Manganese(II) complexes with thiosemicarbazones as potential anti-Mycobacterium tuberculosis agents

Article abstract of DOI:10.1016/j.jinorgbio.2013.10.011

Through a systematic variation on the structure of a series of manganese complexes derived from 2-acetylpyridine-N(4)-R-thiosemicarbazones (Hatc-R), structural features have been investigated with the aim of obtaining complexes with potent anti-Mycobacter

Full text of DOI:10.1016/j.jinorgbio.2013.10.011

Journal of Inorganic Biochemistry 132 (2014) 21–29
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Manganese(II) complexes with thiosemicarbazones as potential
anti-Mycobacterium tuberculosis agents
Carolina G. Oliveira a, Pedro Ivo da S. Maia b, Paula C. Souza c, Fernando R. Pavan c, Clarice Q.F. Leite c,
Rommel B. Viana a, Alzir A. Batista d, Otaciro R. Nascimento e, Victor M. Deflon a,
⁎
a
Instituto de Química de São Carlos, Universidade de São Paulo, 13566-590 São Carlos, SP, Brazil
Departamento de Química, Universidade Federal do Triângulo Mineiro, 38025-440 Uberaba, MG, Brazil
Faculdade de Ciências Farmacêuticas, Universidade Estadual Paulista, 14801-902 Araraquara, SP, Brazil
Departamento de Química, Universidade Federal de São Carlos, 13565-905 São Carlos, SP, Brazil
b
c
d
e
Instituto de Física de São Carlos, Universidade de São Paulo, CP 369, 13560-970 São Carlos, SP, Brazil
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 30 May 2013
Received in revised form 18 October 2013
Accepted 19 October 2013
Available online 25 October 2013
Through a systematic variation on the structure of a series of manganese complexes derived from
2-acetylpyridine-N(4)-R-thiosemicarbazones (Hatc-R), structural features have been investigated
with the aim of obtaining complexes with potent anti-Mycobacterium tuberculosis activity. The analytical
methods used for characterization included FTIR, EPR, UV–visible, elemental analysis, cyclic voltammetry,
magnetic susceptibility measurement and single crystal X-ray diffractometry. Density functional theory
(DFT) calculations were performed in order to evaluate the contribution of the thiosemicarbazonate
ligands on the charge distribution of the complexes by changing the peripheral groups as well as to verify
the Mn–donor atoms bond dissociation predisposition. The results obtained are consistent with the mono-
anionic N,N,S-tridentate coordination of the thiosemicarbazone ligands, resulting in octahedral complexes
of the type [Mn(atc-R)2], paramagnetic in the extension of 5 unpaired electrons, whose EPR spectra are
consistent for manganese(II). The electrochemical analyses show two nearly reversible processes, which
are influenced by the peripheral substituent groups at the N4 position of the atc-R1− ligands. The minimal
inhibitory concentration (MIC) of these compounds against M. tuberculosis as well as their in vitro
cytotoxicity on VERO and J774A.1 cells (IC50) was determined in order to find their selectivity index
(SI) (SI = IC50 / MIC). The results evidenced that the compounds described here can be considered as
promising anti-M. tuberculosis agents, with SI values comparable or better than some commercial drugs
available for the tuberculosis treatment.
Keywords:
Manganese(II)
Thiosemicarbazones
Anti-Mycobacterium tuberculosis activity
Catalase peroxidase
© 2013 Elsevier Inc. All rights reserved.
1. Introduction
defending against oxidative stress generated by aerobic oxygen
metabolism or phagocytic attack from macrophages or neutrophils
The World Health Organization (WHO) estimates that tuberculosis
(TB) has become one of the deadliest global emergencies due to the
widespread existence of multiple drug resistant strains of Mycobacterium
tuberculosis (MTB), frequently caused by limited health care system
resources combined with the increasing numbers of HIV/AIDS [1].
In addition, a number of anti-TB drugs may be discontinued in a
few years due to the expanding number of multidrug resistant TB
(MDR-TB) and extensively drug-resistant (XDR-TB) cases [2]. Therefore,
there is a serious need for new anti-MTB agents that target new
biochemical pathways and treat drug resistant forms of the disease [3,4].
Catalase–peroxidase (CP or KatG) is a heme protein that exhibits
both catalase and peroxidase activities and plays a crucial role in
[5,6]. In MTB the KatG is responsible for the conversion of Mn(II) to
Mn(III) in metalloenzymes containing manganese in active sites and
has attracted special attention due to its participation in mycobacterial
virulence [7]. On the other hand, KatG is also important for the
activation of the pro-drug isoniazid (INH) [8]. Most of the INH resistance
is associated with KatG structural gene alterations resulting in KatG
mutant enzymes with reduced ability to form activated INH-
intermediates [8,9]. It has been shown that both KatG and Mn
complexes are able to oxidize isoniazid and form the active
isonicotinoyl–NAD adduct [7]. Therefore, manganese complexes
have been proposed as an alternative oxidant mimicking the
activity of KatG and thus providing a non-enzymatic isoniazid
activation method [6,7].
In this context, it was demonstrated that the redox reversible metal
complex [FeII(CN)5(INH)]3− [10] containing the pro-drug was able
to overcome the drug resistance. This complex presents a minimal
⁎
Corresponding author.
E-mail address: deflon@iqsc.usp.br (V.M. Deflon).
0162-0134/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jinorgbio.2013.10.011

22
C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
inhibitory concentration (MIC, defined as the lowest concentration
resulting in 90% of growth inhibition of the bacteria) value similar
to that of free isoniazid as well as a good selectivity index. Its
mechanism of action is supposed to be independent of the KatG
enzyme since it is active against INH-resistant MTB [10]. Similar
to iron complexes, manganese species also take part in many
biological redox processes and can easily achieve different oxi-
dation states.
were carried out at room temperature in dichloromethane or dimethyl
sulfoxide containing 0.1 M tetrabutylammonium perchlorate (PTBA)
(Fluka Purum) as supporting electrolyte, using an electro-chemical
analyzer μAutolab III or a Bioanalytical Systems Inc. (BAS), model
100BW. The working and auxiliary electrodes were stationary Pt
and the reference electrode was Ag/AgCl, a medium in which
ferrocene is oxidized at 0.5 V (Fc/Fc+), carried out with a rate
sweep of 100 mV s−1
.
Thiosemicarbazones (TSCs) are of great interest both in chemistry and
biology, especially due to their antiparasital [11], antitumor [12–14] and
antibacterial activities [15]. Moreover, the biological properties of the
TSCs are often referred to their complex formation since it can increase
the biological activity by forming chelates with metal ions [12,16,17].
Previous studies showed that V(IV,V) complexes [18] and octahedral
Ni(II) complexes [19] derived from 2-acetylpyridine thiosemicarbazones
possess anti-MTB activity that can be increased by structural mod-
ifications on the thiosemicarbazone moiety.
Literature data dealing with manganese complexes with thiosemi-
carbazones [20,21] have evaluated the biological activity of these
compounds against cancer cells and some bacteria other than MTB.
Believing in the high potential of such compounds as anti-MTB
agents and also contributing to their full characterization, here
we describe the preparation of manganese complexes derived
from 2-acetylpyridine-thiosemicarbazones, their characterization
by diverse methods as well as the study of their anti-MTB activity
and cytotoxicity.
2.3. Preparations
The Mn(II) complexes were synthesized by adding 0.25 mmol
MnCl2·2H2O to solutions of 0.5 mmol of the desired ligands in
15 mL of MeOH containing 3 drops of Et3N. The resulting solutions
were stirred for 4 h under reflux. The brown solids precipitated
during this time were filtered off, washed with water and dried
under vacuum. By slow evaporation of the mother solution of
3, orange crystals suitable for X-ray diffraction analysis were
obtained.
[Mn(atc)2]·H2O (1·H2O): Yield 0.057 g (48%). Analysis: Found: C,
41.51; H, 4.31; N, 24.55%. Calc. for C16H20N8OS2Mn: C, 41.82; H,
4.39; N, 24.40%. IR (νmax/cm−1): 3373, 3124 ν(N\H), 1624, 1593,
1548 ν(C_N) + ν(C_C), 989 ν(N\N), 773 ν(C\S). UV–Vis in
3.39 × 10−5 M CH2Cl2 solution [λmax ( , M−1 cm−1)]: 392.00 nm
(18466), 296.00 nm (20324). Molar conductivity (1 × 10−3
dichloromethane): 0.24 μS cm−1. μeff: 6.03BM.
M
2. Experimental
[Mn(atc-Me)2] (2): Yield 0.062g (53%). Analysis: Found: C, 45.42; H,
4.72; N, 23.84%. Calc. for C18H22N8S2Mn: C, 46.05; H, 4.73; N, 23.88%.
IR (νmax/cm−1): 3329 ν(N\H), 1591, 1548, 1506 ν(C_N) +
2.1. Materials
ν(C_C), 972 ν(N\N), 781 ν(C\S). UV–Vis in 1.06 × 10−5
M
2-Acetylpyridine, thiosemicarbazide, 4-methyl-3-thiosemicarbazide,
4-ethyl-3-thiosemicarbazide, 4-phenyl-3-thiosemicarbazide and analyt-
ical reagent grade chemicals and solvents were obtained commercially
and used without further purification. 4-Cyclohexyl-3-thiosemicarbazide
and 3-morpholynylthiosemicarbazide were prepared as previously de-
scribed [3,22]. The ligands Hatc, Hatc-Me, Hatc-Et, Hatc-Ch, Hatc-
Ph and Hatc-Mf were prepared by refluxing equimolar ethanolic
solutions containing the desired thiosemicarbazide (10 mmol)
and 2-acetylpyridine (10 mmol) for 1 h, as reported elsewhere
[18,23].
CH2Cl2 solution [λmax ( , M−1 cm−1)]: 397.5 nm (78301), 309.0 nm
(63962). Molar conductivity (1 × 10−3
M dichloromethane):
0.24 μS cm−1. μeff: 5.90 BM.
[Mn(atc-Et)2] (3): Yield 0.080 g (64%). Analysis: Found: C, 48.15; H,
5.31; N, 22.51%. Calc. for C20H26N8S2Mn: C, 48.28; H, 5.27; N, 22.54%.
IR (νmax/cm−1): 3203 ν(N\H), 1593, 1550, 1517 ν(C_N) +
ν(C_C), 972 ν(N\N), 780 ν(C\S). UV–Vis in 2.21 × 10−5
M
CH2Cl2 solution [λmax ( , M−1 cm−1)]: 399.00 nm (20814),
309.50 nm (27511). Molar conductivity (1 × 10−3 M dichloro-
methane): 0.08 μS cm−1. μeff: 6.09BM.
2.2. Instruments
[Mn(atc-Ch)2] (4): Yield: 0.092g (61%). Analysis: Found: C, 55.67; H,
6.40; N, 18.76%. Calc. for C28H38N8S2Mn: C, 55.52; H, 6.33; N, 18.51%.
IR (νmax/cm−1): 3290 ν(N\H), 1591, 1543, 1516 ν(C_N) +
FTIR spectra were measured as KBr pellets on a Shimadzu IR
Prestige-21 spectrophotometer between 400 and 4000cm−1. Elemental
analyses were determined using a Perkin-Elmer CHN 2400 equipment.
EPR spectra were collected on an X-band BrukerER-580 spectrometer
equipped with an Oxford low temperature system at 6 K. The spectra
were simulated with the Symphonia program from BRUKER. The
measurement conditions include microwave frequency of 9.476 GHz,
modulation frequency and amplitude of 100 kHz and 0.4 mT,
respectively, magnetic field scan range of 5–605 mT, gain of 45 dB,
microwave power of 2.5mW, time constant of 20.48ms and conversion
time of 81.92 ms. The quantification of the two species in the EPR
spectra of some of the complexes was estimated by double integration
of each spectrum. Magnetic susceptibilities were measured on a
JOHNSON MATTHEY MSB balance at 298 K and converted into the
corresponding molar susceptibilities in the usual way [24,25].
Diamagnetic corrections, applied to the molar susceptibilities of
the paramagnetic substances, are reported as χdiam. The latter
corrections were calculated using the standard Pascal's constants
[24,25]. The conductivities of the complexes were measured in
CH2Cl2 solutions using an Orion Star Series conductometer. UV–
visible (UV–vis) spectra were measured with a Shimadzu UV-1800
spectrophotometer in CH2Cl2 solutions. The electrochemical experiments
ν(C_C), 989 ν(N\N), 773 ν(C\S). UV–Vis in 3.30 × 10−5
M
CH2Cl2 solution [λmax ( , M−1 cm−1)]: 400.50 nm (25894);
311.50 nm (21121). Molar conductivity (1 × 10−3 M dichloro-
methane): 0.22 μS cm−1. μeff: 6.10BM.
[Mn(atc-Ph)2] (5): Yield 0.118g (80%). Analysis: Found: C, 56.37; H,
4.80; N, 18.88%. Calc. for C28H26N8S2Mn: C, 56.65; H, 4.42; N, 18.89%.
IR (νmax/cm−1): 3313 ν(N\H), 1593, 1537, 1494 ν(C_N) +
ν(C_C), 987 ν(N\N), 777 ν(C\S). UV–Vis in 1.76 × 10−5
M
CH2Cl2 solution [λmax ( , M−1 cm−1)]: 415.50 nm (13851),
369.00 nm (26136), 315.50 nm (34545). Molar conductivity
(1× 10−3 M dichloromethane): 0.03 μS cm−1. μeff: 6.30BM.
[Mn(atc-Mf)2] (6): Yield: 0.080g (55%). Analysis: Found: C, 49.21; H,
4.95; N, 18.75%. Calc. for C24H30N8O2S2Mn: C, 49.56; H, 5.20; N,
19.28%. IR (νmax/cm−1): 1593, 1543, 1450 ν(C_N) + ν(C_C), 985
ν(N\N), 788 ν(C\S). UV–Vis in 2.06 × 10−5 M CH2Cl2 solution
[λmax ( , M−1 cm−1)]: 407.50 nm (1 817), 313.00 nm (12718).
Molar conductivity (1 × 10−3 M dichloromethane): 0.14 μS cm−1
eff: 6.55BM.
.
μ

C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
23
2.4. Crystal structure determination
polarizable continuum model IEFPCM [39]. The solvation energy
was evaluated by the difference between Gibbs free energy in solvent
Orange crystals were grown by slow evaporation of a MeOH solution
of [Mn(atc-Et)2] at room temperature. The data collection was
performed using Mo-Kα radiation (λ = 71.073 pm) on a BRUKER
APEX II Duo diffractometer. Standard procedures were applied for
data reduction and absorption correction. The structures were
solved with SHELXS97 using direct methods [26] and all non-
hydrogen atoms were refined with anisotropic displacement
parameters with SHELXL97 [27]. The hydrogen atoms were cal-
culated at idealized positions using the riding model option of
SHELXL97 [27]. Table 1 presents more detailed information about
the structure determination.
and the free energy in vacuum at 298 K, i.e., ΔGsolvation = ΔG°solvent −
ΔG°vacuum, where ΔG°solvent and ΔG°vacuum is the Gibbs free energy
obtained after the optimization in solvent and vacuum, respectively. In
addition, the isotropic polarizability (α) refers to the three diagonal
elements of the polarizability tensor [α = (αxx + αyy + αzz) / 3].
Transition energies and oscillator strengths were carried out
with time-dependent-DFT (TD-DFT) method [40–44], using M06-2X
functional [34]. TD-DFT approach was performed with optimized
geometries in the solvent. TD-DFT was accomplished considering the
first sixty excited states, although only selected electronic transitions
were reported. Natural bond orbital (NBO) analysis was performed
using the NBO6.0 program [45]. The natural population analysis (NPA)
was employed in the charge distribution [46].
2.5. Computational methodology
2.6. Determination of MICs
The computational simulations were performed using the Gaussian
09 program [28]. Geometry optimization procedure was carried out
with density functional theory (DFT), using the PBE1PBE functional
(also known as PBE0) [29,30] which is a hybrid generalized gradient
approximation (GGA) method with a Hartree–Fock (HF) contribution
of 25%. PBE1PBE functional was employed with triple-ζ basis set
LANL2TZ+, which presents an addition diffuse d function, with Hay
and Wadt effective core potential (ECP) for the Mn atom [31,32] and
6-31G(d,p) for all other atoms. Nevertheless, to assess the variations
between the different DFT methodologies, some tests were performed
on complex 1. This set of tests also used a GGA functional method,
BP86 [33], a hybrid meta-GGA, M06 [34], a double hybrid functional,
B2-PLYP [35] and also the HF method. In addition, different basis sets
were applied to the Mn atom, such as double-ζ LANL2DZ with ECP
[31,32], the correlation consistent valence triple-ζ basis set with
a Stuttgart–Dresden–Bonn fully relativistic effective core potential
ECP10MDF [which will be referred in the text as VTZ-(ecp-10mdf)]
[36,37] and also Weigend and Ahlrichs Def2-TVZP basis set [38] to
whole system (see Supplementary material). Dichloromethane sol-
vation effect was calculated with the continuous surface charge
The anti-MTB activity of the compounds was determined by the
REMA (Resazurin Microtiter Assay) method according to Palomino
et al. [47]. Stock solutions of the tested compounds were prepared in
DMSO and diluted in Middlebrook 7H9 broth (Difco) supplemented
with oleic acid, albumin, dextrose and catalase (OADC enrichment —
BBL/Becton-Dickinson), to obtain final drug concentration ranges of
0.09–25 μg/mL. The isoniazid was dissolved in distilled water and used
as a standard drug. A suspension of the MTB H37Rv ATCC 27294 was
cultured in Middlebrook 7H9 broth supplemented with OADC and
0.05% Tween 80. The culture was frozen at −80 °C in aliquots. After
two days the CFU/mL of an aliquot was carried out. The concentration
was adjusted by 5 × 105 UFC/mL and 100 μL of the inoculum was
added to each well of a 96-well microtiter plate together with 100 μL
of the compounds. Samples were set up in triplicate. The plate was
incubated for 7 days at 37 °C. After 24 h 30 μL of 0.01% resazurin
(solubilized in water) was added. The fluorescence of the wells was
read after 24 h by TECAN Spectrafluor®.
2.7. Cytotoxicity assay
In vitro cytotoxicity assays (IC50) were performed first on VERO
epithelial cells (ATCC CCL81). This lineage is widely used for phenotypic
screening of drugs to be regarded as a normal cell derived from normal
human epithelial tissue. Following this approach, compounds with low
cytotoxicity were investigated on the J774A.1 (ATCC TIB-67) murine
macrophage cell line. The macrophages are the first cells of the immune
response and also serve as reservoirs of MTB that can survive within
these cells. Both studies are recommended by Pavan et al. [3]. The cells
were routinely maintained in complete medium (DMEM for VERO and
RPMI-1640 (VitroCell®) J774A.1) supplemented with 10% heat-
inactivated fetal bovine serum (FBS) plus gentamicin (50 mg/L) and
amphotericin B (2 mg/L), at 37 °C, in a humidified 5% CO2 atmosphere.
After reaching confluence, the cells were detached and counted. For
the cytotoxicity assay, 1 × 105 cells/mL were seeded in 200 μL of
complete medium in 96-well plates (NUNCtm). The plates were
incubated under the same conditions for 24 h, to allow cell adhesion
prior to drug testing. The compounds were dissolved in DMSO (5%)
and subjected to two-fold serial dilution from 500 to 3.9 μg/mL. The
cells were exposed to the compounds at various concentrations for
24 h. Resazurin solution was then added to the cell cultures and
incubated for 6 h. Cell respiration, as an indicator of cell viability, was
detected by reduction of resazurin to resorufin, whose pink color and
fluorescence indicate cell viability. A persistent blue color of resazurin
is a sign of cell death. The fluorescence measurements (530 nm
excitation filter and 590 nm emission filter) were performed in a
SPECTRAfluor Plus (Tecan) microfluorimeter. The IC50 value was
defined as the highest drug concentration at which 50% of the cells are
viable relative to the control. Each test was set up in triplicate.
Table 1
Crystal data and structure refinement for [Mn(atc-Et)2] (3).
Empirical formula
Formula weight
Temperature
Wavelength
Crystal system
Space group
C20H26MnN8S2
497.55
293(2) K
0.71073 Å
Monoclinic
P21
a = 8.83170(10) Å α = 90°
b = 14.7684(3) Å β = 102.3330(10)°
c = 9.3837(2) Å γ = 90°
1195.67(4) Å3
Unit cell dimensions
Volume
Z
Density (calculated)
Absorption coefficient
F(000)
Crystal size
Theta range for data collection
Index ranges
2
1.382 Mg/m3
0.751 mm−1
518
0.55 × 0.23 × 0.18 mm3
2.36 to 25.14°
−10 ≤ h ≤ 10, −17 ≤ k ≤ 17,
−11 ≤ l ≤ 10
7973
4142 [R(int) = 0.0158]
99.5%
Semi-empirical from equivalents
0.8767 and 0.6830
Full-matrix least-squares on F2
4142/1/284
Reflections collected
Independent reflections
Completeness to theta = 25.14°
Absorption correction
Max. and min. transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
1.026
Final R indices [I N 2sigma(I)]
R indices (all data)
Absolute structure parameter
Largest diff. peak and hole
R
R
1 = 0.0219, wR2 = 0.0562
1 = 0.0228, wR2 = 0.0567
0.037(11)
0.290 and −0.165 e.Å−3

24
C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
2.8. Selectivity Index
The crystal of 3 is stabilized by intermolecular hydrogen bonds, as
shown in Fig. 2. The nitrogen atom N(8) is H-bonded to the sulfur
atom S(2b) of a symmetry related molecule, while N(7) interacts with
a hydrogen atom bonded to atom N(4b) building a zigzag alignment of
the complex molecules toward the direction [010].
The selectivity index (SI) was calculated by dividing IC50 for the
VERO cells by the MIC for the pathogen; if the SI is ≥10, the compound
is then investigated further [48].
3. Results and discussion
3.3. Quantum chemical results
3.1. Synthesis of the complexes
DFT calculations, based on the crystallographic data of complex
3, were also performed. After optimization of structure 3, the R
groups on the thiosemicarbazone ligands were substituted and new
optimizations were executed. For comparison with the XRD data, values
for selected bond lengths and angles of the optimized structure of 3
were also inserted in Table 2. The perspective view of the optimized
structure of 3 is presented in Fig. 1. The optimized structures of the
other complexes can be found as supplementary material (Fig. SF1),
while Table ST1 (Supplementary material) shows the geometric aspects
and molecular properties. In regard to the comparison with the
available experimental data, a difference of around 0.2 Å to Mn\S and
0.3 Å to Mn\Nim and Mn\Npy bond lengths between computed and
experimental values is observed. These smaller bond lengths indicate
an electron density increase in the metal–ligand bond orbitals, which
may be a consequence of the quantum chemical simulations performed
in vacuum. An explanation for this difference between the DFT and
experimental values on metal–ligand bond lengths may be due to the
hydrogen bonds formed in the crystal (see Fig. 2). Other DFT methods
were also employed to verify the differences between the experimental
and computational data. However an increase in the quality of the metal
basis did not change the results considerably. These data are better
discussed in the supplementary material quantum chemical analysis.
In order to understand the behavior of atomic charge distribution,
the NPA method was employed [46]. Among the six thiosemicarbazone
complexes, the manganese charge (qMn) ranges from 0.41 to 0.42. This
charge sign is an important indicative of a σ donation from ligands to
metal. Another interesting aspect is the small metal charge variation,
which points out that the structural differences in the thiosemi-
carbazonate ligands used in this study did not affect the metal charge
substantially. This absence of ligand effect in the atomic charge was
also extended to the nitrogen and sulfur atoms, where the atomic
charge was almost unaltered with the coordination of the different
ligands. These results indicate that the atomic charge distribution
toward the coordination sphere may not be an important aspect to
understand the anti-MTB activity. On the other hand, observing the
molecular electrostatic potential map of the six complexes (see
Fig. SF2), slight changes are detected in the peripheral functional
Reactions of Hatc-R with MnCl2·2H2O in the presence of Et3N under
reflux in MeOH afforded analytically pure microcrystalline precipitates
of the manganese complexes 1–6 in reasonable yields (Scheme 1). The
products are quite soluble in CH2Cl2 and DMSO but less soluble in
methanol or ethanol. Elemental analyses are consistent with the
formation of neutral complexes [Mn(atc-R)2], in accordance with the
molar conductivity values found, nearby 0μS cm−1 in CH2Cl2.
3.2. Crystal structure of [Mn(atc-Et)2] (3)
An ORTEP representation of 3 along with the numbering scheme
is presented in Fig. 1. The crystal structure of the complex exhibits a
6-coordinated manganese(II) center bonded to two monoanionic
atc-Et− ligands in N,N,S-tridentate mode. The metal is coordinated
to the ligands through the pyridine nitrogen atoms N(1A) and
N(1B), as well as the azomethine N(2A) and N(2B) atoms and the
sulfur atoms S(1A) and S(1B), forming four six-membered chelate
rings including the Mn(II) atom. The negative charge of the
monoanionic thiosemicarbazonate is delocalized over the ligand
moiety, through the conjugated double bond system, consistent
with the single bond predominant character for the S\C bond the
considerable double bond character observed for the C_N distances.
Comparing the bond lengths found for complex 3 with those of the
previously reported cationic complex [Mn(Hatc)2]·(ClO4)2 [49], in
which the thiosemicarbazone ligand (Hatc) remains protonated
upon coordination, some differences are observed. A predominantly
double C_S bond, bond distances of 1.671(8) and 1.679(8) Å, is
found in [Mn(Hatc)2]·(ClO4)2, while in 3 these values are 1.732(19)
and 1.743(2) Å. The enlargement of the C\S bond in the monoanionic
ligand is due to the deprotonation at N(3) followed by the
thioenolization of the CS bond [21,49]. Furthermore, the tridentate
ligands are nearly planar, almost perpendicular to each other, with
N(1A)\Mn\N(1B) close to 90°. The S(1A)\Mn\S(1B) angle around
102° shows a clear distortion of the octahedral geometry. Additional
information about bond lengths and angles for 3 can be found in Table 2.
Scheme 1. Synthesis of the manganese(II) complexes.

C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
25
Fig. 1. ORTEP view (50% probability) and optimized structure of [Mn(atc-Et)2] (3) with atom labels.
groups, with the exception of complex 6 where a large difference is
observed, which may lead to a distinct biological activity.
are observed for the complexes in CH2Cl2 which display two intense
bands within the regions 295–315 nm and 390–415 nm, similar to
Groni et al. results [50]. The first band (295–315 nm) can be assigned
as having predominantly a ligand–ligand charge transfer (LLCT)
character, showing a ligand π → π* internal transition. Complex 4 is a
distinct case, in which a minor metal to ligand charge transfer (MLCT),
around 12%, is detected, referring to a 3dt2g → π*CN / π*CC transition.
The second band among the complexes (390–415 nm) consists of a
MLCT with particularly small contribution, ranging from 12 to 17%. For
5, a third band is observed at 369.00 nm, corresponding to a major
contribution from phenyl ring π–π* internal transitions. As expected,
spin-forbidden d–d bands were not observed in the spectra of the
high-spin d5 Mn(II) complexes presented in this work.
Understanding the second order perturbation energy (E2) allows the
determination of the strength between donor and acceptor NBO, and
gives some insights into the σ ligand donation to metal. An analysis of
E2 values shows that σ donation from nitrogen lone pair electrons to
manganese (LPN → Mn) is approximately 418 kcal mol−1, while in the
case of the donation from sulfur lone pair electrons to manganese
(
LPS → Mn) it is around 35kcal mol−1, which is in accordance with the
HSAB (hard and soft acids and bases) theory. Despite the fact that
LPN → Mn represents a contribution close to 8% of the total energy
necessary to stabilize each thiosemicarbazone complex, LPS → Mn
consists of less than 1%. These values did not show high variations
among the six complexes, which only points to the slight changes in
the atomic charge distribution. Nevertheless, considering the strength
of these interactions, a larger susceptibility for Mn\S bond dissociation
can be emphasized in comparison to the Mn\Npy and Mn\Nim ones.
A TD-DFT analysis was also performed to understand the electronic
transitions (see Table ST5). In experimental conditions, orange solutions
3.4. Infrared spectroscopy
The IR data are consistent with the N,N,S-coordination of the
monoanionic ligands to the Mn(II) centers in the neutral complexes.
The spectra of Hatc-R are characterized by two strong broad ν(NH)
absorptions in the range of 3373–3203 cm−1. For the complexes,
only one band is observed in this region, according to the single
deprotonation of the ligands. The very strong and sharp bands around
1580 cm−1, assigned to ν(C_N) stretches in the ligands, appear at
around 1590 cm−1 in the spectra of the complexes. The ν(C_S) bands
at 846–800 cm−1 of the spectra of the free thiosemicarbazones are
shifted to 773–788 cm−1 in the complexes.
Table 2
Selected bond lengths and angles (°) refined from X-ray and the optimized data in gas
phase for 3 [PBE1PBE/Lanl2TZ+/6-31G(d,p)].
XRD data for 3
Calculated data for 3
Bond lengths
Mn\N(2A)/N(2B)
Mn\N(1A)/N(1B)
Mn\S(1A)/S(1B)
S(1A)\C(8A)/S(2A)\C(8B)
N(4A)\C(8A)/N(4B)\C(8B)
N(3A)\C(8A)/N(3B)\C(8B)
2.264(16)/2.254(15)
2.264(17)/2.280(15)
2.527(6)/2.521(5)
1.732(19)/1.743(2)
1.336(3)/1.347(3)
1.335(2)/1.321(3)
1.95011/1.93605
1.98394/1.98147
2.33386/2.33590
1.73553/1.72885
1.34581/1.34417
1.32209/1.32887
3.5. Magnetic measurements and EPR studies
The oxidation state of the manganese central atom in the complexes
was confirmed by measurements of the magnetic susceptibility. The
values of magnetic moments measured at 20 °C, in the 5.90–6.55 BM
range, are consistent with the calculated value of 5.9 BM, characteristic
to d5 high spin Mn(II) complexes.
The EPR spectra of Mn(II) complexes 1–6, recorded from frozen
DMSO solutions are depicted in Fig. 3. The spectra show the presence
of two components for complexes 1, 4 and 6. The main component
generates similar profiles for all complexes. The simulated spectrum
shown in Fig. 3 corresponds to this species. The EPR parameters
used to simulate it are g = [2.05 2.05 1.96] and D = 53.0 mT. The
broad lines appear as a result of weak interactions between metal
Bond angles
N(2A)\Mn(1)\N(2B)
N(1B)\Mn(1)\N(1A)
S(1A)\Mn(1)\S(1B)
N(2A)\Mn(1)\N(1A)
N(2B)\Mn(1)\N(1B)
N(1A)\Mn(1)\S(1A)
N(1B)\Mn(1)\S(1B)
S(1A)\Mn(1)\N(2A)
S(1B)\Mn(1)\N(2B)
C(8A)\S(1A)\Mn(1)
C(8B)\S(1B)\Mn(1)
159.18(5)
90.48(6)
102.451(19)
71.73(6)
71.64(6)
145.24(5)
145.50(5)
76.01(4)
75.12(4)
97.56(6)
97.46(7)
177.02241
88.50337
95.37605
79.89498
79.72190
162.53870
162.30577
83.24206
83.22916
93.29756
93.26687

26
C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
Fig. 2. Crystalline and molecular structure of [Mn(atc-Et)2]. [N(8)⋯S(2b)=3.3762(19)Å, N(8)\H(8)⋯S(2b)=168.6°], [N(4)⋯N(7c)=3.191(3)Å, N(4)\H(4)⋯N(7c)=158.0°]. Symmetry
operations used: (b) −x + 2, y − 1 / 2, −z + 2; (c) −x + 2, y + 1 / 2, −z + 2.
centers from neighbor molecules, interconnected by H-bonds, that
are probably partially maintained in DMSO solution, as observed in
the crystal structure of 3 (see Fig. 1), as described also for 5 [51].
Moreover, the minor components correspond to a monomeric species
with D = 0 mT, showing the hyperfine structure of nuclear spin of
manganese (I = 5 / 2), suggesting the binding of DMSO instead of
donor atoms of the TSC ligands in the coordination sphere of the
manganese centers. According to the quantum mechanical calculations
performed, it is most probably that the DMSO molecule coordinates in
positions originally occupied by sulfur atoms of the TSC ligands (see
Section 3.3).
voltammetry experiment with 2 was carried out in DMSO solution at
room temperature and the redox potentials did not change, excluding
the possibility of DMSO coordination, forming new species. Aiming to
prove the presence of only one electron in the manganese compound
processes, the processes were referenced to the standard ferrocene
(Fig. 4C). The complexes presented here have an electrochemical
behavior comparable to that observed for other manganese(II)
complexes reported in the literature [52–54].
Through the values summarized in Table 3, it is possible to
connect the redox potential with the R group bonded to the N(4)
atom of the thiosemicarbazone ligand. It can be seen that the
potential relative to the Mn(II)/Mn(III) couple became more positive
with strongly electron-withdrawing R groups, according to the order:
-methyl b -morpholinyl b -cyclohexyl b -ethyl b -H b phenyl, showing
their influence on the Mn(II) center. Therefore it can be concluded that
the phenyl group better stabilizes the oxidation state +II, while the
methyl group, with electron donating capacity, better stabilizes the
oxidation state +III. On the other hand, for the Mn(III)/Mn(IV) couple
the potential order is somehow changed in relation to the first couple
-methyl b -cyclohexyl b -ethyl b -H b -phenyl b -morpholinyl. It shows
that the complex with the methyl group reaches the oxidation state
+IV more easily than the other groups.
3.6. Electrochemical studies
Cyclic voltammetry (CV) and differential pulse voltammetry (DPV)
of the complexes were performed in dichloromethane with PTBA as
supporting electrolyte, in the potential range from 1.8 to −1.7 V. Due
to the higher sensibility, the DPV results are shown in Table 3. All
complexes presented the same behavior. Fig. 4A and B summarizes
the CV and DPV, respectively, for compound 2, exemplarily. For this
complex, two well-defined quasi-reversible (ipa / ipc ≈ 1) waves are
detected. The two anodic processes correspond to the Mn(II)/Mn(III)
and Mn(III)/Mn(IV) couples, while the two cathodic processes cor-
respond to the Mn(IV)/Mn(III) and Mn(III)/Mn(II) couples. A cyclic
3.7. Biological activity
The biological activity of the compounds was evaluated by
determining the values of MIC against the strains of M. tuberculosis
H37Rv ATCC 27294. Interesting results of MIC for synthetic compounds
are values ≤12.5 μg/mL. They can then be further evaluated in
cytotoxicity tests. VERO cells are more indicated to evaluate the
cytotoxicity, which are epithelial and said to be normal. Tests against
macrophage cell J774 complete the cytotoxicity tests, since they are
cells of the immunologic system.
Table 3
Differential pulse data for the redox couples Mn(II)/Mn(III) and Mn(III)/Mn(IV) for all six
complexes, measured in CH2Cl2 with 0.1 M PTBA as the electrolyte (all potentials
referenced to the Fc/Fc+ couple).
II
III
III
II
III
IV
IV
III
Complexes
EMn
EMn
E1/2(V)
EMn
EMn
E1/2(V)
/Mn
/Mn
/Mn
/Mn
1
2
3
4
5
6
0.36
0.19
0.30
0.27
0.37
0.24
0.38
0.20
0.26
0.25
0.40
0.24
0.37
0.19
0.28
0.26
0.39
0.24
0.96
0.83
0.96
0.93
0.97
1.00
0.99
0.86
0.96
0.88
1.01
1.00
0.98
0.84
0.96
0.91
0.99
1.00
Fig. 3. Simulated and experimental EPR spectra of the [Mn(atc-R)2] complexes from frozen
DMSO solutions.

C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
27
Fig. 4. (A) Cyclic voltammogram of [Mn(atc-Me)2] (scan rate 100 mV s−l) of and (B) differential pulse of [Mn(atc-Me)2]. (C) Cyclic voltammogram of [Mn(atc-Me)2] with the internal
standard ferrocene. All measurements were carried out under argon atmosphere in solutions of CH2Cl2 (0.1 M [PTBA]) at a platinum electrode.
The biological results are shown in Table 4. From six manganese
complexes, three showed MIC values lower than 10 μg/mL. As we
know, structural changes in the ligand can alter significantly the activity
[18]. Complexes 3 and 4 showed similar MIC (3.12 and 3.30 μg/mL,
respectively) with low cytotoxicity of VERO cells (IC50 = 250 and
93.8 μg/mL, respectively) which characterizes these compounds as low
toxic as seen at the high SI values. However, complex 4 was the only
highly cytotoxic of J774A.1 cell. After the coordination of the ligands to
the Mn(II) center, no clear trend in biological activity was observed.
Most of the complexes show remarkable inhibitory effects. In the
case of 1–4 and 6, there was almost no significant change for MIC
with respect to those of the previously published non-coordinated
thiosemicarbazones [3]. However, complex 5 presents a considerably
smaller MIC compared with the free ligand Hatc-Ph (15.6 μg/mL) [3].
This compound showed the lowest MIC and the highest IC50 among all
tested compounds in this work. The calculation of the SI against VERO
cells (SI N 641) shows the promising selectivity of this compound. We
can compare these early results found for complex 5 with first line
drugs such as isoniazid and rifampicin [55] as well as with new
promising compounds such as PA-824, SQ-109 and even TMC 207
(Sirturo) [56], the first drug approved by FDA against MDR TB in the
last 40years. A classification of the manganese complexes in agreement
with the increasing potential as anti-MTB agent, considering the SI
values referred for VERO cells is 6 → 2 → 1 → 4 → 3 → 5. As previously
observed for vanadium(IV) thiosemicarbazone compounds [18], the
most stable manganese complex, considering its participation in an
oxidative process, was also the most active. Complex 5 is the most active
complex and also presents the highest oxidation potential (see Table 4).
The EPR results discussed above showed that in general the more stable
complexes, considering the inertness about binding DMSO in co-
ordination sites originally occupied by donor atoms of the thiosemi-
carbazonate ligands, are the most active. Besides, the manganese salt
MnCl2·2H2O was not active up to 25μg/mL.
Table 4
Anti-MTB activity (MIC), cytotoxicity (IC50), and selectivity index (SI) of the complexes.
Compounds
MIC
IC50
SIa
VERO
J774A.1
μg/mL μM
μg/mL μM
μg/mL μM
IC50/MIC
[Mn(atc)2] (1)
18.2
23.8
3.12 6.27
41.22 500
50.69 125
1132.65 31.3
70.90
n.d.
156.97 80.1
b3.30
65.83
n.d.
27.5
5.3
[Mn(atc-Me)2] (2)
[Mn(atc-Et)2] (3)
[Mn(atc-Ch)2] (4)
[Mn(atc-Ph)2] (5)
[Mn(atc-Mf)2] (6)
MnCl2·2H2O
266.24
502.47
154.85
n.d.
78.1
b2
250
3.3
5.44
93.8
N500
28.4
N641
1.6
0.78 1.31
19.7
N25
N842.27 39.1
33.87 31.3
53.81
n.d.
–
–
–
–
–
–
Fig. 5 is arranged in a decreasing order of MIC values (increasing
anti-MTB activity), which is an easy way to observe how the activity
changes with the modifications of the R groups. Nickel complexes of
a
The selectivity index (SI) was calculated by the ratio IC50(VERO)/MIC. n.d.: not
determined.

28
C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
TSC
DFT
thiosemicarbazone
density functional theory
TD-DFT time-dependent-DFT
MLCT
LLCT
metal to ligand charge transfer
ligand–ligand charge transfer
ferrocene/ferricinium
Fc/Fc+
Acknowledgments
The authors thank FAPESP (Grants 2009/54011-8, 2011/11593-7,
2011/16380-1), CNPq and CAPES for supporting this work. This work
is also a collaboration research project of members of the Rede Mineira
de Química (RQ-MG) supported by FAPEMIG (Project: REDE-113/10).
R.B. Viana is grateful to the NCC/GridUNESP and CENAPAD/SP for the
provision of computational facilities.
Appendix A. Supplementary data
Fig. 5. Increasing order of anti-MTB activity (MIC) in relation to the peripheral
thiosemicarbazonate groups (R).
CCDC 937108 contains the supplementary crystallographic data for
3. These data can be obtained free of charge via http://www.ccdc.cam.
ac.uk/conts/retrieving.html, or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44)
1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk. Supplementary
data associated with this article can be found, in the online version at
http://dx.doi.org/10.1016/j.jinorgbio.2013.10.011.
the type [Ni(atc-R)2] (R = Me, Ph) reported [19] presented similar
MIC values as the analogous manganese complex [Mn(atc-Ph)2] (5)
presented here, but showing invariance of the activity with the different
ligand substituents. High SI indexes are observed for analog complexes
of both metals.
The use of a thiosemicarbazone manganese complex in a cocktail
containing INH could possibly enhance the activation process of INH,
since Mn(III) species formed by oxidation of Mn(II) species by KatG
act as oxidants of the prodrug isoniazid [7], besides having its own
activity, which could be effective in the treatment of tuberculosis.
References
[1] R.O. Pinheiro, E.B. de Oliveira, G. dos Santos, G.M. Sperandio da Silva, B.J. de Andrade
Silva, R.M.B. Teles, A. Milagres, E.N. Sarno, M.P. Dalcolmo, E.P. Sampaio, Clin. Exp.
Immunol. 171 (2013) 210–219.
[2] L.T. Daum, J.D. Rodriguez, S.A. Worthy, N.A. Ismail, S.V. Omar, A.W. Dreyer, P.B.
Fourie, A.A. Hoosen, J.P. Chambers, G.W. Fischer, J. Clin. Microbiol. 50 (2012)
3831–3837.
4. Conclusions
[3] F.R. Pavan, P.I.S. Maia, S.R.A. Leite, V.M. Deflon, A.A. Batista, D.N. Sato, S.G. Franzblau,
C.Q.F. Leite, Eur. J. Med. Chem. 45 (2010) 1898–1905.
[4] S. Sivendran, V. Jones, D. Sun, Y. Wang, A.E. Grzegorzewicz, M.S. Scherman, A.D.
Napper, J.A. McCammon, R.E. Lee, S.L. Diamond, M. McNeil, Bioorg. Med. Chem. 18
(2010) 896–908.
[5] H.I. Lee, J.H. Yoon, J.S. Nam, Y.M. Kim, Y.T. Ro, J. Biochem. 147 (2010) 511–522.
[6] R.S. Magliozzo, J.A. Marcinkeviciene, J. Biol. Chem. 272 (1997) 8867–8870.
[7] M. Nguyen, A. Quémard, S. Broussy, J. Bernadou, B. Meunier, Antimicrob. Agents
Chemother. 46 (2002) 2137–2144.
[8] R.F. Zabinski, J.S. Blanchard, J. Am. Chem. Soc. 119 (1997) 2331–2332.
[9] Z. Rahim, C. Nakajima, R. Raqib, K. Zaman, H.P. Endtz, A.G.M. van der Zanden, Y.
Suzuki, Tuberculosis 92 (2012) 529–534.
A set of manganese compounds with thiosemicarbazones as ligands,
varying the N(4) substituent group by H, methyl, ethyl, cyclohexyl,
phenyl and morpholinyl, were synthesized in satisfactory yields and
were fully characterized both in solution and in the solid state. The
biological assays of these compounds showed early results better than
first line drugs and even new promising anti-TB drugs. The high anti-
MTB activity and the low cytotoxicity against eukaryotic cells show
us a perfect “magic bullet”. The coordination of thiosemicarbazones
to the Mn(II) center may result in the enhancement of their anti-
M. tuberculosis activities. Complex 5 is the one with the highest
oxidation potential, suggesting that redox processes might also be
related to the biological activity. The excellent activity of complex 5
may represent a novel strategy to prepare new antitubercular agents
using manganese as metal center. Finally, these promising results
encourage us to continue with this systematic investigation, modifying
the ligand structure as well as the metal center in order to evaluate their
role in the biological activity. Additionally, in vivo studies are underway
in our laboratories to further evaluate these promising compounds. On
the other hand, new investigations must be performed in order to
know exactly the participation of the manganese in the bacterial
inhibitory effect and to understand their mechanism of action.
[10] J. Oliveira, E.H.S. de Sousa, O.N. de Souza, I.S. Moreira, D.S. Santos, L.A. Basso, Curr.
Pharm. Des. 12 (2006) 2409–2424.
[11] D. da G.J. Batista, P.B. da Silva, D.R. Lachter, R.S. Silva, R.Q. Aucelio, S.R.W. Louro, H.
Beraldo, M. de N.C. Soeiro, L.R. Teixeira, Polyhedron 29 (2010) 2232–2238.
[12] P.I.S. Maia, A. Graminha, F.R. Pavan, C.Q.F. Leite, A.A. Batista, D.F. Back, E.S. Lang, J.
Ellena, S.S. de Lemos, H.S. Salistre-de-Araujo, V.M. Deflon, J. Braz. Chem. Soc. 21
(2010) 1177–1186.
[13] P.I.S. Maia, H.H. Nguyen, D. Ponader, A. Hagenbach, S. Bergemann, R. Gust, V.M.
Deflon, U. Abram, Inorg. Chem. 51 (2012) 1604–1613.
[14] X.F. Zhu, Y.H. Fan, Q. Wang, C.L. Chen, M.X. Li, J.W. Zhao, J. Zhou, Russ. J. Coord. Chem.
38 (2012) 478–483.
[15] R.A. de Souza, A. Stevanato, O. Treu-Filho, A.V.G. Netto, A.E. Mauro, E.E. Castellano,
I.Z. Carlos, F.R. Pavan, C.Q.F. Leite, Eur. J. Med. Chem. 45 (2010) 4863–4868.
[16] D. Santos, B. Parajón-Costa, M. Rossi, F. Caruso, D. Benítez, J. Varela, H. Cerecetto, M.
González, N. Gómez, M.E. Caputto, A.G. Moglioni, G.Y. Moltrasio, L.M. Finkielsztein,
D. Gambino, J. Inorg. Biochem. 117 (2012) 270–276.
[17] D.R. Richardson, D.S. Kalinowski, V. Richardson, P.C. Sharpe, D.B. Lovejoy, M. Islam,
P.V. Bernhardt, J. Med. Chem. 52 (2009) 1459–1470.
[18] P.I. da S. Maia, F.R. Pavan, C.Q.F. Leite, S.S. Lemos, G.F. de Sousa, A.A. Batista, O.R.
Nascimento, J. Ellena, E.E. Castellano, E. Niquet, V.M. Deflon, Polyhedron 28 (2009)
398–406.
[19] P.I.S. Maia, F.R. Pavan, C.Q.F. Leite, U. Abram, E.S. Lang, A.A. Batista, V.M. Deflon, in: L.
Pele, J.J. Powell, S. Kinrade, R. Jugdaohsingh, P. Collery, I. Maymard, A. Badawi (Eds.),
Metal Ions in Biology and Medicine and Aqueous Chemistry and Biochemistry of
Silicon, John Libbey Eurotext, Paris, 2011, pp. 164–171.
Abbreviations
IC50
INH
half maximal inhibitory concentration
isoniazid
CP (KatG) catalase peroxidase
MDRTB multidrug resistant tuberculosis
MIC
MTB
SI
minimal inhibitory concentration
Mycobacterium tuberculosis
selectivity index
[20] M.X. Li, C.L. Chen, D. Zhang, J.Y. Niu, B.S. Ji, Eur. J. Med. Chem. 45 (2010) 3169–3177.
[21] M.X. Li, D. Zhang, L.Z. Zhang, J.Y. Niu, B.S. Ji, Inorg. Chem. Commun. 13 (2010)
1572–1575.
[22] M. Joseph, V. Suni, C.R. Nayar, M.R.P. Kurup, H.-K. Fun, J. Mol. Struct. 705 (2004)
63–70.
TB
tuberculosis

C.G. Oliveira et al. / Journal of Inorganic Biochemistry 132 (2014) 21–29
29
[23] M.A. Soares, J.A. Lessa, I.C. Mendes, J.G. Da Silva, R.G. dos Santos, L.B. Salum, H.
[38] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297.
[39] G. Scalmani, M.J. Frisch, J. Chem. Phys. 132 (2010) 114110.
[40] M.E. Casida, C. Jamorski, K.C. Casida, D.R. Salahub, J. Chem. Phys. 108 (1998)
4439–4449.
[41] R.E. Stratmann, G.E. Scuseria, M.J. Frisch, J. Chem. Phys. 109 (1998) 8218–8224.
[42] C. Van Caillie, R.D. Amos, Chem. Phys. Lett. 308 (1999) 249–255.
[43] F. Furche, R. Ahlrichs, J. Chem. Phys. 117 (2002) 7433.
[44] G. Scalmani, M.J. Frisch, B. Mennucci, J. Tomasi, R. Cammi, V. Barone, J. Chem. Phys.
124 (2006) 094107.
Daghestani, A.D. Andricopulo, B.W. Day, A. Vogt, J.L. Pesquero, W.R. Rocha, H.
Beraldo, Bioorg. Med. Chem. 20 (2012) 3396–3409.
[24] A. Earnshaw, Introduction to Magnetochemistry, Academic Press, New York, 1968.
[25] O. Kahn, Molecular Magnetism, VCH Publishers, 1993.
[26] G.M. Sheldrick, SHELXS97, Program for the Solution of Crystal Structures, University
of Göttingen, Germany, 1997.
[27] G.M. Sheldrick, SHELXL97, Program for the Refinement of Crystal Structures,
University of Göttingen, Germany, 1997.
[28] M. Frisch, G. Trucks, H. Schlegel, G. Scuseria, M. Robb, J. Cheeseman, J. Montgomery,
T. Vreven, K. Kudin, J. Burant, J. Millam, S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H.
Nakai, M. Klene, X. Li, J. Knox, H. Hratchian, J. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R. Stratmann, O. Yazyev, A. Austin, R. Cammi, C. Pomelli, J.
Ochterski, P. Ayala, K. Morokuma, G. Voth, P. Salvador, J. Dannenberg, V.
Zakrzewski, S. Dapprich, A. Daniels, M. Strain, O. Farkas, D. Malick, A. Rabuck, K.
Raghavachari, J. Foresman, J. Ortiz, Q. Cui, A. Baboul, S. Clifford, J. Cioslowski, B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. Martin, D. Fox, T. Keith,
A. Laham, C. Peng, A. Nanayakkara, M. Challacombe, P. Gill, B. Johnson, W. Chen,
M. Wong, C. Gonzalez, J. Pople, Gaussian 03, Revision C.02, 2003.
[29] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868.
[30] C. Adamo, V. Barone, J. Chem. Phys. 110 (1999) 6158.
[31] P.J. Hay, W.R. Wadt, J. Chem. Phys. 82 (1985) 299.
[32] L.E. Roy, P.J. Hay, R.L. Martin, J. Chem. Theory Comput. 4 (2008) 1029–1031.
[33] J.P. Perdew, Phys. Rev. B 33 (1986) 8822–8824.
[34] Y. Zhao, D.G. Truhlar, Theor. Chem. Acc. 120 (2007) 215–241.
[35] S. Grimme, J. Chem. Phys. 124 (2006) 034108.
[45] E.D. Glendening, J.K. Badenhoop, A.E. Reed, J.E. Carpenter, C.M. Bohmann, C.M.
Morales, C.R. Landis, F. Weinhold, NBO 6.0, Theoretical Chemistry Institute,
University of Wisconsin, Madison, 2013.
[46] A.E. Reed, R.B. Weinstock, F. Weinhold, J. Chem. Phys. 83 (1985) 735.
[47] J.C. Palomino, A. Martin, M. Camacho, H. Guerra, J. Swings, F. Portaels, Antimicrob.
Agents Chemother. 46 (2002) 2720–2722.
[48] T.D.S. Program, Antimicrob. Agents Chemother. 45 (2001) 1943–1946.
[49] L. Liu, W. Shi, X.Y. Chen, Y.L. Chen, P. Cheng, Synth. React. Inorg. Met. 36 (2006)
549–554.
[50] S. Groni, C. Hureau, R. Guillot, G. Blondin, G. Blain, E. Anxolabéhère-Mallart, Inorg.
Chem. 47 (2008) 11783–11797.
[51] T.D. Tzima, G. Sioros, C. Duboc, D. Kovala-Demertzi, V.S. Melissas, Y. Sanakis,
Polyhedron 28 (2009) 3257–3264.
[52] A. Neves, S.M.D. Erthal, I. Vencato, A.S. Ceccato, Y.P. Mascarenhas, O.R. Nascimento,
M. Horner, A.A. Batista, Inorg. Chem. 31 (1992) 4749–4755.
[53] R.A. Lewis, G. Wu, T.W. Hayton, Inorg. Chem. 50 (2011) 4660–4668.
[54] N.E. Chakov, M. Soler, W. Wernsdorfer, K.A. Abboud, G. Christou, Inorg. Chem. 44
(2005) 5304–5321.
[55] G. Alliance, T.B.D. Development, B. Street, Handbook of Anti-tuberculosis Agents,
N. York, 88, 2008, pp. 85–170.
[56] R. Osborne, Nat. Biotechnol. 31 (2013) 89–91.
[36] M. Dolg, U. Wedig, H. Stoll, H. Preuss, J. Chem. Phys. 86 (1987) 866.
[37] J.M.L. Martin, A. Sundermann, J. Chem. Phys. 114 (2001) 3408.