Salen?manganese complexes for controlling ROS damage: Neuroprotective effects, antioxidant activity and kinetic studies

Article abstract of DOI:10.1016/j.jinorgbio.2019.110918

A new manganese(III) complex [MnL¹(DCA)(H₂O)](H₂O)_, 1 [H₂L¹ is the chelating ligand N,N′-bis(2-hydroxy-3-methoxybenzylidene)-1,2-diaminopropane, and DCA is dicyanamide], has been prepared a

Full text of DOI:10.1016/j.jinorgbio.2019.110918

JournalofInorganicBiochemistry203(2020)110918
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Salen‑manganese complexes for controlling ROS damage: Neuroprotective
effects, antioxidant activity and kinetic studies☆
Lara Roucoa, Andrea Liberatob, M. Jesús Fernández-Trujillob, Angeles Máñezb,
⁎
⁎
Manuel G. Basalloteb, , Rebeca Alvariñoc, Amparo Alfonsoc, Luis M. Botanac,
,
⁎
Marcelino Maneiroa,
a Departamento de Química Inorgánica, Facultade de Ciencias, Campus Terra, Universidade de Santiago de Compostela, Lugo, Spain
b Departamento de Ciencia de los Materiales e Ingeniería Metalúrgica y Química Inorgánica, Facultad de Ciencias, Universidad de Cádiz, Puerto Real, Cádiz, Spain
c Departamento. de Farmacología, Facultade de Veterinaria, Campus Terra, Universidade de Santiago de Compostela, Lugo, Spain
A R T I C L E I N F O
A B S T R A C T
Keywords:
A new manganese(III) complex [MnL1(DCA)(H2O)](H2O), 1 [H2L1 is the chelating ligand N,N′-bis(2-hydroxy-3-
methoxybenzylidene)-1,2-diaminopropane, and DCA is dicyanamide], has been prepared and characterized by
different analytical and spectroscopic techniques. The tetragonally elongated octahedral geometry for the
manganese coordination sphere was revealed by X-ray diffraction studies for 1. The antioxidant behavior of this
complex and other manganese(III)-salen type complexes was tested through superoxide dismutase and catalase
probes, and through the study of their neuroprotective effects in SH-SY5Y neuroblastoma cells. In this human
neuronal model, these model complexes were found to improve cell survival in an oxidative stress model. During
studies aimed to getting a better understanding of the kinetics of the processes involved in this antioxidant
behavior, an important effect on the solvent in the kinetics of reaction of the complexes with H2O2 was revealed
that suggests a change in the mechanism of reaction of the complexes. The kinetic data in methanol and buffered
aqueous solutions correlate well with the results of the test of catalase activity, thus showing that the rate
determining step in the catalytic cycle corresponds to the initial reaction of the complexes with H2O2.
Manganese
Oxidative stress
Neuroprotection
Schiff bases
1. Introduction
alterations [5]. For instance, AD has already become a healthcare
challenge facing our society today. According to the new report, the loss
Reactive oxygen species (ROS) like superoxide radical anion
(O2%−), hydrogen peroxide (H2O2) or hydroxyl radical (OH%), are
generated as by-products of mitochondrial electron transport of aerobic
respiration [1]. ROS levels are regulated by antioxidant enzymes as
superoxide dismutases (SOD), catalases (CAT), and glutathione perox-
idases (GPx) [2]. An imbalance in this oxidant-antioxidant status could
determine the extent of cell damage, which is particularly risky for
brain cells that are susceptible to oxidative damage due to their high
rate of energy consumption, and other factors as the relative low levels
of brain antioxidant defences or the high concentrations of poly-
unsaturated fatty acids in this organ. Excessive ROS levels have been
associated with neurodegenerative disorders like Alzheimer's Disease
(AD), Parkinson's Disease (PD), amyotrophic lateral sclerosis (ALS) or
Huntington's Disease [3,4]. All these neurodegenerative diseases may
shorten life span, but, most notably, they affect the life quality of aging
populations causing cognitive deficits, memory loss, and behavioral
of cognitive abilities provoked by AD currently accounts for 50 million
cases worldwide, but this number will be more than triple to 150 mil-
lion by 2050 [6]. In this context, efforts have been directed toward the
search of antioxidant enzyme mimics [7–9]. This research determina-
tion has often focused on manganese complexes with Schiff bases as Mn
ions do not induce Fenton chemistry [10–16]. It is well established that
these complexes protect cells from oxidative damage in animal models
and lead to benefits in AD and PD, stroke, motor neuron disease,
multiple sclerosis, and excitotoxic neural injury [17–19]. However, the
mechanism of action of these biomimetic models remain exiguously
understood. They have been reported as antioxidant agents which can
behave as SOD, catalases or peroxidases mimics, but more studies are
needed for a more comprehensive understanding of underlying mole-
cular mechanisms and associated pathophysiological pathways.
We have previously reported a number of these biomimetic models
and tried to delineate the functional roles of the bridging ligands and
☆ In memory of the late Professor Juan Manuel Salas, an inspiring force for Spanish bioinorganic chemists.
⁎ Corresponding authors.
E-mail address: marcelino.maneiro@usc.es (M. Maneiro).
https://doi.org/10.1016/j.jinorgbio.2019.110918
Received 17 July 2019; Received in revised form 4 November 2019; Accepted 10 November 2019
Availableonline14November2019
0162-0134/©2019ElsevierInc.Allrightsreserved.

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
Scheme 1. Structures for H2L1-H2L3.
structural motifs that play a key role in their antioxidant activity
[20–23]. Thus, we have found how the combination of hydrogen
bonding networks and π-aryl interactions would lead to active self-as-
sembled μ-aquo dimeric systems [20–24], and also we have established
an increase of the catalytic activity for this type of biomimetic models
in function of the tetragonal elongation (ratio between the manganese-
axial donor atom distances and the manganese-equatorial donor atom
distances) of the octahedral manganese complexes [21]. This elonga-
tion occurs when the degeneracy is broken by the stabilization (low-
ering in energy) of the d orbitals with a z component, while the orbitals
without a z component are destabilized (higher in energy). The tetra-
gonal elongation enhances the lability of an axial position where the
substrate molecule can be subsequently accommodated in this site,
following an inner-sphere electron-transfer mechanism. As a result,
complexes with square-pyramidal or tetragonally elongated octahedral
geometries turn into efficient biomimetic catalysts for ROS control
[22–25].
hydroxybenzaldehyde and the corresponding diamine in methanol, and
their synthesis and characterization have been already reported [21].
Complexes 1–3 were also synthesized according to the literature; com-
plexes 2, Mn2L22(H2O)2(DCA)2, and 3, MnL3(H2O)2(DCA), have been pre-
viously obtained [29,30], where DCA is the dicyanamide ion. Experi-
mental procedure and characterization data for 1 are collected below:
MnL1(H2O)2(DCA) (1) To a methanolic solution of 0.675 mmol
(0.25 g) of H2L1, 0.675 mmol (0.17 g) of Mn(CH3COO)2 was added and
the stirred solution changed its initial yellow color to brown. After
30 min of stirring with gentle heating, 0.675 mmol (0.06 g) of NaN
(CN)2 in 10 mL of methanol was added. The solution was then con-
centrated by slow evaporation. The complex was obtained as brown
crystals, which were isolated by filtration, washed with diethyl ether
and dried in air. Yield 58%. Anal. Calcd. for C21H24MnN5O6 (497.4): C,
50.7; H, 4.8; N, 14.1. Found: C, 50.5; H, 4.6; N, 14.5%. MS ES (m/z) 395
[MnL]+, 462 [MnL(DCA)]+; IR (KBr, cm−1): ν(OeH) 3422, ν(C]N)
1624, ν(CeO) 1260, νsym(C^N) 2149, νasym(C^N) 2251; 1H NMR
In the work described here, we further explored this scheme. For this
purpose, we selected the dianionic H2L1-H2L3 Schiff base ligands
(Scheme 1; N,N′-bis(2-hydroxy-3-methoxybenzylidene)-1,2-diaminopro-
(DMSO‑d6, ppm): δ −17.3 (H4), −24.7 (H5). μ = 4.9 BM. E1/
2 = −72 mV. ΔEp = 110 mV. ΛM = 45 μS.
pane,
H2L1;
N,N′-bis(2-hydroxy-3-methoxybenzylidene)-1,2-diami-
2.2. Physical and analytical measurements for compound characterization
noethane, H2L2; and N,N′-bis(2-hydroxy-3-methoxybenzylidene)-1,2-
diamino-2-methylpropane, H2L3), which contains four potential donor
atoms: two imine nitrogen atoms and two phenolic oxygen atoms. The
insertion of the outer methoxy groups would favour the aggregation of
neighboring complexes through hydrogen bonding in order to get di-
meric systems. The inductive electron donating effect of the methoxy
group also may facilitate to achieve high-valent manganese centres
during catalysis. Moreover, we selected a bulky ancillary ligand like di-
cyanamide ions in order to favour a higher tetragonal elongation for the
manganese complex, which is expected to increase the lability thus fa-
vouring reaction with oxidants. We examined the cytoprotective effect of
the resulting complexes 1–3 against H2O2-induced oxidative stress in
human neuroblastoma cells. Human neuroblastoma SH-SY5Y cells have
been widely used as a cell model system for studying oxidative stress
[26,27], even though this paper report for the first time the neuropro-
tective results of Mn(III)-Schiff base complexes for this human neuronal
model. Finally, we performed kinetic studies of the reaction of complexes
1–3 versus H2O2 and tertbutyl hydroperoxide (TBHP), aimed at un-
ravelling some clues for their mode of action in their antioxidant activity.
Elemental analyses were performed on a Carlo Erba Model 1108
CHNS-O elemental analyser. The IR spectra were recorded as KBr pel-
lets on a Bio-Rad FTS 135 spectrophotometer. 1H NMR spectra were
recorded on a Bruker AC-300 spectrophotometer using DMSO‑d6 as
solvent and SiMe4 as an internal reference. The electrospray mass
spectra of the compounds were obtained on a Hewlett-Packard model
LC-MSD 1100 instrument. Room-temperature magnetic susceptibilities
were measured using a digital measurement system MSB-MKI, cali-
brated using mercury tetrakis(isothiocyanato)cobaltate(II) as a sus-
ceptibility standard. Electronic spectra were recorded on a Cary 230
spectrometer. Conductivities of 10−3 M solutions in DMF were mea-
sured on a Crison microCM 2200 conductivemeter. Electrochemical
measurements were performed using an EG&G Princeton Applied
Research model 273 potentiostat using three electrode configuration.
The working electrode was a Metrohm model 6.1204.000 graphite disc
while a Pt wire and a saturated calomel electrode served as counter and
reference electrodes, respectively. Measurements were made with ca.
10−3
M
solutions of complexes in DMF using 0.1 M tetra-
ethylammonium perchlorate as a supporting electrolyte.
2. Experimental section
2.3. Studies on catalase-like function
2.1. Materials for synthesis and chemical reactions
A 10 mL flask containing the solution of the complexes in methanol
(3 mL, 1 mM) was sealed with septum and connected to a gas-mea-
suring burette (precision of 0.1 mL) through double-ended needle. The
solution was stirred at constant temperature on a water bath. The cat-
alysis was initiated by introducing H2O2 solution (1 mL, 2.5 M) using
syringe, and the evolved dioxygen was volumetrically measured.
All solvents, 3-methoxy-2-hydroxybenzaldehyde, ethylenediamine,
1,2-diaminepropane, 1,2-diamine-2-methylpropane, manganese(II)
acetate, sodium dicyanamide, hydrogen peroxide and TBHP are com-
mercially available and were used without further purification. The
manganese complex EUK-134, with chemical name chloro[[2,2′-[1,2-
ethanediylbis[(nitrilo-κN)methylidyne]]bis[6-methoxyphenolato-
κO]]]‑manganese, was also purchased and used without further pur-
ification.
2.4. Cell culture
Schiff base ligands H2L1-H2L3 for complexes 1–3 were prepared ac-
cording to the literature [28] by condensation of 3-methoxy-2-
All the experiments were performed as previously described [27].
Human neuroblastoma SH-SY5Y cell line was purchased from the
2

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
American Type Culture Collection (ATCC), number CRL2266. The cells
were maintained in Dulbecco's modified Eagle's medium: Nutrient Mix
F-12 (DMEM/F-12) supplemented with 10% foetal bovine serum (FBS),
glutamax, 100 U/mL penicillin and 100 μg/mL streptomycin at 37 °C in
a humidified atmosphere of 5% CO2 and 95% air. Cells were dissociated
weekly using 0.05% trypsin/EDTA. All reagents were provided by
Thermo Fisher Scientific (Waltham, MA, USA).
provided with a multicell accessory. The stopped-flow experiments
were carried out with an Applied Photophysics SX18-MV instrument
provided with a photodiode array detector. Experiments in acetonitrile
and methanol solvents were carried out at 25.0
0.1 °C by mixing a
solution of the Mn complexes with a solution containing an excess of
H2O2; the final concentrations in the resulting mixture were
5 × 10−5 M for the complex and (0.6–5.0) × 10−2 M for the oxidant.
Experiments in aqueous solution were carried out at 37.0
0.1 °C by
2.5. Cell viability assay
mixing 0.1 mL of a solution of the Mn complexes in DMSO
(1 × 10−6 M) with 2.4 mL of an aqueous solution buffered at
pH = 7.25 containing an excess of H2O2 or TBHP; the final con-
centration of the complex was 1 × 10−6 M and the oxidant was in the
range (0.08–9.0) × 10−3 M. In all cases, the spectral changes were
measured over a wide wavelength range and analysed with the program
Specfit [33] using the kinetic models indicated in the next sections.
One day prior to experiments, cells were seeded in 96-well plates at
a density of 5 × 104 cells per well. SH-SY5Y cells were treated with
biomimetic models 1–3 at different concentrations (0.001, 0.01, 0.1, 1
and 10 μM) for 24 h. The effect of compounds on cell viability was
evaluated by MTT (3-(4, 5-dimethyl thiazol-2-yl)-2,5-diphenyl tetra-
zolium bromide) assay [31,32].SH-SY5Y cells were rinsed with saline
solution and 200 μL of MTT (500 μg/mL) dissolved in saline buffer were
added to each well. Following 1 h of incubation at 37 °C, SH-SY5Y cells
were disaggregated with 5% sodium dodecyl sulphate. Absorbance of
formazan crystals was measured at 595 nm with a spectrophotometer
plate reader. Saponin from quillaja bark was used as a death control and
its absorbance was subtracted from the other data.
2.9. Crystallographic studies
Single crystals of complex 1, suitable for X-ray diffraction studies,
were obtained by slow evaporation of the methanolic solution at room
temperature.
Table 1 collects detailed crystal data collection and refinement for
1. Measurements were made on a Bruker X8 APEXII CCD diffractometer
employing graphite-monochromated Mo-Kα radiation at 100 K. The
structures were solved by direct methods [34] and finally refined by
full-matrix least- squares base on F2. An empirical absorption correction
was applied using SADABS [35]. All non-hydrogen atoms were assigned
anisotropic displacement parameters in the refinement. The hydrogen
atoms were included in the model at geometrically calculated positions.
Refinement of F2 against ALL reflections. The weighted R-factor wR
is based on F2, conventional R-factors R are based on F, with F set to
zero for negative F2. The threshold expression of F2 > 2σ(F2) is used
only for calculating R-factors(gt) etc. and is not relevant to the choice of
reflections for refinement. R-factors based on F2 are statistically about
twice as large as those based on F, and R-factors based on ALL data will
be even larger.
2.6. Superoxide dismutase (SOD) activity assay
The SODs levels were analysed with SOD assay kit (Sigma-Aldrich).
SH-SY5Y were seeded in 24 well plates at a density of 1 × 105 cells per
well. Cells were co-treated with biomimetic models 1–3 and EUK-134 at
different concentrations (0.001, 0.01, 0.1, 1 and 10 μM) for 24 h. Cells
were washed with ice-cold PBS (phosphate-buffer saline) and 50 μL of
lysis buffer (0.1 M Tris- HCl, pH 7.4 containing 0.5% Triton X-100,
5 mM β-mercaptoethanol and 0.1 mg/mL phenylmethylsulfonyl
fluoride) were added to each well. The SOD assay was performed fol-
lowing manufacture's protocol. SODs activity for each compound was
quantified with commercial SOD and signal was normalized by protein
concentration, which was quantified by Bradford method. Half maximal
effective concentration (EC50) values were calculated using a non-
linear regression model with Graph Pad Prism 6 software.
2.7. Neuroprotection and mitochondrial membrane potential assays
Table 1
The ability of compounds to protect cells from TBHP damage was
also evaluated with MTT probe. Cells were co-treated with complexes
1–3 at non-toxic concentrations (0.001, 0.01, 0.1, 1 and 10 μM) and
75 μM TBHP for 6 h, and the assay was carried out as described above.
All experiments were performed at least three times. The mitochondrial
membrane potential (ΔΨm) was assessed using the tetra-
methylrhodamine methyl ester (TMRM) probe. SH-SY5Y cells were
seeded in 96-well plates at a density of 5 × 104 cells per well and al-
lowed to attach for 24 h. After this time, cells were treated with 75 μM
TBHP and biomimetic models 1–3 at different concentrations (0.001,
0.01, 0.1 and 1) for 6 h. Cells were washed twice with saline solution
and 1 μM TMRM was added to each well for 30 min at 37 °C. After in-
cubation, cells were solubilised with DMSO and H2O at 50% and the
fluorescence was monitored with a spectrophotometer plate reader
(535 nm excitation and 590 nm emission). All experiments were per-
formed at least three times. The endogenous antioxidant vitamin E
(vitE) at 25 μM was used to validate the in vitro model in all the assays.
Crystal data and structure refinement for complex 1.
Complex 1
Chemical formula
Formula weight
Crystal system
Space group
Temperature (K)
a (Å)
C21H22MnN5O5
479.34
Monoclinic
P21/c
100(2)
13.3896(18)
12.8009(17)
13.8265(18)
90
b (Å)
c (Å)
α (°)
β (°)
112.767(6)
90
γ (°)
Volume (Å3)
Z
2185.2(5)
4
Radiation type
Mo Kα
μ (mm−1
)
0.647
Crystal size (mm)
Theta range for data collection
F (000)
0.40 × 0.17 × 0.03
1.65 to 25.41°
992
Data are presented as mean
SEM. Statistical differences were eval-
uated by Student's t-tests with Graph Pad Prism 6 software. Statistical
significance was considered at p < 0.05.
No. of measured reflections
No. independent reflections
Goodness-of-fit on F2
R [F2 > 2σ(F2)]
30,054
4005
1.031
0.0759
2.8. Kinetic experiments
wR(F2)
0.1670
No. of parameters
373
Δ)max, Δ)min (e Å−3
)
0.727, −0.797
The kinetic experiments using a conventional spectrophotometer
were carried out with a Cary 50 Bio UV/Vis spectrophotometer
3

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
3. Results and discussion
catalytic purposes. Previous reported complexes incorporating DCA
show Mn-Oax/Mn-Oeq ratios of about 1.22 [24,30], which are sig-
nificantly higher than values of about 1.15 obtained for other similar
Mn-Schiff base compounds [21]. In the case of 1, this ratio of about
1.20 implies a significant tetragonal elongation but the value is not as
high as expected for the effect derived from the introduction of the DCA
ligand [21,30]. In this sense, a key point to achieve high tetragonal
elongations is the short two‑carbon chain between imine groups in the
Schiff base ligand (like in H2L1-H2L3) as this short chain constricts the
chelate ring once nitrogen atoms coordinate to the metal.
3.1. Synthesis and characterization of complex 1
Complex 1 was obtained in high yield as detailed in the experi-
mental section. Metal complexation was evidenced by the change in the
color of the solution, from yellow to brown. Elemental analysis estab-
lishes a formula MnL1(DCA)(H2O)2, other analytical and spectroscopic
techniques support such formula and give insights into the structure of
the complex. Thus, conductivity measurements in 10−3
M di-
methylformamide solutions are in agreement with non-electrolyte be-
havior [36]; magnetic moment value at room temperature are very
close to the spin-value of 4.9 B.M., as expected for a high-spin mag-
netically diluted d4 manganese(III) ion; ESI-mass spectrum shows the
formation of the manganese complex.
Both Schiff base ligand and dicyanamide ligand show disordered
positions. The aliphatic chain of (L1)2− presents two different config-
urations. The main configuration, labelled as C3A–C5A, has a para-
meter occupancy of 78(2) %, and the complementary configuration
present a value of 22(2) %. For the case of the dicyanamide, also with
two different configurations, their respective values are 59.6(8) % for
N31A–N35A, and 40.4(8) % for N31B–N35B.
Paramagnetic 1H NMR spectrum with two upfield proton resonances
due to the isotropically shifting of the ligand protons for high-spin
manganese(III) complexes in an octahedral field. The data interpreta-
tion of this technique was based on previous findings for manganese(III)
complexes with related Schiff base ligands [20–22]. The signals must
arise from the H4 and H5 protons of the aromatic phenoxy rings. Sig-
nals in ortho positions to donor atoms are not observable for these kinds
of complexes by paramagnetic 1H NMR spectroscopy [37].
The superstructure of 1 involves associations via a combination of π-
aryl offset interactions [41] and hydrogen bonding. Hydrogen bonding
represents a class of weaker, non-covalent interactions that not only
induce supramolecular arrays through self-organization of molecules,
but also plays a crucial role in fundamental biological processes, such as
the expression and transfer of genetic information, and is essential for
molecular recognition between receptors and substrates [42]. Hy-
drogen bonds between capping water molecules and both phenoxy and
methoxy oxygen atoms of the neighboring Schiff base ligand are
formed. As result of all these non-covalent interactions complex 1
present a μ-aqua dimeric structure (Fig. 2), with Mn···Mn distances of
about 4.9 Å are short for monomeric compounds.
IR spectrum for 1 exhibits a strong band at 1624 cm−1 characteristic
of the ν(C]N) stretching mode, which is shifted 9 cm−1 lower respect
to the free Schiff base ligand, indicating the coordination to the man-
ganese through the nitrogen atoms of the imine group. The band at-
tributed to the ν(CeO) mode at 1260 cm−1 is shifted 10 cm−1 to higher
frequencies respect to the free ligand. Moreover, two new bands at
2149 and 2251 cm−1 are assigned, respectively, to the symmetric and
asymmetric dicyanamide modes [38].
3.2. Catalase and superoxide dismutase activities
The electronic spectrum indicates that the Mn(III) complex behaves
as high-spin octahedral d4 system that probably suffers a significant
Jahn-Teller distortion. The band at around 330–350 nm can be assigned
to an intraligand π-π transition [15]. A broad band at 440 nm is attri-
butable to the phenolate → Mn(III) charge-transfer [39].
We have checked the CAT activity of 1–3 in methanolic solutions by
measuring the stoichiometry of the disproportionation of H2O2 cata-
lysed by 1–3 by volumetric determination of the evolved O2. Using this
standard method, with a large excess of oxidant (> 830), catalase ac-
tivity of different model compounds can be compared. The experi-
mental setup was evaluated by the measurements for the catalytic de-
composition of H2O2 (2.5 × 10−3 mol) with MnO2: ca. 30 mL of
dioxygen evolution was measured by a burette, which agrees with the
calculated value based on the assumption that catalytic decomposition
of H2O2 to H2O and O2 takes place. Table 3 collects the results of the
CAT activities. The time course of oxygen production is similar for each
of the complexes. The reaction proceeds quickly during the first
5–10 min and over time the total amount of oxygen evolved approaches
an asymptotic limit. Complexes 1–3 achieve percentages of H2O2 dis-
mutation from 23 to 27%, although this type of CAT assay only gives a
semiquantitative approach about the activity of the complexes. A de-
tailed kinetic study for the reaction of 1–3 with hydrogen peroxide is
discussed in Section 3.4 of this paper. Turnover numbers (TON; calcu-
lated as the number of moles of hydrogen peroxide that a mole of
catalyst can convert before becoming inactivated) of about 200 are
obtained with these experimental conditions, which clearly indicates
the catalytic nature of the process. Moreover, higher TONs can be surely
achieved by changing the experimental conditions.
The electrochemical properties of complexes 1–3 were investigated
by cyclic voltammetry (data is collected in Table 3, and figures of the
voltammograms are available in the Supporting Information section,
Fig. S4). The three complexes exhibit one MnIII/MnII quasi-reversible
wave, with peak to peak separation in the range 110 to 196 mV. The
low E1/2 values in the range −72 to −141 mV are in accordance with
the electron-donor character for the methoxy substituents on the phe-
nolato rings. The potentials are lower by 200–400 mV than the poten-
tials of the corresponding complexes without methoxy substituents or
including electron-withdrawing substituents [21]. Thus, the electron-
donor character of the methoxy groups on the phenyl rings of the Schiff
base facilitates to achieve higher oxidation states for the manganese ion
during catalysis [7,40].
Single crystals of complex 1 suitable for X-ray diffraction studies
were obtained by slow evaporation of the methanolic solution at room
temperature. The crystal structure is shown in Fig. 1 and main bond
distances and angles are shown in Table 2.
The monomeric unit is composed by a distorted octahedron, where
the (L1)2− ligand is tightly bound to the metal ion through the inner
N2O2 compartment by the Nimine and Ophenol atoms (Mn-Nimine bond
lengths of 1.972(6)–1.974(19) Å and Mn-Ophenol of 1.868(5)–1.877(4)
Å, which are typical of such complexes [21,22]), occupying the equa-
torial positions. The axial positions are occupied by a monodentate
dicyanamide ligand and a water molecule. The axial distances of
2.238(4) Å (for Mn-O(1W)) and 2.28(3) Å (for Mn-N(31)) are con-
siderably longer than the equatorial MneO bond lengths quoted earlier,
resulting in an expected tetragonal elongation for the solid state
structure already suggested by the electronic spectrum in solution. The
authors' intent was to achieve such distorted geometry in order to get
labile positions that subsequently facilitate vacancy positions for
The effect of complexes and EUK-134 on SOD activity was assessed
in human neuroblastoma cells. The concentration of O2%− was de-
termined with WST-1 [2-(4-lodophenyl)-3-(4-nitrophenyl)-5-(2,4-dis-
ulfophenyl)-2H-tetrazolium, monosodium salt], which reacts with the
ROS to produce a water soluble formazan dye, that was determined by a
colorimetric method. Therefore, SODs activity is inversely related to
WST signal. Cells were treated with different concentrations of com-
pounds for 24 h, and their effect on SODs enzymatic activity was
evaluated. Then, EC50 values were determinated (Fig. S5 collects SOD
activity curves). EUK-134 presented an EC50 of 28.45 nm (R2 = 0.95),
an activity lower than complexes 1 and 2 (Table 3), but higher than
4

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
Fig. 1. ORTEP view of the environment around the manganese ion in complex 1.
activity. In addition, the capability of the complexes to act both as SOD
and catalase mimics anticipate that they can show neuroprotective ef-
fects, as their catalase activity will lead to disappearance of the H2O2
generated in their behavior as SOD mimics.
Table 2
Selected bond lengths (Å) and angles (°) for complex 1.
Mn(1)-O(27)
1.866(5)
1.973(6)
2.28(3)
Mn(1)-O(17)
Mn(1)-N(6)
Mn(1)-O(1W)
1.878(4)
1.973(5)
2.238(4)
Mn(1)-N(2)
Mn(1)-N(31A)
Mn(1)-N(31B)
2.28(3)
3.3. Protective effect of complexes 1–3 on TBHP-treated SH-SY5Y cells
O(27)-Mn(1)-O(17)
O(27)-Mn(1)-N(2)
O(17)-Mn(1)-N(2)
O(27)-Mn(1)-O(1W)
O(1W)-Mn(1)-M(31A)
92.89(18)
174.6(2)
92.1(2)
O(27)-Mn(1)-N(6)
O(17)-Mn(1)-N(6)
N(2)-Mn(1)-N(6)
91.9(2)
174.9(2)
83.0(2)
At first, the cytotoxicity of complexes was determined in SH-SY5Y
cells, which were treated with different concentrations of complexes
1–3 (Fig. 3) (0.001–10 μM) for 24 h. None of the compounds presented
cytotoxic effects at these concentrations.
92.11(18)
172.1(11)
O(17)-Mn(1)-O(1W)
O(27)-Mn(1)-N(31A)
91.80(17)
95.2(12)
For oxidative stress studies, TBHP was used as oxidant instead of
hydrogen peroxide (a oxidant that has been widely used to produce
oxidative injury in SH-SY5Y cells [43]) to induce cell damage due to its
higher stability in physiological media [44]. Cells were co-treated with
1–3 at four concentrations (0.001–1 μM) and 75 μM TBHP for 6 h, and
the effect of the compounds on cell viability was evaluated by MTT
(Fig. 4). Cell survival decreased about 50% when cells were treated
with 75 μM TBHP alone (p < 0.01). In order to validate the oxidative
stress model, vitE was used as positive control and, as Fig. 4 shows, it
increased cell survival until untreated cells levels (100
ment with complexes 1–3 significantly augmented the cell survival
reduction produced by TBHP, with levels between 80 5% and
95 8%, compared to TBPH control cells.
6%). Treat-
The effects of the biomimetic models 1–3 on mitochondrial func-
tioning were determined with TMRM assay. TMRM is a fluorescent dye
that enters in mitochondria and accumulates in inverse proportion to
ΔΨm, so depolarized organelles will produce
a lower signal.
Fig. 2. Stick diagram for the dimeric unit for complex 1 (manganese ion in
magenta, oxygen in red, nitrogen in blue, and carbon in grey). (For inter-
pretation of the references to color in this figure legend, the reader is referred to
the web version of this article.)
Neuroblastoma cells were co-treated with the compounds (0.001–1 μM)
and 75 μM TBHP for 6 h. TBHP addition produced a decrease in ΔΨm of
30% compared to control cells (p < 0.01) (Fig. 5). Complexes 1 and 2
had no effect on the restoration of ΔΨm, and complex 3 had effect at the
lowest concentration tested, with an increase of 106
3% at
0.001 μM. This percentage is greater than the levels reached by cells co-
treated with 25 μM vitE.
complex 3. Complex 1 showed the best results (9.876 nM), a SOD ac-
tivity value three times better than the commercial EUK-134. All
compounds studied manifested SOD activities about 0.05 U/mg, values
that almost duplicate control cells data, that is, the basal activity of the
enzyme (0.032 U/mg). These results indicate that the Schiff base ligand
would have a higher incidence than the DCA co-ligand in the SOD
Our results indicate that 3 is the most promising compound, being
able to protect cells from oxidative damage and ameliorate the mi-
tochondrial function in a human neuronal model, so it may be a good
candidate for further studies in neurodegenerative disorders such as
AD.
5

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
Table 3
Redox potentials, SOD and catalase activities for 1–3. (a) half wave potential; (b) peak-to-peak separation; (c) catalase activity expressed as percentage of H2O2
decomposed after 60 min (conversion) and turnover number (TON), (d) SOD activity expressed as half maximal effective concentration (EC50) values; R2 is the
coefficient of determination which gives the goodness-of-fit for the regression model; (e) SOD activity expressed as enzyme unit (U) per milligram (1 U is defined as
the amount required for the transformation of 1 μmol of substrate per minute).
Complex
E1/2 (mV)a
ΔEpb
Catalase activityc
SOD activity
Conversion
TON
EC50 (nM)d
R2
Activity at EC50 (U/mg)e
1
2
3
−72
−141
−75
110
111
196
27
23
24
3
2
2
225
192
200
25
17
17
9.876
10.540
0.99
0.93
0.98
0.050
0.059
0.049
155.100
Fig. 3. Cell viability of cells treated with 1–3. Human neuroblastoma cells were
Fig. 5. Effect of 1–3 on ΔΨm levels in human neuroblastoma cells. TBHP
(75 μM) and compounds (0.001–1 μM) were added to the cells for 6 h. The
fluorescence dye TMRM was used to evaluate their effect on ΔΨm.
treated with compounds for 24 h and their cytotoxic effects were determined
with MTT assay. Values are mean
SEM of three independent replicates,
expressed as percentage of untreated control cells.
Mean
SEM of three independent experiments. Data are expressed as per-
centage of control cells and compared to cells treated with TBHP alone.
*p < 0.05, **p < 0.01.
3.4. Kinetic studies
Since the three complexes 1–3 can meaningfully improve the mi-
tochondrial function, kinetic studies of their reaction with H2O2 and
TBHP have been performed in different conditions.
In previous studies we found that the reaction of H2O2 and different
Mn(III)-Schiff base biomimetic models, including EUK-134 (reported as
a SOD, peroxidase and CAT mimic, tested against different oxidative
pathologies and commercialized as a potent antioxidant [10–13,19]), in
acetonitrile solutions are strongly dependent on the illumination con-
ditions [23], suggesting the occurrence of a photochemical process.
Similar studies have been now performed for complex 1, and again the
time scale of the spectral changes observed in the ABTS test for 1 in
acetonitrile is significantly different when the kinetics is monitored
with a stopped-flow instrument and a conventional spectrophotometer.
In addition, no spectral changes are observed when the stopped-flow
experiments are carried out using a 455 nm cut-on long pass filter. The
effect of the illumination conditions is also observed for the reaction of
the complex with H2O2, which occurs in the time scale of thousands of
minutes with the conventional spectrophotometer and hundreds of
seconds with the stopped-flow instrument (Fig. S6). Photochemical
processes prevent obtaining reliable kinetic data for the reaction in
acetonitrile even from the spectral changes recorded with the conven-
tional spectrophotometer, where the amount of incident light is mini-
mized. In those cases, the spectral changes show frequently an induc-
tion period followed of slower and irreproducible spectral changes.
Fig. 4. Neuroprotective effect of 1–3 in an oxidative stress model. SH-SY5Y
cells were co-treated with 75 μM TBHP and compounds (0.001–1 μM) for 6 h.
Then, their effect on cell viability was assessed by MTT test. Data are
mean
SEM of three experiments performed by triplicate. Values are ex-
pressed as percentage of control cells and compared to cells treated with TBHP
alone by Student's t-test. *p < 0.05.
6

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
compound.
ab[Ox]
kobs
=
1 + b[Ox]
(4)
Similar kinetic studies were carried out in water solution, in this
case using both H2O2 and TBHP as oxidant. To make the results more
comparable, the pH and temperature were similar to those used in the
tests of biological activity (37 °C, pH = 7.25). Although the lower so-
lubility of the complexes in water forces the use of lower concentration
of the complexes, significant spectral changes could be measured and
they showed in all cases the disappearance of the bands of the starting
complex with the appearance of new bands at 350 nm and 260 nm, thus
indicating the occurrence of the same type of reaction for all the
complexes. Typical changes and calculated spectra for the reaction of
complex 2 with H2O2 are included in Figs. S9–S10, and for the reaction
of 2 with TBHP in Figs. S11–S12. In all cases, the spectral changes can
be satisfactorily fitted with a single kinetic step, although at long re-
action times there are additional small and irreproducible changes that
lead to partial disappearance of the bands formed in the previous step.
It is important to note that the complexes do not react with the buffer
solution in the time in which the reaction with the oxidants takes place.
As observed in methanol, the values derived for the rate constant
change with the concentration of oxidant with saturation behavior
(Figs. 7 and S13), and the fit of kobs to Eq. (4) leads to the values of a
and b in Table 5. Again there are no large differences in the kinetics of
reaction of the different Mn complexes.
Fig. 6. Plot of the [H2O2] dependence of the first observed rate constant in the
reaction of the Mn complexes with H2O2 in methanol at 25 °C: 1 (circles), 2
(triangles), 3 (squares), EUK-134 (diamonds).
At that point, and given the SOD, peroxidase and CAT activity of
EUK-134 and complexes 1–3 in other solvents, we decided to check if
the behavior observed in acetonitrile changes with the solvent. For this
reason, studies parallel to those described above were carried out in
methanol and in water, in the latter case using buffered solutions si-
milar to those used in the biological studies. These experiments did not
show the striking effect of the illumination conditions observed in
acetonitrile solutions, the results obtained with the stopped-flow and
the conventional spectrophotometer being comparable. However, as
the reaction is slow, it is not completed in the stopped-flow time scale
and the kinetics was measured from the spectral changes observed with
the conventional spectrophotometer.
The operation of free radical processes in acetonitrile solution has
been reported previously [23], but some comment must be made at this
point about the reaction mechanism and the nature of the intermediates
and products formed in methanol and water solution in the reaction
with H2O2 and TBHP. Previous reports suggest that the reaction of this
kind of MnIII complexes with H2O2 leads initially to the formation of
MnV(O) intermediates [10,45] but those oxo species can evolve to other
species depending on the reaction conditions, and the band observed at
350 nm could correspond to a new Mn(III) complex formed from an
initial Mn(V) = O product [46].In any case, although further work is
required to ascertain the nature of the species formed, the present ki-
netic results clearly indicate the reaction of the Mn complexes with
H2O2, and the experiments on their CAT activity indicate the operation
The spectral changes observed for the reaction of the Mn complexes
with H2O2 in methanol show the disappearance of the bands typical of
the starting complex, apparently without the appearance of any new
band. An adequate fit of these spectral changes requires of a kinetic
model with two consecutive steps, with rate constants k1obs and k2obs
(A → B → C, see Figs. S7–S8 for an example of spectral changes and
calculated spectra for A, B and C). The dependence of k1obs with the
oxidant concentration is illustrated in Fig. 6, which indicates a sa-
turation behavior, and a fit of the data to Eq. (4) yields the values of a
and b in Table 4. Although the spectrum calculated for intermediate B
shows an absorption band at c.a. 420 nm (Fig. S8), it cannot be ruled
out that this band corresponds to some amount of remaining starting
complex. The spectral changes in the next step are smaller and the
values derived for k2obs are independent of the concentration of H2O2,
except for the case of complex 2. The values of a and b in Table 4 in-
dicate that there are no large differences in the kinetic parameters for
the different systems studied, including the reference EUK-134
Table 4
Summary of kinetic data for the reaction of the Mn complexes with H2O2 in
methanol solution at 25 °C. For the meaning of a and b see Eq. (4).
Complex
k1obs
k2obs/s−1
a/s−1
b/M−1
1
(3.6
(3.8
(5
0.6) × 10−3
0.2) × 10−3
2) × 10−3
(1.0
(0.9
(0.3
(0.7
0.5) × 102
0.1) × 102
0.2) × 103
0.2) × 103
(5.1
0.9) × 10−5
a
2
3
(2.5
(1.2
0.3) × 10−5
Fig. 7. Plot of the [H2O2] dependence of the observed rate constant in the re-
action of the Mn complexes with H2O2 in water solution at 37 °C and
pH = 7.25: 1 (empty circles), 2 (triangles), 3 (filled circles) and EUK-134
(squares).
EUK-134
(1.3
0.2) × 10−3
0.3) × 10−5
a
Curvature is observed in the plots of k2obs vs. [H2O2] and fitting to Eq. (4)
leads to a = (4.2
0.5) × 10−4 s−1 and b = (0.8
0.2) × 102 M−1
.
7

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
Table 5
Summary of kinetic data for the reaction of the Mn complexes with H2O2 and TBHP in water solutions at 37 °C and pH = 7.25. For the meaning of a and b see Eq. (4).
Complex
H2O2
TBHP
a/s−1
b/M−1
a/s−1
b/M−1
1
(1.1
(2.1
(2.6
(1.7
0.1) × 10−3
0.2) × 10−3
0.1) × 10−3
0.2) × 10−3
(6
1) × 103
(2.0
(3.0
(2.0
(3.0
0.1) × 10−4
0.1) × 10−4
0.1) × 10−4
0.1) × 10−4
(9
2) × 103
2
(4
1) × 103
0.2) × 103
1) × 103
(5.2
(1.3
(6
0.5) × 103
0.3) × 103
1) × 103
3
(1.0
(5
EUK-134
explanation but one possibility is that acetonitrile coordination stabi-
lizes the MnIII state hindering oxygen transfer from H2O2 but allowing
free radical oxidation to MnIV. Another possibility is that the process
defined by k1a require in some way participation of a protic solvent to
favour hydrogen bonding or proton transfer. It is evident that further
work is required to shed light on the intimate mechanistic features of
these reactions.
4. Conclusions
Scheme 2. The proposed mechanism of the catalase-like catalytic reaction of 1-
3.
A carefully selection and design of the ligands is a key issue to get
active antioxidant biomimetic models. For the case of the salen-type
ligands, the 2Nimine2Ophenoxy core forms an inner cavity that can ac-
commodate the manganese ion. Extra methoxy atoms, which point to-
ward the outside of the inner cavity, facilitates an extension of the di-
mensionality of the structure. The tetragonal elongation of the
octahedral structure may be forced by the selection of the appropriate
ligands.
of a catalytic cycle for H2O2 disproportionation, which probably occurs
through the previously proposed mechanism [10]. In that mechanism,
H2O2 attacks to the Mn(III) complex with formation of a Mn(V) inter-
mediate that reacts with a second molecule of H2O2 to regenerate the
starting complex with O2 release. Moreover, the observation that most
of the gases evolve in the CAT studies in the first 5–10 min agrees well
with the half times calculated from the limiting rate constant (a in Eq.
(4)) in the kinetic studies in methanol and water (3–10 min for the
different complexes), thus suggesting that the values of a can be asso-
ciated to the rate-determining step in the CAT cycle. The absence of a
detectable MnV(O) intermediate also suggests that the rate determining
step is the initial reaction with H2O2; although the resulting MnV(O)
intermediate would not accumulate because it is very reactive toward
H2O2 and so it would be formed under steady-state conditions. How-
ever, as the catalytic cycle regenerates the Mn(III) complex, an addi-
tional process must be included in the mechanism to account for the
disappearance of the starting complex. The resulting mechanism is in-
cluded in Scheme 2, and application of the steady-state approximation
to the Mn(V) intermediate leads to the rate law in Eq. (5), which is of
the same form that Eq. (4) with the equivalencies in Eqs. (6)–(7). At this
point, it is important to note that the k2b step is the responsible of the
inactivation of the complexes as catalysts for H2O2 disproportionation.
Nevertheless, the TON values in Table 3 clearly indicate that all com-
pounds behave as quite good catalysts, in agreement with the large
values of b (=k2a/k2b) in Table 5.
In this work, we demonstrated for the first time the neuroprotective
effects of this type of biomimetic models against oxidative stress in a
human neuronal model. Complexes 1–3 significantly improved the cell
viability loss induced by TBHP. The reaction mechanism for 1–3 versus
a common oxidant as hydrogen peroxide is strongly dependent on the
conditions. Solvent media, illumination conditions and other factors
may vary not only the antioxidant activity but also the reaction me-
chanism itself.
Kinetic data in buffered solutions are in line with the results of the
oxidative stress studies on SH-SY5Y cells and also with the results of the
CAT activity test. All the complexes studied, including the commercial
EUK-134, show a rate determining step in the catalytic cycle that cor-
responds to the initial reaction of the complexes with H2O2. Rate con-
stants obtained for complexes 1–3 and EUK-134 being comparable,
which could allow the use of the new complexes in antioxidant appli-
cations similar to the commercial compound.
Acknowledgements
The research leading to these results has received funding from the
following FEDER cofunded-grants. From Consellería de Cultura,
Educación e Ordenación Universitaria, Xunta de Galicia, 2017 GRC GI-
1682 (ED431C 2017/01), 2018 GRC GI-1584 (ED431C 2018/13),
MetalBIO Network (ED431D 2017/01). From CDTI and Technological
k1ak2b [H2O2]
kobs
a =
b =
=
k
+ k2a [H2O2 ]
(5)
(6)
(7)
2b
k1ak2b
k2a
Funds, supported by Ministerio de Economía, Industria
y
Competitividad, CTQ2015-65707-C2-2-P, AGL2016-78728-R (AEI/
FEDER, UE), ISCIII/PI16/01830 and RTC-2016-5507-2, ITC-20161072.
From European Union POCTEP 0161-Nanoeaters-1-E-1, Interreg
AlertoxNet EAPA-317-2016, Interreg Agritox EAPA-998-2018, and
H2020 778069-EMERTOX.
k
2a
k
2b
Finally, one might wonder why the reaction with H2O2 in acetoni-
trile behaves so different from that in methanol and water. In acet-
onitrile the reaction is dominated by free radical processes initiated by
photochemical processes, and the reaction under conditions similar to
those used in methanol and water show an induction period that in-
dicates that the k1a step in Scheme 1 (k1a step) is inhibited and the
reaction only starts when free radicals are generated. With the in-
formation currently available it is not possible to give a definitive
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.jinorgbio.2019.110918.
8

L. Rouco, et al.
JournalofInorganicBiochemistry203(2020)110918
References
R. Pedrido, M.J. Rodríguez-Doutón, M. Maneiro, Inorg. Chem. 54 (2015)
2512–2521.
[23] A. Liberato, M.J. Fernández-Trujillo, Á. Máñez, M. Maneiro, L. Rodríguez-Silva,
M.G. Basallote, Inorg. Chem. 57 (2018) 14471–14475.
[1] F.A. Villamena, Chemistry of reactive species, in: F.A. Villamena (Ed.), Molecular
Basis of Oxidative Stress, Ed. John Wiley & Sons, New Jersey, 2013, pp. 1–48.
[2] M. Oszajca, M. Brindell, L. Orzel, J.M. Dabrowski, K. Spiewak, P. Labuz, M. Pacia,
A. Stochel-Gaudyn, W. Macyk, R. van Eldik, G. Stochel, Coord. Chem. Rev. 143
(2016) 327–328.
[24] M.A. Vázquez-Fernández, M.I. Fernández-García, A.M. González-Noya, M. Maneiro,
M.R. Bermejo, M.J. Rodríguez-Doutón, Polyhedron 31 (2012) 379–385.
[25] C. Palopoli, J. Ferreyra, A. Conte-Daban, M. Richezzi, A. Foi, F. Doctorovich,
E. Anxolabéhère-Mallart, C. Hureau, S.R. Signorella, ACS Omega 4 (2019) 48–57.
[26] S.L. Huang, H.B. He, K. Zou, C.H. Bai, Y.H. Xue, J.Z. Wang, J.F. Chen, J. Pharm.
Pharmacol. 66 (2014) 844–854.
[3] M.T. Lin, M.F. Beal, Nature 443 (2006) 787–795.
[4] P. Sharma, P. Srivastava, A. Seth, P.N. Tripathi, A.G. Banerjee, S.K. Shrivastava,
Prog. Neurobiol. 174 (2019) 53–89.
[27] R. Alvariño, E. Alonso, M.A. Tribalat, S. Gegunde, O.P. Thomas, L.M. Botana,
Neurotox. Res. 32 (2017) 368–380.
[5] B. DeStrooper, E. Karran, Cell 164 (2016) 603–615.
[6] C. Patterson, World Alzheimer Report 2018, New Frontiers, London, 2018, pp.
32–36.
[28] U. Casellato, P. Guerriero, S. Tamburini, P.A. Vigato, C. Benelli, Inorg. Chim. Acta
207 (1993) 39–58.
[7] S. Signorella, C. Palopoli, G. Ledesma, Coord. Chem. Rev. 365 (2018) 75–102.
[8] B.A. Baglia, J.P.T. Zaragoza, D.P. Goldberg, Chem. Rev. 117 (2017) 13320–13352.
[9] H.Y.V. Ching, I. Kenkel, N. Delsuc, E. Mathieu, I. Ivanovic-Burrnazovic, C. Policar,
J. Inorg. Biochem. 160 (2016) 172–179.
[29] G. González-Riopedre, M.I. Fernández-García, E. Gómez-Fórneas, M. Maneiro,
Catalysts 3 (2013) 232–246.
[30] M.R. Bermejo, R. Carballido, M.I. Fernández-García, A.M. González-Noya,
G. González-Riopedre, M. Maneiro, L. Rodríguez-Silva, J. Chem. 2017 (2017)
54658901-10.
[10] S.R. Doctrow, K. Huffman, C.B. Marcus, G. Tocco, E. Malfroy, C.A. Adinolfi,
H. Kruk, K. Baker, N. Lazarowych, J. Mascarenhas, B. Malfroy, J. Med. Chem. 45
(2002) 4549–4558.
[31] J.M. McCord, I. Fridovich, J. Biol. Chem. 244 (1969) 6049–6055.
[32] E. Vega-Avila, M.K. Pugsley, Proc. West. Pharmacol. Soc. 54 (2011) 10–14.
[33] R.A. Binstead, B. Jung, A.D. Zuberbühler, SPECFIT-32, Spectrum Software
Associates, Chappel Hill, 2000.
[11] M. Baudry, S. Etienne, A. Bruce, M. Palucki, E. Jacobsen, B. Malfroy, Biochem.
Biophys. Res. Commun. 192 (1993) 964–968.
[12] J.C. Kash, Y. Xiao, A. Sally Davis, K.A. Walters, D.S. Chertow, J.D. Easterbrook,
R.L. Dunfee, A. Sandouk, B.W. Jagger, L.M. Shwartzman, R.E. Kuestner, N.B. Wehr,
K. Huffman, R.A. Rosenthal, A. Ozinsky, R.L. Levine, S.R. Doctrow,
J.K. Taubenberger, Free Radic. Biol. Med. 67 (2014) 235–247.
[13] S.R. Doctrow, K. Huffman, C.B. Marcus, W. Musleh, A. Bruce, M. Baudry,
B. Malfroy, Adv. Pharmacol. 38 (1997) 247–269.
[34] G.M. Sheldrick, SHELX-97 (shelxs 97 and shelxl 97), Programs for Crystal Structure
Analyses, University of Göttingen, Germany, 1998.
[35] G.M. Sheldrick, SADABS, Program for Scaling and Correction of Area Detector Data,
University of Göttingen, Germany, 1996.
[36] W.J. Geary, Coord. Chem. Rev. 7 (1971) 81–122.
[37] M. Zampakou, V. Tangoulis, C.P. Raptopoulou, V. Psycharis, A.N. Papadopoulos,
G. Psomas, Eur. J. Inorg. Chem. (13) (2015) 2285–2294.
[14] C. Palopoli, G. Gómez, A. Foi, F. Doctorovich, S. Mallet-Ladeira, C. Hureau,
S. Signorella, J. Inorg. Biochem. 167 (2017) 49–59.
[38] A. Majumder, G. Pilet, M.T. Garlan, S. Mitra, Polyhedron 25 (2006) 2550–2558.
[39] M. Collomb, C. Mantel, S. Romain, C. Duboc, J. Lepetre, J. Pecaut, Eur. J. Inorg.
Chem. (2007) 3179–3187.
[15] G.N. Ledesma, E. Anxolabéhère-Mallart, L. Sabater, C. Hureau, S.R. Signorella, J.
Inorg. Biochem. 186 (2018) 10–16.
[16] M. Keener, M. Peterson, R.H. Sanchez, V.F. Oswald, G. Wu, G. Menard, Chem. Eur.
J. 23 (2017) 11479–11484.
[40] M. Navarro, C.A. Smith, M. Li, S. Bernhard, M. Albrecht, Chem. Eur. J. 24 (2018)
6386–6398.
[17] D. Bani, A. Bencini, Curr. Med. Chem. 19 (2012) 4431–4444.
[18] S. Bahramikia, R. Yazdanparast, Med. Chem. Res. 21 (2012) 3224–3232.
[19] S.R. Doctrow, M. Liesa, S. Melov, O.S. Shirihai, P. Tofilon, Current Inorg. Chem. 2
(2012) 325–334.
[41] E.C. Constable, C.E. Housecroft, P. Kopecky, E. Schonhofer, J.A. Zampese,
CrystEngComm 13 (2011) 2742–2752.
[42] D.H. Williams, E. Stephens, D.P. O'Brien, M. Zhou, Angew. Chem. Int. Ed. 43 (2004)
6596–6616.
[20] M.R. Bermejo, M.I. Fernández, E. Gómez-Fórneas, A. González-Noya, M. Maneiro,
R. Pedrido, M.J. Rodríguez, J.C. García-Monteagudo, B. Donnadieu, J. Inorg.
Biochem. 100 (2006) 1470–1478.
[43] H. Zhang, M. Gao, L. Zhang, Y. Zhao, L.-L. Shi, B. Chen, Y. Wang, S. Wang, G. Du,
Biochem. Biophys. Res. Commun. 421 (2012) 479–483.
[44] W. Zhao, H. Feng, W. Sun, K. Liu, J.-J. Liu, X. Chem, Redox Biol. 11 (2017)
524–534.
[21] M.A. Vázquez-Fernández, M.R. Bermejo, M.I. Fernández-García, G. González-
Riopedre, M.J. Rodríguez-Doutón, M. Maneiro, J. Inorg. Biochem. 105 (2011)
1538–1547.
[45] V. Lanza, G. Vecchio, J. Inorg. Biochem. 103 (2009) 381–388.
[46] K. Srinivasan, P. Michaud, J.K. Kochi, J. Am. Chem. Soc. 108 (1986) 2309–2320.
[22] G. González-Riopedre, M.R. Bermejo, M.I. Fernández-García, A.M. González-Noya,
9