Investigation of a novel trinuclear μ-oxo ruthenium complex as a potential nitric oxide releaser for biological purposes

Article abstract of DOI:10.1016/j.ica.2015.01.038

The chemical properties of the trinuclear ruthenium complex [Ru₃O(CH₃COO)₆(3-pic)₂(NO)]PF₆ (1, were 3-pic = 3-picoline or 3-methylpyridine), its photochemical behavior and its ability as a vasorelaxing agent are reported in this work. It was shown that the unpaired electrons of NO and of the metal center [Ru₃O]⁺ are tightly coupled, promoting an intermediate NMR profile between the oxidized (paramagnetic) and reduced (diamagnetic) [Ru₃O]^1+/0 species. From IR measurements, it was suggested that in trinuclear complexes, unlike other ruthenium compounds, the interaction of those unpaired electrons overcomes the effect of peripheral ligands and NO-backbonding on the ν(NO) frequency, which shows no dependence on the pK_a values of pyridinic ligands. The photoinduced NO release was investigated by light irradiation at 337, 447, 532 and 660 nm in phosphate buffer solution (pH 7.4), leading to the generation of gaseous NO⁰ and the solvated species [Ru₃O(CH₃COO)₆(3-pic)₂(H₂O)]⁺ (2). Through amperometric and quimioluminescence measurements, the amount of NO(g) released were determined, showing dependence on the wavelength of irradiation. The ability of compound 1 as a vasorelaxing agent was also addressed. It was shown that, under ambient luminosity, compound 1 promotes 89% of relaxation in pre-contracted rat aorta. Molecular modeling of the novel nitrosyl, as well as characterization of the complexes [Ru₃O(CH₃COO)₆(3-pic)₂(L)]ⁿ, L = H₂O, n = +1 (2), L = 3-pic, n = +1 (3), L = CO, n = 0 (4), are also reported.

Full text of DOI:10.1016/j.ica.2015.01.038

Inorganica Chimica Acta 429 (2015) 114–121
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Investigation of a novel trinuclear l-oxo ruthenium complex as a
potential nitric oxide releaser for biological purposes
Natacha Cacita a, Bruna Possato a, Camila F.N. da Silva a, Michele Paulo b, André Luiz Barbosa Formiga c,
Lusiane M. Bendhack b, Sofia Nikolaou a,
⇑
a Departamento de Química, Faculdade de Filosofia Ciências e Letras de Ribeirão Preto da Universidade de São Paulo, Av. Bandeirantes, 3900, CEP 14040-901, Monte Alegre,
Ribeirão Preto, SP, Brazil
b Faculdade de Ciências Farmacêuticas de Ribeirão Preto da Universidade de São Paulo, Av. do Café s/n°, CEP 14040-903, Monte Alegre, Ribeirão Preto, SP, Brazil
c Instituto de Química da Universidade Estadual de Campinas – UNICAMP, P.O. Box 6154, CEP 13083-970, Campinas, São Paulo, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
The chemical properties of the trinuclear ruthenium complex [Ru3O(CH3COO)6(3-pic)2(NO)]PF6 (1, were
3-pic = 3-picoline or 3-methylpyridine), its photochemical behavior and its ability as a vasorelaxing agent
are reported in this work. It was shown that the unpaired electrons of NO and of the metal center [Ru3O]+
are tightly coupled, promoting an intermediate NMR profile between the oxidized (paramagnetic) and
reduced (diamagnetic) [Ru3O]1+/0 species. From IR measurements, it was suggested that in trinuclear
complexes, unlike other ruthenium compounds, the interaction of those unpaired electrons overcomes
Received 2 September 2014
Received in revised form 18 December 2014
Accepted 26 January 2015
Available online 16 February 2015
Keywords:
the effect of peripheral ligands and NO-backbonding on the
m(NO) frequency, which shows no
Trinuclear ruthenium complex
Nitric oxide releaser
Photolysis
dependence on the pKa values of pyridinic ligands. The photoinduced NO release was investigated by
light irradiation at 337, 447, 532 and 660 nm in phosphate buffer solution (pH 7.4), leading to the
generation of gaseous NO0 and the solvated species [Ru3O(CH3COO)6(3-pic)2(H2O)]+ (2). Through
amperometric and quimioluminescence measurements, the amount of NO(g) released were determined,
showing dependence on the wavelength of irradiation. The ability of compound 1 as a vasorelaxing agent
was also addressed. It was shown that, under ambient luminosity, compound 1 promotes 89% of relax-
ation in pre-contracted rat aorta. Molecular modeling of the novel nitrosyl, as well as characterization
of the complexes [Ru3O(CH3COO)6(3-pic)2(L)]n, L = H2O, n = +1 (2), L = 3-pic, n = +1 (3), L = CO, n = 0 (4),
are also reported.
Vasorelaxing property
Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction
increasing the interest on the development of a variety of nitrosyl
ruthenium complexes [5,8–16].
In recent years, the possible applications of metal-nitrosyl com-
pounds in biological systems as potential NO scavengers or relea-
sers, has attracted much attention [1–6]. Light irradiation and
electrochemical reduction of coordinated NO+ are alternatives for
the selective release of NO from coordination compounds. Both
strategies take into account the low affinity of NO0 by some metal
centers [7].
Due to its similarities to iron, ruthenium is considered to have
low toxicity, which in general is related to the available oxidation
states. In this regard, ruthenium ions constitute an interesting
alternative to iron, because it can achieve various oxidation states
(II, III, IV) in physiological environment [8]. Also, its complexes
have the ability to act both as NO scavengers or releasers,
Also of interest to this work, the trinuclear complexes of general
formula [M3O(CH3COO)6(L)3]n (M = transition metal, L = solvent
or N-heterocyclic ligands) are part of an important class of
compounds of transition metals [17–22]. These complexes have a
triangular structure in which the metal ions are held together by a
l-oxo and carboxylate bridges (Fig. 1) [19,23]. In such compounds,
the proximity of the ruthenium atoms leads to a strong metal–metal
interaction, both magnetic and electronic, enhancing the electron
delocalization over the [Ru3O]+ core, which behaves like a single
metal center [24]. Besides the rich redox, electrochromic and cat-
alytic properties, species containing labile and/or bridging ligands
can act as building blocks to assemble supramolecular structures
[18,19,24–28]. The [Ru3O(CH3COO)6)] moiety also appears as an
appealing unit to react with NO0. It has been shown that the strong
interaction between the unpaired electrons of NO and [RuI3II,III,IIIO]
unit accounts for interesting properties [9,10].
⇑
Corresponding author.
E-mail address: sofia@ffclrp.usp.br (S. Nikolaou).
http://dx.doi.org/10.1016/j.ica.2015.01.038
0020-1693/Ó 2015 Elsevier B.V. All rights reserved.

N. Cacita et al. / Inorganica Chimica Acta 429 (2015) 114–121
115
Table 1
Recently, our group reported the very first example of a
trinuclear ruthenium complex which acts against cancer cells
[29]. Complex [Ru3O(CH3COO)6(4-pic)(NO)]PF6 lowers up to 90%
the viability of B16F10 melanoma cell when irradiated with visible
light (532 nm). It also presented some dark activity, which works
in synergism with NO release triggered by light, potentiating the
compound biological activity.
In this context, this work presents the synthesis and charac-
terization of the novel complex [Ru3O(CH3COO)6(3-pic)(NO)]PF6
(1), where 3-pic = 3-methylpyridine, Fig. 1. Although [Ru3O] com-
plexes have been extensively studied, there are very few examples
of this sort of compounds coordinated to NO. In this context, one of
our aims is to introduce in literature a new set of characterization
data in order to expand the number of described [Ru3O]-nitrosyls
monomeric units. Besides that, the presence of methyl groups pro-
vides some hydrophobicity to the whole system, an important
aspect to be addressed for biological applications. We also intend
to advance on the investigation of photoinduced NO release by
verifying the ability of 1 to delivery NO in physiological like medi-
um under irradiation with visible light. Finally, we present its
vasorelaxing capacity, probed by relaxation of pre-contracted rat
aorta. Characterization of the precursors [Ru3O(CH3COO)6(3-
pic)(L)]n, L = H2O, n = +1 (2), L = 3-pic, n = +1 (3), L = CO, n = 0 (4),
is also reported.
Absorption and electrochemical data for compounds 1–4 collected from acetonitrile
solutions.
Electronic spectra data
kmax (nm)/loge )
(molꢀ1 L cmꢀ1
IC
CLCT
CLCT
CLCT
[Ru3O(CH3COO)6(3-pic)2(NO)]+(1) 709/
542/
3.21
521
456/
3.28
460
372/
3.39
372
2.86*
753
Molecular modeling prediction
for (1)
[Ru3O(CH3COO)6(3-
679/
3.45
688/
3.73
587/
3.51
312/
3.76
313/
4.05
327/
3.75
pic)2(H2O)]+(2)
[Ru3O(CH3COO)6(3-pic)3]+ (3)
[Ru3O(CH3COO)6(3-pic)2(CO)] (4)
Electrochemical data
E1/2 (V vs. Ag/AgCl)
[Ru3O]ꢀ1/0
[Ru3O]0/1+
[Ru3O]1+/2+
[Ru3O(CH3COO)6(3-pic)2(NO)]+(1)
[Ru3O(CH3COO)6(3-pic)2(H2O)]+(2)
[Ru3O(CH3COO)6(3-pic)3]+ (3)
ꢀ0.77
ꢀ1.21
ꢀ1.31
ꢀ0.85
ꢀ0.06
0.039
ꢀ0.049
0.65
1.38
1.04
0.98
1.27
[Ru3O(CH3COO)6(3-pic)2(CO)] (4)
*
In the case of compound 1, the intra-cluster band has contribution of NO levels,
having some character of charge transfer as well.
2. Results and discussion
2.1. Electronic spectroscopy, molecular modeling and electrochemical
measurements
The data collected from the absorption spectra of compounds
1–4 are reported in Table 1. Compounds 2–4 behaves as expected,
following the well described characteristics of the trinuclear ruthe-
nium clusters [17,19]. More relevant to this discussion are the fea-
tures of compound 1.
The electronic spectrum of the complex [Ru3O(CH3COO)6(3-
pic)2(NO)]PF6 is shown in Fig. 2, being consistent with the profile
observed earlier for trinuclear ruthenium complexes coordinated
to nitric oxide [9,10]. The classical description of the electronic
structure of a trinuclear ruthenium complex has been done on a
qualitatively basis, in terms of the molecular orbital diagram pro-
posed by Cotton and Norman [30]. This diagram addresses the
Fig. 2. Electronic spectrum of complex [Ru3O(CH3COO)6(3-pic)2(NO)]PF6, collected
from a 10ꢀ3 M acetonitrile solution.
description, the electronic spectra of trinuclear ruthenium com-
plexes of +1 charge (a Ru3(III)(III)(III)O unit) are dominated by clus-
ter-to-ligand charge transfer bands (CLCT) in the UV and visible
region (below 500 nm), and a strong low-energy broad band
named intra-cluster (IC), around 700 nm, in which a metallic char-
acter predominates [17–19,23,31,32].
Although this description has worked well for a very large num-
ber of compounds [19], it seems to fail to describe the profile
observed in Fig. 2. It is well known that for reduced ([Ru3O]0) com-
pounds, both CLCT and IC transitions experience a large bathochro-
mic shift, the last one moving further than 900 nm [19]. The
[Ru3O]
p system under D3h symmetry for complexes of general for-
mula [Ru3O(CH3COO)6(L)3]n, and C2v for asymmetric complexes,
such as compound 1. The central oxygen atom is considered to
have sp2 hybridization, leaving a pz orbital of
p
symmetry free to
interact with the ruthenium dp orbitals [17,30]. Following this
coordination of a strong p-acceptor ligand such a CO to a reduced
[Ru3O]0 core promotes a noticeable hypsochromic shift of the IC
band, which usually lays below 600 nm [18,19,33,34]. NO+ has a
stronger
of -backbonding is attenuated by the presence of the unpaired p-
p
-acceptor character than CO, although for NO0 the effect
p
electron, allowing the formation of a stable complex with the oxi-
dized [Ru3O]+ unit, such as compound 1 [9,20,35]. Indeed, Zhou
et al. have discussed on this matter in their work [35]. They have
shown the possibility to exchange a CO by a NO ligand in the struc-
ture of a trinuclear ruthenium compound by electrochemical con-
trol of the charge of the [Ru3O] unit. The oxidation of a CO-[Ru3O]0
cluster to the CO-[Ru3O]+ species in the presence of NO leads to
ligand substitution, with formation of the stable NO-[Ru3O]+
moiety.
Fig. 1. Structure of [Ru3O(CH3COO)6(3-pic)2(NO)]PF6, and labeling for 1H NMR
assignment.
The spectrum displayed in Fig. 2 is not entirely compatible nei-
ther with a [Ru3O]1+ nor with a [Ru3O]0 core, since it displays

116
N. Cacita et al. / Inorganica Chimica Acta 429 (2015) 114–121
Table 2
m
Regarding the electrochemical behavior of compounds 1–4, the
(NO) values for complexes [Ru3O(CH3COO)6(L)2(NO)]+, and pKa values of ligands L.
data collected from cyclic voltammetry measurements are depict-
ed in Table 1. All compounds display three waves attributed to the
successive redox couples RuIIIRuIIRuII/RuIIIRuIIIRuII/RuIIIRuIIIRuIII/
RuIIIRuIIIRuIV. The electrochemical behavior is quasi-reversible,
showing current ratios close to 1, and about 70 mV for the anod-
ic/cathodic peak separation. The behavior is fully consistent with
that reported for analogous compounds [9,10,19].
L
m
(NO) (cmꢀ1
)
pKa
Refs.
4-Acpy
Py
3-Pic
4-Pic
1874
1865
1883
1874
3.51
5.25
5.63
5.98
unpublished data
[9]
this work
[10]
4-Pic = 4-methylpyridine; 4-acpy = 4-acethylpyridina.
Worth mentioning is the fact that for compounds 1 and 4 (nitro-
syl and carbonyl clusters, respectively) the difference between suc-
cessive E1/2 values is not as regular as for the pyridinic complexes
absorption bands within all visible region (400–800 nm), and no
other absorptions up to 1100 nm. In previous works, Toma et al.
made the calculations of the molecular orbital diagram for the
complexes [Ru3O(CH3COO)6(L)2(NO)]+, where L = pyridine or 4-
methylpyridine, showing that the NO contribution on different
energy levels of the cluster is rather high, particularly to those
levels involving the dxz and dyz orbital of the ruthenium ions
[9,10]. In this context, the usual spectra assignment in terms of
CLCT and IC does not strictly apply, since the NO-[Ru3O] orbital
mixing is high, even for the levels where metallic character
predominates.
In our case, it is reasonable to assume a parallelism of the spec-
troscopic behavior of 1 with that of compound [Ru3O(CH3COO)6(4-
pic)2(NO)]+, (4-pic = 4-methylpyridine) [10], since the ligand 3-pic
is a position isomer of ligand 4-pic and presumably will not intro-
duce a strong perturbation on the electronic structure of com-
pound 1.
This is actually seen in the semiempirical theoretical study of 1,
performed by a combination of PM3(tm) and ZINDO/S investiga-
tion (the optimized structure of 1 and the relative contributions
of the [Ru3O] and NO orbitals on compound 1 levels are available
as Supporting information). As one can see in Table 1, the theore-
tical and experimental transitions are in good agreement. The geo-
metry obtained with PM3(tm) for compound 1 is equivalent to the
reported earlier, and for all complexes the final structure is com-
posed of a planar isosceles Ru3O triangle with two equivalent Ru-
[19]. In the case of the carbonyl cluster 4, the strong p-backbond-
ing between CO and [Ru3O] unit stabilizes the reduced form of the
latter, shifting the E1/2 value of the process [Ru3O]0/1+ to more posi-
tive values. One could expect the same behavior for the nitrosyl
cluster 1, however the presence of the unpaired electron on ligand
NO0 and its strong coupling with the unpaired electron of the
[Ru3O]1+ seems to favor the formation of the +1 species, shifting
the E1/2 value of the process [Ru3O]0/1+ to more negative values.
These observations corroborate the analysis made from UV–Vis
spectroscopy.
2.2. Infrared spectroscopy
The infrared spectra of complexes 1–4 were assigned by com-
parison to similar complexes previously reported in the literature
[9,10,19] (Supplementary information). The NO stretching was
observed at
m
(NO) = 1883 cmꢀ1. This peak is quite symmetrical
and is not observed in the precursor [Ru3O(CH3COO)6(3-pic)2(H2-
O)]PF6. The value occurs within the typical range for nitrosyl com-
plexes with a linear M-NO coordination, close to the vibrational
frequency of free NO (1876 cmꢀ1) [44]. The remaining peaks are
associated with the vibration characteristics of the acetate, the
picoline and the PFꢀ6 species.
Aiming to address the effect of co-ligands in the infrared spectra
of trinuclear ruthenium nitrosyls, Table 2 presents the values of
l3O bonds of 1.905 Å and a longer one (2.005 Å) trans to NO. The
nitrosyl is linearly coordinated to the trinuclear cluster and also
the NO distance is essentially the same for both complexes
(1.145 Å). All these structural parameters are in accordance with
experimental geometries for related complexes [19]. The elonga-
m(NO) frequencies for few different complexes of general formula
[Ru3O(CH3COO)6(L)2(NO)]+, where L are pyridinic ligands with dif-
ferent pKa. Nitric oxide coordinated to ruthenium ions generally
has an intense band in the range of 1800–1970 cmꢀ1, which fre-
quency depends on the metal oxidation state and the stereochem-
istry of the NO bond with the ion [1,37,38]. It is also known that,
typically, several classes of ruthenium nitrosyl show a correlation
tion of the Ru-l3O bond trans to NO can be related to a strong
interaction between this non-innocent ligand and the ruthenium
ion resulting in an antiferromagnetic coupling of the unpaired
electrons on Ru3O and NO and a consequent singlet ground state.
ZINDO/S calculations also revealed the same theoretical spectral
profile as reported before, the only difference being the shift
of the transitions, which are red-shifted in the order: pyridine ꢁ 3-
picoline < 4-picoline. Mulliken population analysis based on the
ZINDO/S wavefunctions revealed that the charge on NO is essen-
tially zero (0.02) for the complex, corroborating with the previous
NO0 assignment for this class of cluster in the RuI3IIO oxidation state.
Although Mulliken population analysis is known to be basis set
dependent and should be used with caution, these qualitative
results are indicative that the electronic coupling between NO
and the metal center can explain the linear NO coordination mode
to Ru3O, even if NO is still regarded as a formal zero charge ligand.
Indeed, Ford and co-workers have recently shown that NO can
coexist as linear and bent coordination isomers depending on the
spin state of a MnII(NO)porphyrinate complex [36]. They have
found that singlet ground state is linearly coordinated but for
another very close triplet state NO is bent with essentially the
same configuration for Mn(II), i.e., the formal oxidation states of
the metal center and NO are not changed. Therefore, even consid-
ering the present semiempirical calculations from a qualitative
viewpoint, our results are indicative that such a behavior cannot
be ruled out for compound 1.
between the value of
co-ligands. This correlation reflects the weakening of the nitro-
gen–oxygen bond of NO due to the strengthening of -backbond-
m(NO) and the pKa value of the respective
p
ing with the metal center which, in turn, is influenced by the pKa
of other ligands [39–42].
In the case of the trinuclear ruthenium complex this typical
behavior does not seem to manifest. It is observed that even for
ligands with very different pKa values such as 4-acethylpyridine
and 4-methylpyridine, the values of m(NO) do not suffer significant
displacements, as well as do not show any clear trend of depen-
dence on co-ligand pKa. Possibly this finding corroborates the fact
that, in the case of trinuclear ruthenium complexes the behavior of
the metal center and the NO ligand is more dependent on the
strong interaction between the two units (through orbital mixing
and interaction between their unpaired electrons) than on the
influence of peripheral ligands L. However it is important to note
that this is a preliminary analysis, since the series of ligands L
exploited so far is still small [9,10, thiswork].
2.3. 1H NMR and EPR
The 1H NMR spectra of compounds 1–4 were assigned by com-
parison with analogous complexes [9,10], with the free ligands and

N. Cacita et al. / Inorganica Chimica Acta 429 (2015) 114–121
117
from the correlations observed on COSY spectra (Supplementary
information). The NMR spectra are available as Supplementary
material, and the relevant data for our discussion are reported in
Table 3.
Regarding the spectra profile changes, the behavior of 1 during
irradiation at 377, 447 and 532 nm is the same, showing a gradual
decrease in the 452 nm band, a bathochromic shift of the 542 nm
transition and the formation of a band at 680 nm, generating a
spectrum profile for the photoproduct that is fully compatible with
the spectrum of precursor 2, [Ru3O(CH3COO)6(3-pic)2(H2O)]+. Isos-
bestic points around 400 nm and 550 nm, with slight variations for
each irradiation wavelength were observed, indicating that the
It is well described that the signals of ligands coordinated to the
[Ru3O] moiety are dependent on the oxidation state of the rutheni-
um ions. Reduced complexes ([Ru3O]0) having the formal oxidation
states RuIIIRuIIIRuII are diamagnetic and the signals of coordinated
ligands are near to the chemical shift values observed for the cor-
responding free ligands [34,43–45]. On the other hand, [Ru3O]1+
complexes (formal oxidation states RuIIIRuIIIRuIII) have one
unpaired electron, being paramagnetic. Therefore, the signals of
their ligands are not only sensitive to inductive effects, but also
to the paramagnetism of the [Ru3O] unit [19,45,46]. The effect of
paramagnetic anisotropy can manifest itself in two different ways:
by the pseudocontact mechanism, which involves a through space
dipole interaction, decreasing with the distance. And/or the contact
mechanism, that depends on the orientation of a nucleus in rela-
tion to the paramagnetic center. The latter promotes shifts either
to higher or lower fields [46]. Typically, for an oxidized cluster such
reaction that occurs during photolysis is the same for all kirrad
.
From the spectroscopic investigation, apparently NO cannot be
released by irradiation within the therapeutic window
(kirrad = 660 nm), what would be more interesting for bioinorganic
purposes.
In order to accomplish the specific detection of NO, photolysis
was performed with irradiation at the same wavelengths, 377,
447, 532 and 660 nm, but using a selective electrode for NO(g)
detection at this time. The chronoamperograms corresponding to
this photolysis are presented in Fig. 4.
Immediately after exposing the solution to laser irradiation, the
chronoamperogram indicated substantial increase in current
detected by the selective electrode, indicating the photochemical
production of NO(g). We estimated the amount of NO(g) released
by the photoinduced reaction, using an analytical curve (available
as Supplementary information), and the values are shown in
Table 4.
It is observed that at wavelengths lower than 450 nm, NO
release is higher. Furthermore, NO release is still photoinduced in
visible region (532 nm), although with a much lower efficiency.
Again, we failed to detect NO within the sensitivity of the
amperometric experiment using 660 nm as irradiation wavelength.
These observations could easily be rationalized in terms of the
selective irradiation of charge-transfer transitions (region
addressed by the 377, 447 and 532 nm) and irradiation of a transi-
tion of metallic character (660 nm), particularly because it is the
typical behavior observed for other ruthenium nitrosyls, which
generally deliveries NO from irradiation of the CT excited states
[47]. However, this argument should not entirely apply to com-
pound 1, taking into account the high orbital mixing between NO
and the metallic unit proposed from the molecular modeling
results, which is consistent with the other data reported in this
work. The orbital mixture occurs with a greater extent in the ultra-
violet region, but also is considered to be relevant for the lower
energy levels [9,10,thiswork].
as precursor 2, the protons at
a position of the pyridinic ligand
shifts to higher field in relation to free ligand and the CH3 groups
of the acetates exhibit shifts to lower field (Table 3). The shifts
on the pyridinic hydrogen signals decrease with distance, being
consistent with the pseudocontact mechanism of paramagnetic
interaction (Table 3). This typical behavior expresses the paramag-
netic effect on the unit [Ru3O]1+
.
Nevertheless, one can observe in Table 3 that the chemical shift
values observed for the hydrogen nucleus of the nitrosyl 1, a
paramagnetic complex at principle, are somewhat different from
the values observed for compound 2. In fact, compound 1 displays
intermediate values of chemical shift between the values observed
for the free ligands (diamagnetic species) and the values observed
for oxidized clusters (paramagnetic species). This result confirms
that the interaction of the unpaired electrons of the [Ru3O]1+ and
the NO0 ligand is so severe that partially removes the paramagnetic
anisotropy effect, promoting in complex 1 an intermediary behav-
ior between that of paramagnetic (2) and diamagnetic (free
ligands) species.
Actually, the EPR spectra of compound 1, both in solid state and
in acetonitrile solution, are silent, displaying no signals (Support-
ing information). This means that compound 1, despite of having
two unpaired electrons, is a diamagnetic species, supporting the
hypothesis of having antiferromagnetic coupling between the
[Ru3O]+1 center and NO0 ligand.
2.4. Photolysis
Photolysis of the nitrosyl 1 were performed in phosphate buffer,
pH 7.4, to simulate physiological environment and spectra changes
were monitored as a function of time for different irradiation
wavelengths (377 nm, 447 nm and 660 nm, Supplementary infor-
mation). Fig. 3 presents the result for irradiation at 532 nm.
Table 3
Data obtained from 1H NMR analysis for compounds [Ru3O(CH3COO)6(3-pic)2L]PF6
(L = NO and H2O, 1 and 2, respectively), from acetonitrile-d3 solutions.
d (ppm)
1
2
3-Pica
Acetatea
CH3(a)
CH3(b)
3.93 (12H)
3.25 (6H)
4.32 (2H)
4.28 (2H)
8.04 (2H)
5.51 (2H)
1.78 (6H)
4.73 (12H)
4.83 (6H)
ꢀ1.65 (2H)
ꢀ1.65 (2H)
ꢀ1.49 (2H)
5.85 (2H)
5.88 (6H)
–
–
2.1
2.1
–
–
–
H
H
a
8.42
8.44
7.16
7.45
2.32
a0
Hb
H
c
–
–
Fig. 3. Electronic spectra of the complex [Ru3O(CH3COO)6(3-pic)2(NO)]PF6
(5.5 ꢂ 10ꢀ5 M, pH 7.4) in phosphate buffer solution during photolysis with
irradiation at 532 nm.
Hd
a
13C and 1H NMR collection from Aldrich.

118
N. Cacita et al. / Inorganica Chimica Acta 429 (2015) 114–121
Fig. 4. Cronoamperogram of NO release in phosphate buffer, pH 7.4.
Therefore, we further tried to detect NO release by stimulation
with kirrad = 660 nm performing a NOA experiment (Nitric Oxide
Analyzer), which detects NO(g) under a helium flow, pushing the
experiment limits of sensitivity to the range of 1 picomolar to
lM concentrations [48]. It was also verified the NO(g) release in
the presence of ascorbic acid, to simulate the reaction with
endogenous reducing agents. The results are shown in Fig. 5.
It is seen that compound 1 in fact deliveries NO as a conse-
quence of irradiation within the therapeutic window, although
with less efficiency than when irradiated with higher energy light.
It is also observed that the release is much more pronounced in the
presence of a reducing agent such as ascorbic acid. This fact is not
surprising, since it has been recently described the release of NO
from the analog [Ru3O(CH3COO)6(4-pic)2NO]PF6, triggered by a
redox reaction with ascorbic acid [29].
Fig. 5. NO release detected by NOA from
a [Ru3O(CH3COO)6(3-pic)2(NO)]PF6
solution (5.89 ꢂ 10ꢀ5 M, phosphate buffer, pH 7.4) (A) irradiated at 660 nm; (B):
in absence of light after reaction with a 45 mM solution of ascorbic acid.
In compound (1), formally the [Ru3O] unit presents +1 formal
charge, while coordinated NO is neutral. One possibility to explain
the release of NO mediated by ascorbic acid, is that the reductant
acts on the metallic unit, not on the ligand. The implication here
is that lowering the oxidation state of the metal center decreases
the affinity with the coordinated NO0 as much as reducting a coor-
dinated nitrosonium ion decreases the affinity with a metal center
of lower oxidation states.
However, the reactivity of [Ru3O] compounds with ascorbic acid
might be more complicated. Recently it was reported the reactivity
of the analog [Ru3O(CH3COO)6(H2O)3]+ with ascorbic acid at pH 3.8
(acetate buffer) [49]. The authors have shown that the first reaction
step occurs with the selective reduction of one Ru(III) ion concomi-
tantly with the cleavage of one acetate bridge. This leads to the for-
mation of aquo/hydroxo species. After that, further reduction
reactions take place with decomposition of the [Ru3O] unit.
We have performed a qualitative evaluation of the interaction
between 1 and ascorbic acid, aiming to simulate the excess of
ascorbic acid present in the experimental conditions of the NOA-
nalyser experiment. Thus, a 2 M solution of ascorbic acid was pre-
pared in phosphate buffer, and a 100
of a 0.2 mM complex solution (2% of DMSO, phosphate buffer), giv-
ing a proportion of 0.2 mmol of reductant to 0.6 mol of complex.
lL aliquot was added to 3 mL
l
As a result, very little modification on the spectrum profile of 1 was
observed after reaction with ascorbic acid (spectra available as
Supporting information). Besides that, the final spectrum does
not match the profile observed either for the species described
by Dasgupta et al. [49], or for the reducted nitrosilated analog [Ru3-
O(CH3COO)6(4-pic)2NO]0, generated electrochemically [10]. In our
case, despite the fact that the spectra changes very little, one can
observe the onset of an absorption in between 800 and 1000 nm,
characteristic of solvated species [Ru3O(CH3COO)6(L)2S]0. There-
fore, in this particular case most probably it is occurring a reduc-
tion reaction centered on the metal unit [Ru3O], leading to
release of the previously coordinated NO0.
Even though we have calculated the amount of NO released
from different experiments (Table 4), it is possible to propose a
qualitative discussion of the dependence of these values with the
wavelength of irradiation. From the spectroscopic analysis, one
can verify that all the transitions irradiated with the wavelengths
used in photolysis experiments imply in molecular orbitals with
different extents of NO character [9,10, thiswork]. However, it does
not display an obvious correlation with the observed yields pre-
sented in Table 4, since orbital mixing is high. On the other hand,
it is described that the [Ru3O]1+ unit constitute an excellent final
acceptor of excitation energy in different polynuclear systems,
mostly because it performs rapid photoinduced electron transfer
reactions [50–52]. Also its decay is very fast, since the low lying
energy levels associated with the IC transitions promotes a very
rapid path for vibrational decay. In our case, presumably these
low lying energy levels must quench the excited state responsible
for NO release, due to the fast vibrational decay, leading to the
observed lower value of delivered NO when irradiation takes place
Table 4
Quantum yields and amount of NO(g) released (mol/L) from compound [Ru3O(CH3-
COO)6(3-pic)2NO]PF6 in phosphate buffer solution (pH 7.4) for different irradiation
wavelengths.
kirrad (nm)
/
[NO]b (10ꢀ6 M)
[NO]c (10ꢀ12 M)
377
447
532
660
–
0.10
0.006
2.15
2.10
0.95
–
–
–
–
0.004
a
19
–
–
144d
a
Quantum yields were not calculated for kirrad = 660 nm because there were not
observable changes in the spectral profile of (1) during irradiation at this
wavelength.
b
Obtained from the chronoamperograms.
Obtained by NOA.
In the presence of the reducing agent ascorbic acid, [ascorbic acid] = 45 mM,
c
d
PBS solution 0.01 M.

N. Cacita et al. / Inorganica Chimica Acta 429 (2015) 114–121
119
at 660 nm. Time-resolved measurements are under way in our
laboratories, in order to properly describe this assumption.
Finally, values of quantum yields follow the same trend of the
amount of NO released obtained using the NOmeter and NOA.
Namely, for minor amounts of NO released, the quantum yield
was also lower, depending on the irradiation wavelength (Table 4).
If there was a significant rate of recombination, the observed
dependence of amount of NO released on kirrad would not follow
the same trend of quantum yields.
This preliminary result leads us to two important conclusions.
First, although the liberation of NO seems to occur mediated by a
reduction reaction as well (presumably by phenylephrin), light
plays an important role to maximize the effect of vasodilatation.
Secondly, even with the apparent low yields of NO release (Table 4),
compound 1 is still able to perform vasorelaxation with high effi-
ciency, by stimulation with visible light.
3. Conclusions
2.5. Preliminary essay on the vasorelaxing properties of compound 1
This work presented the synthesis and characterization of the
novel nitrosyl [Ru3O(CH3COO)6(3-pic)2(NO)]PF6 and of compounds
2–4 as well. Particularly the 1H NMR measurements showed that
complex (1) displays an average behavior between a reduced and
an oxidized trinuclear complex. This reflects the significant
increase in electron density of the [Ru3O]1+ unit, due to the exten-
sive orbital mixing of the metallic and NO0 levels and the conse-
quent interaction between their unpaired electrons. The mixing
Nitrosyl ruthenium compounds have already been studied as
NO donor that induces rat aorta relaxation [53–56]. Taking it
into account, we also evaluate the induced vasodilatation by
[Ru3O(CH3COO)6(3-pic)2(NO)]+ in rat aortic rings pre-contracted
with phenylephrin, which is a reducing agent.
As shown in Fig. 6(A), the relaxation induced by this compound
in the presence of visible light (ambient light, the system set in a
glass vessel) was concentration-dependent in denuded rat aortas
pre-contracted with phenylephrine. The absence of light
significantly reduced the relaxation induced by complex 1: the
ME (maximal effect) induced by the complex in the presence of
visible light (89.9 4.1%) was greater than ME induced by the
complex in the absence of visible light (25.25 2.24%), Fig. 6(B).
also impacts the m(NO) frequency observed in the infrared spec-
trum, which does not display the typical correlation with the co-li-
gands pKa. The hypothesis of having the pairing of electrons is
corroborated by a silent EPR result.
The photolysis results showed that compound 1 is able to
release NO in physiological pH by irradiation with visible light
and, more interesting, within the therapeutic window (660 nm).
Its ability to induce vasodilatation was also probed both under
ambient light and in the absence of light, under a reducing envi-
ronment. Our results demonstrated that under ambient light the
compound induced more than three times the rate of relaxation
than in the absence of light. However, the dark activity reveals
the possibility of using this kind of compound in therapies other
than those mediated by light-induction.
Overall, our results show that compound 1 is a suitable candi-
date for the development of novel pro-drugs based on the
[Ru3O(CH3COO)6] moiety, both as a carrier and a controlled NO
releaser.
4. Experimental
The synthesis of precursors [Ru3O(CH3COO)6(3-pic)2(H2O)]PF6,
[Ru3O(CH3COO)6(3-pic)3]PF6 and [Ru3O(CH3COO)6(3-pic)2(CO)]
were performed as previously described in the literature
[17,19,57]. Complex [Ru3O(CH3COO)6(3-pic)2(H2O)]PF6 was puri-
fied in a neutral alumina column, eluted with a mixture of 70%
CH3CN:30% CH3OH. ESI-HRMS: [2ꢀH2OꢀPF6]+ m/z 860,9;
[3ꢀPF6]+ m/z 953,9 and [4+H+]+ m/z 889.9.
4.1. Synthesis of [Ru3O(CH3COO)6(3-pic)2(NO)]PF6
A solution of dichloromethane, 33 mL, containing 0.35 g of
[Ru3O(CH3COO)6(3-pic)2(H2O)]PF6 complex was degassed with
argon for 20 minutes, then it was bubbled NO for 2 h (generated
by the reaction of nitric acid and copper pads, bubbled through a
concentrated solution of NaOH to remove traces of NO2), and again
deaerated with Ar for 20 min. The product was precipitated by
adding petroleum ether. The purple solid was collected on a filter,
washed with petroleum ether and dried under vacuum in a desic-
cator with silica gel. Yield: 92 %. Elemental analysis: C, 27.8; H, 3.1;
N, 4.1. Found: C, 26.5; H, 3.4; N, 5.1. ESI-HRMS: [1 - PF6]+ m/z 890,9.
4.2. Physical measurements
Fig. 6. (A) Effect of [Ru3O(CH3COO)6(3-pic)2(NO)]PF6 on rat thoracic aorta pre-
contracted with phenylephrine 0.1 M in the presence and absence of visible light.
(B) Maximum effect (ME) induced by [Ru3O(CH3COO)6(3-pic)2(NO)] PF6 in aorta
rings. Data are means SEM of experiments performed on 4 preparations obtained
from different animals. ⁄P < 0.05.
Infrared spectra were obtained from samples dispersed in KBr
pellets in the region 400–4000 cmꢀ1 with resolution of 4 cmꢀ1 in
an IR spectrophotometer Shimadzu Prestige 21. 1H NMR and COSY

120
N. Cacita et al. / Inorganica Chimica Acta 429 (2015) 114–121
spectra were recorded on a Bruker Avance DRX- 500 500 MHz;
from 1 ꢂ 10ꢀ2 M of the complexes in acetonitrile-d3. High resolu-
tion mass spectra (HRMS) were collected in positive mode on a
micrOTOF II-ESI-TOF Mass Spectrometer, with a flow rate of
proportional to each reacted molecule. Through this reaction, the
level of light produced is proportional to the NO concentration in
the sample. Determination of NO moles is done based on the
straight line obtained from the analytical curve (Supplementary
information) of the device and the full area of the release curve.
300
lL pump/h, mobile phase MeOH:H2O 1v:1v. EPR spectra were
recorded in
a
Bruker EMX instrument operating at X-band
(9.48 GHz frequency, 20 mW power, 100 kHz modulation frequen-
cy, 5–15 G modulation amplitude), at 77 K, in solid state and in
acetonitrile solution, using Wilmad quartz tubes.
4.3. Experimental procedure of aortic ring preparation
Male Wistar rats (180–200 g) were killed by decapitation in
accordance with the Ethical Animal Committee, Ribeirão Preto
Campus at the University of São Paulo, Brazil. The aorta was quick-
ly removed, dissected free, and cut into 4 mm long rings. In order
to avoid possible influence of endothelial factors, the endothelium
was mechanically removed by gently rolling the lumen of the ves-
sel on a thin wire. The aortic rings were placed between two stain-
less-steel stirrups and connected to an isometric force transducer
(Letica Scientific Instruments) to measure tension in the vessels.
The rings were placed in a 10 mL organ chamber containing Krebs
solution with the following composition (mM): NaCl 130, KCl 4.7,
KH2PO4 1.2, MgSO4 1.2, NaHCO3 14.9 glucose 5.5, CaCl2 1.6. The
solution was maintained at pH 7.4 gassed with 95% O2 and 5%
CO2 at 37 °C. The rings were initially stretched to a basal tension
of 1.5 g (previously determined by length–tension relationship
experiments), before allowing them to equilibrate for 60 min in
the bath fluid, which was changed every 15–20 min. The aortic
Electronic spectra were recorded on an Agilent 8453
spectrophotometer, in the region 190–1100 nm, using quartz
cuvettes with optical path length of 1 cm. The cyclic voltammetry
experiments were performed in a conventional three electrodes
cell (platinum as working electrode; platinum wire as auxiliary
electrode and Ag/AgCl as reference electrode; the reported E1/2
values are uncorrected for junction potential values). A 0.1 M
acetonitrile solution of tetrabutylammonium hexafluorophosphate
(TBAPF6) was prepared in 0.1 mol Lꢀ1 acetonitrile. The complexes
solutions were prepared to have final concentrations around
1 ꢂ 10ꢀ3 M.
The photolysis of the [Ru3O(CH3COO)6(3-pic)2(NO)]PF6 complex
was performed from 5.5 ꢂ 10ꢀ5 M solutions in aqueous phosphate
buffer solution, pH 7.4. Since complex 1 is not fully soluble in
water,
a stock solution was prepared in the acetonitrile
(5.5 ꢂ 10ꢀ3 Mol L-1) and 30
lL aliquots were added to 2970 lL of
buffer solution in quartz cuvette of 1 cm optical path, with a suit-
able magnetic stirrer. The solutions were stirred continuously dur-
ing the irradiation time intervals (2 min) at different wavelengths
(377, 447, 532 and 660 nm). The irradiation was performed with
a laser tag Colibri Quantum Laser Tech. Three types of analyzes
were performed during the photolysis experiments: first, the
absorption spectra of the samples were recorded after each
irradiation period until the original spectrum converges to the
spectra of the solvated species 2. Light intensities were determined
by chemical actinometry (ferrioxalate and Reinecke salts) in order
to calculate quantum yields.
rings were stimulated continuously with 0.1 lM phenylephrine
until reproducible contractile responses were obtained. The
absence of endothelium was confirmed by the lack of relaxation
response to 1
lM acetylcholine in aortic rings pre-contracted with
0.1 M phenylephrine.
l
4.4. Relaxation induced by complex [Ru3O(CH3COO)6(3-pic)2(NO)]PF6
in pre-contracted arteries with phenylephrine
A 0.1 mM stock solution of 1 was prepared in DMSO. It was per-
formed serial dilutions of the stock solutions in water. We have
analyzed the maximal effect (ME) of the complex in concentra-
tion-effect curves constructed in rat aortic rings. To examine
whether [Ru3O(CH3COO)6(3-pic)2(NO)]PF6 induces smooth muscle
The second experiment was performed using a selective elec-
trode for NO(g) positioned inside the cuvette, but outside the light
beam way, in order to avoid any photoelectric interference. The
detection of NO was taken using this selective electrode for NO
Noxímetro Brand Insight. NOmeter calibration was done monitor-
ing the amount of dissolved NO(g) in a standard solution of NO gas
[58]. This solution was prepared by bubbling NO for 1 h, to ensure
saturation of 8 mL of phosphate buffer solution (pH 7.4) previously
degassed to remove oxygen. The nitric oxide was generated from a
solution of 50% nitric acid and metallic copper. Before bubbling
through the buffer solution, the generated gas passed through a
concentrated solution of NaOH to remove traces of NO2. The con-
cell relaxation, aortic rings were pre-contracted with 0.1
phenylephrine and when the contraction had reached a plateau,
the complex (0.1 nM–10 M) was cumulatively added. The experi-
lM
l
ments were conducted in the presence or absence of visible light.
4.5. Statistical analysis
Data are expressed as mean SEM. In each set of experiments, n
indicates the number of rats studied. Maximal effect (ME) was
centration of NO of the final saturated solution is 2 ꢂ 10ꢀ3
M
obtained
from
the
concentration–response
curves
for
[59]. The analytical curve (Supporting information) was obtained
by varying the concentration of NO adding known aliquots of the
saturated solution to 6 mL of previously degassed phosphate buffer
solution, pH 7.4, and monitoring the corresponding current value.
The third experiment was performed using a NOA 280i Nitric
Oxide Analyzer. A 5.89 ꢂ 10ꢀ5 M solution of [Ru3O(CH3COO)6
(3-pic)2(NO)]PF6 was prepared in phosphate buffer (pH 7.4) and
2 mL of it was irradiated under 660 nm light during 30 min in a
[Ru3O(CH3COO)6(3-pic)2(NO)]+, being considered as the maximal
amplitude response reached on those curves. The mean of indi-
vidual values for each tissue was obtained using the GraphPad
Prism version 3.0 (GraphPad Software Corporation San Diego, CA).
Statistical significance was tested by one way ANOVA (post-test:
Newman–Keuls) and Student’s t test, and values of P < 0.05 were
considered to be significant.
fluorescence cuvette. A 100
lL aliquot was injected in NOA, under
4.6. Molecular modeling
an N2(g) flow in order to detect the amount of NO released. Aiming
to evaluate the NO release by chemical reduction stimulation, a
50 lL of the same solution (without exposure to light), was inject-
ed over 5 mL of a 45 mM solution of ascorbic acid under the N2(g)
flow. NOA has the ability to measure concentrations in the range
of 1 pM and has proven to be a tool widely used by researchers
being cited in over 700 publications [48]. Due to the instability of
the NO molecule, it reacts with ozone producing a quantity of light
Geometries were obtained with RHF wavefunctions assuming
singlet ground states with the PM3(tm) implementation of
HYPERCHEM [60]. A conjugate gradient was used to achieve a
10ꢀ3 kcal molꢀ1 Åꢀ1 convergence for geometry. For the SCF calcula-
tions we used a 10ꢀ5 kcal molꢀ1 criterium for convergence of the
wavefunction. Using these PM3(tm) geometries, molecular orbi-
tals, vertical excitations and population analysis were obtained

N. Cacita et al. / Inorganica Chimica Acta 429 (2015) 114–121
121
[21] S.D. Glover, J.C. Goeltz, B.J. Lear, C.P. Kubiak, Coord. Chem. Rev. 254 (2010) 331.
[22] J. Qin, Y. Liao, X.-Y. Chen, C.-X. Qian, J.-L. Zuo, X.-Z. You, J. Organomet. Chem.
716 (2012) 275.
with ZINDO/S wavefunctions obtained single excitations over an
active space of 20 molecular orbitals (10 occupied plus 10 virtual).
[23] A.D.P. Alexiou, S. Dovidauskas, H.E. Toma, Quim. Nova 23 (2000) 785.
[24] S. Nikolaou, A.L.B. Formiga, H.E. Toma, Inorg. Chem. Commun. 13 (2010) 1032.
[25] S.H. Toma, J.J. Santos, S. Nikolaou, K. Araki, H.E. Toma, Inorg. Chim. Acta 390
(2012) 148.
[26] S. Nikolaou, H.E. Toma, Eur. J. Inorg. Chem. 14 (2008) 2266.
[27] S. Ye, W. Zhou, M. Abe, T. Nishida, L. Cui, K. Uosaki, M. Osawa, Y. Sasaki, J. Am.
Chem. Soc. 126 (2004) 7434.
[28] M. Abe, T. Michi, A. Sato, T. Kondo, W. Zhou, S. Ye, K. Uosaki, Y. Sasaki, Angew.
Chem., Int. Ed. Engl. 42 (2003) 2912.
[29] Z.A. Carneiro, J.C. Biazzotto, A.D.P. Alexiou, S. Nikolaou, J. Inorg. Biochem 134
(2014) 36–38.
[30] F.A. Cotton, J.G. Norman, Inorg. Chim. Acta 6 (1972) 411.
[31] H.E. Toma, C.J. Cunha, C. Cipriano, Inorg. Chim. Acta 154 (1988) 63.
[32] M. Abe, Y. Sasaki, Y. Yamada, K. Tsukahara, S. Yano, T. Ito, Inorg. Chem. 34
(1995) 4490.
[33] J.C. Salsman, S. Ronco, C.H. Londergan, C.P. Kubiak, Inorg. Chem. 45 (2006) 547.
[34] M. Abe, Y. Sasaki, Y. Yamada, K. Tsukahara, S. Yano, T. Yamaguchi, M.
Tominaga, I. Taniguchi, T. Ito, Inorg. Chem. 35 (1996) 6724.
[35] W. Zhou, Y. Zhang, M. Abe, K. Uosaki, M. Osawa, Y. Sasaki, S. Ye, Langmuir 24
(2008) 8027.
Acknowledgments
The authors thank Professor Ana Maria da Costa Ferreira, form
IQ – USP – São Paulo, for the acquisition of EPR spectra. We also
thank Dr. Juliana Cristina Moraes Biazzotto and Clovis Junior for
their collaboration to this work, Dr. José Carlos Tomaz for the
acquisition of high resolution mass spectra, Prof. Dr. Roberto
Santana da Silva for allowing the use of some equipments and
the financial support from CAPES , CNPq and FAPESP agencies.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2015.01.038.
[36] T.S. Kurtikyan, V.A. Hayrapetyan, G.G. Martirosyan, R.K. Ghazaryan, A.V.
Iretskii, H. Zhao, K. Pierloot, P.C. Ford, Chem. Commun. 48 (2012) 12088.
[37] B.L. Westcott, J.H. Enemark, Transition Metal Nitrosyls, John Wiley & Sons, Inc.,
1999.
[38] P.C. Ford, B.K. Miranda, B. Lee, I. Lorkovic, S. Boggs, S. Kudo, L. Laverman, Coord.
Chem. Rev. 171 (1998) 185.
[39] C.W. Bezerra, S.C. da Silva, M.T.P. Gambardella, R.H.A. Santos, L.M.A. Plicas, E.
Tfouni, D.W. Franco, Inorg. Chem. 38 (1999) 5660.
[40] M.L. Bento, E. Tfouni, Inorg. Chem. 27 (1988) 3410.
[41] E. Tfouni, P.C. Ford, Inorg. Chem. 18 (1980) 72.
References
[1] R.G. de Lima, M. Gama, D. Bonaventura, A.C. Tedesco, R.F. Vianna, Inorg. Chim.
Acta 358 (2005) 2643.
[2] M.G. Sauaia, R.G. de Lima, A.C. Tedesco, R.S. da Silva, Inorg. Chem. 44 (2005)
9946.
[3] R.D.L. Barreto, R.D. Correia, M.N. Muscará, Quim. Nova 28 (2005) 1046.
[4] E.C. Fornari, M.S.P. Marchesi, A.E.H. Machado, S. Nikolaou, Polyhedron 28
(2009) 1121.
[5] Z.N. da Rocha, R.G. de Lima, F.G. Doro, E. Tfouni, R.S. da Silva, Inorg. Chem.
Commun. 11 (2008) 737.
[42] E. Tfouni, M. Krieger, B.R. McGarvey, D.W. Franco, Coord. Chem. Rev. 236
(2003) 57.
[43] A.D.P. Alexiou, H.E. Toma, J. Chem. Res. (S) (1997) 338.
[44] S. Nikolaou, H.E. Toma, Polyhedron 20 (2001) 253.
[45] S. Nikolaou, M. Uemi, H.E. Toma, Spectrosc. Lett. 34 (2001) 267.
[46] I. Bertini, C. Luchinat, Coord. Chem. Rev. 150 (1996) 1.
[47] U. Schatzschneider, Eur. J. Inorg. Chem. 2010 (2010) 1451.
[48] <www.geinstruments.com/products-and-services/nitric-oxide-analyzer>, n.d.
2014, 134, 36.
[49] M.A.W. Lawrence, P.T. Maragh, T.P. Dasgupta, Transition Met. Chem. 37 (2012)
505.
[50] S. Nikolaou, H.E. Toma, J. Chem. Soc., Dalton Trans. (2002) 352.
[51] T. Koshiyama, M. Shirai, T. Hikage, H. Tabe, K. Tanaka, S. Kitagawa, T. Ueno,
Angew. Chem., Int. Ed. Engl. 50 (2011) 4849.
[52] A.J. Morris-Cohen, K.O. Aruda, A.M. Rasmussen, G. Canzi, T. Seideman, C.P.
Kubiak, E.A. Weiss, Phys. Chem. Chem. Phys. 14 (2012) 13794.
[53] D. Bonaventura, Nitric Oxide 10 (2004) 83.
[54] M.S.P. Marchesi, S.A. Cicillini, A.C.L. Prazias, L.M. Bendhack, A.A. Batista, R.S.
Silva, Transition Met. Chem. 37 (2012) 475.
[55] F.C. Munhoz, S.R. Potje, A.C. Pereira, M.G. Daruge, R.S. da Silva, L.M. Bendhack,
C. Antoniali, Nitric Oxide 26 (2012) 111.
[56] G.J. Rodrigues, S.A. Cicillini, R.S. Silva, L.M. Bendhack, Nitric Oxide 25 (2011)
331.
[6] D.C.A.S. de Santana, T.T. Pupo, M.G. Sauaia, R.S. da Silva, R.F.V. Lopez, Int. J.
Pharm. 391 (2010) 21.
[7] G. Stochel, A. Wanat, E. Kulis, Z. Stasicka, Coord. Chem. Rev. 171 (1998) 203.
[8] B.C.S. Allardyce, P.J. Dyson, Platinum Met. Rev. 45 (2001) 62.
[9] H.E. Toma, A.D.P. Alexiou, S. Dovidauskas, Eur. J. Inorg. Chem. 2002 (2002)
3010.
[10] H.E. Toma, A.D.P. Alexiou, A.L.B. Formiga, M. Nakamura, S. Dovidauskas, M.N.
Eberlin, D.M. Tomazela, Inorg. Chim. Acta 358 (2005) 2891.
[11] R.C.L. Zampieri, G. Von Poelhsitz, A.A. Batista, O.R. Nascimento, J. Ellena, E.E.
Castellano, J. Inorg. Biochem. 92 (2002) 82.
[12] S.A. Cicillini, A.C.L. Prazias, A.C. Tedesco, O.A. Serra, R.S. da Silva, Polyhedron 28
(2009) 2766.
[13] F. Marquele-Oliveira, D.C.D.A. Santana, S.F. Taveira, D.M. Vermeulen, A.R.M. de
Oliveira, R.S. da Silva, R.F.V. Lopez, J. Pharm. Biomed. Anal. 53 (2010) 843.
[14] E. Tfouni, F.G. Doro, A.J. Gomes, R.S. Da Silva, G. Metzker, P.G.Z. Benini, D.W.
Franco, Coord. Chem. Rev. 254 (2010) 355.
[15] M.G. Sauaia, F.D.S. Oliveira, R.G. De Lima, A.D.L. Cacciari, E. Tfouni, R.S. Da Silva,
Inorg. Chem. Commun. 8 (2005) 347.
[16] R.G. de Lima, M.G. Sauaia, C. Ferezin, I.M. Pepe, N.M. José, L.M. Bendhack, Z.N.
da Rocha, R.S. da Silva, Polyhedron 26 (2007) 4620.
[17] J.A. Baumann, D.J. Salmon, S.T. Wilson, T.J. Meyer, W.E. Hatfield, Inorg. Chem.
17 (1978) 3342.
[57] A. Spencer, G. Wilkinson, J. Chem. Soc., Dalton Trans. (1972) 1570.
[58] S. Kudo, J.L. Bourassa, S.E. Boggs, Y. Sato, P.C. Ford, Anal. Biochem. 247 (1997)
193.
[59] V. Mori, M. Bertotti, Analyst 125 (2000) 1629.
[60] ‘‘HyperChem Professional 8.0, Hypercube, Inc. 1115 NW 4th St. Gainesville, FL
32601, USA’’, n.d.
[18] J.A. Baumann, S.T. Wilson, D.J. Salmon, P.L. Hood, T.J. Meyer, J. Am. Chem. Soc.
101 (1979) 2916.
[19] H.E. Toma, K. Araki, A.D.P. Alexiou, S. Nikolaou, S. Dovidauskas, Coord. Chem.
Rev. 221 (2001) 187.
[20] M. Abe, A. Sato, T. Inomata, T. Kondo, K. Uosaki, Y. Sasaki, J. Chem. Soc., Dalton
Trans. (2000) 2693.