The role of ancillary ligand substituents in the biological activity of triruthenium-NO complexes

Article abstract of DOI:10.1016/j.jinorgbio.2018.05.021

Two novel triruthenium clusters, [Ru₃(μ₃-O)(μ-OOCCH₃)₆(NO)L₂]PF₆ (L = 4?acetylpyridine, 1, or 4?tert?butylpyridine, 2) release NO. Their spectroscopic and electrochemical characterization confirmed their structure. These complexes efficiently deliver NO in solution under irradiation at λ_irrad = 377 nm and/or through chemical reduction with ascorbic acid. Clusters 1 and 2 elicit vasodilation and, at concentrations of 10^?5 M, can relax up to 100% of pre-contracted rat aorta. Complex 2 is more cytotoxic to murine melanoma B16F10 cells than complex 1: at 50 times lower concentration than complex 1, complex 2 decreases cell viability to 50% in the dark or under irradiation with visible light (λ_irrad = 527 nm). The higher cytotoxicity of complex 2 can be assigned to its larger hydrophobicity, promoted by the methylated tert?butylpyridine ancillary ligand in its structure. Investigation into human serum albumin (HSA) fluorescence quenching by clusters 1 or 2 revealed that complex 2 quenches HSA luminescence with a very high Stern-Vomer constant (K_SV = 9.49 × 10⁷ M^?1 at T = 298 K) and suggested that the nature of the interaction between complex 2 and HSA is hydrophobic (ΔH = 80.81 kJ/mol and ΔS = 334.71 J/K mol). HSA lifetime and circular dichroism data pointed to a static quenching mechanism for both complexes. Together, our results show that a hydrophobic substituent in the cluster ancillary ligand improves NO release ability, cytotoxicity, and interaction with a bio-target.

Full text of DOI:10.1016/j.jinorgbio.2018.05.021

Accepted Manuscript
The role of ancillary ligand substituents in the biological activity
of triruthenium-NO complexes
Camila Fontes Neves da Silva, Bruna Possato, Lilian Pereira
Franco, Loyanne Carla Barbosa Ramos, Sofia Nikolaou
PII:
DOI:
Reference:
S0162-0134(18)30052-7
doi:10.1016/j.jinorgbio.2018.05.021
JIB 10511
To appear in:
Journal of Inorganic Biochemistry
Received date:
Revised date:
Accepted date:
26 January 2018
28 May 2018
28 May 2018
Please cite this article as: Camila Fontes Neves da Silva, Bruna Possato, Lilian Pereira
Franco, Loyanne Carla Barbosa Ramos, Sofia Nikolaou , The role of ancillary ligand
substituents in the biological activity of triruthenium-NO complexes. Jib (2017),
doi:10.1016/j.jinorgbio.2018.05.021
This is a PDF file of an unedited manuscript that has been accepted for publication. As
a service to our customers we are providing this early version of the manuscript. The
manuscript will undergo copyediting, typesetting, and review of the resulting proof before
it is published in its final form. Please note that during the production process errors may
be discovered which could affect the content, and all legal disclaimers that apply to the
journal pertain.

ACCEPTED MANUSCRIPT
The role of ancillary ligand substituents in the biological activity of triruthenium-
NO complexes
Camila Fontes Neves da Silva,[a] Bruna Possato, [a] Lilian Pereira Franco,[a] Loyanne Carla Barbosa
Ramos,[b] and Sofia Nikolaou*[a]
aDepartamento de Química da Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto – Universidade
de São Paulo, Av. Bandeirantes 3900, ZIP-CODE 14040-901, Ribeirão Preto-SP, Brazil, Phone: 55 16
33153748, sofia@ffclrp.usp.br
bFaculdade de Ciências Farmacêuticas de Ribeirão Preto - Universidade de São Paulo, Av. Bandeirantes
3900, ZIP-CODE 14040-901, Ribeirão Preto-SP, Brazil.
*corresponding author: Prof. Sofia Nikolaou, sofia@ffclrp.usp.br.
1

ACCEPTED MANUSCRIPT
Graphical Abstract
Graphical synopsis
[Ru3(μ3-O)(μ-OOCCH3)6(NO)(L)2]PF6 (L = 4-acetylpyridine, 1 or 4-tert-butylpyridine,
2) release NO. Vasodilation activity does not depend on ligand L structure; whereas
methyl groups on the ancillary ligand of 2 modulate its cytotoxicity to B16F10 murine
melanoma cells. Methylation also increases interaction of 2 with HSA (human serum
albumin), through hydrophobic forces.
Highlights
 [Ru3O(OOCCH3)6(L)2NO]+ (4-acetylpyridine, 1; 4-tert-butylpyridine, 2) were
characterized
 Both complexes release NO in solution by photoinduction and/or reductive
stimulus
 Both complexes relax pre-contracted rat aorta with maximum efficiency
 Compound 2 is more cytotoxic to B16F10 melanoma cells due its ancillary
ligand
 2 also interacts with human serum albumin more efficiently, through
hydrophobic forces
2

ACCEPTED MANUSCRIPT
Abstract
Two novel triruthenium clusters, [Ru3(μ3-O)(μ-OOCCH3)6(NO)L2]PF6 (L = 4-
acetylpyridine, 1, or 4-tert-butylpyridine, 2) release NO. Their spectroscopic and
electrochemical characterization confirmed their structure. These complexes efficiently
deliver NO in solution under irradiation at irrad = 377 nm and/or through chemical
reduction with ascorbic acid. Clusters 1 and 2 elicit vasodilation and, at concentrations
of 10-5M, can relax up to 100% of pre-contracted rat aorta. Complex 2 is more cytotoxic
to murine melanoma B16F10 cells than complex 1: at 50 times lower concentration than
complex 1, complex 2 decreases cell viability to 50% in the dark or under irradiation
with visible light (λirrad = 527 nm). The higher cytotoxicity of complex 2 can be
assigned to its larger hydrophobicity, promoted by the methylated tert-butylpyridine
ancillary ligand in its structure. Investigation into Human Serum Albumin (HSA)
fluorescence quenching by clusters 1 or 2 revealed that complex 2 quenches HSA
luminescence with a very high Stern-Vomer constant (KSV = 9.49 x 107 M-1 at T = 298
K) and suggested that the nature of the interaction between complex 2 and HSA is
hydrophobic (ΔH = 80.81 KJ/mol and ΔS = 334.71 J/K mol). HSA lifetime and circular
dichroism data pointed to a static quenching mechanism for both complexes. Together,
our results show that a hydrophobic substituent in the cluster ancillary ligand improves
NO release ability, cytotoxicity, and interaction with a bio-target.
Keywords: triruthenium nitrosyl clusters; NO releasing molecules; methylation;
vasodilation; anti-cancer activity; HSA fluorescence quenching.
3

ACCEPTED MANUSCRIPT
1. Introduction
Alfred W. Mond started to study μ-oxo trinuclear ruthenium clusters in the 1930s [1].
This subject has been reviewed twice in the past years [2, 3]. These clusters have
significant redox and catalytic properties that depend on the substituent present in the
ancillary ligands and which are directly related to the ligands pKa [4, 5]. Various
applications of these clusters have been reported: they can be used in electronic
materials to perform intramolecular electron transfer; they can be employed as
precursors for construction of supramolecular assemblies, 3D architectures on electrode
surfaces, and nanomaterials; and they have potential biological application as anti-tumor
and vasodilation agents [6–17].
Strong electronic mixing based on metal-metal interactions dictates the
chemistry of trinuclear ruthenium carboxylates and leads to large electronic
delocalization within the [Ru3O]+ unit [2]. This property allows clusters to react with π-
acceptor species like CO and NO due to the intrinsic ability of the latter molecules to
perform -backbonding with the [Ru3O]n metal center. This fact has increased interest
in applying these compounds as releasing or scavenging agents of relevant biological
molecules, including CO(g) and NO(g). To date, four literature works have focused on
controlled NO release from complexes with general formula [Ru3(μ3-O)(μ-
OOCCH3)6(NO)L2]PF6 (where L = pyridinic ligands) [16–20], but other structures and
approaches have also been described [21–23].
Both photo and redox stimuli can trigger NO release. Triruthenium complexes
undergo up to three redox processes within the potential window that is accessible in
physiological medium, so these complexes are potential NO-delivery systems even in
the dark. Our group has previously dealt with these aspects without discussing how
ancillary ligands influence biological assays [16–18]. To the best of our knowledge,
investigating structure-activity relationship is the best way to design new drugs.
Medicinal chemistry has already explored the methylation effect [24]. The methyl group
(-CH3) is very important for molecular recognition of organic and inorganic compounds
by bioreceptors [25–28]. Furthermore, the CH3 group increases the overall lipophilicity
of a given molecule and promotes stronger hydrophobic interactions in the case of some
compounds [24–28].
4

ACCEPTED MANUSCRIPT
This study presents the synthesis and characterization of two novel clusters with
general formula [Ru3(μ3-O)(μ-OOCCH3)6(NO)L2]PF6 (where L = 4-acetylpyridine, 4-
acpy (1) or 4-tert-butylpyridine, 4-tbpy (2), Figure 1) and discusses how the substituent
in the ancillary ligand affects the biological activity of clusters 1 and 2. We have
conducted in vitro MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide)
assays [29] to evaluate cluster cytotoxicity to two cell lines, L929 (murine fibroblast, a
non-tumor cell), and B16F10 (murine melanoma). We have also investigated other
biological properties of clusters 1 and 2 including their vasorelaxation capacity and their
interaction with HSA (Human Serum Albumin).
+
+
OH2
NO
O
O
O
O
O
O
O
Ru
O
O
Ru
O
O
O
(g)
NO
Ru
O
Ru
O
O
O
L
L
O
O
O
O
O
O
Ru
Ru
O
O
O
O
L
L
N
N
N
C
N
C
L =
CH3
CH3
H3C
CH3
O
CH3
H3C
4-acpy(1) 4-tbpy(2) 4-pic(3) 3-pic(4)
Figure 1. Reaction of the precursor ([Ru3(μ3-O)(μ-OOCCH3)6(H2O)(L)2]+ with the nitrosyl ligand
(NO(g)), to obtain clusters with the general formula [Ru3(μ3-O)(μ-OOCCH3)6(NO)(L)2]+, where L = 4-
acpy (1), 4-tbpy (2); compounds 3 (4-pic = 4-methylpyridien) and 4 (3-pic = 3-methylpyridine) are from
literature papers [16, 17].
2. Results and discussion
2.1 Chemical characterization
As shown in Figure 1, we prepared clusters 1 and 2 by passing a continuous NO(g) flow
through a solution of the precursor [Ru3(μ3-O)(μ-OOCCH3)6(H2O)(L)2]PF6 (where L =
4-acpy or 4-tbpy) in dichloromethane (DCM) at ambient temperature, which afforded a
5

ACCEPTED MANUSCRIPT
violet solid after purification [16–20]. Elemental analysis, cyclic voltammetry, and UV-
1
vis, IR, H NMR, and EPR spectroscopies helped to determine the structure and purity
of complexes 1 and 2. Table 1 lists part of these data (Full data are provided as
Supplementary Material, Figures S1–S9).
We assigned the data in Table 1 by comparison with similar complexes [17, 19,
20]. Clusters 1 and 2 display the typical profile of the electronic spectra (Figures S3 and
S4) of complexes in which the cluster to ligand charge transfer bands (CLCT) due to
charge transfer from the metallic center [Ru3O]+ to both pyridinic and NO ligands
emerge below 600 nm. The intra-cluster (IC) band, which involves transitions centered
in states that are predominantly metallic in nature, arises between 700 and 800 nm [17].
The cores of the nitrosyl clusters 1 and 2 bear three ruthenium ions in the
oxidation state +3 (RuIIIRuIIIRuIII). Thus, this kind of complex has formal charge +1
([Ru3O]+, oxidized form). Following the electron distribution in their molecular orbital
diagram under D3h symmetry ((a”2)2 (e’)4 (a’1)2 (e’)4 (e”)4 (a’2)1), there is an unpaired
electron, which responsible for the effect named paramagnetic anisotropy [16–20, 30-
33]. In the NMR spectra, this effect disturbs the typical chemical shift values of the
1
different H nuclei, displacing the values of the coordinated ligands as compared to the
free species (chemical shift values for both complexes are available in the experimental
section) [2,16–20, 34–39].
Table 1: Absorption (UV-vis) and electrochemical data collected from acetonitrile
(ACN) solutions and IR data collected from KBr pellets for clusters 1 and 2
λmax (nm) / log ε (L mol-1 cm-1)
Clusters
1
IC*
756/ 2.87
CLCT** 1
553/ 3.24
CLCT 2
456/ 3.32
CLCT 3
354/ 3.74
707/ 3.04
544/ 2.73
458/ 3.30
--------
2
IR data (cm-1)
Clusters
ν(NO)
1874
1870
νas(COO-)
1584
1591
1
2
Electrochemical data E1/2 (V vs. Ag/AgCl)
Clusters
[Ru3O]-1/0
-0.70
[Ru3O]0/+1
-0.06
[Ru3O]+1/+2
1.34
1
-0.81
-0.09
1.36
2
The absorption spectra of clusters 1 and 2 were recorded in ACN (10-3 M). IC* = intra-cluster transition.
CLCT** = cluster-to-ligand charge transfer transition. The cyclic voltammograms of clusters 1 and 2
6

ACCEPTED MANUSCRIPT
were obtained in ACN solution containing 0.1 M TBAPF6 (TBA = tetrabutylamonium cation); all the data
are reported against the Ag/AgCl reference electrode.
In the particular case of nitrosyl trinuclear complexes, the paramagnetic
anisotropy effect is partially removed. This happens because severe orbital mixing
occurs between the metallic unit and NO, to culminate in pairing of the unpaired
electron of the [Ru3O]+ core and the unpaired electron of the NO0 ligand [16–20]. As a
result, nitrosyl clusters and complexes in the reduced form ([Ru3O]0) present similar
magnetic behavior, since chemical reduction causes electron pairing. The NMR and
EPR data collected in this work confirm this fact. The EPR spectra of clusters 1 and 2
do not exhibit any signals, which reinforces strong coupling between the NO0 π*
electron and the ruthenium metal ion dπ electron (1H NMR and EPR spectra of clusters
1 and 2 – Figures S6, S7 and S8, S9 respectively – are shown in Supplementary
Material).
2.2 NO release from clusters 1 and 2 by photo and/or redox stimuli
Complexes 1 and 2 are poorly soluble, so we conducted photolysis in organic medium.
Table 2 summarizes the photoinduced NO release yields; spectral changes observed
during irradiation are available as Supplementary Material, Figures S10 and S11.
Irradiation of clusters 1 or 2 in the region of their charge transfer transition decreases
the bands at 756 and 707 nm, respectively, and intensifies the band around 700 nm, to
give
a
spectral profile typical of the solvated species [Ru3(μ3-O)(μ-
OOCCH3)6(S)(L)2]PF6 [2].
Table 2. Quantum yields (ϕ) and concentration of NO(g) released from clusters 1 and 2
obtained by photolysis and NOA experiments
a
Clusters
ϕNO
[NO]b/[NO]c
10-12M
[NO]d/[NO]e
10-12M
λ377 nm
0.23 ± 0.1
1.65 / 2.26
1.08 / 4.04
1
2
0.61 ± 0.1
2.48/ 4.01
2.82 / 12.4
aQuantum yields (ϕ) of NO release during photolysis of clusters 1 and 2, 1.10-5 M, in ACN solution,
calculated with chemical actnometry. b/cConcentration of NO(g) released from 20- and 40-μL aliquots of
cluster solution (ACN, 1.10-5 M) in the presence of an aqueous ascorbic acid solution (1.10-5 M).
d/eConcentration of NO(g) released from 60-μL aliquots of cluster solution in the presence of an aqueous
ascorbic acid solution (1.10-5 M) and under irradiation (UV lamp) for dt1 = 0 and et2 = 5 s.
7

ACCEPTED MANUSCRIPT
Interestingly, cluster 2 has higher quantum yield than cluster 1. Our group has
recently reported the same observation for analogues containing the CO ligand [40] and
attributed the higher quantum yield to the presence of 4-tert-butylpyridine, a more basic
ancillary ligand with σ-donor characteristics. In the ground state, these clusters have an
18-electron configuration [19], where the [Ru3O]+ unit contributes with 17 electrons,
and NO0 contributes with its unpaired electron. Due to orbital mixing, this ground state
resembles the reduced [Ru3O]0 moiety more closely. Irradiation in the charge transfer
region promotes full electronic density transfer from the [Ru3O] unit to the NO π* level
during the excited state lifetime. This removes the electronic pairing verified in the
ground state, to reduce interaction between the species and consequently prompt NO
release. The [Ru3O] center becomes paramagnetic after NO release, as probed by the
photoproduct typical spectral profile (Figures S10 and S11) [17, 19, 20]. Therefore,
when the complex is in the excited state, the metallic core formal charge is indeed +1,
which is more stabilized by interaction with a σ-donor ancillary ligand such as 4-tbpy as
compared to a more -acid ligand like 4-acpy [17]. Upon return to the ground state, the
oxidized profile is maintained, to afford a complex bearing a coordinated solvent
molecule instead of the NO ligand.
To probe the ability of these complexes to deliver NO under reductive
conditions, we determined the concentration of NO(g) released from cluster 1 or 2 in the
presence of ascorbic acid (Table 2) in the dark and under light irradiation (UV lamp,
continuous light). To this end, we employed a NO analyzer (NOA equipment), which
detects very low NO(g) concentrations under helium flow [41]. The plots of released
NO(g) detected by NOA are available as Supplementary Material, Figures S12 and S13.
Table 2 revealed that even in small concentrations, NO is delivered because
ascorbic acid reduces clusters 1 and 2. As pointed out above, the main feature that
maintains interaction between the [Ru3O]+ unit and the NO0 ligand is pairing of their
unpaired electrons in the 18-electron configuration. Reduction of complexes 1 and 2
disrupts this interaction and lowers the affinity of the coordinated NO for the metallic
center, ultimately leading to its release. NO release mediated by reduction is also more
efficient for complex 2 and, as expected, is boosted by light irradiation (Table 2, third
column).
2.3 Vasorelaxation caused by cluster 1 or 2 in pre-contracted rat aorta
8

ACCEPTED MANUSCRIPT
Knowing that clusters 1 and 2 behave as NO releasers in solution, we investigated two
biological properties (vasodilation and cytotoxicity) mediated by NO release. First, we
evaluated relaxation by incubating pre-contracted rat aorta without endothelium with
cluster 1 or 2. Figure 2 shows that these clusters induce concentration-dependent
relaxation in denuded rat aortas pre-contracted with phenylephrine in the presence of
environment light. The maximal effects (Emax) are 98.08 ± 6.40% and 102.5 ± 5.10% for
clusters 1 and 2, respectively.
NOA experiments demonstrated that NO release follows reduction of these
complexes (with or without light irradiation). In the vasodilation assay, phenylephrine
might act as a reducing agent [17]. In fact, we recorded the spectral profile upon
reduction of cluster 1 or 2 after reaction with ascorbic acid or phenylephrine, to find that
they are essentially the same (Figures S15 and S16). Hence, we attribute the relaxation
effect to NO release triggered by the presence of ambient light and phenylephrine.
Figure 2: Vasodilator effect of (A) [Ru3(μ3-O)(μ-OOCCH3)6(NO)(4-acpy)]PF6 (1) or (B) [Ru3(μ3-O)(μ-
OOCCH3)6(NO)(4-tbpy)]PF6 (2) on rat aorta pre-contracted with phenylephrine (0.1 μM) in the presence
of environment light. Data are the means ± SEM of experiments performed on four preparations obtained
from different animals. *P < 0.05.
Although the vasodilation response is maximum at 1 x 10-5 M, almost 10 times
more cluster 2 (8 x 10-7 M) than cluster 1 (1 x 10-7M) is necessary to achieve 50%
relaxation. In terms of concentration range, the ability of clusters 1 and 2 to provide
50% and 100% relaxation is quite similar (10-7 to 10-6 and 10-5 M, respectively) to the
behavior reported in the literature for related complexes including 4,
[Ru(NH.NHq)(tpy)NO]3+ (tpy = terpyridine), trans-[RuCl([15]aneN4)NO]2+ ([15]aneN4
9

ACCEPTED MANUSCRIPT
= 1,5,8,12-tetraazacyclotetraundecane), and the iron complex sodium nitroprusside
(SNP) [17, 42–44]. Except for the first example (in which the ancillary ligand is 3-
methylpyridine), the other complexes do not bear any structural similarity with the
complexes investigated here. Thus, the substituent, or even better, the ancillary ligands
themselves, does not influence the vasodilation effect.
2.4 Cytotoxicity studies of nitrosyl clusters 1 and 2
We examined the in vitro cytotoxicity of nitrosyl cluster 1 or 2 against B16F10 (Figure
3) or L929 (Figure 4) cells in the absence or in the presence of light irradiation (at λ =
527 nm; dose of 0.25 or 0.41 J/cm2). Overall, both clusters are cytotoxic to melanoma
cells at different concentration ranges. In addition, relatively low concentrations (25
M) of cluster 2 appear to be toxic to healthy cells. Surprisingly, photostimulation
increases the cytotoxicity of complexes 1 and 2 only slightly within the concentrations
investigated here. The dark activity of clusters 1 and 2 can be assigned to NO release
mediated by reductive processes [16].
At each particular concentration, clusters 1 and 2 are less toxic to non-tumor
cells (L929) as compared to the murine melanoma cell line (B16F10) in the dark or
upon irradiation with visible light. At 50 µM, cluster 1 is not toxic to fibroblast cells at
all, but it lowers B16F10 cell viability to 54%. Cluster 2 elicits the same effect on
B16F10 cells (approximately 50% of cell viability) at only 1 µM. This concentration
reduces L929 cell viability to approximately 80%.
Our group was the first to report on the cytotoxicity of a triruthenium cluster;
more specifically, complex 3, toward melanoma cells [16]. In that case, complex 3 at 15
M was necessary to lower B16F10 cell viability to 50%. Light irradiation only raised
the cytotoxicity of complex 3 at higher concentration ranges (30 to 60 M). The results
of the present work are consistent with the aforementioned behavior and allow us to set
the following order of increasing cytotoxicity of triruthenium complexes toward
B16F10 melanoma cells with respect to the ancillary ligand: 4-acpy < 4-pic < 4-tbpy.
The cytoplasmic conditions of tumor cells, such as low O2(g) concentration,
atypical nutrient intake, low pH due to lactic acid production in anaerobic glycolysis,
and presence of higher glutathione levels [37,39], often create a favorable environment
for reduction processes, which favors NO release from these clusters even in the
absence of light irradiation.
10

ACCEPTED MANUSCRIPT
Figure 3. Cellular viability of B16F10 cells after treatment with different concentrations of nitrosyl
cluster 1 (A) or 2 (B), with (black bars) or without (white bars) light irradiation (λ=527 nm; dose of 0.41
J/cm2) for 24 h. Cell viability was assessed by an MTT assay and expressed as percent of the control.
Data are presented as the mean ± SEM of three independent experiments. * is used to represent the
statistically significant differences between the treatments and control, considering p < 0.05 according to
the ANOVA and post hoc Newman–Keuls tests.
Figure 4. Cell viability of L929 cells after treatment with different concentrations of nitrosyl cluster 1 (A)
or 2 (B), with (black bars) aor without (white bars) light irradiation (λ=527 nm; dose of 0.41 J/cm2) for 24
h. Cell viability was assessed by an MTT assay and expressed as a percent of the control. Data are
presented as the mean ± SEM of three independent experiments. * was used to represent the statistically
significant differences between the treatments and control, considering p < 0.05 according to the ANOVA
and post hoc Newman–Keuls tests.
Presumably, this effect is operating here, to contribute to both the
vasorelaxation and cytotoxic activities of complexes 1 and 2. However, this contribution
should be similar for both complexes because the first reduction potentials of clusters 1
and 2 differ only slightly (Table 1), and both are accessible in biological medium.
Therefore, the differences in the cytotoxicity of these triruthenium complexes must be a
consequence of structural aspects, namely the presence of distinct substituents in the
11

ACCEPTED MANUSCRIPT
ancillary pyridinic ligands. More specifically, we realized that higher degree of
methylation in the ancillary ligand of complexes 1, 2, and 3 [16] results in higher
cytotoxicity.
The importance of methylation in modulating biological activity is well
documented [45–47]. The methyl group presents many functions and can alter the
pharmacological properties of a molecule [24, 45]. Introduction of this group into
bioactive compounds makes them more lipophilic/hydrophobic. Hence, the structural
differences between clusters 1 (L = 4-acpy) and 2 (L = 4-tbpy) might directly impact
processes like internalization, subcellular localization, and interaction with several
biomolecules (proteins and/or DNA) or bio-structures including cellular membranes
[46].
2.5 Interaction of cluster 1 or 2 with Human Serum Albumin (HSA)
To illustrate our point that increased hydrophobicity through methylation promotes
interaction with biologically relevant targets, we investigated interaction of cluster 1 or
2 with HSA.
Biological activity is related to association between bioactive compounds and
biological targets. This directly affects compound bioavailability, distribution,
metabolism, toxicity, and in vivo half-life. Therefore, studying the interaction of cluster
1 or 2 with HSA (the major transport protein in blood plasma) is mandatory. We
evaluated this interaction by means of HSA fluorescence quenching by cluster 1 or 2.
Here, the assay was considered an indirect method to probe the hydrophobic nature of
the complexes under investigation [45–48, 50]. It has recently been shown that
compound 4 interacts with HSA through hydrophobic forces; the observed Kb suggests
that HSA can efficiently store and transport this complex in the body [18].
Figures 5A and 5B depict HSA fluorescence quenching in the presence of
increasing concentrations of cluster 1 or 2, respectively. Fluorescence quenching may
occur through two distinct mechanisms: static and/or dynamic [51–53]. It is rather
difficult to assign which mechanism takes place exclusively on the basis of fluorescence
measurements. Usually, it is possible to assign one of the mechanisms by analyzing how
KSV constants vary with temperature. The KSV quenching constant increases with rising
temperature when interaction is dynamic: because this interaction has a collisional
nature, it predominates at higher temperature. Nevertheless, static contribution
12

ACCEPTED MANUSCRIPT
(formation of a HSA-quencher adduct in the ground state) might be significant even
when Ksv increases with temperature. Hence, the quenching mechanism can be
determined by comparing the bimolecular quenching constant (kq) and limiting
diffusion rate constant of biomolecules (kd ≈ 2.0 ×1010M−1 S−1) [50, 53]. If kq is smaller
than kd, dynamic quenching is the main mechanism; otherwise, static quenching
prevails [50, 53].
To treat our data, we firstly employed the dynamic Stern-Volmer equation (1).
F0/F = KSV[Q] + 1 = 0kq[Q] + 1
(1)
where F0 and F are the fluorescence intensities in the absence or in the presence of the
quencher, respectively; KSV is the Stern–Volmer quenching constant; kq is the
bimolecular quenching constant; and τ0 is the average lifetime of the excited
biomolecule in the absence of a quencher (for HSA, τ0 is about 10−8 s) [48, 53].
Figure 5: Intensity of HSA solution emission with increasing concentration of cluster 1 (A) or cluster 2
(B). ([HSA] = 1.0 μM) (on the basis of the quencher concentration (0.0 M to 10-5 M) during 5-min
incubation with HAS).
We obtained the KSV of systems HSA-1 and HSA-2 at different temperatures
from the slopes of F0/F versus [Q] plots (Figure S19) and the kq values from the slopes
of F0/F versus [Q]xτ0 plots (Figure S20). Table 3 presents these values.
KSV increases slightly and significantly with temperature for clusters 1 and 2,
respectively. However, kq is greater than the diffusion rate constant of the biomolecule
(kd) for both complexes, which indicates that static quenching happens between the
13

ACCEPTED MANUSCRIPT
clusters and HSA. To confirm this observation, we performed lifetime measurements
(Figure S21 and S22). We did not observe variation in the HSA excited state lifetime at
any of the investigated temperatures, which attested to static quenching.
Table 3. Stern-Volmer quenching constants (Ksv and kq) of the systems HSA-1 and HSA-2 at different
temperatures
Cluster
1
T(K)
298
305
312
298
305
312
Ksv (M-1)
6.79 x 103
7.06 x 103
8.01 x 103
9.49 x 107
1.22 x 108
1.12 x 108
R
kq (M-1 S-1)
6.79 x 1011
7.06 x 1011
8.01 x 1011
9.49 x 1015
1.22 x 1016
1.12 x 1016
R
0.9923
0.9931
0.9951
0.9986
0.9987
0.9981
0.9923
0.9931
0.9955
0.9985
0.9986
0.9981
2
Furthermore, cluster 2 has atypically high KSV, even greater than the value
obtained for complex 4 and other coordination complexes [18, 55, 56]. Therefore,
cluster 2, which bears the methylated ligand 4-tbpy in its structure, interacts more
efficiently with HSA and is a more potent quencher.
Because we concluded that static quenching takes place for the systems HSA-1
and HSA-2, we calculated the binding constant Kb by using the modified Stern-Volmer
equation:
log(F0-F)/F = logKb + nlog[Q]
(2)
where F0 and F are the fluorescence intensities in the absence or in the presence of the
quencher, respectively; Kb is the binding constant; and n is the average number of
interacting clusters (or drug) per protein molecule. Plots of log[(F0−F)/F] versus log[Q]
yields log Kb as the intercept and n as the slope (Figure S23 and Table S1). For both
clusters, the number of interacting clusters per protein unit is close to 1, which suggests
that a single binding site is present in the HSA structure in these complexes, and that the
stoichiometry is 1:1. On the other hand, Kb is rather small (~103 M-1) despite the
efficiency with which complexes 1 and 2 quench HSA fluorescence [18, 54].
14

ACCEPTED MANUSCRIPT
Together, these results suggests that the nature of the interaction leading to the
formation of a ground state adduct between HSA and the clusters must be weak even for
complex 2 and presumably does not involve covalent bonds. Alternatively, quenching
might occur due to unspecific interactions on the protein surface rather than through
binding to specific sites. Indeed, the circular dichroism spectra collected for HSA do not
reveal significant changes upon addition of increasing amounts of complexes 1 or 2
(Figure S24), which shows that interaction between these species does not modify HSA
conformation upon binding.
The nature of the interaction forces between HSA and a molecule may comprise
hydrophobic interactions, hydrogen bonds, van der Waals forces, and electrostatic
interactions [18, 54]. Positive enthalpic and entropic changes indicate hydrophobic
interactions; negative variations indicate that hydrogen bonds and/or van der Waals
forces take place. Finally, negative ΔH and positive ΔS values attest to electrostatic
interaction [52]. We determined the thermodynamic parameters for the interactions
between cluster 1 or 2 and HSA according to the following equations:
lnKb = -H/RT + S/R (3)
G = -RTlnKb
(4)
where Kb is the binding constant at a temperature T, and R is the gas constant. ΔH and
ΔS are calculated from the slope and intercept values of a lnKb versus 1/T plot,
respectively (Figure S25). ΔG, which addresses interaction spontaneity, was determined
according to equation 4. The obtained values are available in Table S2. Positive ΔH and
ΔS suggest that hydrophobic forces direct binding between cluster 1 or 2 and HSA. For
cluster 2, values are more positive (H = 80.81 KJ/mol; S = 334.71 J/Kmol) as
compared to complexes 1 (H = 48.10 KJ/mol; S = 238.02 J/Kmol) and 4 [18]. This
fact highlights the more hydrophobic character of cluster 2 due to the presence of the
tert-butyl substituent in the structure of the ancillary ligands. For both clusters studied
herein, processes are spontaneous (ΔG > 0).
3. Conclusion
We have presented the synthesis and characterization of two novel triruthenium clusters
with general formula [Ru3(μ3-O)(μ-OOCCH3)6(NO)L2]PF6, where L = 4-acetylpyridine
15

ACCEPTED MANUSCRIPT
(4-acpy, 1) or 4-tert-butylpyridine (4-tbpy, 2)). Chemical characterization by
spectroscopic and electrochemical experiments has demonstrated the proposed structure
of the new nitrosyls. In addition, complexes 1 and 2 can release NO in solution under
irradiation at irrad = 377 nm and through chemical reduction with ascorbic acid.
More interestingly, these compounds have proven to be useful vasodilation
agents: they relax pre-contract rat aorta up to 100%. This ability apparently does not
depend on the structure of the complexes in our experimental conditions. In contrast, the
cytotoxicity of complexes 1 and 2 toward murine melanoma (B16F10 cells) and murine
fibroblasts (L929) is related to the structure of the ancillary ligands 4-acpy and 4-tbpy.
Cluster 2 has three methyl groups in the ancillary ligands 4-tbpy. This certainly
influences the increased cytotoxicity as compared to the results obtained for clusters 1
and 3 [16]. The presence of methyl groups also promotes stronger hydrophobic
interactions between cluster 2 and HSA as compared to cluster 1, which confirms that
methylation increases interactions with biomolecules.
All these results allow us to conclude that the choice of ancillary ligands is
critical to define the cytotoxic potential of the candidate drug when dealing with the
design of metallo drugs. However, the search for compounds with higher cytotoxicity
obtained through introduction of hydrophobic CH3 groups is a controversial issue—
methylation also increases cytotoxicity to healthy cells, as demonstrated here for cluster
2 at concentrations above 25 M. At this point, our data do not allow us to discuss
whether internalization of the whole complex or of the released NO molecule only
underlies the observed cytotoxicity. In other words, it is not yet known whether NO is
released outside or inside the cells. Nevertheless, our results constitute an interesting
example of how a molecule with increased hydrophobicity (through methylation)
enhances its interaction with biotargets and its cytotoxicity.
4. Experimental Section
4.1 Preparation and Purification
Complexes 1 and 2 were synthesized according to modifications of the procedure
published by Toma et al. [18, 19].
[Ru3(μ3-O)(μ-OOCCH3)6(NO)(L)2]PF6 (L = 4-acpy (1) and 4-tbpy (2)). The precursors
of [Ru3(μ3-O)(μ-OOCCH3)6(H2O)(4-acpy)2]PF6 (0.40 g) (1a) and [Ru3(μ3-O)(μ-
16

ACCEPTED MANUSCRIPT
OOCCH3)6(H2O)(4-tbpy)2]PF6 (0.40 g) (2b) were dissolved in DCM, and the solution
was degassed with Ar(g) for 20 min. The system was saturated with NO(g) for 24 h, and
the blue solution became violet. After this reaction time, the reaction medium was
evaporated to dryness, and the obtained purple solid was washed with petroleum ether
and dried under vacuum in desiccators. Further purification of cluster (2) was performed
by using alumina as stationary phase and a mixture of 10% ACN and 90% DCM as
mobile phase. In this system, NO(g) was generated by reaction of nitric acid and copper
pellets; two solutions were used as trapping system: concentrated NaOH(aq) solution (1
-
M), to remove NO2 (g) traces, and DCM solution, to saturate the reaction system solvent.
[Ru3(μ3-O)(μ-OOCCH3)6(NO)(4-acpy)2]PF6 (1). Yield: 0.26 g, 0.24 mmol, 65% (MW
= 1090.72 g/mol). Elemental Analysis Calcd for (Ru3O15C19H25N2PF6) C: 28.63% H:
2,96% N: 3.85% Found: C: 28.41% H: 3.01% N: 3.90%. 1H NMR (ACN-d3, 500 MHz):
δ (ppm) = 4.37 (12 H, s, CH3 (a)); 3.48 (6 H, s, CH3 (b)); 8.64 (4 H, d, J = 6.5 Hz, Ha);
4.03 (4 H, d, J = 6.5 Hz, Hβ); 2.70 (6 H, d, J = 4.0 Hz, Hγ). IR (KBr pellets, cm-1): 568
δ (FPF), 690 π (Ru3O), 838 νas(PF6), 1255 δs (CH3) 4-acpy, 1415 ⱱs (COO)Ac, 1589
v(pyridine ring), 1700 (C=O) 4-acpy, 3442 ⱱ (OH).
[Ru3(μ3-O)(μ-OOCCH3)6(NO)(4-tbpy)2)]PF6 (2). Yield: 0.30 g, 0.27 mmol, 75%; (MW
= 1118.96 g/mol). Elemental Analysis Calcd for (Ru3O14C21H31N2PF6) C: 32.20% H:
3.96% N: 3.76% Found: C: 33.40% H: 3.99% N: 3.96%. 1H NMR (ACN-d3, 500 MHz):
δ (ppm) = 3.79 (12 H, s, CH3 (a)); 3.15 (6 H, s, CH3 (b)); 8.22 (4 H, d, J = 6.6 Hz, Ha);
4.81 (4 H, d, J = 6.6 Hz, Hβ); 1.53 (18 H, s, Hγ). IR (KBR pellets, cm-1): 548 δ (FPF),
693 π (Ru3O), 830 νas(PF6), 1430 ⱱs (COO)Ac, 1591 v(pyridine ring), 2940 ⱱ (CH3) 4-
tbpy, 2971 δs (CH3) 4-tbpy, 3432 ⱱ (OH).
Warning: in the solid state, especially with exposure to light, we observed some degree
of decomposition for this compound, with substitution of NO ligands by solvent
molecules of crystallization, which is acetonitrile in this case (from the mobile phase of
the chromatographic column). This effect accounts for the poor match of the calculated
and experimental values of elemental analysis. The experimental values reported above
corresponds to a mixture of the original nitrosyl compound and the solvated species in
1
an approximated ratio of 60% nitrosyl : 40% solvated complex. We performed a H
NMR measurement in CD3OD in order to probe the presence of acetonitrile in our
product. The spectrum profile was the same as the one reported in CD3CN, with an
additional signal at 2.1 ppm ascribed to acetonitrile present in the sample. We also
17

ACCEPTED MANUSCRIPT
performed a control IR experiment in which we measured the spectrum of a freshly
prepared KBr pellet of compound 2 and measured it again after 2 days of exposure of
compound 2 to ambient light and heat. It was verified, by integration of the nNO band
area, the loss of 20% of NO.
4.2 Physical measurements
UV–Vis spectra were recorded on a UV Agilent 8453 spectrophotometer (190-1100
nm) equipped with xenon and tungsten lamps and a quartz cuvette with path length of
1.0 cm. IR spectra were acquired on an IR spectrophotometer Shimadzu Prestige 21
1
(400-4000 cm-1 with resolution 4 cm-1), from KBr pellets. H NMR spectra were
obtained on a Bruker Avance DRX- 500 500 MHz from 10-2 M CD3CN solutions. EPR
spectra were recorded in a Bruker EMX instrument operating at the X-band (frequency
= 9.48 GHz, power = 20mW, modulation frequency =100 kHz, modulation amplitude =
5–15 G), at 77 K, in the solid state and in ACN solution; Wilmad quartz tubes were
used. Cyclic voltammetry experiments were performed in a conventional three-electrode
cell (platinum as working electrode, platinum wire as auxiliary electrode, and Ag/AgCl
as reference electrode). The supporting electrolyte was tetrabutylammonium
hexafluorophosphate (TBAPF6) 0.1 M. Cluster solutions (10-3 M) were prepared in
ACN. Lifetime measurements were recorded on Tsunami - Spectra Physics; the selected
excitation wavelength was 281 nm at three incubation temperatures (298, 305, or 310
K). To a HSA solution (1.0 × 10−6 M), successive aliquots of cluster 1 or cluster 2
solutions (0, 1.4 x 10-5, 1.8 x 10-5, 2.2 x 10-5, 2.6 x 10-5, or 3.0 x 10-5 M), both prepared
in Trizma® buffer, were added. The signal detected as pulse excitation is called IRF
(instrument response function). The resulting curves were fitted by an exponential
adjust of the program Origin 8.0®. Circular Dichroism spectra were registered with a
Jasco 810 Spectropolarimeter (Jasco, Hachioji City, Tokyo, Japan). HSA (1.0 × 10−6 M)
CD spectra in the presence of complex 1 or 2 were acquired in a 1-cm quartz cuvette in
Trizma® buffer solutions (complex concentration: 1.5 x 10-5, 2.5 x 10-5, or 3.0 x 10-5 M).
4.3 NO release experiments
NO release from clusters 1 or 2 was evaluated by two methods. Photolysis of cluster 1
or 2 was accomplished from 10-5 M solutions in ACN in quartz cuvette with optical path
of 1 cm. Solutions were stirred continuously during the irradiation time (2 min, λ = 377
18

ACCEPTED MANUSCRIPT
nm). Irradiation was performed with a diode laser Colibri QuantumTech®. Light
intensities were determined by chemical actinometry (ferrioxalate salts) in order to
calculate quantum yields. The second method was based on a General Electric model
NOA-280i nitric oxide analyzer (NOA) under helium flow, which helped to analyze the
generated NO. NO was detected in the absence of ambient light; an ascorbic acid
aqueous solution (10-5 M) was used as reducing agent. First, 20 or 40 μL of cluster 1 or
2 in ACN solution (10-5 M) was added to the system. NO was detected in the presence
of UV irradiation. To this end, 60 μL of cluster 1 or 2 in ACN (10-5 M) was added to the
system, followed by irradiation with a UV lamp for 5 s (The Graphs are in SI-1 and SI-
2). The number of NO moles was based on the straight line obtained from the analytical
curve (Supplementary Data) of the device; the full area was based on the release curve.
4.4 Vasodilation assay and aortic ring preparation
Experiments were conducted in accordance with the Animal Ethics Committee,
Ribeirão Preto Campus, University of São Paulo, Brazil, under environment light. The
vasodilation assay was conducted in a bath system for isolated organs; the vascular
reactivity procedure was employed. First, Male Wistar rats (180-200 g) were stunned by
inhalation of isoflurane, euthanized by decapitation, and exsanguinated. The descending
thoracic aorta was rapidly dissected and cut into rings (4-mm length). Endothelium was
mechanically removed by gently rubbing the vessel surface of the rings with a thin wire.
Rings were maintained in a 10-mL organ chamber containing Krebs buffer solution
with the following composition (mM): NaCl 130, KCl 4.7, KH2PO4 1.2, MgSO4 1.2,
NaHCO3 14.9, glucose 5.5, and CaCl2 1.6. The solution was maintained at pH 7.4 and
gassed with 95% O2 and 5% CO2 at 37 °C. Initially, rings were stretched to a basal
tension of 1.5 g (previously determined by length–tension relationship experiments),
before they were allowed to equilibrate for 60 min in the bath fluid, which was changed
every 15–20 min. Since our studies required endothelium-denuded aorta, endothelial
integrity was qualitatively assessed by the degree of relaxation caused by acetylcholine
(ACh, 1 M) in the presence of a contractile tone induced by phenylephrine (0.1 µM).
Absence of endothelium was confirmed by lack of relaxation response to acetylcholine
1 μM in aortic rings pre-contracted with phenylephrine 0.1 µM.
Complex solutions were prepared in water (0.1 nM to 10 µM) by diluting the
stock concentration that had been prepared in DMSO (0.1 mM). To examine how the
19

ACCEPTED MANUSCRIPT
complexes affected smooth muscle cell relaxation, aortic rings were pre-contracted with
phenylephrine 0.1 µM. When the contraction had reached a plateau, complex solution
(0.1 nM to 10 µM) was cumulatively added.
4.5 Cell Culture
B16F10 (melanome murine) and L929 (fibroblast murine) cells were obtained from the
cell bank from the city of Rio de Janeiro. Cells were cultured in 75-cm2 flasks
containing RPMI medium (Roswell Parking Memorial Institute Medium – Gibco®) with
10% defined supplement fetal bovine serum plus penicillin G 100 IU/mL, streptomycin
100 mg/mL, and amphotericin 1 μg/mL. Cells were cultured until 75–90% confluence
in a humidified incubator at 37.0 °C with 5.0% CO2(g) for 24 h.
4.6 Cell viability assay
Cytotoxicity of the nitrosyl clusters was evaluated against two cell lines, murine
melanoma, B16F10, and murine fibroblast, L929. To this end, 2 x 104 cells were
incubated in 96-well cell culture plates for 24 h. After this period, treatment with cluster
1 (2.0, 5.0, 10, 25, 50, or 100 μM) or cluster 2 (0.1, 0.5, 1.0, 2.5, 5.0, or 25 μM)
previously dissolved in DMSO was conducted with or without light irradiation (λ = 527
nm; dose of 0.25 or 0.41 J/cm2). Graphs obtained at 0.25 J/cm2 are available as
Supplementary Data (S15). After the incubation time, cell viability was evaluated by the
MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-ditetrazolium bromide) colorimetric assay; wells
that did not contain any of the complexes were used as control [29]. The blue MTT
formazan precipitate was then dissolved in 100 μL of DMSO, and absorbance at 492 nm
was measured on a VARIAN CARY-50MPR multi-well plate reader. Cell viability is
expressed as the percentage of these values in treated cells as compared to non-treated
(control) cells. All experiments were carried out in triplicates on different occasions.
Data are presented as the mean ± standard error of triplicate cultures. For statistical
significance, differences were considered at the 0.05 level of significance with p ≤ 0.05;
One-way ANOVA- Newman-Keuls Multiple Comparison Test was used to compare
results from different treatments. Complexes were dissolved in DMSO and then in the
culture medium and serially diluted to the appropriate concentration, to give a final
DMSO concentration of 1%.
20

ACCEPTED MANUSCRIPT
4.7 Complex interaction with HSA
The interaction of the complexes with human serum albumin (HSA) was assayed in the
absence of ambient light, to ensure that the investigated species were the nitrosyls. HSA
was purchased from Sigma-Aldrich. Stock HSA solutions were prepared in aqueous
tris-buffer (pH = 7.2, 10-4 M) solutions on the basis of its molecular weight (66,463Da)
and kept refrigerated at 10 °C. A stock solution (10-3 M) of cluster 1 or 2 was prepared
in ACN (0 M – 1 x 10-5 M). Measurements were carried out at three incubation
temperatures (298, 305, or 312 K). Initially, the spectrum of the solution containing
HAS only was recorded after incubation for 5 min to adapt the conformation of the
macromolecule to the monitoring temperature. Sequentially, aliquots (30 uL) of on of
the complexes were added, and 2-min incubation was allowed before the luminescence
spectrum was recorded.
4.8 Fluorescence measurements
HSS fluorescence spectra at different temperatures were recorded on a Shimadzu
fluorescence spectrophotometer, model RF-5301PC, in a quartz cell with optical path of
1.0 cm. The fluorescence spectrophotometer was set up with excitation and emission slit
widths of 5 nm. The wavelength 280 nm was used for sample excitation. The
fluorescence intensities were corrected for absorption of the exciting light and inner
filter effects by means of the equation:
Fcorr = Fobs e(Aexc+Aem)/2
where Fcor and Fobs represent the corrected and observed fluorescence intensities, and
Aex and Aem are the absorbance of protein and ligand at the excitation and emission
wavelengths, respectively.
Acknowledgements
This work was supported by the Brazilian agencies Fundação de Amparo a Pesquisa do
Estado de São Paulo (FAPESP, grant number 2015/20302-7 and 2014/25561-8) and
Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, grant number
305415/2015-8). We thank Dra. Ana Maria da Costa Ferreira (IQ, USP) for the EPR
analysis; Prof. Lusiane Bendhack and PhD Michele Paulo (FCFRP, USP) for the
21

ACCEPTED MANUSCRIPT
vasorelaxation experiments; Prof. Amando S. Ito (FFCLRP, USP) for the use of the
Tsunami - Spectra Physics spectrophotometer; and Dr. Juliana Cristina Moraes
Biazzotto, Clovis Junior, and Prof. Dr. Roberto Santana da Silva (FCFRP, USP) for
their help with the acquisition of in vitro anti-cancer data.
Abbreviation List
4-acpy = 4-acetylpyridine
4-tbpy = 4-tert-butylpyridine
4-pic = 4-methylpyridine
3-pic = 3-methylpyridine
ACN = acetonitrile
CLCT = cluster-to-ligand charge transfer
DCM = dichloromethane
DMSO = dimethylsulfoxide
IC = intra-cluster transition
HSA = human serum albumin
MTT = 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide
NOA = NO analyzer
TBA = tetrabutylammonium
Appendix A. Supplementary data
Supplementary data related to this article can be found at http:// dx.doi.org
References
[1] A.W. Mond, The acetates of ruthenium, J. Am. Chem. Soc. (Resumed) (1930) 1247-
1249.
[2] H.E. Toma, K. Araki, A.D.P Alexiou, S. Nikolaou, S. Dovidauskas. Monomeric and
extended oxo-centered triruthenium clusters. Coord. Chem. Rev. 221 (2001) 187–234.
[3] G.D. Starla, J.C. Goeltz, B.J. Lear, C.P. Kubiak. Inter-or intramolecular electron
transfer between triruthenium clusters: we’ll cross that bridge when we come to it.
Coord. Chem. Rev 254 (2010) 331-345.
[4] G.C. John, E.E. Benson, C.P. Kubiak. Electronic structural effects in self-exchange
reactions. J. Phys. Chem. B. 114 (2010) 14729-14734.
22

ACCEPTED MANUSCRIPT
[5] A.D.P. Alexiou, H.E toma. NMR Spectroscopic Correlations for a Series of
Triangular μ-Oxoruthenium Acetate Clusters containing Substituted Pyridine Ligands.
J. Chem. Research (S). 9 (1997) 338-339.
[6] S.H. Toma, J.J. Santos, K. Araki, H.E. Toma. Supramolecular Approach to Gold
Nanoparticle/Triruthenium Cluster Hybrid Materials and Interfaces. Eur. J. Inorg.
Chem. 10 (2011) 1640-1648.
[7] S. Nikolaou, A.L.B. Formiga, H.E. Toma. Probing the electronic delocalization in a
cyclic pyrazine ruthenium cluster hexamer. Inorg. Chem. Commun. 9 (2010) 1032-
1035.
[8] S. H. Toma, J. J. Santos, S. Nikolaou, K. Araki, H. E. Toma. Direct assembly of a
metallodendrimer encompassing seven triruthenium clusters units. Inorganica Chim.
Acta. 390 (2012) 148-153.
[9] H. Zhang, Y. Sasaki, M. Abe, Y. Zhang, S. YE, M. Osawa, K. Uosaki.
Electrochemical and infrared spectroscopic study of the self-assembled monolayer of a
cyano-bridged dimeric triruthenium complex on gold surface. J. Electroanal. Chem.
(2014) 51–55.
[10] Z. Hua-Xin, Y. Sasaki, Y Zhang, S. Ye, M.O. Masaaki Abe, K. Uosaki. Synthesis
and properties of the cyano complex of oxo-centered triruthenium core [Ru3(μ3-O)(μ-
CH3COO)6(pyridine)2(CN)]. Inorg. Chem. 3 (2013)1288-1294.
[11] S. Nikolaou, H.E. Toma. A Convergent Approach for the Generation of
Dendrimers Containing the [Ru3O(CH3COO)6] Electroactive Core. Eur. J. Inorg. Chem
14 (2008) 2266-2271.
[12] L. Mimi, Z. J. Lim, Y. Y. Gwee, A. Levina, P. A. Lay. Characterization of a
Ruthenium (III)/NAMI‐A Adduct with Bovine Serum Albumin that Exhibits a High
Anti‐Metastatic Activity. Angew. Chem. Int. Ed. 9 (2010) 1661-1664.
[13] T. Jiří, L.E.H Paul, J. Furrer, B. Therrien, G. Süss-Fink. Cationic triruthenium (III)
oxo complexes of the type [Ru3O(OAc)6L3]+ containing imidazole, pyrazole, thiazole
and oxazole ligands: Synthesis, molecular structure, and cytotoxicity. Inorganica Chim.
Acta 423 (2014) 16-20.
[14] O. Hiroyoshi, N. Fujiwara, T. Yamaguchi. Mixed-Valence State of a Pyrazine-
Bridged Dimer of Oxocarboxylatotriruthenium Complexes with a Nitrosyl Ligand.
Inorg. Chem. 16 (2011) 7382-7384.
[15] L. Benjamin, C. P. Kubiak. Charge gating and electronic delocalization over a
denderimeric assembly of trinuclear ruthenium clusters. Inorg. Chem 18 (2006) 7041-
7043.
23

ACCEPTED MANUSCRIPT
[16] Z.A. Carneiro, J.C. Biazzotto, A.D. P Alexiou, S. Nikolaou. Nitric oxide
photorelease from a trinuclear ruthenium nitrosyl complex and its in vitro cytotoxicity
against melanoma cells. J. Inorg. Biochem. 134 (2014) 36-38.
[17] N. Cacita. B. Possato, C.F.N. Silva, M. Paulo, A. L. B. Formiga, L. M. Bendhack,
S. Nikolaou. Investigation of a novel trinuclear μ-oxo ruthenium complex as a potential
nitric oxide releaser for biological purposes. Inorg. Chim. Acta. 429 (2015) 114-121.
[18] N. Cacita, S. Nikolaou. Studying the interaction between trinuclear ruthenium
complexes and human serum albumin by means of fluorescence quenching. J. Lumin.
169 (2016) 115-120.
[19] H.E. Toma, A.D.P Alexiou, S. Dovidauskas. Extended Electronic Interactions in a
Triangular μ‐Oxotriruthenium Acetate Cluster Containing Nitric Oxide. Eur. J. Inorg.
Chem 11 (2002) 3010-3017.
[20] H. E. Toma, A.D.P Alexiou, A.L.B. Formiga, M. Nakamura, S. Dovidauskas, M.
N. Eberlin, D. M. Tomazela. A nitric oxide releaser based on the μ-oxo-hexaacetate-bis
(4-methylpyridine) triruthenium nitrosyl complex. Inorganica Chim. Acta 358 (2005)
2891-2899.
[21] Z. Wei, Y. Zhang, M. Abe, K. Uosaki, M. Osawa, Y. Sasaki, S. Ye. Surface
coordination of nitric oxide to a self-assembled monolayer of a triruthenium cluster: An
in situ infrared spectroscopic study. Langmuir 15 (2008) 8027-8035.
[22] Z. Yi, Y. Tong, M. Abe, K. Uosaki, M. Osawa, Y. Sasaki, S. Ye. Fabrication of
photochemical pattern on a self-assembled monolayer (SAM) of a ruthenium cluster
under electrochemical control. J. Mater. Chem. 2 (2009) 261-267.
[23] S. Yoichi, M. Abe. Ligand–ligand redox interaction through some metal‐cluster
units. Chem. Rec. 5 (2004) 279-290.
[24] E J. Barreiro, A. E. Kꢀmmerle, C.A.M. Fraga. The methylation effect in medicinal
chemistry. Chem. Rev. 9 (2011) 5215-5246.
[25] D. Bonaventura, R.G. Lima, J. A. Vercesi, R. S. Silva, L. M. Bendhack.
Comparison of the mechanisms underlying the relaxation induced by two nitric oxide
donors: sodium nitroprusside and a new ruthenium complex. Vasc. Pharmacol. 3 (2007)
215-222.
[26] E.J. Barreiro, C.A.M. Fraga. Química Medicinal-: As bases moleculares da ação
dos fármacos. Artmed Editora, 2014.
[27] K.C. Chu, In The Basis of Medicinal Chemistry/Burger’s Medicinal Chemistry;
John Wiley: New York, 1980.
[28] P. Bazzini, C. G. Wermuth, In The Practice of Medicinal Chemistry; Academic
Press: San Diego, 2008.
24

ACCEPTED MANUSCRIPT
[29] T. Mosmann. Rapid colorimetric assay for cellular growth and survival: application
to proliferation and cytotoxicity assays. J. Immunol. Methods. 65 (1983) 55-63.
[30] J.A. Baumann, D.J. Salmon, T. Stephen, T. J. Meyer, W. E. H. Field. Electronic
structure and redox properties of the clusters [Ru3O(CH3CO2)6L3]n+.Inorg. Chem. 12
(1978) 3342-3350.
[31] A. F. Cotton, J. G. Norman. Structural characterization of a basic trinuclear
ruthenium acetate. Inorganica Chim. Acta. 6 (1972) 411-419.
[32] H. E. Toma, F. M. Matsumoto, C. Cipriano. Spectroelectrochemistry of the
hexanuclear cluster [Ru3O(acetate)6-μ-(pyrazine)3-{Fe(CN)5}3]n− and of its modified
nickel electrode in aqueous solution. J. Electroanal. Chem. 2 (1993) 261-270.
[33] A. Masaaki, Y. Sasaki, A. Nagasawa, T. Ito. Kinetic Studies on the Terminal
Pyridine Exchange Reactions of Some Oxo-Acetate Bridged Trinuclear Ruthenium
Complexes. Influence of the Metal Oxidation State and Non-Leaving Terminal Ligand
(CO). Bull. Chem. Soc. Jpn. 5 (1992) 1411-1414.
[34] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds. 5th ed. John Wiley & Sons INC, 1997.
[35] D. Teng‐Yuan, H. Lee, T. Lee, C. Hsieh. Magnetic and Spectral Properties of
Heterotrinuclear Acetates [Ru2Co(μ3‐O)(μ‐CH3CO2)6(Py)3]n+ (n= 0 or 1). Journal of the
Chinese Chemical Society. 5 (1992) 393-399.
[36] O. Akihiro, A. Tokiwa-Yamamoto, M. Abe, T. Ito, Y. Sasaki, K, Umakoshi, R. D.
Cannon. Structure and Vibrational Spectra of Trinuclear Metal Cluster Complexes. A
Question of Symmetry. Chem. Lett. 2 (1995) 97-98.
[37] S. Dovidauskas, H. E. Toma, K. Araki, H. C. Sacco, Y. Iamamoto. (5, 10, 15, 20-
Tetra (4-pyridil) porphinato) manganese (III) acetate modified by four μ3-oxo-
triruthenium acetate clusters: synthesis, characterization, electrochemical behavior and
catalytic activity. Inorganica Chim. Acta. 2 (2000) 206-213.
[38] K, Araki, S. Dovidauskas, H. Winnischofer, A.D.P, Alexiou, H. E. Toma. A new
highly efficient tetra-electronic catalyst based on a cobalt porphyrin bound to four μ3-
oxo-ruthenium acetate clusters. J. Electroanal. Chem. 1 (2001) 152-160.
[39] H.E. Toma, A.DP Alexiou, S. Nikolaou, S. Dovidauskas. 13C NMR spectra of
triangular -oxoruthenium acetate clusters revisited: HETCOR study of two pyridine and
4, 4'-bipyridine derivatives. Magn. Reson. Chem. 37 (1999) 322-324.
[40] M. Moreira, C.F.N. Silva, R.B. Pesci, V. M. Deflon, S. Nikolaou. "Revisiting oxo-
centered carbonyl-triruthenium clusters: investigating CO photorelease and some
spectroscopic and electrochemical correlations. Dalton Trans. 42 (2016) 16799-16809.
25

ACCEPTED MANUSCRIPT
[41]
Sievers
Nitric
Oxide
Analyzer
Overview.
http://
www.geinstruments.com/products-and-services/nitric-oxide-analyzer/. (accessed 14
August 2016).
[42] R. G. De lima, M. G. Sauaia, D. Bonaventura, A. C. Tedesco, L. M. Bendhack, R.
S. Silva. Influence of ancillary ligand L in the nitric oxide photorelease by the
Ru(L)(tpy) NO]3+ complex and its vasodilator activity based on visible light irradiation.
Inorganica Chim. Acta. 8 (2006) 2543-2549.
[43] M. S. P. Marchesi, S. A. Cicillini, A. C. L. Prazias, L. M. Bendhack, A. A. Batista,
R. S. Silva. "Chemical mechanism of controlled nitric oxide release from trans-
[RuCl([15]aneN4)NO](PF6)2 as a vasorelaxant agent. Transition Met. Chem. 5 (2012)
475-479.
[44] R. G. De Lima, B. R. Silva, R. S. Silva, L. M. Bendhack. Ruthenium complexes as
NO donors for vascular relaxation induction. Molecules 7 (2014) 9628-9654.
[45] A. E Sales, L. Breydo, T.S. Porto, A.L.F, Porto, V.N Uversky. Hydrophobicity-
dependent effects of polymers on different protein conformations. RSC Advances 49
(2016) 42971-42983.
[46] C. S. Leung., S. F. Siegfried, J. Tirado-rives, W. L. Jorgensen. Methyl Effects on
Protein−Ligand Binding. J. Med. Chem. 55 (2012) 4489−4500.
[47] N. Bijari,Y. Shokoohinia, M.R.Ashrafi-Shk,S. S. Ranjbar, M. Parvaneh, M.
Moieni-Arya, R. Khodarahmi. Spectroscopic study of interaction between osthole and
human serum albumin: Identification of possible binding site of the compound. J.
Lumin. 143 (2013) 328-336.
[48] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd ed, Kluwer Aca-
demic Publisher, New York, Boston, Dordrecht, London, Moscow, 2006.
[49] E. Alessio. Thirty Years of the Drug Candidate NAMI‐A and the Myths in the
Field of Ruthenium Anticancer Compounds: A Personal Perspective. Eur. J. Inorg.
Chem. 12 (2017) 1549 - 1560.
[50] W.R Ware. Oxygen quenching of fluorescence in solution: an experimental study
of the diffusion process. J. Phys. Chem. 3 (1962) 455-458.
[51] J. Min, X. Meng-xia, Z. Dong, L. Yuan, L. Xiao-yu, C. Xing. Spectroscopic studies
on the interaction of cinnamic acid and its hydroxyl derivatives with human serum
albumin. J. Mol. Struct. 1 (2004) 71-80.
[52] R. Liu, X. Yu, W. Gao, D. Ji, F. Yang, X. Li, J. Chen, H. Tao, H. Huang, P. Yi.
Study on the interaction between salvianic acid A sodium and bovine serum albumin by
spectroscopic methods. Spectrochimica Acta Part A: Molecular and Biomolecular
Spectroscopy. 5 (2011) 1535-1539.
26

ACCEPTED MANUSCRIPT
[53] T.G. Dewey. Biophysical and Biochemical Aspects of Fluorescence Spectroscopy,
first ed.,Springer, US, NewYork, 1991.
[54] N. Bijari, Y. Schokoohinia, M.R. Ashrafi-kooshk, S. Ranjbar, S. Parvaneh, M.
Moieni-arya, R. Khodarahmi. Spectroscopic study of interaction between osthole and
human serum albumin: Identification of possible binding site of the compound. J.
Lumin. 143 (2013) 328-336.
[55] P.D. Ross, S Subramanian. Thermodynamics of protein association reactions:
forces to stability. Biochemistry 20 (1981) 3096 - 3102.
[56] J.T Peters. All About Albumin: Biochemistry, Genetics, and Medical Applications,
Academic Press: San Diego, 1996.
27