Beyond H2S and NO interplay: Hydrogen sulfide and nitroprusside react directly to give nitroxyl (HNO). A new pharmacological source of HNO

Article abstract of DOI:10.1021/jm3012036

Hydrogen sulfide (H₂S) has been increasingly recognized as an important signaling molecule that regulates both blood pressure and neuronal activity. Attention has been drawn to its interactions with another gasotransmitter, nitric oxide (NO). Here, we provide evidence that the physiological effects observed upon the application of sodium nitroprusside (SNP) and H₂S can be ascribed to the generation of nitroxyl (HNO), which is a direct product of the reaction between SNP and H₂S, not a consequence of released NO subsequently reacting with H₂S. Intracellular HNO formation has been confirmed, and the subsequent release of calcitonin gene-related peptide from a mouse heart has been demonstrated. Unlike with other thiols, SNP reacts with H₂S in the same way as rhodanese, i.e., the cyanide transforms into a thiocyanate. These findings shed new light on how H₂S is understood to interact with nitroprusside. Additionally, they offer a new and convenient pharmacological source of HNO for therapeutic purposes.

Full text of DOI:10.1021/jm3012036

Article
pubs.acs.org/jmc
Beyond H2S and NO Interplay: Hydrogen Sulfide and Nitroprusside
React Directly to Give Nitroxyl (HNO). A New Pharmacological Source
of HNO
Milos R. Filipovic,*,† Mirjam Eberhardt,†,‡ Vladimir Prokopovic,† Ana Mijuskovic,§
Zorana Orescanin-Dusic,§ Peter Reeh,‡ and Ivana Ivanovic-Burmazovic*,†
†Department of Chemistry and Pharmacy and ‡Department of Physiology and Pathophysiology, Univeristy of Erlangen−Nuremberg,
Erlangen, Germany
§Institute for Biological Research Sinisa Stankovic, University of Belgrade, Belgrade, Serbia
S
* Supporting Information
ABSTRACT: Hydrogen sulfide (H2S) has been increasingly
recognized as an important signaling molecule that regulates both
blood pressure and neuronal activity. Attention has been drawn to its
interactions with another gasotransmitter, nitric oxide (NO). Here,
we provide evidence that the physiological effects observed upon the
application of sodium nitroprusside (SNP) and H2S can be ascribed
to the generation of nitroxyl (HNO), which is a direct product of the
reaction between SNP and H2S, not a consequence of released NO
subsequently reacting with H2S. Intracellular HNO formation has
been confirmed, and the subsequent release of calcitonin gene-related peptide from a mouse heart has been demonstrated.
Unlike with other thiols, SNP reacts with H2S in the same way as rhodanese, i.e., the cyanide transforms into a thiocyanate. These
findings shed new light on how H2S is understood to interact with nitroprusside. Additionally, they offer a new and convenient
pharmacological source of HNO for therapeutic purposes.
been investigated, and the intermediate formation of S-
nitrosothiol was postulated to account for the vasodilatory
effects of SNP.11−14 Studies on the reaction of SNP with H2S
were performed under anaerobic and alkaline conditions,11,15,16
but nothing is definitively known about the products of this
reaction under physiological conditions.17
Some of us have recently characterized the smallest S-
nitrosothiol, thionitrous acid (HSNO), and its subsequent
chemistry in its biological milieu, showing that this molecule
can serve as a NO, NO+, and HNO donor, depending on the
reaction condition.18 Coordinated NO+ in SNP represents an
interesting feature of this molecule, making the formation of
HSNO in a direct reaction of SNP with HS− feasible and
subsequent generation of HNO possible. All this prompted us
to study in detail the reaction of H2S and SNP under
physiologically relevant conditions using both chemical and
physiological/pharmacological tools.
INTRODUCTION
■
Evidence accumulated over the past decade shows that H2S
plays an important role in regulating vascular contractility and
neuronal activity.1,2 Experiments using knockout mice lacking
cystathionine γ-lyase, an enzyme involved in H2S biosynthesis,
unambiguously demonstrated its essential role in blood
pressure regulation.3,4 A strong pharmacological potential for
both H2S and newly developed H2S donors to minimize the
effects of ischemia-reperfusion injuries and atherosclerosis has
been documented using numerous disease models.5
The similarities between the physiological effects of NO and
H2S suggest a possible crosstalk between these two gaseous
species.1 The first such proposal came from Ali et al., who
showed that H2S inhibits the SNP-induced relaxation of
isolated aorta rings.6 They proposed the intermediate formation
of a stable S-nitrosothiol (RSNO). More recently, Yong et al.
suggested that the product of NO and H2S interactions should
be a thiol-modifying agent with a positive inotropic effect on
the heart.7 In a prior study, they had found similarities between
this product and nitroxyl (HNO).8
RESULTS AND DISCUSSION
■
General Notes on the Reaction of SNP with Thiols. As
summarized in Scheme 1, the general mechanism for the
reactions of SNP with thiols involves the initial formation of a
coordinated S-nitrosothiol, which has the characteristic pink
color that has long been used as a proof for thiols in Brand’s
In all of these studies, SNP was used as the NO donor.
However, SNP releases NO only when exposed to light and its
“NO-like” physiological effects are still not well understood.9 In
SNP, Na2[Fe(CN)5(NO)], NO coordinated to iron has an
NO+ character and undergoes chemistry that is best described
as direct nitrosation biochemistry.10 The reactions of SNP with
low molecular weight thiols, like cysteine and glutathione, have
Received: August 17, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Scheme 1. Proposed Products for the Reaction between SNP
a
and Different Thiols
Figure 1. SNP efficiently removes H2S from the solution without any
NO release. (A) The NO electrode trace of the reaction between H2S
and SNP. The upper panel represents the response of the NO
electrode (in the dark) to the addition of 100 μM SNP in a buffered
solution (black) and the response upon the addition of 100 μM of
both Na2S and CuCl2 (pink). The middle panel shows the effect of
H2S on the ambient light-induced release of NO from SNP (red). The
electrode response to SNP in the absence of the light (black) is shown
for comparison. The lower panel depicts the effect of H2S on the light-
induced NO release from SNP. The addition of Na2S to a buffered
solution exposed to a 500−550 nm light source (green line) or in the
absence of light (red line) prior to the addition of SNP completely
abolished the NO release. The black lines represent the corresponding
responses upon the addition of only SNP (black arrow). (B) H2S
electrode traces for the reaction of H2S with SNP. A 50 μM Na2S
solution was injected into a 300 mM phosphate buffer at a pH 7.4
under aerobic conditions, followed by the injection of an equimolar
amount of SNP.
a
Low molecular weight thiols (cysteine and glutathione) are thought
to react with SNP to yield either NO14 (pathway a) and/or the
corresponding S-nitrosothiols12 (pathway b). The reaction with H2S is
thought to yield NH3 if the pH of the solution is less than 1116
(pathway c) or N2O when the pH values are above 1116 (pathway d).
Throughout all of these studies, the release of free cyanide was
proposed.
assay.19 The RSNOs formed account for the “NO-like”
physiology of SNP because they can be either homolytically
cleaved to give NO12,14 or released to further react with the
corresponding physiological targets9−13 (pathways a and b in
Scheme 1). A chemical study of the reactions of H2S with SNP
has recently appeared16 and describes the reaction mechanism
as initially being similar to that of other thiols (although at pH
9−11) with the resulting pink-colored product characteristic of
the HSNO/SNO− coordinated adduct of SNP. The end
products of this reaction were found to be NH3 (at pH 8.5−11)
and N2O (at pH > 11) (pathways c and d in Scheme 1), but we
questioned the mechanism and characterization of the
intermediates reported in this study.17
However, our first qualitative experiment has already shown
that, under physiologically relevant conditions (aerobic
conditions at pH = 7.4), the H2S and SNP reaction mixture
forms a blue product solution (Supporting Information Figure
S1). This result is obviously different from the pink/purple
products reported for reactions with cysteine, glutathione, or
H2S (under anaerobic and alkaline conditions), which implies
that a different reaction mechanism occurred.
No NO Released from SNP during the Reaction with
H2S. In the literature dealing with the physiology and
pharmacology of H2S, there is a prevailing opinion6,20 that
H2S and NO react to form some type of stable S-nitrosothiol
that decomposes upon the addition of copper ions to release
NO. This conclusion is primarily based on the pharmacological
evidence that H2S diminishes SNP-induced vasodilation, which
is restored by the addition of copper ions.
To both test the hypothesis of stable S-nitrosothiol formation
and determine the actual source of the NO release, we followed
the reaction between SNP and H2S using an NO sensitive
electrode both in the total absence of light and under ambient
light. As expected, there was no release of NO from SNP in the
dark, and the addition of H2S had no effect. Minor amounts of
NO were released over time when SNP was exposed to
ambient light; however, H2S addition completely suppressed
further NO release (Figure 1A). Because it is known that light
induces homolytic S−N bond breakage in S-nitrosothiols,21 we
additionally irradiated an equimolar mixture of SNP (1 mM)
and H2S (1 mM) under a visible light source (500−550 nm).
However, no NO release was observed relative to the control,
which consisted of only 1 mM SNP and released substantial
amounts of NO (Figure 1A). Furthermore, the addition of
copper ions into the reaction mixture did not stimulate NO
release (Figure 1A) as suggested6,20 by some authors. If the
product were an S-nitrosothiol, the immediate release of NO
would be expected upon both the addition of copper ions and
irradiation.21−23 These results strongly indicate that the
reaction produces neither a free S-nitrosothiol nor NO.
Next, we used an H2S-selective electrode to follow the effect
SNP has on H2S removal from the solution. As shown in Figure
1B, the equimolar addition of SNP completely removed H2S,
which implies H2S was consumed rather quickly. In addition,
the effects of H2S, SNP, and their combination on the
spontaneous and Ca2+-induced contraction of isolated uterus
preparations were tested. This particular pharmacological
model was chosen because, unlike other models, it is relatively
resistant (at high millimolar concentration) to SNP-induced
dilatation24 but still sensitive to other, direct NO donors,25
which makes it suitable for testing the potential NO release
from the SNP and H2S reaction mixture. H2S caused a
concentration-dependent inhibition of both the spontaneous
and calcium-induced smooth muscle contraction (Figure 2A)
with 1 mM of Na2S, yielding ∼95% inhibition. Conversely, SNP
had little effect over a wide range of concentrations (1−20
mM) for the spontaneous inhibition with a somewhat stronger
effect, IC50 = 10 mM, for the Ca2+-induced contractions (Figure
2B). When mixed together, 1 mM of each SNP and H2S caused
an effect with the same magnitude as the SNP alone for both
the spontaneous and Ca2+-induced contractions (Figure 2C).
Surprisingly, even a 5-fold excess for H2S over SNP proved to
be ineffective for evoking a greater inhibition of uterus
contractions, which implies that any excess of H2S is consumed
under physiological (aerobic) conditions. As a positive control,
we used a direct NO donor, diethylamine NONOate, which
caused 85 12% of inhibition of spontaneous contractions at a
B
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
presented in Supporting Information Figure S3, clearly show
the characteristic mass spectrum for N2O formation (m/z 44)
(inset in Supporting Information Figure S3).26 The amount of
N2O formed over 15 min increased as the H2S/SNP molar
ratio increased and was highest (0.15 μmol in the gas phase, i.e.,
5% of the stoichiometricaly expected yield) in the reaction
mixture containing a 5-fold excess of H2S over SNP. No
additional peak corresponding to NO was observed, which
agrees with the results obtained via amperometric and isolated
organ bath experiments.
To further confirm this N2O formation, we performed a 14
N
NMR study by collecting NMR spectra over 10 h at 10 °C. As
shown in Figure 3A, the spectrum for 50 mM of SNP consists
Figure 2. SNP blocks H2S induced contractile activity of rat uteri.
Both (A) and (B) show the spontaneous and Ca2+ induced contractile
activity of rat uteri treated with H2S (A) or SNP (B). The contractile
activity was expressed as the relative ratio between the mean height
peak of the untreated control and the treated uteri. The data are
expressed as the mean SEM (n = 8). (C) The effects of 1 mM H2S,
1 mM SNP and the combination of both on the spontaneous and Ca2+
induced uterus contractions.
Figure 3. 14N NMR spectroscopic analysis of the reaction between
200 mM Na2S and 100 mM SNP at pH 7.4. (A) The entire range of
the 14N NMR spectra for 100 mM SNP in 300 mM KPi at pH 7.4
(blue line), and the spectrum of the reaction mixture containing 100
mM of both SNP and Na2S in 300 mM KPi at pH 7.4 (red line). (B)
The time-dependent detection of N2O formation at 10 °C based on its
characteristic chemical shifts at −142 and −225 ppm.
concentration of 100 μM. This further confirmed that NO was
not the intermediate released in the reaction of SNP and H2S.
Product Identification. To identify the reaction products,
we first analyzed the reaction mixture using ultrahigh resolution
ESI-TOF MS (maXis, Bruker) with a low-temperature
ionization (+4 °C) to prevent any potential temperature-
induced decomposition of the products. A negative mode mass
spectrum of SNP revealed a molecular ion peak at m/z 238.9,
which was assigned to Na[Fe(CN)5(NO)]− (Supporting
Information Figure S2). However, upon addition of H2S, all
of the peaks for SNP disappeared, which indicates the blue
reaction product either does not ionize or fully decomposes
under experimental conditions.
To gain further insight into the reaction mechanism and
products, we performed GC-MS and 14N NMR measurements.
A 2 mM SNP solution at pH 7.4 (corresponding to 6 μmol of
SNP) was mixed with H2S in a 1:1, 1:2, and 1:5 ratio, and the
gas phase of the reaction mixture was analyzed (15 min after
the reaction began) by gas chromatography with a triple-
quadrupole mass spectrometer detector. Monitoring was
performed in the selected ion mode for ions 44 and 30
(corresponding to N2O and NO, respectively). The results,
of two broad peaks with chemical shifts at −3.7 and −90 ppm
that we assigned, using the 15N NMR table for metal
complexes,27 to the nitrogen atoms from NO and CN−
coordinated to paramagnetic Fe3+, respectively. The appearance
of additional peaks was noticed 60 min after the addition of
H2S. Two of them, −142 and −225 ppm, were assigned to N2O
(Figure 3B).28 The N2O signals increased over time and finally
disappeared after 15 h. The peak disappearance was presumably
caused by the fact that the measurement was done under air in
an unsealed tube. These results are in agreement with the GC-
MS data. More importantly, the third signal in the NMR
spectra was present in the more negative region (−355 ppm),
which is characteristic of N-coordinated thiocyanate (SCN−)
ligands (Figure 3A).27
Mechanistic Insights. To obtain additional mechanistic
insights, we followed the reaction in solution amperometrically
using a selective H2S electrode and via both time-resolved UV/
vis spectroscopy and real-time FTIR spectroscopy. When
C
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 4. Kinetic analysis of the reaction between SNP (200 μM) and Na2S (25 mM) in 300 mM KPi pH 7.4. (A) The monitoring of this reaction
(under aerobic conditions) using time-resolved UV/vis spectroscopy coupled to a stopped-flow device (total observation time 2.6 s, integration time
2.6 ms). Inset: spectra of the intermediates (1 = [(CN)5FeN(O)SH]3−, (2 = mixture of a Prussian blue type compounds, transient
[(CN)5Fe(HNO)]3−), and the final product (3 = [FeII(CN)5−x(SCN)x(H2O)]3−) over the course of the reaction based on the global spectra
analysis (SpecFit). (B) The second reaction step (the entire reaction is presented in Figure 5A), which shows constant decrease and shifting of the
535 nm peak and the appearance of an absorbance at ∼720 nm. The first reaction step, i.e., the formation of the 535 nm species, was too fast to be
detected. (C) The third reaction step during which a further shift and buildup of the 575 nm species is observed with a decrease in the absorbance
maximum at 720 nm. (D) A kinetic trace of the peak at 575 nm clearly shows the induction period for the catalytic reactions.
injected into a buffered solution, H2S disappears over time due
to its slow oxidation in ambient air. The presence of SNP in the
buffered solutions (pseudo-first-order conditions: 50 μM H2S
and 500 μM−10 mM SNP) led to the rapid removal of H2S,
which was determined to be concentration dependent
(Supporting Information Figure S4). An obvious decrease in
the initial electrode signal with increasing SNP concentration
indicates the reaction is faster than the electrode response time.
Therefore, the observed rates could have been underestimated.
However, the linear dependence of the rate of H2S removal on
the SNP concentration can at least qualitatively imply a first-
order reaction with respect to SNP.
To gain further mechanistic insights and detect possible
intermediate species in the reaction pathway, we used rapid
scan UV/vis stopped-flow measurements. The reaction was
performed under pseudo-first-order conditions by applying an
excess of H2S. Unlike the studies described in the
literature,11,15,16 we used a physiologically relevant pH 7.4
and worked under aerobic conditions.
iron(II) observed as the pink product of SNP reactions with
other thiols (Supporting Information Figure S1).11−13 This
intermediate can be defined as the HSNO−iron adduct,
[(CN)5FeN(O)SH]3−, assuming the pKa value for the
coordinated HSNO is 10.5, as previously reported.16 This
reaction showed a dependence on the H2S concentration. The
intensity of the UV/vis absorption band at ca. 535 nm was
strongly dependent on the H2S concentration, which suggests
that the [(CN)5FeN(O)SH]3− adduct formation is a reversible
process in accordance with previous observations.16
Next, we focused on the fate of this adduct, which proved to
be short-lived at pH 7.4, as it rapidly transformed into a final
blue species with an absorbance maximum at 575 nm and a
broad UV/vis band (shoulder) at ca. 390 nm (Figure 4B,C).
Additional transient species appeared on the transformation
pathway (which were not previously observed) that exhibited
absorbance maximum at ca. 720 and 440 nm (inset in Figure
4A).
Although a species with an absorbance maximum at 575 nm
was ascribed to [(CN)5FeN(O)S]4− (deprotonated form of the
initial [(CN)5FeN(O)SH]3− adduct) in the recent literature,16
it should be noted that some previous work29 on light-induced
SNP reactions with SCN− reported the exact same spectral
features (absorption maximum at ∼575 nm). This finding
The time-resolved spectra (Figure 4A) revealed the rapid
formation (within the dead time of the stopped-flow device, ca.
3 ms) of an intermediate with an absorption maximum at 535
nm for the first reaction step (Figure 4B). This absorption
maximum is characteristic of the RSNOs coordinated to
D
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
implies a thiocyanate coordination occurred as a result of the
photoinduced NO release.29 In general, the resulting blue−
violet color and an absorbance maximum at 575 nm are typical
of Fe(II)−thiocyanato species,30 especially those with more
SCN− ligands (between 4 and 6) that can even be maintained
under aerobic conditions (also while starting from the Fe(III)
species).31 The same product (Supporting Information Figure
S5) could be obtained by mixing [FeII(CN)5(H2O)]3− with an
excess of SCN−. Furthermore, our 14N NMR studies contained
a signal at −355 ppm that is characteristic of N-coordinated
thiocyanate (Figure 3A). The consistent, continuous shifting of
the 535 nm band toward longer wavelengths during the
transformation of the initial [(CN)5FeN(O)SH]3− adduct
(Figure 4B,C) could imply a consecutive increase in the
number of coordinated SCN− ions on the Fe(II) center. Similar
behavior has been reported for the binding of thiocyanate to
another type of iron center.32
The 720 nm band for the above-mentioned transient species
is characteristic of a mixed-valent cyanobridged complex (a
Prussian blue type compound) that could be independently
obtained either from the reaction of a qua
[FeII(CN)5−x(SCN)x(H2O)]3− complexes in the presence of
oxygen or upon addition of the corresponding Fe(III)
complexes.30 The UV/vis band at approximately 440 nm can
be attributed to transient [(CN)5Fe(HNO)]3− (inset in Figure
4A).33
The transformation of the initial [(CN)5FeN(O)SH]3−
adduct into the final thiocyanato species was faster with
increasing H2S and oxygen concentrations. However, the
concentration dependence proved to be rather complicated
−
because of (a) the cross-interaction between H2S (i.e., HSx /
•2−
HS(x+1)
and oxygen, (b) the involvement of oxygen in the
Figure 5. Real-time IR spectroscopic study of the reactions between
150 mM Na2S and 100 mM SNP at pH 7.4. (A) IR spectral changes
observed upon the addition of Na2S into a buffered SNP solution.
(B,C) The time dependence of the characteristic stretching vibrations
(2058 and 2144 cm−1 on B and 2091 cm−1 on C) upon the addition of
an equimolar concentration of Na2S into a buffered 100 mM SNP
solution as measured by reaction IR spectroscopy.
generation of the transient mixed-valent Fe(II)/Fe(III) species,
−
•2−
and (c) the involvement of the HSx /HS(x+1) species in the
transformation of CN− to SCN−.34 The detailed quantification
of the H2S and O2 concentration-dependent kinetic behavior
would require a separate study. However, the kinetic trace of
the peak at 575 nm clearly shows the induction period,
characteristic for the catalytic reactions (Figure 4D).14
pentafluorobenzyl bromide as the alkylating agent and
tetrabutylammonium sulfate as the phase-transfer catalyst36
(Supporting Information Figure S6). The data shown in Table
1 demonstrate that SCN− could be clearly detected in all of the
The proposed mechanism is in good agreement with our
solution in real-time FTIR measurements, which monitored the
characteristic vibrations of SNP in the 1700−2500 cm−1 region.
Namely, the IR spectrum of SNP in the solution contained
three particularly intense bands at 1924, 2143, and 2258 cm−1
assigned to the NO, radial CN−, and axial CN− stretching
frequencies, respectively, as described previously.29 The
addition of H2S (in 4-fold excess) caused an immediate change
that led to the disappearance of the NO (1924 cm−1) and radial
CN− (2144 cm−1) stretching bands with a concomitant
formation of two new peaks at 2056 and 2091 cm−1 (Figure
5A). While the first of these peaks reached a maximum after a
few min and remained in solution for over a few hours (Figure
5B), the latter completely disappeared after 2 min (Figure 5C).
On the basis of the overall results, we propose that the first
band corresponds to the coordinated thiocyanate, as reported
previously,35 while the latter could originate from the formation
of an intermediate species with an absorbance maximum at 720
nm. These results imply a thiocyanate-generating activity
whenever [Fe(CN)5−x(SCN)x(H2O)]3− with x ≤ 5 is formed,
while HNO is released concomitantly.
Table 1. Detection of SCN− Formed during the Reaction of
a
1 mM SNP with 2, 5, or 10 mM H2S
SNP: H2S ratio
5 min (μM)
24 h (μM)
1:2
200
467
511
5
7
7
315 12
856 11
1:5
1:10
1038
8
a
The samples were analyzed both 5 min and 24 h after the reaction
had started. SCN− was detected using an extractive alkylation protocol
involving the GC-MS detection of the final pentafluorobenzyl
thiocyanate.
samples with a yield dependent on the H2S concentration. The
relatively low amounts of SCN− detected are in an agreement
with the fact that only free thiocyanate anions that transfer into
the ethylacetate phase can be detected. The total amount of
SCN− increased significantly when the reaction mixtures were
left overnight, which indicates the labile Fe(II) complex
decomposed and formed freer SCN− in solution.
To further prove that SCN− formation occurs, 1 mM SNP
samples treated with a 2-, 5-, and 10-fold excess of H2S were
analyzed by GC-MS via an extractive alkylation with
E
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Intracellular HNO Detection and ex Vivo CGRP
Release Induced by the SNP/H2S Reaction Mixture.
Our final goal was to confirm the formation of HNO from the
SNP and H2S mixture in a biological milieu. To accomplish this
goal, we used a recently developed Cu2+-based fluorescent
sensor, CuBOT1, for the intracellular HNO detection.37
Human umbilical vein cells were loaded with a 10 μM dye
for 15 min, washed three times, and treated with 100 μM SNP,
100 μM H2S, or the mixture of both for 15 min. The
intracellular formation of the fluorescent Cu+BOT1 was
visualized by fluorescence microscopy. Neither H2S nor SNP
caused any detectable change in the fluorescent intensity;
however, the combination of these two molecules increased the
fluorescence (Figure 6). For cells previously loaded with NO
Figure 7. HNO-induced release of CGRP from an isolated mouse
heart upon stimulation with a reaction mixture containing 500 μM
SNP and 1 mM Na2S.
logical pH and under aerobic conditions, and based on the
results presented herein, the overall reaction mechanism can be
depicted as shown in Scheme 2. SNP reacts very rapidly with
H2S to form the intermediate [(CN)5FeN(O)SH]3−, which is
further transformed into the thiocyanate product(s). During the
second reaction step, H2S plays a catalytic role. Upon reducing
the coordinated HSNO/SNO−, it initially oxidizes to disulfide,
which can oxidize to form the polysulfides used in the reaction
progress. As a result of this step, a [(CN)5Fe(HNO)]3−
intermediate formed, as indicated by the time-resolved UV/
vis spectral analysis (440 nm band, see inset in Figure 4A). The
reaction of thiols (RSH) and H2S with S-nitrosothiols
(R′SNO) to generate HNO and disulfides (RSSR′) is a well-
established process.18,42 [(CN)5Fe(HNO)]3− is prepared by
following the literature protocol33 using SNP and 2 equiv of
dithionite. This complex proved to be stable toward H2S
(Supporting Information Figure S7); however, it could react
with S−S bond containing compounds and convert the CN−
into SCN− as for any cyanide species.34 We have shown
through independent experiments that [(CN)5Fe(HNO)]3−
reacts with polysulfides to generate this thiocyanate product
with an absorption maxima of ca. 575 nm and shoulder at ca.
390 nm (Supporting Information Figure S7). This process
most likely labilizes HNO, which can be released and/or further
dimerize into N2O. We were able to demonstrate the
production of both N2O (Figure 3 and Supporting Information
Figure S3) and HNO (Figure 6), whereas no evidence for the
release of NO could be found. Upon release of HNO, the aqua
complex [(CN)4(SCN)Fe(H2O)]3− is formed and is a source
of mixed-valent cyanobridged adducts due to the inevitable
existence of some Fe(III) species under aerobic conditions.
Both of these adducts, as well as any other cyanide containing
species, further react with the disulfide/polysulfides generated
during the CN− to SCN− conversion of the ligands. Although a
thorough kinetic analysis would require a separate study, the
kinetic traces, which correspond to the formation of the final
product, are indicative of an induction period (Figure 4D). This
period is related to the catalytic nature of the process and the
preformation of mixed-valent cyanobridged adducts. Therefore,
the duration of the induction period also depends on the H2S
and O2 concentrations with shorter durations corresponding to
an increased H2S concentration and longer durations to an
increased O2 concentration.
Figure 6. Intracellular fluorescence in HUVECs, caused by HNO
formation. After loading with the HNO sensor, CuBOT1, the cells
were treated with either 100 μM Na2S, 100 μM SNP, or a combination
of both. The photomicrographs were taken at a ×40 magnification.
The representative photomicrograph for the cells treated with a
combination of Na2S and SNP are also shown at a ×100 magnification.
fluorescent sensor, DAF-FM-DA, treatment with the SNP and
H2S reaction mixture caused no significant changes in the
fluorescence intensity relative to the control (not shown). It is
worth mentioning that the reaction of HNO with oxygen gives
a product that could induce DAF fluorescence but only in the
absence of the physiological scavengers.38 To avoid this, we
treated the cells by adding the activators in a complete cell
medium to allow only potentially generated NO to be detected.
The results are in good agreement with our overall data and
support the observation that coordinated NO is not released.
One of the main biological markers for HNO formation is
the increased release of calcitonin gene-related peptide
(CGRP), a potent vasodilator.39 We tested our hypothesis by
monitoring the release of this peptide from an isolated mouse
heart. CGRP is a neuropeptide for chemosensory primary
afferent nerve fibers located in the epicardial surface of the
heart. Using a previously described experimental model,40 we
showed that 1 mM of H2S had no effect on CGRP release,
while 500 μM SNP induced a small increase that could have
originated from the reaction of SNP with intracellular H2S. The
combination of both H2S and SNP significantly increased the
CGRP release (p = 0.028, n = 6, Wilcoxon matched pairs test,
Figure 7), which strongly confirmed the HNO formation.
Reaction Mechanism and Its Physiological Conse-
quences. Despite the long interest in the reactions of H2S with
SNP (since 1850),41 the reaction mechanism has eluded
researchers. This is the first study to combine chemical tools
with pharmacological/physiological experiments at a physio-
Overall, all these data are indicative of the operation of a new
previously undescribed mechanism (Scheme 2) with physio-
F
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 2. Mechanistic Depiction of the Reaction between SNP and H2S at pH 7.4 under Aerobic Conditions
a
SNP reacts very rapidly with H2S to form an intermediate [(CN)5FeN(O)SH]3−. In the second reaction step, H2S reduces the coordinated HSNO/
SNO− and initially becoming oxidized to disulfide, which can be further oxidized to polysulfides as the reaction progresses. As a result of this step,
the intermediate [(CN)5Fe(HNO)]3− is formed and is unreactive towards H2S but does react with polysulfides to form a thiocyanate adduct upon
the elimination of HNO. HNO can dimerize to yield N2O, while the other coordinated cyanides are consecutively transformed into thiocyanates
(main reaction path, a). Oxygen can oxidize the iron(II) center and create the mixed-valent bridged complexes known as Prussian blue, which are
short-lived and further react with polysulfides to give the final thiocyanate products as well (minor reaction route, b).
logically important features: the release of HNO at a
physiological pH = 7.4 and the conversion of cyanides into
thiocyanates with the latter generally being catalized in vivo by
rhodaneses.43,44 Rhodanese is the common name for a
thiosulfate:cyanide sulfurtransferase class of enzymes that use
thiosulfate to catalyze the detoxification of cyanide.44 This
enzyme is found in mitochondria, and the mechanism of its
action is similar to what we observed for the reaction of SNP
with H2S:persulfides (formed during the reaction of thiosulfate
with the free protein thiols in the enzyme) reacting with
cyanide to form thiocyanate.44
effective, and less toxic therapeutic alternative to the combining
of SNP with thiosulfate. Furthermore, by accounting for the
recently revised plasma levels of H2S (in the micromolar
range),59 one can speculate that both the lack of toxicity and
part of the vasodilatory effects of SNP could be due to reactions
with circulatory H2S via a rhodanese-like mechanism.
Finally, these data shed new light on prevailing assumptions
about the interplay of NO/H2S and explain the nitroxyl-like8
and thiol-modifying effects7 observed in pharmacological
studies using SNP as an NO donor. However, the nature of
the direct reaction of NO with H2S has yet to be characterized.
The formation of HNO is important because it is considered
to be a very promising therapeutic agent for cardiovascular
diseases.45,46 As a redox congener of NO, HNO reacts in the
cell orthogonally to NO, which stimulates CGRP release47
among other things and accounts for the positive inotropic
heart effects observed when an HNO donor was used.39,48 It
has been shown that, unlike NO donors, HNO donors do not
induce a nitrate-tolerance.49 Because the intracellular produc-
tion of HNO is still a matter of debate, with evidence showing
its production from NO via NO synthase,50 cytochrome c,51
and MnSOD52 or from hydroxylamine by heme-containing
proteins,53 the synthesis of donors that could release HNO
under physiological conditions is one of a current pharmaco-
logical goals.54−56
SNP is still used in acute hypertensive crises to regulate
blood pressure; however, it must be combined with thiosulfate
to minimize the toxic effects of cyanide through natural
rhodanese activity.57,58 Our results offer a new perspective on
the clinical application of SNP. The combination of SNP with
H2S, which causes the HNO-induced release of CGRP and
transforms toxic cyanide into thiocyanate, could be a new, more
EXPERIMENTAL SECTION
■
Material. Unless stated otherwise, all chemicals were purchased
from Sigma Aldrich. Na2S was purchased in the anhydrous form and
was both opened and stored in a glovebox (<1 ppm O2 and <1 ppm
H2O). The solutions were prepared and handled as previously
described.60
Amperometric Detection of NO and H2S. The fate of H2S and
NO was monitored throughout the course of the reaction using 2 mm
shielded H2S sensors and ISO-NO probes connected to a Free Radical
Analyzer (World Precision Instruments).18
GC-MS Detection of N2O. GC-MS analyses were performed on a
Bruker GC 450 TQ MS 300. The gas chromatograph was equipped
with a Varian VF-5m capillary column. Headspace gas samples 50 μL
in volume were injected in the splitless mode. The following oven
temperature program was used with helium as the carrier gas: ramped
from 50 to 155 °C at a rate of 10 °C/min and then to 260 °C at a rate
of 30 °C/min. The positive electron impact ionization mode was used.
The detector multiplier voltage was set to 1400 V, and the detection
was performed by selected ion monitoring of m/z 44 (N2O) and m/z
30 (NO) with a dwell time of 50 ms and SIM scan width of 0.7 au The
areas under the peaks were determined using software provided by the
manufacturer.
G
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
GC-MS Detection of SCN−. Detection of thiocyanate formation
was performed following the extractive alkylation procedure.36 Briefly,
tetrabutylammonium sulfate (TBA) was used as phase-transfer catalyst
with pentafluorobenzyl bromide (PFB-Br) as the derivatizing agent.
2,5-Dibromotoluene (DBT) was used as an internal standard for the
quantitation of SCN. The products were analyzed by gas
chromatography−mass spectrometry on a Bruker GC 450 TQ MS
300.
ASSOCIATED CONTENT
* Supporting Information
■
S
Color changes of SNP solution (300 mM KPi, pH 7.4) upon
addition of H2S, cysteine and gluathione. Ultra-high resolution
cryospray (4 °C) ESI-TOF-mass spectrometry in negative
mode of SNP and reaction mixture. GC-MS detection of N2O
generation from SNP and H2S reaction mixture (1:5).
Representative traces of H2S electrode responses upon addition
of 50 μM Na2S into buffered solution containing increasing
concentrations of SNP (0−10 mM). The spectrum of
[Fe(H2O)(CN)5]3− before and after the addition of the excess
of potassium thiocyanate. GC-MS analysis of the thiocyanate
coupled to pentafluorobenzyl bromide in extractive alkylation
process. The spectrum of [Fe(CN)5HNO]3− generated by
total reduction of SNP with 2 equiv of dithionite at pH 7.4.
Schematic representation of the experimental setup for CGRP
release from isolated heart as described in Materials and
Methods. This material is available free of charge via the
Internet at http://pubs.acs.org.
14N NMR. A 25 mM solution of SNP both with and without the
addition of 25 mM Na2S in 300 mM KPi pH 7.4 was placed in a 10
mm NMR tube, and the spectra were recorded using a Bruker 400
MHz spectrometer using the 15N signal of nitromethane as a reference.
All of the spectra were recorded at 10 °C and at 50.67 MHz. Two
hundred transients were collected with a 35° pulse width and a 5 s
relaxation delay.
UHR ESI TOF MS. Samples were analyzed using a maXis (Bruker
Daltonics), ultrahigh-resolution (UHR) ESI-TOF mass spectrometer
operating in both the negative and positive ion modes. Cryospray
ionization was used to prevent sample decomposition during
ionization. The data analysis, mass convolution, and spectral
simulations were performed using the Data Analysis software provided
by the manufacturer. The instrument parameters are listed in the figure
legends.
AUTHOR INFORMATION
Corresponding Author
*Phone, +49-(0)9131-8527391; E-mail, milos.filipovic@
chemie.uni-erlangen.de (M.R.F.). E-mail, ivana.ivanovic@
chemie.uni-erlangen.de (I.I.-B.).
■
Stopped-Flow Spectrometry. Kinetic data were obtained by
recording the time-resolved UV/vis spectra using a modified μSFM-20
Bio-Logic stopped-flow module combined with a Huber CC90
cryostat and equipped with a J&M TIDAS high-speed diode array
spectrometer containing both deuterium and tungsten lamps (200−
1015 nm wavelength range) as previously described.60
Author Contributions
Cell Culture and NO• and HNO Bioimaging Studies. Human
umbilical vein endothelial cells (HUVEC, passage 2−3) were obtained
from PromoCell GmbH (Heidelberg, Germany) and cultured in 35
mm μ-dishes (ibidi, Martinsried, Germany) using a commercially
available cell growth medium (PromoCell GmbH) at 37 °C and 5%
CO2. For NO detection, the cells were preloaded for 10 min with 10
μM DAF-FM. For HNO detection, the HUVECs were incubated for
20 min with 10 μM CuBOT1.18,37 Fluorescence microscopy was
performed using an inverted microscope (Axiovert 40 CLF, Carl
Zeiss) equipped with green fluorescent filters and an AxioCam. The
images were processed using ImageJ software.
CGRP Release from the Heart. The release of CGRP from the
isolated heart of C57BL/6 mice was measured as previously
reported.40 Briefly, the mice were suffocated in a pure CO2 atmosphere
and the complete heart was excised by cutting the central blood
vessels. The procedures were approved by the animal protection
authorities (of the Regierung von Mittelfranken, Ansbach, Germany).
The hearts were passed through a series of synthetic interstitial fluid
incubations both with and without the addition of the test compounds
and spent 5 min in each test tube at 37 °C (for details see Supporting
Information Figure S8). The CGRP concentration was measured via
enzyme immunoassay (Bertin Pharma, Montigny, France).
Organ Bath Studies on Isolated Uterus. Isolated uteri from
virgin Wistar rats (200−250 g) in the oestrus phase of the estrous
cycle, as determined by examining a daily vaginal lavage, were used.61
All of the rats were sacrificed by cervical dislocation following a
procedure approved by the animal protection authorities (University
of Belgrade, Serbia). The uterine horns were rapidly excised and
carefully cleaned of any surrounding connective tissue and mounted
vertically in a 10 mL volume organ bath containing De Jalon’s solution
(in g·L−1: NaCl 9.0, KCl 0.42, NaHCO3 0.5, CaCl2 0.06, and glucose
0.5.) aerated with 95% oxygen and 5% carbon dioxide at 37 °C. The
uteri, either acting spontaneously or being contracted with Ca2+ (6
mM), were allowed to equilibrate at 1 g tension before adding the
experimental drugs. Both H2S and SNP were added cumulatively. The
myometrial tension was recorded isometrically using a TSZ-04-E
isolated organ bath and transducer (Experimetria, Budapest, Hungary).
Each concentration of the activating substance studied was allowed to
act for 15 min. Overall, 7−10 uteri were used per experiment.
The manuscript was written through the contributions of all
authors. All authors have given their approval of the final
version of this manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to Professor Steven Lippard and Dr. Maksim
Royzen (MIT) for providing the HNO fluorescent sensors,
CuBOT1, a careful review of this manuscript, and their helpful
discussions. This work was supported by an intramural grant
from FAU Erlangen−Nuremberg (Emerging Field Initiative:
Medicinal Redox Inorganic Chemistry). V.P. acknowledges the
fellowship from DAAD. A.M. and Z.O.D. thank the Serbian
Ministry of Education and science grant OI 173014.
ABBREVIATIONS USED
■
SNP, sodium nitroprusside; CGRP, calcitonin gene-related
peptide; RSNO, S-nitrosothiols; DAF-FM-DA, 4-amino-5-
methylamino-2′,7′-difluorofluorescein diacetate; MnSOD, man-
ganese superoxide dismutase
REFERENCES
■
(1) Li, L.; Hsu, A.; Moore, P. K. Actions and interactions of nitric
oxide, carbon monoxide and hydrogen sulfide in the cardiovascular
system and in inflammationa tale of three gases! Pharmacol. Ther.
2009, 123, 386−400.
(2) Li, L.; Rose, P.; Moore, P. K. Hydrogen sulfide and cell signaling.
Annu. Rev. Pharmacol. Toxicol. 2011, 51, 169−187.
(3) Yang, G.; Wu, L.; Jiang, B.; Yang, W.; Qi, J.; Cao, K.; Meng, Q.;
Mustafa, A. K.; Mu, W.; Zhang, S.; Snyder, S. H.; Wang, R. H2S as a
physiologic vasorelaxant: hypertension in mice with deletion of
cystathionine gamma-lyase. Science 2008, 322, 587−559.
(4) Mustafa, A. K.; Sikka, G.; Gazi, S. K.; Steppan, J.; Jung, S. M.;
Bhunia, A. K.; Barodka, V. M.; Gazi, F. K.; Barrow, R. K.; Wang, R.;
Amzel, L. M.; Berkowitzm, D. E.; Snyder, S. H. Hydrogen sulfide as
H
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
endothelium-derived hyperpolarizing factor sulfhydrates potassium
channels. Circ. Res. 2011, 109, 1259−1268.
(5) Szabo, C. Hydrogen sulfide and its therapeutic potential. Nature
Rev. Drug Discovery 2007, 6, 917−935.
(24) Hennan, J. K; Diamond, J. Evidence that spontaneous
contractile activity in the rat myometrium is not inhibited by NO-
mediated increases in tissue levels of cyclic GMP. Br. J. Pharmacol.
1998, 123, 959−967.
(25) Demirkoprulu, N.; Cetin, M.; Bagcivan, I.; Kaya, T.; Soydan, A.
S.; Karadas, B.; Cetin, A. Comparative relaxant effects of YC-1 and
DETA/NO on spontaneous contractions and the levels of cGMP of
isolated pregnant rat myometrium. Eur. J. Pharmacol. 2005, 517, 240−
245.
(6) Ali, M. Y.; Ping, C. Y.; Mok, Y. Y.; Ling, L.; Whiteman, M.;
Bhatia, M.; Moore, P. K. Regulation of vascular nitric oxide in vitro and
in vivo; a new role for endogenous hydrogen sulfide? Br. J. Pharmacol.
2006, 149, 625−634.
(7) Yong, Q. C.; Cheong, J. L.; Hua, F.; Deng, L. W.; Khoo, Y. M.;
Lee, H. S.; Perry, A.; Wood, M.; Whiteman, M.; Bian, J. S. Regulation
of heart function by endogenous gaseous mediatorscrosstalk
between nitric oxide and hydrogen sulfide. Antioxid. Redox Signaling
2011, 14, 2081−2091.
(8) Yong, Q. C.; Hu, L. F.; Wang, S.; Huang, D.; Bian, J. S. Hydrogen
sulfide interacts with nitric oxide in the heart: possible involvement of
nitroxyl. Cardiovasc. Res. 2010, 88, 482−491.
(9) Butler, A. R.; Megson, I. L. Non-heme iron nitrosyls in biology.
Chem. Rev. 2002, 102, 1155−1165.
(10) Roncaroli, F.; Videla, M.; Slep, L. D.; Olabe, J. A. New features
in the redox coordination chemistry of metal nitrosyls {M − NO+; M
− NO•; M − NO−(HNO)}. Coord. Chem. Rev. 2007, 251, 1903−
1930.
(11) Butler, A. R.; Calsy-Harrison, A. M.; Glidewell, C.; Sorenses, P.
E. The pentacyanonitrosylferrate ion-V. The course of the reactions of
nitroprusside with a range of thiols. Polyhedron 1988, 7, 1197−1202.
(12) Grossi, L.; D’Angelo, S. Sodium nitroprusside: mechanis, of NO
release mediated by sulfhydryl-containing molecules. J. Med. Chem.
2005, 48, 2622−2626.
(13) Johnson, M. D.; Wilkins, R. G. Kinetics of the primary
interaction of pentacyanonitrosylferrate (2−) (nitroprusside) with
aliphatic thiols. Inorg. Chem. 1983, 23, 231−235.
(14) Szaciłowski, K.; Wanat, A.; Barbieri, A.; Wasielewska, E.; Witko,
M.; Stochel, G.; Stasicka, Z. Reactions of the [Fe(CN)5NO]2−
complex with biologically relevant thiols. New J. Chem. 2002, 26,
1495−1502.
(15) Rock, P. A.; Swinehart, J. H. The kinetics of the aqueous
hydrogen sulfide−nitroprusside system. Inorg. Chem. 1965, 5, 1078−
1079.
(26) Wink, D. A.; Feelisch, M. Formaton and Detection of Nitroxyl
and Nitrous Oxide. In Methods in Nitric Oxide Research; Feelisch, M.,
Stamler, J. S., Eds.; John Wiley & Sons, Ltd: Chichester, England,
1996; pp 403−412.
(27) Mason, J.; Larkworthy, L. F.; Moore, E. A. Nitrogen NMR
spectroscopy of metal nitrosyl and related compounds. Chem. Rev.
2002, 102, 913−934.
(28) Strelenko, Yu. A.; Sergeyev, N. M. Isotope effect due to 15N/14N
source substitution on 14N−14N coupling constants in nitrous oxide. J.
Mol. Struct. 1996, 378, 61−65.
(29) Swinehart, J. H. The nitrprusside ion. Coord. Chem. Rev. 1967, 2,
385−402.
(30) Roncaroli, F.; Olabe, J. A.; van Eldik, R. Kinetic and mechanism
of the formation of nitroprusside from aquapentacyanoferrate(III) and
NO: complex formation controlled by outer-sphere electron transfer.
Inorg. Chem. 2002, 41, 5417−5425.
(31) Kratochvil, B.; Long, R. Iron (III)−(II)couple in acetonitrile.
Oxidation of thiocyanate by iron (III). Anal. Chem. 1970, 42, 43−46.
(32) Sarauli, D.; Popova, V.; Zahl, A.; Puchta, R.; Ivanovic-
Burmazovic, I. Seven-coordinate iron complex as a ditopic receptor
for lithium salts: study of host−guest interactions and substitution
behaviour. Inrog. Chem. 2007, 46, 7848−7860.
(33) Montenegro, A. C.; Amorebieta, V. T.; Slep, L. D.; Martín, D.
F.; Roncaroli, F.; Murgida, D. H.; Bari, S. E.; Olabe, J. A. Three redox
states of nitrosyl: NO+, NO−, and NO−/HNO interconvert reversibly
on the same pentacyanoferrate(II) platform. Angew. Chem. 2009, 48,
4213−4216.
(34) Luthy, R. G.; Bruce, S. G., Jr. Kinetics of reaction of cyanide and
reduced sulfur species in aqueous solution. Environ. Sci. Technol. 1979,
13, 1481−1487.
(16) Quiroga, S. L.; Almaraz, A. E.; Amorebieta, V. T.; Perissinotti, L.
L.; Olabe, J. A. Addition and redox reactivity of hydrogen sulfides
(H2S/HS−) with nitroprusside: new chemistry of nitrososulfide
ligands. Chem.Eur. J. 2011, 17, 4145−4156.
(35) Ivanovic-Burmazovic, I.; Hamza, M. S. A.; van Eldik, R. A
detailed mechanistic study of the substitution behaviour of an unusual
seven-coordinate iron(III) complex in aqueous solution. Inorg Chem.
2002, 41, 5150−5161.
(17) Filipovic, M. R.; Ivanovic-Burmazovic, I. The kinetics and
character of the intermediates formed in the reaction between sodium
nitroprusside and hydrogen sulfide need further clarification. Chem.
Eur. J. 2012, 18, 13538−13540.
(36) Paul, B. D.; Smith, M. L. Cyanide and thiocyanate in human
saliva by gas chromatography−mass spectrometry. J. Anal. Toxicol.
2006, 30, 511−515.
(37) Rosenthal, J.; Lippard, S. Direct Detection of Nitroxyl in
Aqueous Solution using a Tripodal Copper(II) BODIPY Complex. J.
Am. Chem. Soc. 2009, 132, 5536−5537.
(18) Filipovic, M. R.; Miljkovic, J.Lj.; Nauser, T.; Royzen, M.; Klos,
K.; Shubina, T.; Koppenol, W. H.; Lippard, S. J.; Ivanovic-Burmazovic,
́ ́
I. Chemical Characterization of the Smallest S-Nitrosothiol, HSNO;
Cellular Crosstalk of H2S and S-Nitrosothiols. J. Am. Chem. Soc. 2012,
134, 12016−12027.
(19) Brand, E.; Harris, M. M.; Biloon., S. Cystinuria. The excretion of
cystine complex which decomposes in the urine with the liberation of
free cystine. J. Biol. Chem. 1930, 86, 315−331.
(20) Whiteman, M.; Li, L.; Kostetski, I.; Chu, S. H.; Siau, J. L.; Bhatia,
M.; Moore, P. K. Evidence for the formation of a novel nitrosothiol
from the gaseous mediators nitric oxide and hydrogen sulphide.
Biochem. Biophys. Res. Commun. 2006, 343, 303−310.
(21) Stamler, J. S.; Feelisch, M. Preparation and Detection of S-
Nitrosothiols. In Methods in Nitric Oxide Research; Feelisch, M.,
Stamler, J. S., Eds.; John Wiley & Sons, Ltd: Chichester, England,
1996; pp 521−539.
(22) Williams, D. L. H. The chemistry of S-nitrosothiols. Acc. Chem.
Res. 1999, 32, 869−876.
(23) Dicks, A. P.; Swift, H. R.; Williams, D. L. H.; Butler, A. R.; Al-
Sadoni, H. H.; Cox, B. G. J. Identification of Cu+ as the effective
reagent in nitric oxide formation from S-nitrosothiols (RSNO). J.
Chem. Soc., Perkin. Trans. 2 1996, 481−487.
(38) Espey, M. G.; Miranda, K. M.; Thomas, D. D.; Wink, D. A.
Ingress and reactive chemistry of nitroxyl-derived species within himan
cells. Free Radic. Biol. Med. 2002, 33, 827−834.
(39) Paolocci, N.; Saavedra, W. F.; Miranda, K. M.; Martignani, C.;
Isoda, T.; Hare, J. M.; Espey, M. G.; Fukuto, J. M.; Feelisch, M.; Wink,
D. A.; Kass, D. A. Nitroxyl anion exerts redox-sensitive positive cardiac
inotropy in vivo by calcitonin gene-related peptide signaling. Proc.
Natl. Acad. Sci. U. S. A. 2001, 98, 10463−10468.
(40) Strecker, T.; Reeh, P. W.; Weyand, M.; Messlinger, K. Release of
calcitonin gene-related peptide from the isolated mouse heart:
methodological validation of a new model. Neuropeptides 2006, 40,
107−113.
(41) Sidgwick, N. V. The Chemical Elements and Their Compounds II;
Oxford Univeristy Press: London, 1950; p 1345.
(42) Arnelle, D. R.; Stamler, J. S. NO+, NO and NO− donation by S-
nitrosothiols: implications for regulation of physiological functions by
S-nitrosulation and acceleration of disulfide formation. Arch. Biochem.
Biophys. 1995, 318, 279−285.
(43) Mueller, E. G. Trafficking in persulfides: delivering sulfur in
biosynthetic pathways. Nature Chem. Biol. 2006, 2, 185−194.
I
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(44) Cipollone, R.; Ascenzi, P.; Visca, P. Common themes and
variations in the rhodanese superfamily. IUBMB Life 2007, 59, 51−59.
(45) Flores-Santana, W.; Salmon, D. J.; Donzelli, S.; Switzer, C. H.;
Basudhar, D.; Ridnour, L.; Cheng, R.; Glynn, S. A.; Paolocci, N.;
Fukuto, J. M.; Miranda, K. M.; Wink, D. A. The specificity of nitroxyl
chemistry is unique among nitrogen oxides in biological systems.
Antioxid. Redox Signaling 2011, 14, 1659−1674.
(46) Irvine, J. C.; Ritchie, R. H.; Favaloro, J. L.; Andrews, K. L.;
Widdop, R. E.; Kemp-Harper, B. K. Nitroxyl (HNO): the Cinderella
of the nitric oxide story. Trends Pharmacol. Sci. 2008, 29, 601−608.
(47) Wink, D. A.; Miranda, K. M.; Katori, T.; Mancardi, D.; Thomas,
D. D.; Ridnour, L.; Espey, M. G.; Feelisch, M.; Colton, C. A.; Fukuto,
J. M.; Pagliaro, P.; Kass, D. A.; Paolocci, N. Orthogonal properties of
the redox siblings nitroxyl and nitric oxide in the cardiovascular
system: a novel redox paradigm. Am. J. Physiol.: Heart Circ. Physiol.
2003, 285, H2264−2276.
(48) Paolocci, N.; Jackson, M. I.; Lopez, B. E.; Miranda, K.;
Tocchetti, C. G.; Wink, D. A.; Hobbs, A. J.; Fukuto, M. The
pharmacology of nitroxyl (HNO) and its therapeutic potential: not
just the Janus face of NO. Pharmacol. Ther. 2007, 113, 442−458.
(49) Bullen, M. L.; Miller, A. A.; Andrews, K. L.; Irvine, J. C.; Ritchie,
R. H.; Sobey, C. G.; Kemp-Harper, B. K. Nitroxyl (HNO) as a
vasoprotective signaling molecule. Antioxid. Redox Signaling 2011, 14,
1675−1686.
(50) Schmidt, H. H.; Hofmann, H.; Schindler, U.; Shutenko, Z. S.;
Cunningham, D. D.; Feelisch, M. No NO from NO synthase. Proc.
Natl. Acad. Sci. U. S. A. 1996, 93, 14492−11197.
(51) Sharpe, M. A.; Cooper, C. Reactions of nitric oxide with
mitochondrial cytochrome c: a novel mechanism for the formation of
nitroxyl anion and peroxynitrite. Biochem. J. 1998, 332, 9−19.
(52) Filipovic, M. R.; Stanic, D.; Raicevic, S.; Spasic, M.; Niketic, V.
Consequences of MnSOD interactions with nitric oxide: nitric oxide
dismutation and the generation of peroxynitrite and hydrogen
peroxide. Free Radic. Res. 2007, 41, 62.
(53) Donzelli, S.; Espey, M. G.; Flores-Santana, W.; Switzer, C. H.;
Yeh, G. C.; Huang, J.; Stuehr, D. J.; King, S. B.; Miranda, K. M.; Wink,
D. A. Generation of nitroxyl by heme protein-mediated peroxidation
of hydroxylamine but not N-hydroxy-L-arginine. Free Radic. Biol. Med.
2008, 45, 578−584.
(54) DuMond, J. F.; King, S. B. The chemistry of nitroxyl-releasing
compounds. Antioxid. Redox Signaling 2011, 14, 1637−1648.
(55) Shoman, M. E.; DuMond, J. F.; Isbell, T. S.; Crawford, J. H.;
Brandon, A.; Honovar, J.; Vitturi, D. A.; White, C. R.; Patel, R. P.;
King, S. B. Acyloxy Nitroso Compounds as Nitroxyl (HNO) Donors:
Kinetics, Reactions with Thiols, and Vasodilation Properties. J. Med.
Chem. 2011, 54, 1059−1070.
(56) Keefer, L. K. Fifty Years of Diazeniumdiolate Research. From
Laboratory Curiosity to Broad-Spectrum Biomedical Advances. ACS
Chem. Biol. 2011, 6, 1147−1155.
(57) Bisset, W. I. K.; Butler, A. R.; Glidewell, C.; Reglinski, J. Sodium
nitroprusside and cyanide release: reasons for re-appraisal. Br. J.
Anaesth. 1981, 53, 1015−1018.
(58) Pasch, T.; Schulz, V.; Hoppelshauser, G. Nitroprusside-Induced
̈
Formation of Cyanide and Its Detoxication with Thiosulfate During
Deliberate Hypotension. J. Cardiovasc. Pharmacol. 1983, 5, 77−86.
(59) Shen, X.; Pattillo, C. B.; Pardue, S.; Bir, S. C.; Wang, R.; Kevil,
C. G. Measurement of plasma hydrogen sulfide in vivo and in vitro.
Free Radic. Biol. Med. 2011, 50, 1021−1031.
(60) Filipovic, M. R.; Miljkovic, J.; Allgauer, A.; Chaurio, R.; Shubina,
̈
T.; Herrmann, M.; Ivanovic-Burmazovic, I. Biochemical insight into
physiological effects of H2S: reaction with peroxynitrite and formation
of a new nitric oxide donor, sulfinyl nitrite. Biochem. J. 2012, 441,
609−621.
(61) Appiah, I.; Milovanovic, S.; Radojicic, R.; Nikolic-Kokic, A.;
Orescanin-Dusic, Z.; Slavic, M.; Trbojevic, S.; Skrbic, R.; Spasic, M. B.;
Blagojevic, D. Hydrogen peroxide affects contractile activity and anti-
oxidant enzymes in rat uterus. Br. J. Pharmacol. 2010, 158, 1932−1941.
J
dx.doi.org/10.1021/jm3012036 | J. Med. Chem. XXXX, XXX, XXX−XXX