b
a r t i c l e i n f o
a b s t r a c t
Hypobromous acid (HOBr) is an inorganic acid produced by the oxidation of the bromide anion (Br−). The blood
Article history:
Received 27 November 2014
Received in revised form 20 February 2015
Accepted 20 February 2015
Available online 2 March 2015
plasma level of Br− is more than 1,000-fold lower than that of chloride anion (Cl−). Consequently, the endoge-
nous production of HOBr is also lower compared to hypochlorous acid (HOCl). Nevertheless, there is much
evidence of the deleterious effects of HOBr. From these data, we hypothesized that the reactivity of HOBr could
be better associated with its electrophilic strength. Our hypothesis was confirmed, since HOBr was significantly
more reactive than HOCl when the oxidability of the studied compounds was not relevant. For instance: anisole
Keywords:
2
−1 −1
= 7.3 × 106
−1 −1
(
HOBr, k
2
= 2.3 × 10
M
s
, HOCl non-reactive); dansylglycine (HOBr, k
2
M
s
, HOCl,
Reactive electrophilic species
Reactive oxygen species
Hypobromous acid
Hypochlorous acid
Myeloperoxidase
2
−1 −1
4
−1 −1
5.2 × 10
(HOBr, k
M
s
); salicylic acid (HOBr, k
= 5.9 × 10 M
2
= 4.0 × 10
= 1.1 × 10 M
M
s
, non-reactive); 3-hydroxybenzoic acid
4
−1 −1
1
−1 −1
3
−1 −1
2
s
, HOCl, k
2
s
); uridine (HOBr, k
2
= 1.3 × 10 M
s
, HOCl
non-reactive). The compounds 4-bromoanisole and 5-bromouridine were identified as the products of the
reactions between HOBr and anisole or uridine, respectively, i.e. typical products of electrophilic substitutions.
Together, these results show that, rather than an oxidant, HOBr is a powerful electrophilic reactant. This chemical
property was theoretically confirmed by measuring the positive Mulliken and ChelpG charges upon bromine and
chlorine. In conclusion, the high electrophilicity of HOBr could be behind its well-established deleterious effects.
We propose that HOBr is the most powerful endogenous electrophile.
©
2015 Elsevier Inc. All rights reserved.
1
. Introduction
including the killing of pathogens, recruitment of PMNs, apoptosis in-
duction in PMNs and production of regulatory molecules inside
phagosomes of macrophages [10]. Similarly, eosinophil peroxidase
(EPO) is stored in the specific granules of eosinophils, which are special-
ized human phagocytes that play a critical beneficial role to eliminate
tissue-invasive parasites [11]. MPO is able to catalyze the oxidation of
There is much evidence that harmful endogenous chemicals are in-
volved in the aetiology and/or progression of chronic inflammatory
and degenerative diseases [1–3]. One of the most important primary
sources of these chemicals is the one-electron reduction of molecular
oxygen to a superoxide anion radical (O
chemicals include hydrogen peroxide (H
−
●
−
−
2
) [4]. Other reactive
both chloride (Cl ) and bromide (Br ), whereas EPO, at least at
•
−
O
2 2
), hydroxyl radical (HO ),
pH 7.0, is much more effective to Br [12]. This particularity of MPO is
•
−
peroxyl radical (ROO ), peroxynitrite (ONOO ), hypochlorous acid
related to the standard reduction potential (E′°) of the redox couple
compound I/ferric MPO (1.16 V), which is higher compared with EPO
(compound I/ferric EPO, 1.10 V) [13].
(
HOCl) and hypobromous acid (HOBr) [5–9]. Altogether, these sub-
stances are named reactive oxygen or reactive nitrogen species (ROS/
RNS) and their harmful chemicals are associated with the ubiquitous
presence of easily oxidizable moieties in proteins, lipids and nucleic
acids.
Myeloperoxidase (MPO) is an enzyme present in the azurophilic
granules of polymorphonuclear leukocytes (PMNs) and plays a funda-
mental role in the inflammatory and anti-inflammatory processes,
Although the production HOCl and HOBr is an essential part of the
innate immune defence, there is increasing experimental and clinical
studies showing the correlation between hypohalous acid-mediated al-
teration in biomolecules and the progression of inflammatory-based
diseases. Some important findings include the determination of the
levels of 3-chlorotyrosine and 3-bromotyrosine in proteins and the
identification of higher concentrations in the bronchoalveolar lavage
of cystic fibrosis patients [14]; the concentration of 3-chlorotyrosine in
circulating HDL has been shown to be an indicator of the risk of cardio-
vascular disease [15]; the increased level of 3-chlorotyrosine and
3-nitrotyrosine in HDL of rheumatoid arthritis patients [16]; the urinary
⁎
Corresponding author at: Department of Chemistry, Faculty of Science, São Paulo State
University (UNESP), CEP 17033–360, Bauru, São Paulo, Brazil. Tel.: +55 14 31036088;
fax: +55 14 31036071.
E-mail address: vfximenes@fc.unesp.br (V.F. Ximenes).
http://dx.doi.org/10.1016/j.jinorgbio.2015.02.014
0162-0134/© 2015 Elsevier Inc. All rights reserved.

62
V.F. Ximenes et al. / Journal of Inorganic Biochemistry 146 (2015) 61–68
level of 3-bromotyrosine can be used as a biomarker for asthma control
and for prediction of the risk of future asthma exacerbation in children
(kobs) was obtained by fitting the absorbance decay of the nucleoside in
a single exponential decay equation, as follows:
[
17].
Bromine is a trace element in the human body. Indeed, the Br− plas-
−kobsꢀ t
−
F ¼ F0 ꢀ e
ma level (2–100 μM) is more than 1,000-fold lower than that of Cl
(
100–140 mM) [18]. However, there is plenty of evidence of the endog-
enous generation of HOBr and its deleterious effects. For instance, the
detection of 3-bromotyrosine can be as effective as 3-chlorotyrosine as
a biomarker of MPO/EPO involvement in inflammatory disorders [14,
From the kobs values obtained at various nucleoside concentrations,
the bimolecular rate constants (k ) were calculated from the slope of
2
the linear regression as follows:
1
1
7]. Moreover, the reactivity of HOBr with amino acids can be 30- to
00-fold faster than that of HOCl. Particularly for tyrosine, the ring halo-
2
Reaction rate = k *[A]*[B]
genation by HOBr was 5,000-fold faster than by HOCl [19].
As stated above, tissue damage associated with ROS/RNS is related to
the susceptibility of biomolecules to oxidation. However, the oxidizing
capacity does not seem to be the major factor that influences the delete-
rious properties of HOBr. Indeed, though HOBr reacts faster with several
biomolecules [19], HOCl is the stronger two-electron oxidant (E′° HOCl/
where [A] is the hypohalous acid and [B] is the studied compound.
If [A] NN [B] → reaction rate = kobs*[B].
2
Then, kobs = k *[A].
The k is the slope of the linear fit of kobs versus [A].
2
−
−
Cl = 1.28 V, E′° HOBr/Br = 1.13 V) [20]. On the other hand, taking
into account the electrophilicity of these chemicals, an inversion is
observed, since HOBr is more reactive than HOCl under non-
physiological experimental conditions [21]. Hence, considering the
importance of bromination reactions in the deleterious pathways that
underline chronic inflammatory diseases and the low concentration of
2.2.2. Dansylglycine
As above with the following modifications: dansylglycine 50 μM and
HOCl 520 to 3200 μM, and dansylglycine 5 μM and HOBr 50 to 125 μM.
The reactions were monitored by the intrinsic fluorescence of
dansylglycine (excitation using a 360 nm LED and emission using a
475 nm cut-off filter).
−
Br in the physiological medium, we hypothesized that the electrophi-
licity of HOBr could be mainly responsible by its effects. Hence, we
selected and studied endogenous and non-endogenous compounds
for which the relative electrophilic potency of the hypohalous acids
could be compared without, or with less, interference of oxidative pro-
cess. We found that HOBr was able to act as an authentic electrophile
and with higher efficacy than HOCl.
2.2.3. Salicylic acid
As above with the following modifications: salicylic acid 50 μM and
HOBr 500 to 1250 μM. The reactions were monitored by the intrinsic
fluorescence of salicylic acid (excitation using a 280 nm LED and
emission using a 375 nm cut-off filter).
2.2.4. 3-Hydroxybenzoic acid
2
. Materials and methods
As above with the following modifications: 3-hydroxybenzoic acid
0 μM and HOBr 125 to 500 μM, and 3-hydroxybenzoic acid 50 μM
5
2
.1. Chemicals
and HOCl 250 to 2000 μM. The reactions were monitored by its intrinsic
fluorescence (excitation using a 280 nm LED and emission using a
375 nm cut-off filter).
Dansylglycine, anisole, acetophenone, salicylic acid, 3-hydroxybenzoic
acid, tryptophan, adenine, cytosine, guanine, thymine, uracil, adenosine,
cytidine, guanosine, thymidine, uridine, 5-bromouridine and 4-
bromoanisole were purchased from Sigma-Aldrich Chemical Co.
2.2.5. Tryptophan
As above with the following modifications: tryptophan 50 μM and
HOCl 125 to 375 μM, and tryptophan 5 μM and HOBr 10 μM. The
reactions were monitored by the intrinsic fluorescence of tryptophan
(excitation using a 280 nm LED and emission using a 325 nm cut-off
filter).
(
St. Louis, MO, USA). Stock solutions of anisole 4-bromoanisole and
acetophenone (10 mM) were prepared in ethyl alcohol. Stock solution
of dansylglycine (5 mM) was prepared in 10 mM hydrochloric acid.
Stock solution of amino acids, nitrogen bases and nucleosides (5 mM)
were prepared in 50 mM phosphate buffer at pH 7.0. The concentration
of the commercial HOCl solution was determined spectrophotometri-
2.2.6. Anisole
cally after dilution in 0.01
3
to give an aqueous stock solution of 100 mM. HOBr was synthesized by
combining 100 mM HOCl and 200 mM NaBr in water immediately
before the assays [23].
M
NaOH at pH 12 (ε292nm
=
As above with the following modifications: anisole 50 μM and HOBr
125 to 375 μM, and anisole 50 μM and HOCl 500 μM. The measurements
were performed by its absorbance using a 280 nm LED.
50 M cm−1) [22]. HOCl was prepared immediately before the assays
−
1
2.3. HPLC studies
2
.2. Determination of rate constants
The nitrogenous bases, nucleosides and anisole (1 mM) were sub-
mitted to the reactions with the hypohalous acids (1 mM) in phosphate
buffer 0.05 mM, pH 7.0 and 25 °C, and the consumption after 30 min
was monitored by HPLC analysis. The consumption of nitrogenous
bases and nucleosides were chromatographically evaluated by HPLC
(Jasco, Easton, MD, USA) in line with a UV-visible (UV–vis) detector
set at 280 nm. The analyses were carried out isocratically on a Synergi
Polar reversed-phase column (150 mm × 4.6 mm, 4 μm), with 0.1%
formic acid in water and 0.1% formic acid in acetonitrile (95:5, v:v) as
mobile phase, and flow rate of 1.0 mL/min. The analyses of consumption
of anisole and formation of 4-bromoanisole were carried out
isocratically on a Shim-pack C18 CLC-ODS reversed-phase column
(150 mm × 6.0 mm, 4 μm), with 0.1% formic acid in water and 0.1%
formic acid in acetonitrile (70:30, v:v) as mobile phase, and flow rate
of 1.0 mL/min.
The fast-kinetic experiments were performed using a single-mixing
stopped-flow system equipped with high intensity LED source and cut-
off filters (SX20/LED Stopped-Flow System, Applied Photophysics, UK).
The photophysical properties of the studied compounds and their
intrinsic reactivity determined the experimental conditions used to
monitor each reaction.
2
.2.1. Uridine
The measurements were performed by its absorbance using a
80 nm LED. The experiments were performed using pseudo-first-
2
order conditions where the hypohalous acids were used in excess
500–1250 μM) compared with nucleoside (50 μM) in 50 mM phos-
phate buffer, pH 7.0 at 25 °C. The observed pseudo-first-order constant
(

V.F. Ximenes et al. / Journal of Inorganic Biochemistry 146 (2015) 61–68
63
2
.4. Theoretical studies
All the computer simulations were done in the GridUnesp super-
computer facilities, which is composed of 256 SUN X4150 servers with
048 cores (Intel Xeon 2.83 GHZ), with 4096 GB of RAM memory
2 GB per core) and an infiniband 4X DDR (20 Gbps) connection. The
storage capacity of these system is 36 TB through DAS optical fibres
StorageTek 6140) and 96 TB at four SUN X4500 servers. The
2
(
(
Gaussian09 [24] suite of programs was utilized to obtain the geometri-
cal and the Mulliken and ChelpG charges on HOCl and HOBr molecules.
The Mulliken population [25] analysis was obtained from the diago-
nal elements of PS, where P is the density matrix and S the overlap ma-
trix. The net charge associated with an atom is given by the following
equation:
qA ¼ Z –∑iεAðPSÞii
A
A
where Z is the charge of atomic nucleus A; the index of summation in-
dicates that we only sum over the basis functions centred on A.
The Charges from Electrostatic Potentials Using a Grid-Based
Method (CHELPG) is a method in which the atomic charges are fitted
to reproduce the MEP (Map of Electrostatic Potential) at a certain num-
ber of points around the molecule of interest [26]. As a first step of the
fitting procedure, the MEP is calculated at a number of grid points
spaced 3.0 pm apart and distributed regularly in a cube. The dimensions
of the cube are chosen such that the molecule is located at the centre of
the cube, adding 28.0 pm between the molecule and the end of the box
in all three dimensions. All points falling inside the Van der Waals radius
of the molecule are discarded from the fitting procedure. After evaluat-
ing the MEP at all valid grid points, the atomic charges were obtained in
a manner that reproduces the MEP in the most optimum way. The addi-
tional constraint in the fitting procedure is that the sum of all atomic
charges must be equal to that of the overall charge of the system.
Fig. 1. Molecular structures of the compounds used in this study.
While the acetyl group deactivated the aromatic ring in
acetophenone, the electron-donating methoxy group increases the re-
activity of the anisole [27]. This compound was also chosen because
its oxidation is precluded by the methyl substituent, and hence the elec-
trophilic attack on the aromatic ring is the more favourable reaction.
Corroborating this, HOBr was able to brominate the benzene ring lead-
ing to 4-bromoanisole. Fig. 2a shows the HPLC analysis of the reaction
mixture composed of anisole and HOBr (3) and, for comparison, pure
anisole (1) and 4-bromoanisole (2). As could be expected for a typical
electrophilic aromatic substitution, the anisole was totally converted
to 4-bromoanisole, which can be explained taking into account the
ortho/para directing nature of the methoxy group [27]. Fig. 2b
shows the kinetic study of the interaction between HOBr and anisole.
In these experiments, the concentration of anisole was kept constant
at 50 μM and HOBr varied from 125 to 350 μM. From these pseudo-
first-order experimental conditions, the apparent bimolecular rate
3
. Results and discussion
3
.1. Reactivity with acetophenone and anisole
Electrophilicity is a kinetic parameter related to the capacity of mol-
ecules bearing electron-poor moieties (electrophiles) to interact with
molecules bearing electron-rich moieties (nucleophiles) [27]. Obvious-
ly, there are many biomolecules bearing nucleophilic moieties. Howev-
er, most of these nucleophilic sites are also oxidizable through hydrogen
atom and/or electron abstraction, e.g. sulfhydryl and aromatic hydroxyl
and amino moieties. Consequently, the discrimination between these
distinct chemical features in endogenous biomolecules is not an easy
task. For this reason, and aiming to study and compare the electrophilic-
ity of HOBr and HOCl, we selected several compounds for which their
susceptibility to electrophilic attack could be studied and compared
without, or with only small, interference of the oxidative process
−1
−1
constant was calculated (k
2
= 2.3 × 102
M
s
, Fig. 2c). Impor-
(
Fig. 1).
tantly, Fig. 2b also shows that HOCl was not reactive with anisole.
This was the first direct evidence of its lower electrophilic capacity
compared to HOBr. As can be observed, while 125 μM of HOBr pro-
voked the consumption of anisole in less than 25 s, HOCl, as high as
500 μM, was totally unreactive.
Our first targets were acetophenone and anisole. These compounds
are susceptible to electrophilic attack, but not to oxidation. In this
regard, anisole (Epa = 1.80 V) and acetophenone (Epa = 1.68 V)
have a significantly higher oxidation peak potential compared to easily
oxidizable phenol derivatives, e.g. caffeic acid (Epa = 0.45 V) [28–30].
As is well known from basic organic chemistry, the presence of the
electron-withdrawing acetyl group in the aromatic ring makes
acetophenone less susceptible to electrophilic substitution [27]. In
agreement with that, neither HOBr nor HOCl were able to react with
acetophenone under the mild experimental conditions used here
3.2. Reactivity with dansylglycine
Dansylglycine is a fluorescent probe used for the characterization of
binding sites in albumin [31]. Here, its choice was based on previous re-
sults that showed that HOCl was not reactive with anisole. Hence,
aiming to demonstrate a quantitative difference between HOBr and
HOCl, we used dansylglycine, which could be still more susceptible to
electrophilic substitution due to the presence of the naphytalene ring.
It is of note that the naphytalene ring is typically more susceptible to
(
3
1 mM, 25 °C, 0.05 M phosphate buffer, pH 7.0 and incubation for
0 minutes) (results not shown). Obviously, these reactions may take
place with concentrated reagents and heating – however, this would
not be relevant in the biological context.

64
V.F. Ximenes et al. / Journal of Inorganic Biochemistry 146 (2015) 61–68
in agreement with the previous results, salicylic acid was reactive with
HOBr, but not with HOCl. As can be observed, while 500 μM of HOBr de-
4
−1 −1
pleted salicylic acid in less than 0.3 seconds (k
2
= 4.03 × 10 M
s
),
the addition of 2000 μM HOCl caused only a slight consumption of
salicylic acid after 60 s of reaction.
Although the higher reactivity of HOBr was in agreement with the
previous experiments, the extremely low reactivity of HOCl with
salicylic acid was unexpected. In fact, taking into account its phenolic
character, even if not reactive by electrophilic attack, at least the direct
abstraction of the phenolic hydrogen (oxidation) should occur with
HOCl. One putative explanation for the low oxidability of salicylic acid
is the commitment of its phenolic hydrogen in the intramolecular hy-
drogen bound between the hydroxyl and carboxylate groups (Fig. 5)
[32].
To test this hypothesis, we also studied the reaction of the
hypohalous acids with the isomer of salicylic acid, 3-hydroxybenzoic
acid. In this compound, the greater distance between the hydroxyl
and carboxylate groups weaken the intramolecular hydrogen
bond compared to salicylic acid (2-hydroxybenzoic acid) (Fig. 5)
[
32]. Hence, if this chemical feature affected their reactivities, includ-
ing oxidizability, then HOCl should be more reactive with
-hydroxybenzoic compared to salicylic acid. Corroborating this, 3-
hydroxybenzoic acid has a lower first anodic peak potential
3
(
Epa = 0.83 V) compared to salicylic acid (Epa = 0.94 V) [30].
Taking into account these chemical properties, our proposal was that
3-hydroxybenzoic acid could be more susceptible to oxidation through
hydrogen abstraction and, consequently, the rate constant with HOCl
should increase when compared with salicylic acid. Our expectation
was confirmed because, in contrast to salicylic acid, 3-hydroxybenzoic
4
−1 −1
acid was reactive with both HOBr (k
2
= 5.9 × 10 M
s
) and HOCl
= 1.1 × 101
M
−1
s
−1
). Obviously, HOBr was still more reactive,
(
k
2
which suggests that electrophilic attack and oxidation could take place
simultaneously.
In summary, this small structural alteration in the hydroxylated
benzoic acids helped us to show that when the oxidation was impeded,
the electrophilic character of HOBr was much more relevant than that of
HOCl.
Fig. 2. Reactivity of hypohalous acids with anisole. (a) HPLC analysis of the reaction mix-
ture composed of anisole (1 mM) and HOBr (1 mM) in 0.05 M phosphate buffer, pH 7.0.
The chromatogram traces are (1) anisole standard, (2) 4-bromoanisole standard and
3.4. Reactivity with nitrogenous bases and their nucleosides
(
3) reaction mixture. (b) Kinetic profile of anisole (50 μM) consumption by HOBr for de-
termination of kobs in pseudo-first-order experimental conditions. The figure also shows
that HOCl was not reactive. (c) Determination of bimolecular rate constant (k ) for the re-
There have been many demonstrations that hypohalous acids react
2
with nitrogenous bases and their nucleosides. For instance, the forma-
tion of 5-chlorouracil and 5-bromouracil from MPO catalysed oxidation
of uracil [33]; formation of 5-chloro-2′-deoxycytidine, 8-chloro-2′-
deoxyadenosine and 8-chloro-2′-deoxyguanosine by the treatment of
DNA with HOCl [34]; bromination of deoxycytidine by HOBr pro-
duced by the catalytic action of EPO [35]. For HOCl-mediated interac-
tions, the reactions are initiated by formation of chloramines, which
decay to nucleoside-derived nitrogen-centred radicals [35]. It was
also demonstrated that the propensity for radical formation was
cytidine N adenosine = guanosine N uridine = thymidine, i.e. the
action between anisole and HOBr. The results are the mean of three experiments.
electrophilic attack than the benzene ring [27]. It is also important to
note that the intrinsic fluorescence of dansylglycine at 360/550 nm
makes its consumption easily monitored, i.e. suitable for kinetic studies,
since interference of any inner-filter effect caused by the reactants is
minimised.
The results in Fig. 3 confirmed our expectation since not only HOBr
but also HOCl was able to react with dansylglycine. Corroborating the
anisole results, the reactivity with HOBr was significantly higher, as
2
bases bearing an exocyclic-NH group reacted faster, which was
explained by taking into account the thermodynamic stability of
the exocyclic chloramines [36].
6
−1 −1
can be seen by the bimolecular rate constants, 7.30 × 10 M
s
and
2
−1 −1
5
.20 × 10 M
s
for HOBr and HOCl, respectively. It is of note that,
From these experimental data, it is clear that for nitrogenous bases,
the differentiation between oxidation and electrophilic susceptibility
is not an easy task. Taking into account these experimental findings
and aiming to differentiate oxidation from electrophilic attack, we di-
vided the nitrogenous bases and their nucleosides into two groups as
follows: adenine, cytosine and guanine and their nucleosides adenosine,
cytidine and guanosine (group 1, see Supplementary Material 1). For
these bases, an oxidizable endocyclic NH group is lost in the formation
of the nucleoside derivatives. However, they still have the free and
for the experiments with HOBr, the concentration of dansylglycine
was decreased to 5 μM, which was 10-fold lower than in the experi-
ments with HOCl. This alteration was necessary because otherwise the
precise measurement of the rate constants for HOBr would not be
possible.
3
3
.3. Reactivity with salicylic acid (2-hydroxybenzoic acid) and
-hydroxybenzoic acid
oxidizable exocyclic-NH
and thymine, for which there is no additional NH
transformation to their nucleosides uridine and thymidine (group 2,
2
group. On the other hand, we have uracil
Salicylic acid is another compound for which its consumption can be
monitored by its intrinsic fluorescence at 295/405 nm. Fig. 4 shows that,
2
group after

V.F. Ximenes et al. / Journal of Inorganic Biochemistry 146 (2015) 61–68
65
Fig. 3. Reactivity of hypohalous acids with dansylglycine. (a) Kinetic profile of dansylglycine (5 μM) consumption by HOBr for determination of kobs in pseudo-first-order experimental
conditions. (b) Determination of the bimolecular rate constant (k ) for the reaction between dansylglycine and HOBr. (c) Kinetic profile of dansylglycine (50 μM) consumption by
HOCl. (d) Determination of the bimolecular rate constant (k ) for the reaction between dansylglycine and HOCl. The results are the mean of three experiments.
2
2
see Supplementary Material 1). Considering that, our hypothesis was
that the nucleosides from group 2 would be less susceptible to oxidation
and, consequently, better to show the electrophilic activity of HOBr.
This hypothesis was confirmed, since the nitrogenous bases from
group 1 were promptly consumed in the free and nucleoside forms
when HOCl or HOBr were added to the reaction medium (see Supple-
mentary Material 2). On the other hand, while the free bases from
group 2 reacted promptly with both HOCl and HOBr, their nucleoside
derivatives, uridine and thymidine, only reacted with HOBr (see Supple-
mentary Material 3). This is a demonstration that, when these mole-
cules were susceptible to oxidation, HOCl and HOBr were both
effective reactants. It was not our aim to characterize and differentiate
the products of these reactions, which indeed have been performed
elsewhere [33–36], but only to show that, when oxidation is precluded,
the susceptibility to electrophilic attack gains importance. This was the
case for uridine and thymidine.
electrophilic attack, becomes the dominant reactive site in the molecule,
which explains the higher reactivity of HOBr.
3.5. Reactivity with indole compounds
Hypohalous acids are also highly reactive with indole derivatives, for
instance, the amino acid tryptophan and the neurohormone melatonin
[37,38]. Hence, we also tried to measure and compare their reactivity
with HOCl and HOBr. Considering that the indole ring is much more
susceptive to electrophilic attack than the benzene ring [39], it was no
surprise to verify that tryptophan reacted with both hypohalous acids
(Fig. 7). However, again HOBr was much more reactive than HOCl. In-
deed, aiming to measure the bimolecular rate constant with HOBr, the
tryptophan concentration was ten-fold decreased, but the fluorescence
decay was still too fast to be measured with precision (Fig. 7c). Hence,
the bimolecular rate constant for the reaction between HOBr and
tryptophan was not measured. On the other hand, the bimolecular
rate constant for the reaction between HOCl and tryptophan was easily
Taking into account this qualitative difference, the next step was the
quantitative measurement of the difference between HOCl and HOBr for
uridine. Fig. 6 shows the kinetic study and confirms that only HOBr was
4
−1 −1
measured, i.e. 8.1 × 10 M
s
. It is of note that indole derivatives, in-
3
−1 −1
reactive with uridine (k
2
= 1.3 × 10 M
s
). As can be observed,
cluding tryptophan and melatonin, are substrates of MPO and inhibitors
of the its halogenating activity [40,41]. Hence, the high reactivity of
HOBr with these compounds could also represent a pathway for deple-
tion of these harmful species.
while 2.0 mM of HOCl was not able to consume uridine, 0.5 mM HOBr
provoked almost complete depletion in about 2.0 minutes. The expect-
ed product of the reaction between HOBr and uridine was 5-
bromouridine, a typical electrophilic substitution product, which was
confirmed by chromatographic comparison with the pure compound
3.6. Bromide as a cofactor in HOCl reactions
(
Fig. 6c). Similar findings were obtained for thymidine (results not
shown).
This alteration in the chemical reactivity of uridine and thymidine
So far, we have shown that HOBr is significantly more electrophilic
than HOCl. Hence, the next step was to measure the effect of free Br
−
regarding their reactivity with HOBr is another demonstration of the
strong electrophilic character of this bromination agent. In other
words, the binding of ribose in the uracil and thymine ring decreased
its oxidability, since the hydrogen abstraction from the NH group is no
longer available. Hence, the C = C moiety, which is susceptible to
in the HOCl-mediated reactions. We chose dansylglycine as the sub-
strate for this study because its reaction with HOCl can be measured
quantitatively. In these experimental settings, dansylglycine and Br
were mixed in the same syringe in the stopped-flow system. Hence,
the pre-formation of HOBr was avoided, and once the reaction was
−

66
V.F. Ximenes et al. / Journal of Inorganic Biochemistry 146 (2015) 61–68
Fig. 5. Molecular structure of salicylic acid and 3-hydroxybenzoic acid at pH 7.0. The intra-
molecular hydrogen bond altered their susceptibility to oxidation.
HOCl and HOBr were obtained at the CCSD(T) level of theory with the
aug-cc-pVTZ basis set applying one s, one d and one p diffuse functions
on the hydrogen atoms, and one d, one p, one d and one diffuse
functions on the oxygen, chlorine and bromine atoms. Table 1 shows
the bond distances and angles for the selected molecules along with
the experimental results for comparison for HOCl [42] and HOBr [43],
respectively. Upon analysis of the calculated results for the distance
Fig. 4. Reactivity of hypohalous acids with salicylic acid. (a) Kinetic profile of salicylic acid
50 μM) consumption by HOBr and determination of kobs in pseudo-first-order experi-
mental conditions. (b) Determination of the bimolecular rate constant (k ) for the reaction
(
2
between salicylic acid and HOBr. (c) Consumption of salicylic acid (50 μM) consumption
by HOCl. The results are the mean of three experiments.
−
triggered, HOCl had contact with dansylglycine and Br at the same
−
moment. The concentration of Br was selected as being similar to the
−
physiological level. Fig. 8 shows the effect of 20–80 μM of Br in the re-
action medium with a fixed concentration of HOCl (500 μM) and
dansylglycine (50 μM). As can be observed, the in situ formation of
−
HOBr, through the reaction between HOCl and Br , was enough to
accelerate the reaction rate.
3
.7. Electrophilicity of hypohalous acids: theoretical studies
Fig. 6. Reactivity of hypohalous acids with uridine. (a) Kinetic profile of uridine (50 μM)
consumption by HOBr or HOCl for determination of kobs in pseudo-first-order experimen-
2
tal conditions. (b) Determination of the bimolecular rate constant (k ) for the reaction be-
Together, the above experimental results show that the main chem-
tween uridine and HOBr. The results are the mean of three experiments. (c) HPLC analysis
of the reaction mixture composed of uridine (1 mM) and HOBr (1 mM) in 0.05 M
phosphate buffer, pH 7.0. The chromatogram traces of (1) pure 5-bromouridine, and
(2) reaction mixture.
ical property that differs in HOCl and HOBr is the higher electrophilicity
of the latter. Hence, our next step was to perform computational studies
that could support these findings. Initially, the optimized structures of

V.F. Ximenes et al. / Journal of Inorganic Biochemistry 146 (2015) 61–68
67
Table 1
Optimised and experimental geometry of HOBr and HOCl. Bond distances in angstroms,
and angles in degrees.
CCDS(T)/aug-cc-pVTZ
Experimental [42,43]
HOBr*
103.09
1.839
0.967
102.51
1.709
0.967
102.3
1.834
0.961
102.45
1.691
0.964
#
BrO
#
HO
HOCl*
#
ClO
HO
#
*
Bond angle, #bond distance.
and bond angle, we can infer a good agreement with the experimental
ones.
The atomic charges on the HOCl and HOBr molecules are presented
in Table 2. The results are presented at the same level of theory and basis
set utilised for the optimisation of the molecular geometry. The results
for the Mulliken and ChelpG charges follow the same tendency of neg-
ative charges for the oxygen and positive ones for the hydrogen and hal-
ogen atoms. How can be observed, the positive charge is positioned on
the halogen atoms with the greater value on the bromine atom. In other
words, the higher positive charge on bromine is a theoretical demon-
stration of its higher susceptibility to electron-rich centre, as we have
demonstrated in this report.
3.8. Electrophilicity of HOBr: physiological and pathological implications
Recently, a biochemical reaction was proposed as being the first, in
the animal kingdom, where Br is an essential cofactor: Br− is an essen-
tial cofactor for peroxidasin, a haem peroxidase that catalyzes the for-
mation of sulfilimine crosslinks for reinforcement of collagen IV
scaffolds [44]. This enzymatic reaction involves the formation of HOBr,
which generates a halo sulfonium intermediate. The latter acts as an
electrophilic site for the formation of the sulfilimine crosslink [44]. In
−
−
other words, this putative physiological role for Br involves the devel-
opment of an intermediate electrophilic site. In line with this finding,
−
here we propose that electrophilicity might be the main feature of Br
,
obviously as a source of HOBr. As demonstrated throughout the
study, the electrophilic strength of HOBr is evidenced when the target
substrates are not easily oxidized. In fact, whether direct electron and/
or hydrogen abstraction is a feasible pathway, the difference between
HOBr and HOCl is irrelevant or the latter is the better reactant.
From a pathophysiological point of view the high electrophilicity of
−
HOBr gains importance if we take into account that Br has been dem-
onstrated as a key substrate not only to EPO, but also to MPO at physio-
logical pH. In this concern, Senthilmohan and Kettle demonstrated
Fig. 7. Reactivity of hypohalous acids with tryptophan. (a) Kinetic profile of tryptophan
2
50 μM) consumption by HOCl. (b) Determination of the bimolecular rate constant (k )
for the reaction between tryptophan and HOCl. (c) Kinetic profile of tryptophan (5 μM)
consumption by HOBr (10 μM). The results are the mean of three experiments.
(
that ≈ 25% of the H
2
O
2
used by MPO was converted to HOBr, even
−
though Cl was present in a 1,400-fold excess compared to bromide
[
45]. From a pharmacological point of view, potassium bromide is
considered the drug of tertiary choice in the treatment of children
−
with epilepsy [46]. In these cases, the blood level of Br during the treat-
ment reaches 145 mg/dL (~18 mM) [46], which is about 180-fold higher
compared to the physiological level (~100 μM) [18]. Hence, taking into
account that epilepsy has been associated with brain inflammation [47];
the conditions for a significant increase in the production of HOBr is
Table 2
Mulliken and ChelpG charges on atoms (in atomic units).
HOBr
HOCl
Mulliken
ChelpG
Mulliken
ChelpG
H
O
Br
Cl
0.2503
−0.5413
0.2910
-
0.3525
−0.4104
0.0576
-
0.2527
−0.4357
-
0.3733
−0.3817
-
Fig. 8. Bromide as a co-catalyst for the reaction of HOCl with dansylglycine. Dansylglycine
0.1829
0.0084
5
0 μM, HOCl 500 μM and Br− (20–80 μM).

68
V.F. Ximenes et al. / Journal of Inorganic Biochemistry 146 (2015) 61–68
[
[
2] K. Urbański, M. Nowak, T.J. Guzik, Postepy Biochem. 59 (2013) 424–431.
3] F. Liu-Smith, R. Dellinger, F.L. Meyskens Jr., Arch. Biochem. Biophys. 563 (2014)
1–55.
established and could contribute for the deleterious side effects of this
pharmacological treatment.
5
[
4] J.M. McCord, B.A. Omar, Toxicol. Ind. Health 9 (1993) 23–37.
[
5] M. Katsuyama, K. Matsuno, C. Yabe-Nishimura, J. Clin. Biochem. Nutr. 50 (2012)
4
. Conclusion
9–22.
[
6] S. Dröse, U. Brandt, Adv. Exp. Med. Biol. 748 (2012) 145–169.
As we have demonstrated, when the susceptibility to oxidation is
[7] M. Gardès-Albert, Ann. Pharm. Fr. 64 (2006) 365–372.
[8] D. Vara, G. Pula, Curr. Mol. Med. 14 (2014) 1103–1125.
[
decreased and the interaction with electrophiles becomes a favourable
pathway, HOBr is much more reactive than HOCl. For instance, by
changing 3-hydroxybenzoic acid to its isomer 2-hydroxybenzoic acid
9] G. Proteasa, Y.R. Tahboub, S. Galijasevic, F.M. Raushel, H.M. Abu-Soud, Biochemistry
6 (2007) 398–405.
4
[10] J. Arnhold, J. Flemmig, Arch. Biochem. Biophys. 500 (2010) 92–106.
[11] J. Wang, A. Slungaard, Arch. Biochem. Biophys. 445 (2006) 256–260.
12] P.G. Furtmüller, U. Burner, G. Regelsberger, C. Obinger, Biochemistry 39 (2000)
5578–15584.
(
salicylic acid), it was possible to provoke a small decrease in the sus-
[
ceptibility to oxidation, and then only HOBr was able to interact. Similar
findings were observed for the nitrogenous bases and their nucleosides.
An explanation for the higher electrophilicity of HOBr compared with
1
[
13] P.G. Furtmüller, M. Zederbauer, W. Jantschko, J. Helm, M. Bogner, et al., Arch.
Biochem. Biophys. 445 (2006) 199–213.
+
[14] E. Thomson, S. Brennan, R. Senthilmohan, C.L. Gangell, A.L. Chapman, et al., Free
HOCl is based on the formation of a bromonium ion (Br ) as an inter-
Radic. Biol. Med. 49 (2010) 1354–1360.
mediate in these reactions. Its formation is favoured by the lower elec-
tronegativity and higher polarizability of bromine compared to
chlorine, which makes it more able to accept a positive charge, and, con-
sequently, more electrophilic.
[
[
[
15] B. Shao, C. Tang, A. Sinha, P.S. Mayer, G.D. Davenport, et al., Circ. Res. 114 (2014)
1733–1742.
16] A. Ivekanandan-Giri, J.L. Slocum, J. Byun, C. Tang, R.L. Sands, et al., Ann. Rheum. Dis.
72 (2013) 1725–1731.
17] S.H. Wedes, W. Wu, S.A. Comhair, K.M. McDowell, J.A. DiDonato, et al., J. Pediatr. 159
(2011) 248–255.
18] P.G. Furtmüller, U. Burner, C. Obinger, Biochemistry 37 (1998) 17923–17930.
19] D.I. Pattison, M.J. Davies, Biochemistry 43 (2004) 4799–4809.
20] O. Augusto, S. Miyamoto, in: K. Pantopoulos, H.M. Schipper (Eds.), Principles of free
radical biomedicine, Nova Science Publishers Inc., New York, 2011, pp. 1–23.
[21] J.D. Sivey, J.S. Arey, P.R. Tentscher, A.L. Roberts, Environ. Sci. Technol. 47 (2013)
330–1338.
22] A.J. Kettle, Methods Enzymol. 300 (1999) 111–120.
23] S. Tokunaga, A. Kanayama, Y. Miyamoto, Inflamm. Res. 56 (2007) 479–486.
[24] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, et al., Gaussian 98,
Gaussian Inc., Pittsburgh, PA, 1998.
25] A. Szabo, N.S. Ostlund, Modern quantum chemistry: introduction to advanced
electronic structure theory, Courier Dover Publications, New York, 1996.
[26] K.I. Ramachandran, G. Deepa, K. Namboori, Computational chemistry and molecular
modeling: principles and applications, Springer Science & Business Media, Berlin,
In conclusion, we propose that HOBr should be classified as an en-
dogenous reactive electrophile species (RES), and perhaps the stronger
endogenous electrophile. As is well known, RES comprises a large num-
ber of compounds that have in common the presence of electron-poor
centres [48]. In general, RES are produced from ROS-mediated non-
enzymatic oxidation and nitration of several biomolecules leading to
formation of aldehydes, α-β-unsaturated carbonyls, nitroalkenes,
nitrated cyclic nucleotide, etc. [49,50]. These RES are able to interact
with strong nucleophilic moieties, e.g. cysteine groups in proteins via
Michael addition. However, in contrast to HOBr, they are not able to
attack aromatic rings, which are much less reactive as nucleophilic sites.
[
[
[
1
[
[
[
2
008.
Abbreviations
[
27] M.B. Smith, J. March, March's advanced organic chemistry: reactions, mechanisms
and structure, John Wiley & Sons, New Jersey, 2007.
[
[
[
28] M.R. Ciumaga, T. Tzedakisa, C.A. Barrès, Electrochim. Acta 70 (2012) 142–152.
29] M. Noel, I. Rajini, R.K. Srinivasan, Bull. Electrochem. 4 (1988) 359–363.
30] A. Simić, D. Manojlović, D. Segan, M. Todorović, Molecules 12 (2007) 2327–2340.
PMNs
MPO
EPO
ROS
RNS
RES
HDL
LED
MEP
Epa
Polymorphonuclear leukocytes
Myeloperoxidase
Eosinophil peroxidase
Reactive oxygen species
Reactive nitrogen species
Reactive electrophilic species
High-density lipoprotein
light-emitting diode
[31] F.S. Graciani, V.F. Ximenes, PLoS One 8 (2013) e76849.
[
[
32] G.E. Dunn, T.L. Penner, Can. J. Chem. 45 (1967) 1699–1706.
33] P. Jeffrey, J.B. Henderson, J. Takeshita, J.W. Heinecke , J. Biol. Chem. 278 (2003)
23522–23528.
[34] N.R. Stanley, D.I. Pattison, C.L. Hawkins, Chem. Res. Toxicol. 23 (2010) 1293–1302.
[35] J.P. Henderson, J. Byun, M.V. Williams, M.L. McCormick, W.C. Parks, et al., Proc. Natl.
Acad. Sci. U. S. A. 98 (2001) 1631–1636.
[
36] C.L. Hawkins, M.J. Davies, Chem. Res. Toxicol. 14 (2001) 1071–1081.
Map of electrostatic potential
Anodic peak potential
[37] M.S. Petrônio, V.F. Ximenes, Luminescence 28 (2013) 853–859.
[38] S.M. Dellegar, S.A. Murphy, A.E. Bourne, J.C. DiCesare, G.H. Purser, Biochem. Biophys.
Res. Commun. 257 (1999) 431–439.
[
[
CHELPG Charges from electrostatic potentials using a grid based
method
39] R.K. Parashar, Chemistry of heterocyclic compounds, CRC Press, Florida, 2014.
40] A.J. Kettle, L.P. Candaeis, Redox Rep. 5 (2000) 179–184.
CCSD(T) Coupled cluster singles and doubles
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.jinorgbio.2015.02.014.
[41] S. Galijasevic, I. Abdulhamid, H.M. Abu-Soud, Biochemistry 47 (2008) 2668–2677.
[
[
42] W.D. Anderson, M.C.L. Gerry, R.W. Davis, J. Mol. Spectrosc. 115 (1986) 117–130.
43] Y. Koga, H. Takeo, S. Kondo, S. Sugie, C. Matsumura, et al., J. Mol. Spectrosc. 138
(
1989) 467–481.
44] A.S. McCall, C.F. Cummings, G. Bhave, R. Vanacore, A. Page-McCaw, et al., Cell 157
2014) 1380–1392.
[
(
Acknowledgments
[
[
45] R. Senthilmohan, A.J. Kettle, Arch. Biochem. Biophys. 445 (2006) 235–244.
46] R. Korinthenberg, P. Burkart, C. Woelfle, J.S. Moenting, J.P. Ernst, J. Child Neurol. 22
(2007) 414–418.
This work was supported by grants from the São Paulo Research
Foundation (FAPESP), Coordination for Enhancement of Higher Educa-
tion Personnel (CAPES), the National Council for Scientific and Techno-
logical Development (CNPq) and GridUnesp.
[47] N. Dupuis, S. Auvin, CNS Neurosci. Ther. 21 (2015) 141–151.
[48] E.E. Farmer, C. Davoine, Curr. Opin. Plant Biol. 10 (2007) 380–386.
[49] A.L. Levonen, B.G. Hill, E. Kansanen, J. Zhang, V.M. Darley-Usmar, Free Radic. Biol.
Med. 71 (2014) 196–207.
[
50] T. Sawa, H. Ihara, T. Akaike, Curr. Top. Med. Chem. 11 (2011) 1854–1860.
References
[1] L. Rochette, M. Zeller, Y. Cottin, C. Vergely, Biochim. Biophys. Acta 1840 (2014)
2709–2729.