Synthesis, characterization and in vitro evaluation of anticancer activity of a new water-soluble thiosemicarbazone ligand and its complexes

Article abstract of DOI:10.1016/j.poly.2019.114218

A new water-soluble thiosemicarbazone ligand, 4-hydroxyl-3-({[(methylamino)carbonothioyl] hydrazono}methyl)benzoic acid (H₃L^COOH) and its complexes with Cu(II), Ni(II), Zn(II), Fe(III) and Mn(III) were synthesized and characterized.

Full text of DOI:10.1016/j.poly.2019.114218

Journal Pre-proofs
Synthesis, characterization and in vitro evaluation of anticancer activity of a new
water-soluble thiosemicarbazone ligand and its complexes
Soheila Jenabi Sardroud, Seyed Abolfazl Hosseini-Yazdi, Majid Mahdavi,
Morgane Poupon, Eliska Skorepova
PII:
S0277-5387(19)30663-1
DOI:
https://doi.org/10.1016/j.poly.2019.114218
POLY 114218
Reference:
To appear in:
Polyhedron
Received Date:
Revised Date:
Accepted Date:
29 June 2019
24 October 2019
25 October 2019
Please cite this article as: S.J. Sardroud, S.A. Hosseini-Yazdi, M. Mahdavi, M. Poupon, E. Skorepova, Synthesis,
characterization and in vitro evaluation of anticancer activity of a new water-soluble thiosemicarbazone ligand and
its complexes, Polyhedron (2019), doi: https://doi.org/10.1016/j.poly.2019.114218
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2019 Published by Elsevier Ltd.

Synthesis, characterization and in vitro evaluation of anticancer activity of a
new water-soluble thiosemicarbazone ligand and its complexes
a
a,
b
Soheila Jenabi Sardroud , Seyed Abolfazl Hosseini-Yazdi *, Majid Mahdavi , Morgane
c
c
Poupon , Eliska Skorepova
aDepartment of Inorganic Chemistry, Faculty of Chemistry, University of Tabriz, Tabriz 51666-
1
4766, Iran
bDepartment of Biology, Faculty of Natural Sciences, University of Tabriz, Tabriz, Iran
cInstitute of Physics of the Czech Academy of Sciences, Na Slovance 2, 182 21 Prague 8, Czech
Republic
*
Corresponding author: Department of Inorganic Chemistry, Faculty of Chemistry, University of
Tabriz,
Tabriz
51666-14766
Iran.
Seyed
Abolfazl
Hosseini-Yazdi
(hosseiniyazdi@tabrizu.ac.ir.com , hosseiniyazdi@yahoo.com)
Tel.:+98-4133393134, fax: +98-4133340191
1

Abstract
A new water-soluble thiosemicarbazone ligand, 4-hydroxyl-3-({[(methylamino)carbonothioyl]
hydrazono}methyl)benzoic acid (H
3
LCOOH) and its complexes with Cu(II), Ni(II), Zn(II), Fe(III)
and Mn(III) were synthesized and characterized. These compounds were characterized by
1
13
conductivity, magnetic susceptibility, elemental analysis, FT-IR, H NMR, C NMR, and UV-Vis
spectroscopy. All the compounds have sufficient solubility and stability in water as a safe, green,
and biologic solvent. The compounds were evaluated for their anticancer activities against the
human chronic myelogenous K562 leukemia and human breast carcinoma MCF-7 cell lines. All
the complexes significantly showed more anticancer activity than the ligand alone and the Cu(II)
complex showed considerable cytotoxicity against the tested cell lines. The mechanism of action
of the most potent complex, Cu(HLCOO) against K562 cell line was further explored by
fluorescence microscopy, flow cytometry, cell cycle and reactive oxygen species (ROS) assays.
Hoechst staining and flow cytometry indicated the apoptosis induction ability of Cu(HLCOO) on
K562 cells. Mechanistic studies indicated that Cu(HLCOO) complex caused cell cycle arrest at
G0/G1 phase and this complex was able to inhibit the cell cycle in a time-dependent manner.
Cu(HLCOO) can also increase the reactive oxygen species levels in the K562 cells. The obtained
results suggest that the studied compounds can be helpful candidates for further evaluations as
chemotherapeutic agents in the treatment of cancer.
Keywords: Apoptosis; Cell cycle; Cytotoxicity; Flow cytometry; Schiff base; Thiosemicarbazone
2

1
. Introduction
Nowadays cancer is considered to be a threatening illness and also a top trend in the field of
therapeutic chemistry. Chemotherapy with metal complexes is an effective approach for fighting
against cancer. For example, Cisplatin is an efficient anticancer drug for treating various tumours,
but it causes some problems due to its resistance and critical side effects [1, 2]. Hence, continual
studies are going on to develop other antitumour-active inorganic complexes to conquer these
problems [3-7]. Thiosemicarbazone metal complexes have been interesting for researchers because
of having various biological activities such as antiviral [8], antibacterial, [9-11] and anticancer
[
12-17] activities. These compounds have shown selective and noticeable antitumour activites and
conquering the persistence to chemotherapy [18-24]. For studying the biological activity of the
metal complexes, it is important to provide natural biological conditions and insolubility of the
metal complexes in water has been a disadvantage. So it is of great importance to design new
thiosemicarbazone chelators and developing new metal complexes with them in order to highten
their water solubility character to be studied in the field of cancer treatment [25]. Most of the water
soluble ligands have hydrophilic groups such as carboxylate or sulfonate moieties [26]. Recently
SO -
we have introduced the sulfonate group as a part of thiosemicarbazones (H
2
L
3
) for increasing
the water solubility character (Scheme 1) [27]. Water solubility of the complexes was metal-
dependent and anticancer activity of the complexes and ligands was studied on some leukemia cell
lines. Effect of the substituents on the ligands and the type of metal were investigated on activity
of the complexes. Therefore, in this paper we introduced carboxylate group to increase aqueous
solubility of also the ligand and evaluated the effect of carboxylate group on solubility, stability,
and anticancer activity on the MCF-7 and K562 cell lines. More over, in this work we improved
depth of anticancer study of the compounds with flow cytometry, cell cycle and ROS assays.
3

SO -
Scheme 1. Structure of the ligand H
2
L
3
2
2
. Experimental section
.1. Materials and instrumentation
The reagents and chemicals used in the present study were bought were purchased from
commercial sources and used without purification. 3-(4, 5-Dimathylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT) was bought from Sigma-Aldrich company. Fetal bovine
serum (FBS) was purchased from HyClone. RPMI 1640 was bought from Gibco. The human K562
and MCF-7 cell lines were obtained from Pasteur Institute (Tehran, Iran). Annexin-V FITC
apoptosis kit, propidium iodide (PI), and cell extraction buffer were bought from Invitrogen
1
13
(
United States). The H NMR and C NMR spectra were studied on a Bruker Avance 400
spectrometer in d -DMSO. UV–Vis measurements were performed with a Shimadzu UV–Vis
700 spectrophotometer in H O and DMSO. Using KBr pellets, infrared spectra were recorded on
6
1
2
an FTIR Bruker Tensor 27 spectrometer. C, H, N, and S analyses were done on an Elementar Vario
El III Melting points were measured with an Electrothermal 9300. Magnetic susceptibility were
determined by Evans NMR method in d -DMSO and TMS as a reference material [28]. Molar
6
conductivity were measured using a cond 7110 WTW conductometer.
4

2
.2. Synthesis of 3-formyl-4-hydroxybenzoic acid
3
-Formyl-4hydroxybenzoic acid synthesised as stated by the literature [29] with slight
modification (Scheme 2). 4-hydroxylbenzoic acid (22 mmol, 3.0 g) suspended in 10 mL
trifluoroacetic acid (TFA) was added drop-wise to hexamethylenetetramine (22 mmol, 3.1 g)
dissolved in trifluoroacetic acid and was refluxed by heating under argon atmosphere for 2 h. When
cooled, the obtained solution was stirred using 4 M HCl (300 mL) for 3 h. The yellowish powder
was collected and washed with an excess of brine. Then it was dissolved in ether and dried. Yield:
1
3
.5 g (19%). H NMR (DMSO- d
6
): δ 12.86 (s, 1H, COOH), 11.47 (s, 1H, ArOH), 10.29 (s, 1H,
CHO), 8.23 (d, 1H, ArH), 8.04 (dd, 1H, ArH), 7.07 (d, 1H, ArH).
2
.3. Synthesis of 4-hydroxyl-3-({[(methylamino)carbonythioyl] hydrazono} methyl) benzoic
LCOOH
acid (H
3
)
3
-Formyl-4-hydroxy-benzoic acid (6.01 mmol, 1.0 g) was dissolved in MeOH (50 mL) and
refluxed by heating gradually. A solution of 4-methyl-3-thiosemicarbazide in methanol (6.01
mmol, 0.63 g) was added to the above solution, then the obtained solution was refluxed for 3 h.
By cooling and evaporating the solvent at room temperature, the ligand was formed as a yellow
1
precipitation (Scheme 2). Yield: 1.5 g (98%). Sample decomposes above 300 °C. H NMR (d
6
-
DMSO): δ 12.65 (broad s, 1H, COOH), 11.46 (s, 1H, ArOH), 10.75 (broad s, 1H, NH), 8.47
(quartet, 1H, NH), 8.44 (d, 1H, ArH), 8.39 (s, 1H, ArCHN), 7.79 (dd, 1H, ArH), 6.94 (d, 1H, ArH),
1
3
3
1
.01 (d, 3H, CH
3
) ppm. C NMR (d
6
-DMSO): δ 177.6 (C=S), 167.1 (COOH), 160.1 (CH=N),
). Selected FT IR
38.6 (Ar), 132.3 (Ar), 128.3 (Ar), 122.1 (Ar), 120.5 (Ar), 115.9 (Ar), 31.0 (CH
3
-1
data (KBr disc, cm ): 3438 s, 3344 s, 3136 s, 2938 m, 1670 s (C=O), 1651 s, 1536 s, 1489 s, 1443
s, 1280 s, 763 m, 625 s, 532 m.
5

2
.4. Synthesis procedures of the complexes
methanol solution of the Mn(CH
Cu(CH COO) .H O and Zn(CH COO) .2H O (1 mmol) was added drop-wise to the solution
A
3
COO)
3
.2H
2
O, FeCl3, Ni(CH
3
COO)
2
.4H O,
2
3
2
2
3
2
2
containing the ligand in methanol (1 mmol) and stirred for 4 h. The obtained precipitate was
filtered off and washed with cold methanol.
Mn(HLCOO)(OAc). Brown precipitate, Yield: 0.27 g (64%). Anal. Calc. (%) for
Mn(HLCOO)(OAc)·3H
O (C12
H18MnN
2
3
8
O S; F.W.= 419.291) C 34.37, H 4.33, N 10.02, S 7.65;
-1
Found (%): C 34.08, H 3.98, N 9.78, S 7.36. Selected FT IR data (KBr disc, cm ): 3407 s, 2930
m, 1610 s, 1552 s, 1489 s, 1377 s, 1275 s, 677 m.
Fe(HLCOO)Cl. Dark brown precipitate, Yield: 0.30 g (83%). Anal. Calc. (%). For
Fe(HLCOO)Cl·H
O (C10
H
11ClFeN
2
3
4
O S; F.W.= 360.576) C 33.31, H 3.07, N 11.65, S 8.89; Found
-1
(
%): C 33.04, H 2.83, N 11.32, S 8.73. Selected FT IR data (KBr disc, cm ): 3435 m, 1697 s
(C=O), 1640 s, 1520 s, 1480 s, 1466 s, 1273 s, 672 m.
Ni(HLCOO). Brown precipitate, Yield: 0.28 g (85%). Anal. Calc. (%) for Ni(HLCOO)·H
S; F.W.= 327.972) C 36.62, H 3.38, N 12.81, S 9.78; Found (%): C 36.18, H 3.01,
O
2
(C
10
H11NiN
3
O
4
-1
N 12.38, S 9.46. Selected FT IR data (KBr disc, cm ): 3437 m, 2928 m, 1693 s (C=O), 1609 s,
1
593 s, 1543 s, 1203 s, 670 s, 514 m.
Cu(HLCOO). Green precipitate, Yield: 0.22 g (66%). Anal. Calc. (%). for CuHLCOO·H
O
2
(C
10
H11CuN
3
4
O S; F.W.= 332.824) C 36.09, H 3.33, N 12.63, S 9.63; Found (%): C 36.24, H 3.04,
-1
N 12.42, S 9.56. Selected FT IR data (KBr disc, cm ): 3397 s, 2859 m, 1689 s (C=O), 1602 s,
1
515 s, 1404 s, 1269 s, 670 s, 535 m.
Zn(HLCOO). Yellow precipitate, Yield: 0.31 g (88%). Anal. Calc. (%). For Zn(HLCOO)·2H
O
2
(C
10
H
13
N
3
5
O SZn; F.W= 352.683) C 34.06, H 3.72, N 11.91, S 9.09; Found (%): C 33.52, H 3.64,
6

1
N 11.59, S 8.89. H NMR ( DMSO-d
ArH), 6.67 ( s, 1H, ArH), 2.77 ( s, 3H, CH
m, 1618 s, 1518 s, 1485 s, 1377 s, 1270 s, 685 s, 505 m.
6
): δ 8.35 ( s, 1H, CHN), 7.81 ( s, 1H, ArH), 7.63 ( d, 1H,
-1
3
). Selected FT IR data (KBr disc, cm ): 3221 m, 2869
2
.5. Single crystal X-ray diffraction studies
Crystallographic studies were performed with a four-circle CCD diffractometer at 120 K, Gemini
of Oxford Diffraction, using Cu/K radiation (λ = 1.54184 Ǻ) radiated from a classical sealed tube
α
monochromated by using graphite and collimated by fibre-optics Enhance collimator. A CCD
detector was also utilized. Cell refinement, data collection, and reduction, and absorbance
corrections performed using CrysAlisPro 1.171.39.35 [30]. By means of equivalent radius and
absorption coefficient the spherical absorption correction was done. The recorded crystal was
twinned with the twin law of 180° around c-axis with a ratio per domains 62/38 %. Charge flipping
method using Superflip was used to solve the structure which was refined by least-squares
2
calculations on F with Jana2006 [31, 32]. Difference Fourier maps were studied to descen the
hydrogen atoms in the structure. H atoms which were bonded to C atom were in their ideal position
(
C-H bond, 0.96 A°). But those were bonded to N were refined in a new geometry. In this survey,
iso (H) was fixed to 1.2 times of Ueq of C, N, and O. The other atoms (except hydrogen) were
refined using harmonic refinement.
U
2
.6. Lipophilicity measurement
A shake-flask method was performed to determine the lipophilicity of the compounds [33]. 50 mL
of 1-octanol and 50 mL of water were mixed and shaken for 24 h for saturation of both phases. An
aqueous solution containing 5 µM of the synthesised compounds were prepared. Consequently 3
7

mL of the prepared solutions were added to 3 mL of 1-octanol. The obtained solutions were shaken
for a night and then centrifuged for the separation of the two phases. Finally the concentration of
the target compounds were analysed before and after mixing with 1-octanol using a UV-Vis
spectrophotometry.
2
.7. Cell culture
MCF-7 (human breast carcinoma cell line) and K562 (human chronic myelogenous leukemia) cells
were obtained from National Cell Bank of Iran (Pasteur Institute, Tehran, Iran). The cells were
grown in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) (Sigma),
1
00 U/mL penicillin and 100 mg/mL streptomycin (Sigma) and subsequently were incubated in a
humidified incubator containing 5% CO at 37 °C.
.8. MTT assay
2
2
The cytotoxic effect of the compounds was assessed against human cancer cell lines (K562 and
MCF-7) by utilizing the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT)
assay according to the manufacturer’s protocol [34, 35]. The procedure is on the basis of the
reduction of MTT by the mitochondrial dehydrogenases to the crystals of formazan. Cells (1 × 105
cells/well) were seeded into 96 well plates and treated with different concentrations of the
compounds for 24 and 48 h. After the treatment, 20 µL of the MTT solution (5 mg/mL in PBS)
were added to each well and incubation was continued for 4 h. The purple-blue MTT formazan
precipitate was dissolved in 200 µL of DMSO and the absorbance values at 570 nm were measured
on a multiwall plate reader (Quant Bio-tek Instruments, Winooski, VT, USA). Untreated K562
and MCF-7 cells were used as a negative control. The cytotoxity effect of Doxorubicin and
Paclitaxel drugs were examined against K562 and MCF-7 cells, respectively, as positive control.
The performed experiments were repeated three times.
8

2
.9. Morphological evaluation of the apoptotic cells
Apoptosis induction was examined using Hoechst staining (blue fluorescent dye), which
intercalate with DNA. The K562 cells were cultured in 12 well plates and treated with IC50 dose
of Cu(HLCOO) (as the most potent complex) for 48 h at 37 C in a humidified 5% CO
ᴑ
atmosphere.
2
Finally, after washing the cells with PBS, DNA was stained with Hoechst 33258 (1 mg/mL) in
PBS for 5 min and evaluated by fluorescence microscopy (Olympus BX41, Germany).
2
.10. Apoptosis survey using by Flow Cytometric Analysis
The annexin-V and PI (propidium iodide) dyes were used to differentiate between necrotic and
apoptoic cells. According to the manufacture’s protocol, an annexin-V FITC/PI double staining
method was utilized. Briefly, K562 cells were cultivated after the treatment with Cu(HLCOO
)
complex. The elusion of treated and untreated cells was performed with PBS for two times. Then
6
1
2
× 10 cells were resuspended in binding buffer (10 mM HEPES/NaOH, pH 7.4, 140 mM NaCl,
.5 mM CaCl ). Then, 5 µL of PI and 5 µL of annexin-V FITC were added. The obtained solution
2
was incubated in the dark place at room temperature for 15 min. Finally it was analysed by flow
cytometry (BD FACSCalibur flow cytometer, USA).
2
.11. Analysis of cell cycle
In order to investigate the cell cycle phase distribution of K562 cells in response to the Cu(HLCOO
)
4
complex the DNA content of cells was analysed. For this analysis, the cells (5× 10 cell/well) were
seeded in 6-well plates 24 h prior to treatment. The treated cells with a single dose (at IC50 value)
of the complex were collected after 24 and 48 h, and eluted for 2 times with PBS. After that, cells
ᴑ
were harvested, fixed in 70% ethanol and kept at -20 C overnight. Then the cells were stained
with a solution containing 1% Triton-X 100 in PBS, treated with RNase (20 µg/mL) and exposed
to the DNA binding dye propidium iodide (50 µg/mL) for 30 min in darkness. The samples were
9

then analysed with a flow cytometer (BD FACSCalibur flow cytometer, USA) and the percent of
calculated cells in the sub-G1, G0/G1, S, and G2/M phases were determined.
2
.12. Intracellular ROS measurements
Reactive oxygen species (ROS) have a significant role in apoptosis induction in the pathologic
conditions and physiologic [36, 37]. The amount of ROS present in cells treated with Cu(HLCOO
)
was analysed using 2′,7′-dichloro-dihydro-fluorescin diacetate (DCFH-DA). The dye is
deacetylated to a non-fluorescent compound, by cellular esterases after diffusion into the cell
which is susequently oxidized by ROS into 2’,7’-dichlorofluorescein (DCF). DCF is a fluorescent
chemical with maximum excitation and emission spectra of 495 and 529 nm, respectively. Thus,
the fluorescence intensity is dependent to the amount of peroxide generated by the cells. DCFH-
DA with a concentration of 10 µM, was added to the cell medium 30 min before the end of the
treatment. The cells were collected by centrifugation at 2000 rpm for 5 min at 4 °C, eluted for two
times with PBS to eliminate the extracellular compounds, and DCFH-DA fluorescence was
detected using flow cytometry (BD FACSCalibur, USA).
1
0

3
3
. Results and discussion
.1. Synthesis and characterization
The ligand was obtained by the condensation of 3-formyl-4-hydroxy-benzoic acid and 4-methyl
1
13
thiosemicarbazide (Scheme 2) and characterized by FTIR, H NMR, C NMR, CHN analysis and
1
13
single-crystal X-ray diffraction. Elemental analysis, H NMR and C NMR showed that the ligand
has high purity without any excess peaks corresponding to starting materials or an unflavoured
product. Singlet peak corresponding to the aldehyde proton at 10.29 ppm in 3-formyl-4-hydroxy-
benzoic acid completely disappeared and a singlet peak at 8.38 ppm was found corresponding to
the iminic proton. Peaks of the protons of COOH and NH-N groups were found as broad singlets
1
3
at 12.64 and 10.75 ppm, respectively. In C NMR, 10 peaks were found corresponding to the
expected structure. The FTIR spectrum of KBr disc of the ligand showed a strong absorption band
-1
at 1651 cm corresponding to C=N stretching vibration after condensation of aldehyde and 4-
methyl thiosemicarbazide. In the FTIR spectrum of the ligand, a strong band corresponding to
-1
C=O stretching vibration of COOH group was found at 1670 cm .
Scheme 2: Synthetic pathways for aldehyde and the ligand
The complexes were produced by the reaction between the ligand and metal salts in methanol in
reflux. All complexes are stable on air. The complexes are soluble in methanol, DMF, DMSO, and
water. Solubilities for H
LCOOH, Mn(HLCOO)(OAc), Fe(HLCOO)Cl, Ni(HLCOO), Cu(HLCOO) and
3
Zn(HLCOO) in water are 40, 2.4, 2.2, 4, 3 and 2.6 g/L, respectively. Unfortunately, all endeavours
to achieve a single crystal of complexes were unsuccessful. So, it was very difficult for us to assign
1
1

exact structures for the complexes. However, we have proposed legal structures for complexes by
1
using elemental analysis, FT-IR, HNMR, molar conductivity, magnetic susceptibility and
electronic absorption spectroscopy methods and refer to previous our research in this field [27, 38,
3
9].
In the FT-IR spectrum of Cu(HLCOO) complex a strong band corresponding to C=O appeared at
689 cm higher than 1670 cm in the free ligand, H
LCOOH. So, carboxylate group of the ligand
-
1
-1
1
3
has been coordinated to metal center as a unidentate mode [40]. The measured magnetic moment
of the compound in DMSO by Evaus method was found to be 1.93 BM. The complex exhibits a
broad band from about 500 to 800 nm with a maximum at 518 nm corresponding to a four-
9
coordinate square planar geometry (Fig. 1a) which is strongly favorable for the d configuration
-3
[
41]. Also, 10 M solution of this complex in DMSO is nonelectrolyte. On the other hand, our
recent research with the similar ligand with sulfonate group have showed the coordination of
sulfonate group to the metal center and forming dinuclear complexes [27, 38, 39]. All in all, we
can propose structure shown in Fig. 2 for this complex.
Mn(HLCOO)(OAc) complex shows no signal above 1610 cm in the region of carbonyl group in
FT-IR spectrum. So, in this complex carboxylate group of the ligand and acetate group should be
coordinated to Mn(III) as a chelating or ionic. We think acetate anion is coordinated to metal as a
chelating bidentate ligand because this complex is nonelectrolyte in solution. So acetate anion
remains coordinated in solution. On the other hand, after coordination of the ligand as three-dentate
and acetate anion as bidentate, coordination of carboxylate group as a chelating ligand to the metal
-1
center is difficult to form seven-coordinate complex. However coordination of H O to Mn(III) in
2
order to from a octahedral structure is possible. Magnetic moment of the complex in DMSO is
4
4
.23 BM, corresponding to a d high spin complex. Overall, our proposed structure for
1
2

Mn(HLCOO)(OAc) is shown in the Fig. 2. Manganese(III) six-coordinate complexes commonly
cause to form fairly intense ligand field bands [41]. This probably originates from a coordination
of factors such as; a fairly covalent metal ligand band, low symmetry components to the ligand
field, and fairly low lying CT bands [41]. As a result we can not see a separated band for this
complex because of the above mentioned reasons (Fig. 1).
In the FT-IR spectrum of Fe(HLCOO)Cl, a strong band at 1697 cm is seen which is corresponding
-1
to unidentate coordination mode of carboxylate group in the solid state. Magnetic moment of the
5
complex is 5.20 BM in agreement with a d high spin configuration in DMSO solution. Also, molar
-1
-1 -1
conductivity (Ʌ = 20 cm mol Ω ) is lower than a 1:1 electrolyte solutions [42] namely chloride
m
anion partially separated in DMSO. In electronic absorption spectrum of the complex only a
shoulder can be seen around 500 nm (high energy) related to a ligand to metal charge transfer
transition (Fig. 1a). Overall, based on all data a dinuclear structure in octahedral geometry is
proposed for Fe(HLCOO)Cl complex (Fig. 2).
In the FT-IR spectrum of the Ni(HLCOO) complex, a band corresponding to C=O is appeared at
1
693 cm-1 higher than 1670 cm-1 for H LCOOH. Therefore, similar to Cu(HLCOO) and
3
Fe(HLCOO)Cl complexes the COO group is coordinated to the nickel as a monodentate mode.
-
1
Magnetic moment of the complex in DMSO is slightly higher than zero and the signals at HNMR
8
appeared in a normal range with fairly broadening corresponding to a d diamagnetic
configuration. Also, there is no band beyond 600 nm in the visible region of electronic spectra
(Fig. 1a). Hence, structure of the complex is four-coordinate square planar (Fig. 2). In contrast to
other complexes, electronic spectrum of the Ni(HLCOO) complex in water is different from DMSO
solution. So, a broad band in low energy region (˃750 nm) confirms a six-coordinate octahedral
geometry in water (Fig. 1b).
1
3

1HNMR of diamagnetic Zn(HLCOO) complex showed purity of complex and depicted clear
evidences for deprotonation of carboxylic acid (COOH) and phenolic hydrogen (OH). After
coordination of Zn(II) to the ligand. The peaks at 12.65 and 11.46 ppm correspond to COOH and
OH, respectively, disappeared in the spectrum. Also, all of the aromatic and iminic protonsʼ peaks
shifted to higher fields. In the FT-IR spectrum of Zn(HLCOO) complex similar to
Mn(HLCOO)(OAc) no band has seen above 1618 cm . Thus, we think similar to
Mn(HLCOO)(OAc) complex the carboxylate group of the ligand is ionic in the complex. However,
coordination of water molecules to form higher coordination numbers for zinc complexes is usual.
-1
Fig. 1. a) visible spectra for Mn(HLCOO)(OAc), Fe(HLCOO)Cl, Ni(HLCOO) and Cu(HLCOO
)
complexes in DMSO b) visible spectra for Ni(HLCOO) complex in water and DMSO
1
4

.
Fig. 2. Proposed structures for complexes
3
.2. X-ray crystallography of the ligand
LCOOH is shown in Fig. 3 and main crystallographic data are
Crystal structure of the ligand H
3
summarized in Table 1 and angles and bond lengths in Table 2. The ligand crystallizes in the
monoclinic space group P2 /n together with one acetonitrile molecule in the asymmetric unit. The
1
thion form of the free ligand is supported by the C=S bond distance with a value of 1.693(4) Å and
bond distance is in the range (1.686(3)-1.730(7) Å) found for similar ligands [27, 38, 39].
1
5

Fig. 3. Molecular structure of H
3
LCOOH
LCOOH
Table 1. Crystal data and structure refinement parameters for H
3
Empirical formula
12 14 4 3 1
C H N O S
Formula weight
Temperature/K
294.3
120.00(10)
Crystal system, Space group
a (Å)
Monoclinic, P2
4.8466 (6)
22.2692 (17)
12.664 (2)
91.630 (11)
1366.3 (3)
4
1
/n
b (Å)
c (Å)
β/°
Volume/Å3
Z
μ/mm-1
F(000)
2.24
616
T
min, Tmax
0.647, 1
3
Crystal size/mm
Radiation
Measured reflections
0.38 × 0.07 × 0.06
Cu Kα
3907
Independent reflections
3872
R
int
0.104
0.080
0.143
2
2
R[F >3σ(F )]
wR(F )
2
S
2.42
1
6

Table 2. Some selected bond lengths and angles for H
LCOOH
3
Bonds lengths (Å)
S1–C5
O1–C2
O1–H1o1
O2–C4
O2–H1o3
O3–C4
1.693(4)
1.353(5)
0.82(5)
1.303(5)
0.86(5)
N1–C5
N1–C9
N2–N3
N2–C5
N3–C10
N4–C12
1.312(6)
1.471(5)
1.382(4)
1.359(5)
1.294(5)
1.120(7)
1.240(5)
Bond angles (º)
C2—O1—H1o1 111(3)
C4—O2—H1o3 109(3)
S1—C5—N1
S1—C5—N2
N1—C5—N2
N3—C10—H1c10
124.1(3)
117.8(3)
117.9(4)
119.36
N2—N3—C10
N3—C10—C6
113.7(3)
121.3(4)
The packing of this compound is governed by the combination of H-bonding (Table 3) and a
LCOOH
typical π-π stacking along a axis. H-bonding creates an infinite zigzag chain of H
3
molecules connected through carboxylic acid (O2—H1O3···O3) and thioamide (N2—H1N2···S1)
homodimers (see Fig. 4).
This chain is decorated with the acetonitrile molecules which are connected by partially accepting
two H-bonds from the hydroxyl and second NH of the terminal thioamide of the H
LCOOH
3
molecule, with distances O1…N4 = 2.917(5) Å and N1…N4 = 2.988(6) Å. The hydroxyl hydrogen
also participates in an intramolecular H-bond with the acceptor being the azido nitrogen. The
neighbouring chains are connected by a non-classical CH-N bond between the acetonitrile methyl
group and the carboxyl carbonyl of the ligand distance, with the distance C11…O3 being
3
.269 (5) Å.
1
7

Fig. 4 Non-bonding interactions in the crystal structure H
LCOOH acetonitrile solvate. Right –
3
crystal packing with the two types of interactions highlighted, top left– H-bonding, bottom left–
π-π stacking
The second most prominent feature of the structure is an atypical π-π stacking (Table 4). We can
considered four different ‘delocalization’ rings inside the structure as the origin of the π-stacking:
Q1 = C1-C2-C6-C7-C8-C3, Q2 = C2-O1-H1o1-N3-C10-C6, Q3 = N3-N2-C5-N1-H1N1 and Q4
i
i
i
i
=
O3-C4-O2-H1o1-O3 -C4 -O2 -H1o1 (with Symmetry codes i : -x, -y, -z+2). The distances of ≈
3
4
.5 Å and angles 95-110° (Table 4) are similar to common π-π stacking between phenyl rings [43-
5].
Table 3. Selected hydrogen bonding interactions in the structure of H
LCOOH
3
D–H∙∙∙A
O1—H1O1···N3 intra
O2—H1O3···O3
N2—H1N2···S1
C11—H2C11···O3
O1—H1O1···N4
N1—H1N1···N4
D–H
H∙∙∙A
D∙∙∙A
D–H∙∙∙A
143(4)
166 (4)
165 (3)
145
0.83(5)
0.86 (5)
0.97 (5)
0.96
0.83(5)
0.95 (4)
2.00(5)
1.78 (5)
2.38 (5)
2.43
2.60(5)
2.19 (5)
2.701(4)
2.619 (4)
3.323 (4)
3.269(5)
2.917(5)
2.988 (6)
i
ii
iii
iv
104(4)
141(4)
iv
Symmetry codes: (i) −x, −y, −z+2; (ii) −x+3, −y, −z+1; (iii) x+1/2, −y+1/2, z−1/2; (iv)
x+1, y, z.
1
8

Table 4. π-π non-bonding interactions in H
LCOOH
3
Dummy atom A Dummy atom B Dummy atom C
Distances A…B(Å) Angles (A…B…C)°
Q1
Q4
Q2
Q2’’
Q1’’
Q3’’
Q1’’
Q2’’
Q2’’
3.5380(4)
3.5330(3)
3.5421(5)
110.024(4)
104.401(3)
97.635(3)
3
.3. Lipophilicity of the compounds
’
Lipophilicity is one of the most important factors that illustrates the molecule s ability to pass
through cell membranes [46]. The lipophilicity of the compounds was quantified by calculating
the concentration ratio of the compounds in water and 1-octanol. All compounds were mostly
distributed in the aqueous phase after mixing. The results indicate that these compounds have very
low lipophilicity. Therefore, we can expect, that any possible cytotoxicity of the compounds would
not be related to their high level of lipophilicity.
3
.4. Stability determination
When working with Schiff bases and their complexes, we must consider sensibility of C=N bond
to water and the following hydrolysis. Therefore, the stability of Schiff base ligand and their
complexes in aqueous solution was followed by UV-Vis spectroscopy at acidic, basic and neutral
conditions (pH = 1, 7.4 and 11) for 4 days in the range of 250–500 nm. Almost all compounds
were stable up to 4 days in water at acidic, neutral and basic pHs (Fig. 5). All our further studies
on the compounds took less than two hours and were performed in neutral media two days.
Therefore, there was no problem with instability of these compounds in aqueous solutions.
1
9

Fig. 5. Electronic absorption spectra for the ligand and complexes at different pH values.
2
0

3
3
.5. Anticancer activity
.5.1 Cell growth inhibition
MTT assay was used for determination of the cytotoxicity of the ligand and its complexes on MCF-
and K562 cell lines. The outcome (shown in Table 5) were evaluated according to cell inhibition
7
which are illustrated as IC50 values. The anti-proliferative activity of the studied complexes was
higher than the free ligand, which demonstrates that the chelation of metal ion is the reason of high
cytotoxic character of the complexes. IC50 values for H
LCOOH and its complexes (Table 5) are
3
SO -
much lower than of our reported values for H
2
L
3
and its complexes [27] (Table 6). Thus, the
new compounds have higher cytotoxicity on K562 cell line. The effect of the compounds on MCF-
cell line is more time-dependent than on K562. For the majority of the compounds, except
7
Fe(HLCOO)Cl, the time (24 h to 48 h) had small effect on viability of the K562 cells. Ni(HLCOO),
Zn(HLCOO), Mn(HLCOO)(OAc) and Fe(HLCOO)Cl are more effective against K562 cells and only
H
LCOOH is more effective on MCF-7 cells. The Cu(HLCOO) complex exhibited the highest
3
cytotoxicity against both cell lines, also this complex showed better activity than Taxol as a drug
in breast cancerthrapy. Therefore this complex was selected for further studies.
2
1

Table 5. IC50 (µg/mL) values of the ligand and complexes for the selected cells at 24 and 48 h.
IC50 (μg/mL) on K562
IC50 (μg/mL) on MCF-7
Compounds
Ligand
24 h
48 h
24 h
48 h
>200
>200
107.3±5.86
63.14±2.91
101±3.23
69.00±2.07
46.45±1.45
57.4±2.08
29.72±0.76
12.85±0.93
37.59±1.03
18.1±1.34
Mn(HLCOO)(OAc)
Fe(HLCOO)Cl
21.93±3.21
82.82±3.45
18.14±2.36
13.9±2.09
33.95±2.98
0.27±0.15
18.45±1.85
22.48±3.34
17.18±1.08
10.19±0.17
27.68±2.06
0.13±0.05-
Ni(HLCOO
Cu(HLCOO
Zn(HLCOO
Positive controls
)
77.5±1.45
12.95±1.04
89.15±2.09
59.21±3.02
)
)
Table 6. IC50(µg/mL) values and solubility of H
2
LSO3- and its complexes on K562 cell line.
compound
IC50 for 24h
—
IC50 for 48
Solubility (g/L)
SO
H
NiHL
2
L
3
SO
—
> 150
110
50
50
10
50
3
> 150
130
130
SO
[
Cu(HL
3
)]
2
SO
ZnHL
3
110
3
3
.5.2. Determination of cell death
.5.2.1. Apoptosis detection using Hoechst staining technique
The changes in the nuclear morphology of the cells were investigated by utilizing fluorescent
microscopy (a qualitative method). As shown in Fig. 6, the control cells (untreated cells) showed
large and round form, while the cells after treatment with Cu(HLCOO) complex exhibited
chromatin condensation and fragmentation. The results demonstrated that the Cu(HLCOO
)
complex induced apoptosis in K562 cells.
2
2

Fig. 6. Morphological study of the K562 cells treated with Cu(HLCOO) (at IC50 value) after 48 h.
3
.5.2.2. Evaluation of apoptosis by flow cytometry
A lot of cancer drugs based on metals apply their cytotoxic impact through apoptosis. Therefore,
we used Annexin V/PI staining and flow cytometric method to determine whether the decrease in
human tumour cell growth was attributable to the induction of apoptosis of cancer cells. This
analysis quantifies the level of apoptosis. The obtained results indicated that K562 cells undergo
apoptosis and not necrosis when they are treated for 24 h in the presence of Cu(HLCOO). In the
control population of K562 cells, 99.5 % of cells remain alive as they stained negative for both
annexin V-FITC and PI (lower left quadrant of Fig. 7a). However, after 24 h of Cu(HLCOO
)
treatment (at IC25 value) 7.43% of cell populations doubled (upper right quadrant of Fig. 7b)
illustrating that they are late apoptotic. There is a population of early apoptotic cells (23.7%) that
are annexin V positive and PI negative (lower right quadrant of Fig. 7b). After morphological
study, these data verified that cytotoxic effect of Cu(HLCOO) is associated by induction of
apoptosis and not necrosis.
2
3

Fig. 7. Quantitatively analysis of apoptosis in the K562 cells treated with IC25 value of the Cu(HLCOO
complex was studied by annexin-V/PI double staining assay. The K562 cells treated with (a) and without
b) of the complex for 24 h and assessed by flow cytometry
)
(
3
.5.2.3. Cell cycle analysis
Flow cytometry and PI staining analysis were performed to investigate the cell cycle phase
distribution in Cu(HLCOO) on the K562, after 24 and 48 hours of treatment. After treatment of the
K562 cells with Cu(HLCOO), a significant increase in the sub-G1 cell (apoptotic cells) population
was detected compared to the control cells that showed the strong evidence of inducing apoptosis
(
Fig. 8). An increase in G0/G1phase population after 48 h treatment indicates that Cu(HLCOO
)
arrests cells in the G0/G1 phase. Percentage of untreated cells in the sub-G1, G0/G1, S and G2/M
phases were computed 3.80%, 43.3%, 19.4% and 32.5%, respectively. While these proportion for
the cells treated with Cu(HLCOO) obtained 12.9%, 42.7%, 19.5%, 23.1% for 24 h and 24.0%,
5
0.0%, 18.1%, 6.58% for 48 h, respectively. Cisplatin inhibits cells predominantly in the S-phase
that acts as a DNA covalent-binding agent to block DNA replication [47]. But our results revealed
the mechanism of action of our compounds is different relative to cisplatin. Also the result of cell
2
4

cycle analysis revealed that Cu(HLCOO) critically obstructed cell’s growth in a time dependent
condition.
Fig. 8. Cell cycle analysis of the K562 cells treated with Cu(HLCOO) complex after 24 –48 h. the
percentage of the cells was determined in each phase of cell cycle by flow cytometry. The
percent of K562 cells population in sub-G1 phase in untreated (control) and treated cells for 24
and 48 h were reported 3.80%, 12.9%, and 24.0%, respectively.
3
.5.2.4. Intracellular ROS measurements
Some chemical compounds promote a sudden increase of ROS inside cells that lead to activation
of apoptosis and promote cell death [48]. Therefore, probing the cellular ROS levels is important
to understand mechanism of compound induced apoptosis. Fig. 9 indicated that K562 cells
exhibited gradual increase in the ROS levels when treated with Cu(HLCOO) complex at IC50 dose
for 24 and 48 h compared with the control. Also there is a noteworthy time-dependent raise in
DCF fluorescence (the cellular ROS levels) after 24-48 h treatment. These findings suggeste that
this the antitumor activity of the complex can be due to its ability to generate cytotoxic ROS during
a redox cycle.
2
5

Fig. 9. Flow cytometry analysis of cellular ROS level in K562 cells treated by Cu(HLCOO
)
complex. The K562 cells were treated with IC50 dose of the complex for 24-48 h. Then, cells
were incubated with DCHF-DA and detected by flow-cytometry
2
6

4
. Conclusion
Synthesis and characterization of new water-soluble thiosemicarbazone ligand and its complexes
is presented. Their anticancer activity against two cell lines (K562 and MCF-7) were examined. In
comparison with the free ligand, all complexes demonstrated significantly increased anticancer
activity. Among the complexes, the Cu(HLCOO) complex showed highest cytotoxicity on both cell
lines. Investigation of the anticancer mechanism of action of the Cu(HLCOO) complex confirmed
that it is due to an increase in the apoptotic process. The results of flow cytometry analysis
indicated that apoptosis of cancer cells induced by complex occurred via formation of ROS.
Furthermore, analysis of cell cycle progression revealed that treatment of K562 cells with
Cu(HLCOO) resulted G0/G1 phase cell cycle arrest. These experimental results showed these new
compounds with COOH functional group on the backbone of the ligand increased anticancer
activity compared to similar compounds with SO ‒
functional group.
3
Acknowledgement
We would like to thank University of Tabriz–Iran for supporting this work. The crystallographic
part was supported by the project 18-10504S of the Czech Science Foundation using instruments
of the ASTRA laboratory established within the Operation program Prague Competitiveness -
project CZ.2.16/3.1.00/24510.
Also we are so thankful to Sakha Pezhhanfar because of carefully reading the manuscript, editing
it, and giving helpful coments.
Appendix A. Supplementary material
2
7

CCDC 1921450 contains the supplementary crystallographic data for H
LCOOH. These data can be
3
obtained free of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
2
8

References
[
[
[
1] J. Reedjik, Inorg. Chim. Acta 198 (1992) 873-881.
2] J. Reedjik, Pure Appl. Chem. 59 (1987) 181-192.
3] A. Terenzi, M. Fanelli, G. Ambrosi, S. Amatori, V. Fusi, L. Giorgi, V. T. Liveria, G.Barone,
Dalton Trans. 41 (2012) 4389-4395.
4] M. Frezza , S. Hindo , D. Chen, A. Davenport, S. Schmitt, D. Tomco, Q.P. Dou, Curr. Pharm.
Des. 16 (2010) 1813-1825.
[
[
[
[
[
5] G. Sathyaraj, T. Weyhermuller, B.U. Nair, Eur. J. Med. Chem. 45 (2010) 284-291.
6] D.S. Raja, N. S. P. Bhuvanesh and K. Natarajan, Dalton Trans. 41 (2012) 4365-4377.
7] C.W. Hsu, C.F. Kuo, S.M. Chuang, M.H. Hou, Biometals 26 (2013) 1-12.
8] C.C. Garcia, B.N. Brousse, M.J. Carlucci, A.G. Moglioni, M.M. Alho, G.Y. Moltrasio, D.
Accorso, N.B. Damonte, E.B. Damonte, Antivir. Res. 57 (2003) 161-170.
[
[
9] D.K. Sau, R.J. Butcher, S. Chaudhuri, N. Saha, Mol. Cell. Biochem. 253 (2003) 21-29.
10] A.P. Rebolledo, G.M. de Lima, L.N. Speziali, D.F. Maia, C.B. Pinheiro, J.D. Ardisson, M.E.
Cortes, H. Beraldo, Appl. Organomet. Chem. 17 (2003) 945-951.
11] N.C. Kasuga, K. Sekino, M. Ishikawa, A. Honda, Yokoyama, M. Nakano, S. Shimada, N.
Koumo, C.K.J. Nomiya, Inorg. Biochem. 96 (2003) 298-305.
12] Z. Afrasiabi, E. Sinn, J.N. Chen, Y.F. Ma, A.L. Rheingold, L.N. Zakharov, N. Rath, S.
Padhye, Inorg. Chim. Acta 357 (2004) 271-279.
13] Z. Afrasiabi, E. Sinn, S. Padhye, S. Dutta, C. Newton, C.E. Anson, A.K. Powell, J. Inorg.
Biochem. 95 (2003) 306-315.
14] D. Kovala-Demertzi, M.A. Demertzis, J.R. Miller, C.S. Frampton, J.P. Jasinski, D.X. West,
J. Inorg. Biochem. 92 (2002) 137-140.
[
[
[
[
2
9