Synthesis and computational study of new meta- and para-substituted ferrocenyl thioureas as potent protein kinase inhibitors and cytotoxic agents

Article abstract of DOI:10.1016/j.ica.2018.12.052

The present study describes the synthesis, characterization, DNA binding and in vitro biological evaluation of ferrocene-enhanced thioureas. The new complexes (N1–N6) were prepared by reacting ferrocenyl anilines (A-B) with freshly prepared isothiocyanates in dry acetone. A crystallographic study of N3 revealed a supramolecular structure involving secondary non-covalent interactions (π?H and π?π). The nature and extent of the binding of these complexes with the salmon sperm DNA (SS-DNA) were examined by cyclic voltammetry and UV–Vis spectroscopy. The complexes have strong binding to DNA with binding constants ranging from 9.83 × 10³ to 5.76 × 10⁴ M^?1. The shifts in the cathodic peak potentials with the addition of DNA in the electrochemical studies of the new complexes are attributed to electrostatic interactions. This observation is an indicator of the oxidizable behavior of the complexes in the presence of DNA. The theoretically calculated energies of the frontier molecular orbitals (E_HOMO and E_LUMO) and the Mulliken charge distributions for the optimized structures determined by the DFT/B3LYP method correlate well with the electrochemically determined redox potentials (correlation coefficient, 0.986). The computational measurements also led to a close agreement between the calculated and observed vibrational frequencies. The new complexes proved to be good candidates for protein kinase inhibition and cytotoxicity studies.

Full text of DOI:10.1016/j.ica.2018.12.052

Inorganica Chimica Acta 488 (2019) 8–18
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Research paper
Synthesis and computational study of new meta- and para-substituted
ferrocenyl thioureas as potent protein kinase inhibitors and cytotoxic agents
a,b,c,⁎
d
b,⁎
c
e
Faiza Asghar
b
c
d
e
, Bhajan Lal , Amin Badshah , Ian S. Butler , Muhammad Nawaz Tahir
a
Department of Chemistry, University of Wah, Quaid Avenue, Wah 47000, Pakistan
Department of Chemistry, Quaid-i-Azam University, Islamabad 45320, Pakistan
Department of Chemistry, McGill University, Montreal, QC H3A 2K6, Canada
Institute of Chemistry, Shah Abdul Latif University, Khairpur, Pakistan
Department of Physics, University of Sargodha, Sargodha, Pakistan
A R T I C L E I N F O
A B S T R A C T
Keywords:
The present study describes the synthesis, characterization, DNA binding and in vitro biological evaluation of
ferrocene-enhanced thioureas. The new complexes (N1–N6) were prepared by reacting ferrocenyl anilines (A-B)
with freshly prepared isothiocyanates in dry acetone. A crystallographic study of N3 revealed a supramolecular
structure involving secondary non-covalent interactions (π⋯H and π⋯π). The nature and extent of the binding
of these complexes with the salmon sperm DNA (SS-DNA) were examined by cyclic voltammetry and UV–Vis
Ferrocenyl thioureas
DNA interaction
DFT calculations
Cytotoxicity
Protein kinase inhibitor
3
spectroscopy. The complexes have strong binding to DNA with binding constants ranging from 9.83 × 10 to
4
−1
5
.76 × 10 M . The shifts in the cathodic peak potentials with the addition of DNA in the electrochemical
studies of the new complexes are attributed to electrostatic interactions. This observation is an indicator of the
oxidizable behavior of the complexes in the presence of DNA. The theoretically calculated energies of the
frontier molecular orbitals (EHOMO and ELUMO) and the Mulliken charge distributions for the optimized structures
determined by the DFT/B3LYP method correlate well with the electrochemically determined redox potentials
(
correlation coefficient, 0.986). The computational measurements also led to a close agreement between the
calculated and observed vibrational frequencies. The new complexes proved to be good candidates for protein
kinase inhibition and cytotoxicity studies.
1
. Introduction
medicine [13], and 8-hydroxy quinolone [14], anticancer ligands has
significantly enhanced their selectivity and cytotoxicity towards cancer
Bioorganometallic chemistry has developed rapidly since the mid-
980s and is now closely linked to medicinal chemistry, especially with
cells compared to their organic analogues. Recently, the ferrocenyl
neodymium(III) complex of curcumin was found to be substantially
more photocytotoxic to Hela and MCF-7 cancer cells than is the phenyl-
modified complex [15]. In addition, ferrocene-functionalized pyr-
azaboles [16], naphthalimides [17], perylenediimides [18], and bis-
thiazoles [19], have been reported to possess good optical, electro-
chemical and thermal properties, and these materials may serve as
potential candidates for photovoltaic and NLO applications. Finally,
both meta- and para ferrocene-based guanidines have demonstrated
DNA binding potency in which the para compounds are more active
than are the meta derivatives [20].
1
respect to metallo-therapeutic agents [1,2]. The “sandwich type” fer-
5
rocene moiety, (η -C
5
H
5
) Fe (Cp), has revealed considerable potential
2
in the development of novel organometallic drugs [3]. Ferrocene has
been projected as an arene bioisostere with which it is possible to
modify the pharmacokinetic or pharmacodynamic profiles of drugs
[
4,5]. Several chemical characteristics, such as chemical modification
and functionalization, hydrophobicity, reduced toxicity and stability to
moisture and air have made ferrocene an important molecule for de-
tailed investigations [6,7]. Several compounds, into which the ferro-
cene fragment has been integrated, have been examined as anti-
malarial, anticancer, antifungal, antiviral [8–10], antituberculosis [11]
and anti-inflammatory drugs [12]. Moreover, ferrocene conjugation to
copper(II) complexes of terpyridine, used in traditional Chinese
Many organic compounds and metallodrugs containing a thiourea
functionality are powerful cytotoxic agents [21]. The synthesis of more
lipophilic thioureas has been of interest for many years, because such
compounds can migrate easily across a cell membrane. Halogen
⁎
Corresponding authors at: Department of Chemistry, University of Wah, Quaid Avenue, Wah 47000, Pakistan.
E-mail addresses: faiza994@hotmail.com (F. Asghar), aminbadshah@yahoo.com (A. Badshah).
https://doi.org/10.1016/j.ica.2018.12.052
Received 25 July 2018; Received in revised form 8 November 2018; Accepted 31 December 2018
Available online 02 January 2019
0020-1693/ © 2019 Elsevier B.V. All rights reserved.

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
substitution can influence the pharmacodynamic properties of the
thioureas by making the resulting compounds more lipophilic or less
basic and vice versa. Fluorinated thioureas are a novel class of effective
anti-trypanosomal agents [22] and represent a new class of influenza
virus neuraminidase inhibitors [23]. Recently, the in vitro antifungal
activity of a series of isomeric fluoro-substituted thioureas has been
reported [24]. Acyl/aroyl thioureas are well-known for their superior
pesticidal, fungicidal and plant growth regulatory activity [25], while
others have significant anti-inflammatory and analgesic activities [26].
The symmetrical and unsymmetrical phenethyl thioureas, 5-halo sub-
stituted thiophene pyridyl thioureas and heterocyclic thioureas are in-
hibitors of HIV-1 reverse transcriptase [27].
2.2.1. 1-(2-Chlorobenzoyl)-3-(4-ferrocenylphenyl)thiourea (N1)
The quantities used were 4-ferrocenylaniline (2.22 g, 8 mmol), po-
tassium thiocyanate (0.78 g, 8 mmol) and 2-chlorobenzoyl chloride
(0.81 mL, 8 mmol). Yield 67%; Yellow solid; m.p. 201 °C; FT-IR and
−1
Raman (powder, cm ): 3368–3175 (NeH), 3077, 3038 (C-Haromatic),
−1
1678, 1665 (C]O), 1598, 1604 (C]C), 1262–1149 cm (C]S), 481,
1
485 (Fe-Cp); H NMR (500 MHz, DMSO‑d
6
, ppm) δ 12.23 (s, 1H, NH),
11.42 (s, 1H, NH), 7.92 (d, 2H, J = 7 Hz, ArH), 7.72 (d, 1H, J = 8 Hz,
ArH), 7.60 (d, 2H, J = 7.5 Hz, ArH), 7.44 (s, 1H, ArH), 7.36 (d, 2H,
J = 7 Hz, ArH), 4.81 (s, 2H, C
5
H
4
), 4.37 (s, 2H, C
5
H ), 4.06 (s, 5H,
4
1
3
C
5
H
5
); C NMR (125.81 MHz, DMSO‑d
6
, ppm) δ 179.2, 167.5, 139.1,
136.3, 135.4, 132.2, 130.6, 128.7, 127.3, 122.4, 84.6, 69.9, 69.3, 66.8;
Motivated by the fascinating pharmacological behavior of meta- and
para-ferrocenyl derivatives and thioureas, we now report the synthesis
of some meta- and para-linked bio-conjugates of ferrocene with
thioureas in order to examine whether this type of linkage has a ben-
eficial influence on antitumor activity. Bioassays and a study of the
interaction of the complexes with cells and with DNA were undertaken
for the purpose of obtaining more information on the mechanism of
action of these potential drugs. In addition, a DFT study was also per-
formed on these structures to determine the theoretical energies of the
frontier molecular orbitals, vibrational frequencies and the Mulliken
charge distributions for the molecular structures.
ESI-MS (m/z): 474.79; Elemental anal. Calcd. (%) for C24
H
19ClFeN OS:
2
C, 60.72; H, 4.04; N, 5.91; S, 6.74; Fe, 11.74. Found (%): C, 60.64; H,
4.11; N, 5.98; S, 6.67; Fe, 11.68.
2
.2.2. 1-(3-Chlorobenzoyl)-3-(4-ferrocenylphenyl)thiourea (N2)
The quantities used were 4-ferrocenylaniline (2.22 g, 8 mmol), po-
tassium thiocyanate (0.78 g, 8 mmol) and 3-chlorobenzoyl chloride
(
0.82 mL, 8 mmol). Yield 73%; Brown solid; m.p. 195 °C (decompose);
−1
FT-IR and Raman (powder, cm ): 3385–3230 (NeH), 3092, 3057 (C-
−1
H
aromatic), 1670, 1674 (C]O), 1596, 1599 (C]C), 1256–1130 cm
1
(
C]S), 485, 478 (Fe-Cp); H NMR (500 MHz, DMSO‑d
6
, ppm) δ 12.35
(
s, 1H, NH), 11.68 (s, 1H, NH), 8.12 (s, 1H, ArH), 8.01 (d, 2H,
2
. Experimental
J = 7.5 Hz, ArH), 7.86 (d, 2H, J = 7.5 Hz, ArH), 7.68–7.45 (m, 3H,
1
3
ArH), 4.73 (s, 2H, C
NMR (125.81 MHz, DMSO‑d
34.2, 132.1, 129.9, 128.5, 127.8, 121.4, 86.9, 70.1, 69.6, 66.8; ESI-MS
m/z): 477.19; Elemental anal. Calcd. (%) for C24 19ClFeN OS: C,
0.72; H, 4.04; N, 5.91; S, 6.74; Fe, 11.74. Found (%): C, 60.81; H, 4.01;
N, 5.84; S, 6.82; Fe, 11.81.
5
H
4
), 4.26 (s, 2H, C
5
H
4
), 4.09 (s, 5H, C
5
H
5
);
C
2.1. Materials and methods
6
, ppm) δ 177.6, 165.8, 138.5, 135.7,
1
Melting points were determined using capillary tubes on an electro-
(
H
2
thermal melting point apparatus model MP-D Mitamura Riken Kogyo
6
(
Japan). Infrared spectra were measured on
a
Thermoscientific
NICOLET 6700 FT-IR instrument. H- and C NMR data were recorded
on a Bruker AV500 MHz spectrometer in DMSO with Si(CH being
1
13
)
3 4
−
1
2.2.3. 1-(4-Chlorobenzoyl)-3-(4-ferrocenylphenyl)thiourea (N3)
used as the internal reference. Raman spectra ( ± 1 cm ) were mea-
sured on an InVia Renishaw spectrometer using argon-ion (514.5 nm)
and near-infrared diode (785 nm) lasers. The Renishaw WiRE 2.0 soft-
ware was used for the Raman data acquisition and spectral manipula-
tion. Mass spectra (ESI-MS) were recorded using LCQ Duo and double-
focusing MS25RFA instruments. Elemental analyses were undertaken
using a LECO-932 CHNS analyzer, while the Fe concentrations were
determined on a Perkin-Elmer model 2380 atomic absorption spectro-
photometer.
The quantities used were 4-ferrocenylaniline (2.22 g, 8 mmol), po-
tassium thiocyanate (0.78 g, 8 mmol) and 4-chlorobenzoyl chloride
(
0.82 mL, 8 mmol). Yield 79%; Brown solid; m.p. 212 °C; FT-IR and
−1
Raman (powder, cm ): 3363–3180 (NeH), 3065, 3089 (C-Haromatic),
−1
1
4
1
669, 1677 (C]O), 1602, 1590 (C]C), 1247–1133 cm (C]S), 471,
1
74 (Fe-Cp); H NMR (500 MHz, DMSO‑d
6
, ppm) δ 12.28 (s, 1H, NH),
1.71 (s, 1H, NH), 8.01 (d, 2H, J = 8.0 Hz, ArH), 7.65 (d, 4H,
),
C NMR (125.81 MHz,
, ppm) δ 176.8, 162.5, 140.1, 136.9, 134.2, 131.7, 129.8,
J = 7.5 Hz, ArH), 7.29 (d, 2H, J = 8.0 Hz, ArH), 4.69 (s, 2H, C
H
5 4
1
3
4
.32 (s, 2H, C
5
H
4
), 4.04 (s, 5H, C
5
5
H );
Ferrocene, 4-nitroaniline, 2-methyl-5-nitroaniline, sodium nitrite,
diethyl ether, acetone, Zn dust, DMSO, ammonium formate, HCl, KSCN
and acid chlorides, such as o-chlorobenzoyl chloride, m-chlorobenzoyl
chloride, and p-chlorobenzoyl chloride, were obtained from commercial
sources (Sigma Aldrich/Fluka) and were used as received. All solvents
were dried and purified before use according to established procedures
DMSO‑d
6
1
29.0, 128.2, 124.6, 85.3, 69.9, 69.2, 66.9; ESI-MS (m/z): 478.83;
Elemental anal. Calcd. (%) for C24
.91; S, 6.74; Fe, 11.74. Found (%): C, 60.65; H, 4.13; N, 5.96; S, 6.68;
Fe, 11.66.
H
19ClFeN OS: C, 60.72; H, 4.04; N,
2
5
[
28]. The ferrocenylanilines (A-B) were synthesized by the method
reported earlier by our group (Step 1, Scheme 1) [29–31].
.2. General procedure for the synthesis of ferrocene-substituted thioureas
2.2.4. 1-(2-Chlorobenzoyl)-3-(4-methyl-3-ferrocenylphenyl)thiourea (N4)
The quantities used were 4-methyl-3-ferrocenylphenylaniline
(2.33 g, 8 mmol), potassium thiocyanate (0.78 g, 8 mmol) and 2-chlor-
obenzoyl chloride (0.81 mL, 8 mmol). Yield 75%; Brown solid; m.p.
2
(
N1–N6)
−
1
1
94 °C; FT-IR and Raman (powder, cm ): 3379–3210 (NeH), 3073,
The synthesis of the ferrocene-incorporated thioureas was accom-
3091 (C-Haromatic), 2928, 2946 (C-Haliphatic), 1665, 1674 (C]O), 1592,
−1
1
plished according to the method reported previously by our group with
some modifications [32–35]. To a suspension of potassium thiocyanate
1600 (C]C), 1261–1145 cm
(C]S), 479, 482 (Fe-Cp); H NMR
(500 MHz, DMSO‑d , ppm) δ 12.32 (s, 1H, NH), 11.60 (s, 1H, NH), 8.10
6
(
8 mmol) in dry acetone (40–50 mL), different acid chlorides (8 mmol)
(d, 1H, J = 9 Hz, ArH), 7.85 (s, 1H, ArH), 7.44 (d, 2H, J = 8.5 Hz, ArH),
7.28 (dd, 1H, J = 7 Hz, ArH), 7.09 (d, 1H, J = 9 Hz, ArH), 6.98 (d, 1H,
were introduced under a N atmosphere leading to the formation of the
2
related isothiocyanates. Ferrocenylanilines (A-B) (8 mmol) were added
to the reaction mixtures, which were then stirred for 12 h. The resulting
reaction mixtures were transferred to ice-cold water and stirred well in
order to remove any water-soluble impurities. The solid products were
separated by filtration and washed with deionized water. The residues
were then dissolved in dichloromethane/chloroform solution and kept
for crystallization (Scheme 1).
J = 9 Hz, ArH), 4.65 (s, 2H, C
5
H
4
), 4.30 (s, 2H, C
5
H ), 4.13 (s, 5H,
4
1
3
C
5
5
H ), 2.41 (s, 3H, CH
3
); C NMR (125.81 MHz, DMSO‑d , ppm) δ
6
178.3, 168.2, 140.1, 137.5, 135.6, 133.8, 131.7, 129.4, 125.0, 121.9,
117.1, 115.7, 86.0, 69.7, 68.5, 66.3, 22.6; ESI-MS (m/z): 486.20;
Elemental anal. Calcd. (%) for C25
H
21ClFeN OS: C, 61.44; H, 4.32; N,
2
5.74; S, 6.57; Fe, 11.41. Found (%): C, 61.49; H, 4.36; N, 5.79; S, 6.48;
Fe, 11.47.
9

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Scheme 1. Synthetic scheme for ferrocenylphenyl thioureas. The nitro/amino group is substituted at the para position in N1–N3, and at the meta position in N4–N6,
with respect to ferrocene. X is H for N1-N3, and CH for N4–N6.
3
2
.2.5. 1-(3-Chlorobenzoyl)-3-(4-methyl-3-ferrocenylphenyl)thiourea (N5)
The quantities used were 4-methyl-3-ferrocenylphenylaniline
2.33 g, 8 mmol), potassium thiocyanate (0.78 g, 8 mmol) and 3-chlor-
Mo-Kα (λ = 0.71073 Å) radiation source. Data collection used ω scans
and a multi-scan absorption correction was applied. The structure was
2
(
solved using the SHELXS-97 program. Final refinement on F was car-
obenzoyl chloride (0.82 mL, 8 mmol). Yield 66%; Brown solid; m.p.
ried out by full-matrix least squares using SHELXL-97 software [36].
−
1
1
3
1
96 °C; FT-IR and Raman (powder, cm ): 3365–3205 (NeH), 3089,
063 (C-Haromatic), 2952, 2914 (C-Haliphatic), 1671, 1663 (C]O), 1598,
−
1
1
591 (C]C), 1258–1132 cm
(C]S), 484, 470 (Fe-Cp); H NMR
2.4. DNA binding studies
(
(
500 MHz, DMSO‑d , ppm) δ 12.43 (s, 1H, NH), 11.56 (s, 1H, NH), 8.04
6
d, 1H, J = 8 Hz, ArH), 7.62 (dd, 4H, J = 7.5 Hz, ArH), 7.08 (d, 2H,
2.4.1. Cyclic voltammetry
J = 8 Hz, ArH), 4.71 (s, 2H, C
5
H
4
), 4.39 (s, 2H, C ), 4.10 (s, 5H,
5
H
4
Commercial salmon sperm DNA (SS-DNA) was solubilized in doubly
1
3
−4
C
H
5 5
), 2.45 (s, 3H, CH
3
); C NMR (125.81 MHz, DMSO‑d
6
, ppm) δ
distilled water to prepare a stock solution of 6 × 10 M from which
1
1
77.2, 166.3 140.5, 138.6, 135.7, 132.2, 130.8, 129.0 125.9, 122.3,
working concentrations of DNA could be obtained. The concentration of
the stock solution was determined from the UV absorbance at 260 nm
16.5, 114.2, 86.4, 70.2, 69.6, 66.9, 22.0; ESI-MS (m/z): 492.83;
−1
−1
Elemental anal. Calcd. (%) for C25
H
21ClFeN
2
OS: C, 61.44; H, 4.32; N,
using an epsilon value of 6600 M cm . Analytical grade tertiar-
ybutylammonium perchlorate (TBAP) was used as the supporting
5
.74; S, 6.57; Fe, 11.41. Found (%): C, 61.39; H, 4.26; N, 5.71; S, 6.62;
Fe, 11.34.
electrolyte and N gas (99.9%) was purged through the mixture to
2
avoid any interference from oxygen. For the electrochemical measure-
ments, an ethanol solution containing 1 mM of the compound being
tested was placed in an electrochemical cell and the voltammogram was
recorded in the absence of DNA. The procedure was then repeated for
the same systems with constant concentrations of the drugs and varying
concentrations of SS-DNA. The working electrode was polished with
alumina powder followed by thorough rinsing with distilled water be-
fore each measurement [37–39].
2
.2.6. 1-(4-Chlorobenzoyl)-3-(4-methyl-3-ferrocenylphenyl)thiourea (N6)
The quantities used were 4-methyl-3-ferrocenylphenylaniline
(
2.33 g, 8 mmol), potassium thiocyanate (0.78 g, 8 mmol) and 4-chlor-
obenzoyl chloride (0.82 mL, 8 mmol). Yield 72%; Brown solid; m.p.
−
1
1
3
1
92 °C; FT-IR and Raman (powder, cm ): 3382–3220 (NeH), 3054,
095 (C-Haromatic), 2948, 2964 (C-Haliphatic), 1667, 1678 (C]O), 1589,
−
1
1
596 (C]C), 1267–1129 cm
(C]S), 480, 472 (Fe-Cp); H NMR
(
(
500 MHz, DMSO‑d , ppm) δ 12.51 (s, 1H, NH), 11.75 (s, 1H, NH), 7.87
6
d, 1H, J = 8 Hz, ArH), 7.72 (d, 1H, J = 8.5 Hz, ArH), 7.20 (d, 2H,
2
.4.2. UV–Vis spectrophotometry
J = 8 Hz, ArH), 7.06 (s, 1H, ArH), 6.81 (d, 2H, J = 8.5 Hz, ArH), 4.77
The UV–Vis spectrum of a known concentration of the drug (25 µM
(
s, 2H, C
5
H
4
), 4.33 (s, 2H, C
5
H
4
), 4.13 (s, 5H, C
5
H
5
), 2.43 (s, 3H, CH );
3
1
3
in ethanol) was first recorded without DNA. The spectroscopic response
was then monitored for the same amount of drug on addition of small
aliquots of DNA solution by keeping the volume and concentration of
the compound constant. Equal amounts of DNA were added to both the
reference and sample cells to avoid the appearance of the peak at about
C NMR (125.81 MHz, DMSO‑d , ppm) δ 179.3, 168.7, 140.6, 138.4,
6
1
6
36.8, 135.5, 133.2, 130.6, 126.9, 123.6, 118.3, 116.2, 87.0, 70.6,
9.7, 66.9, 22.3; ESI-MS (m/z): 492.01; Elemental anal. Calcd. (%) for
C
25
H
21ClFeN OS: C, 61.44; H, 4.32; N, 5.74; S, 6.57; Fe, 11.41. Found
2
(
%): C, 61.47; H, 4.39; N, 5.80; S, 6.53; Fe, 11.45.
2
60 nm where ferrocene also absorbs. All samples were allowed to
equilibrate for 15 min prior to each spectroscopic measurement [37].
Based on the variation in absorption maxima (λmax) in the presence of
DNA, the binding constant “K” of drug-DNA complex was calculated
using the Benesi-Hildebrand equation given below: [40]
2.3. Crystallographic analysis
Diffraction data for a crystal of N3 were collected on a Bruker Kappa
APEXII CCD Diffractometer equipped with a graphite-monochromated
10

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Ao
Ao
G
G G
G
1
3
. Result and discussion
=
+
A
H
H
G G K [DNA]
(1)
3.1. Spectroscopic studies
where A
o
and A are the absorbances of the free compound and the
compound-DNA complex, while Ɛ and ƐH-G correspond to the molar
G
The FT-IR and Raman spectra of the compounds N1–N6 indicated
extinction coefficients of the free drug and the extinction coefficient of
the drug in the fully bound form, respectively.
that the main vibrational stretching modes were observed
between 3382 and 3175 (NeH), 3095–3038 (CeHaromatic), 2964–2914
−
1
(
CeHaliphatic), 1678–1665 (C]O), 1604–1589 (C]C), 1267–1129 cm
(
C]S), and 486–470 (Fe-Cp), respectively. The disappearance of the
2.5. DFT study
characteristic band for isothiocyanate (N]C]S) group in the
−1
2
000 cm region and the appearance of a thiocarbonyl (C]S) band in
−1
Computational studies were performed to calculate the energies of
the 1300–1100 cm
region confirmed the formation of the thioureas
the frontier molecular orbitals (EHOMO and ELUMO) and the charge dis-
tributions on the molecular structures using the DFT/B3LYP method
combined with the 6-31G(d,p) basis set. The density functional theory
method (DFT) was used because of its simplicity and less time con-
sumption [41]. Molecules were first optimized using DFT and the en-
ergy calculations were performed on the optimized structures. The
Gaussian 03 W software was used for the calculations [42,43].
(
N1–N6), as a result of coupling of the ferrocenyl anilines (A-B) with the
freshly prepared isothiocyanates. The NH group exhibits a broad signal
2
−
1
between 3382 and 3175 cm attributable to intra- and intermolecular
hydrogen bonding. The carbonyl group appeared as an intense band at
−1
1
678–1665 cm , while all the other bands are present in their normal
characteristic regions.
1
The H NMR data for the synthesized N,N′-disubstituted thioureas
(
N1–N6) indicate that the –NH protons resonate considerably down-
2
field from the other resonances in the spectra at around 12.51–12.23
and 11.75–11.42 ppm. In most cases, the formation of the thioureas
results in slight downfield chemical shifts of the aromatic and ferro-
cenyl protons compared to those for ferrocenyl aniline owing to the
presence of the neighboring thiocarbonyl moiety, which is also at-
2.6. Protein kinase inhibition assays
The kinase inhibition assays were performed using Streptomyces
5E strain according to the previously described protocol with slight
8
1
5
modifications [44]. The microorganisms under examination were first
refreshed in a sterile Trypton soy broth (Merck, Germany) for 24–48 h
and then applied to petri plates containing ISP4 minimal medium.
Then, 6-mm Whatman filter paper discs soaked with 5 µL of the test
sample (20 mg/mL DMSO) were placed on freshly seeded plates. In-
cubating the plates at 28 °C for 72 h was done to permit the growth of
hyphae. The clear or bald zones around the discs, which indicate hy-
phae formation inhibition, were measured to the nearest mm with the
help of a Vernier caliper. Surfactin served as the positive control, while
DMSO impregnated discs were set as the negative control in order to
confirm the non-toxic effect of DMSO.
tracting electron density. The H NMR spectra show that the un-
substituted cyclopentadienyl (η -C
5
H
5
) ring of ferrocene displays an
5
intense singlet at ∼4 ppm. In the case of the substituted (η -C
5
H ) ring,
4
the ortho protons appear at 4.81–4.65 ppm, while the meta protons
occur at 4.39–4.26 ppm. Signals from the aromatic protons were visible
1
3
between 8 and 7 ppm. The C NMR spectral data also support the
coupling of ferrocenyl aniline to the various isothiocynates to yield the
ferrocene-incorporated acyl thioureas (N1–N6). The chemical shifts of
the carbons of the CSNH and CONH groups of the thioureas resonate
around 179.3–176.8 and 168.7–162.5 ppm, respectively.
Elemental analyses of all the new compounds showed that the cal-
culated and found values for carbon, hydrogen, nitrogen, and iron are
in good agreement with one another indicating their purity.
2.7. Cytotoxicity assays
3.2. Single crystal X-ray analysis
The growth inhibitory effects of the synthesized complexes on the
human breast cancer cell line MCF-7 (ATCC# HTB-22) and the breast
non-cancerous cells MCF-10A (ATCC# CRL-10317) were assessed by
the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bro-
mide) assay method in which the metabolically active cells reduce
MTT to yield a DMSO soluble formazan product that can be analyzed
colorimetrically [45]. The cells were maintained and cultured in RPMI
The X-ray data and the structure refinement parameters show that
N3 crystallizes in the triclinic system with P-1 space group. The basic
crystal data and description of the diffraction experiment are sum-
marized in Table 1. The ORTEP diagram for N3 with the atomic
numbering scheme is illustrated in Fig. 1, while selected bond lengths
and bond angles are given in Table 2. A close look at the structure
5
1
640 culture medium supplemented with 10% FBS (fetal bovine
reveals that the (η -C
5
H
5
) rings of ferrocene have adopted a staggered
4
5
serum). Cisplatin was used as the control. Briefly, 1 × 10 cells were
seeded in triplicate into the wells of the flat-bottomed 96-well culture
plates. The plates were incubated at 37 °C for 24 h in a humidified
atmosphere (5% carbon dioxide in air, pH 7.4) to permit the cells to
attach. The samples were then added at different concentrations
conformation. The phenyl rings attached to the (η -C
5
H ) rings are not
5
5
in the same plane as the (η -C
5
H ) moieties, but both phenyl rings are
5
almost planar to each other. The ferrocene moieties of two immediate
molecules are held at a trans position to each other in order to minimize
repulsion (Fig. 2). The H-bond interactions are presented in Table 3. As
the result of intramolecular H-bonding, a pseudo hexagonal ring
O1C18N2C17N1H1A is formed (Fig. 3). Intermolecular H-bonds of the
type NH⋯O and NH⋯S, and secondary non-covalent interactions
(
0–100 µM in ethanol) to triplicate wells. After 72 h of incubation, the
cells were exposed to 20 μL of MTT (5 mg/mL) and incubated for 4 h at
3
7 °C for further cultivation. The yellow formazan crystals produced in
each well were solubilized in 200 μL of DMSO and the absorbance of
the resulting solution was read at 570 nm using a microplate reader
(
π⋯H), also exist between the molecules and these are responsible for
the formation of supramolecular structure shown in Fig. 3. The occur-
rence of such non-bonding interactions enables N3 to interact with
biological macromolecules like DNA and proteins and are indicative of
potentially good bio-activity [46,47].
(
Labsystems Multiskan MS). The sensitivity of the synthesized com-
pounds against the subjected cell lines was expressed in terms of IC50
values (drug concentration that resulted in 50% reduction of cell
growth), calculated on the regression line where the absorbance va-
lues at 570 nm were plotted against logarithm of drug concentration.
The IC50 concentration values were determined by using a Table
curve 2D Ver. 4 software. The experiments were performed in
triplicate.
3.3. DNA interaction study
3.3.1. Cyclic voltammetry
Cyclic voltammetric measurements were performed with the
11

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Table 1
Table 2
Crystal data of N3.
Selected bond lengths [Å] and angles [°] for N3.
N3
Bond Lengths (Å)
C(1)eC(2)
1.375(9)
CCDC Number
Empirical formula
Formula weight
Temperature (K)
Wavelength Å
a [Å]
1527498
C(6)eC(7)
1.393(8)
2.032(6)
2.038(5)
1.475(6)
1.390(6)
1.380(7)
1.418(6)
1.331(6)
1.652(5)
1.398(6)
1.382(6)
1.230(6)
1.377(6)
1.737(5)
1.361(7)
2(C24
H
19ClFeN
2
OS), CH
2
Cl
2
Fe(1)eC(5)
Fe(1)eC(10)
C(10)eC(11)
C(11)eC(12)
C(13)eC(14)
N(1)eC(14)
N(1)eC(17)
S(1)eC(17)
N(2)eC(17)
N(2)eC(18)
O(1)eC(18)
C(19)eC(20)
Cl(1)eC(22)
C(21)eC(22)
1034.47
296(2)
0.71073
7.6453(5)
9.1372(6)
17.0666(12)
79.461(4)°
86.178(4)°
82.270(4)°
1160.37(14)
triclinic
b [Å]
c [Å]
α [deg]
β [deg]
γ [deg]
3
Volume Å
Crystal System
Space group
P-1
Z
1
3
Density (calculated)
Index Ranges
Absorption coefficient (µ)
F(0 0 0)
1.480 g/cm
Bond Angles (°)
−9≤h≤9, −10≤k≤11, −7≤l≤21
C(1)eC(2)eFe(1)
C(6)eC(7)eFe(1)
C(9)eC(10)eC(11)
C(10)eC(11)eC(12)
C(11)eC(16)eC(17)
C(15)eC(16)eC(17)
N(1)eC(14)eC(15)
C(14)eN(1)eC(18)
N(1)eC(18)eS(1)
N(2)eC(18)eS(1)
N(1)eC(18)eN(2)
C(18)eN(2)eC(19)
N(2)eC(19)eO(1)
N(2)eC(19)eC(20)
C(19)eC(20)eC(25)
C(21)eC(22)eC(26)
39.6(2)
0.990
530.0
1.092
69.4(3)
2
127.1(4)
Goodness-of-fit on F (S)
−
3
119.8(15)
123.2(15)
118.2(15)
116.6(15)
127.9(15)
126.4(13)
118.2(12)
115.4(15)
128.2(14)
122.7(16)
115.9(15)
117.1((16)
120.9(19)
Largest diff. peak and hole
Final R indices [1 > 2sigma(I)]
R indices (all data)
0.51 and −0.73 eÅ
R1 = 0.0593, wR2 = 0.1510
R1 = 0.0821, wR2 = 0.1645
2.28–26
θ range for data collection (°)
Fig. 1. Molecular diagram of N3 with ellipsoid displacement, non-hydrogen
atoms represented by 60% probability boundary spheres and hydrogen atoms
are sphere of arbitrary size.
objective of understanding the redox behavior and the DNA binding
affinities of the synthesized compounds. The electrochemical study was
accomplished in a single compartment cell with a three-electrode
system, i.e., working (platinum disc electrode with a geometric area of
2
0
.071 cm ), reference (saturated calomel electrode, i.e., SCE) and aux-
iliary electrodes (platinum electrodes with geometric areas much
greater than that of the working electrode) [48,49]. Changes in peak
current provided information about binding strength, while the mode of
interaction of compound-DNA was judged from the variation in peak
potential. Drug-DNA binding constants were calculated using the fol-
lowing equation: [50]
log(1/[DNA]) = log K + log{I/(Io I)}
(2)
where K is the binding constant, I and I
o
are the peak currents with and
Fig. 2. Packing diagram of N3 along the crystallographic axis “b” designating
without DNA. The following equation was used to determine the
intramolecular H-bonding.
binding site sizes: [51]
Cb/Cf = K [free base pairs]/s
(3)
Table 3
where s is the binding site size in terms of the base pair, K is the binding
The intramolecular and intermolecular hydrogen bond interactions in N3.
constant, C
f
represents the concentration of the free species and C
b
H-Bonding
D
H
A
d(DeH)/Å d(HeA)/Å d(DeA)/Å DeHeA/°
denotes the concentration of the drug-DNA bound species. Considering
the concentration of DNA in terms of nucleotide phosphate, the con-
centrations of the DNA base pairs can be taken as [DNA base pair]/2
and Eq. (3) can be rewritten as: [51]
N1
H1A O1 0.86
0.93
1.92
2.66
2.652(5)
3.236(5)
142.1
121.1
C13 H13 S1
12

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Fig. 3. Secondary non-covalent interactions in N3 leading to supramolecular structure.
the scan rate in V s−1, α is the transfer coefficient, n is the number of
Cb/Cf = K [DNA base pairs]/2s
(4)
electrons involved in the process and D
o
is the diffusion coefficient in
The values of C
b
/C
f
are equal to (I
o
-I)/I, which are the values of the
2
−1
cm s
.
experimental peak currents. The diffusion coefficients of free drugs and
DNA-bound drugs provide the best information about the molecular
masses of the drug-DNA adducts. The following form of the Randles-
Sevcik equation was used for calculating the diffusion coefficients:
The synthesized ferrocene derivatives (N1–N6) display similar vol-
tammetric responses. Fig. 4 illustrates the DNA binding study of N3
using cyclic voltammetry. This compound shows quasi-reversible elec-
trochemical behavior as evidenced by a change in the peak potential at
different scan rates (Fig. 4a). The voltammogram without DNA features
a couple of well-defined redox peaks in the potential range of
[
52,53]
pa = 2.99 × 10 n( n)1/2A C D1/2
5
1/2
I
(5)
−
0.2–0.9 V. The anodic and cathodic peaks appeared at 0.479 and
*
0.353 V, respectively, vs. SCE. In the presence of an increasing con-
where Ipa is the anodic peak current, C
o
is the concentration of the
−
1
2
centration (4–20 μM) of DNA, the oxidation and reduction peak
reductant (mol L ), A is the geometric area of the electrode in cm , ʋ is
Fig. 4. (a) Representative plots of Current vs. Potential/V (SCE) at different scan rates for N3. (b) Cyclic voltammogram of 1 mM N3 in ethanol with 1 mL of 0.5 M
TBAP as supporting electrolyte in the absence and presence of 4–20 µM DNA showing a decrease in I from I
o
and a concentration dependent -ve shift in potential
½
showing electrostatic interactions. (c) Representative plot of current vs. (V/s) , for the determination of diffusion coefficient of free N3 and N3-4 µM DNA. (d)
Representative plot of log (I/I
o
-I) vs. log (1/[DNA] for determination of binding constant of N3. (e) Plot of C
b
/C vs. [DNA]/μM for determination of binding site size
f
of 4–20 µM DNA concentrations N3.
13

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Table 4
3.3.2. UV–Vis spectroscopy
Important parameters for redox behavior and DNA binding studies (CV and
UV).
The interaction of the ferrocene-modified thioureas (N1–N6) with
DNA has also been investigated by UV–Vis spectroscopy in order to
obtain further evidence concerning the mode of interaction and the
binding strength in parallel to the CV studies. Fig. 5a shows the effect of
various concentrations of DNA (4–20 µM) on the electronic absorption
spectra of N3. The characteristic UV spectrum exhibits two peaks in the
Code Binding
CV
UV
Constant K
−
1
[cm s−1]
2
Do [cm s−1]
2
[
M
]
D
o
Binding
site size
Binding
Free drug
drug-DNA
Constant K
−1
‘s’ [bp]
[M
]
*
UV region. One at 242 nm is due to a π-π transition in the cyclo-
5.76 × 104
1.48 × 10
3.43 × 10
6.54 × 10
5.22 × 10
4.36 × 10
7.90 × 10
−7
−7
−7
−7
−7
−7
1.07 × 10
−7
−7
−7
−7
−7
−7
0.590
0.338
0.321
0.174
0.192
0.108
7.04 × 10
3.27 × 10
6.91 × 10
1.95 × 10
2.20 × 10
1.08 × 10
4
4
4
4
4
4
N1
N2
N3
N4
N5
N6
pentadienyl ring of ferrocene, and the other at 284 nm is attributed to
4
*
4.25 × 10
2.84 × 10
the π-π transition of electrons in the phenyl chromophores. Following
4
4
4
4.03 × 10
4.18 × 10
the substantial addition of DNA, the UV–Vis spectra of the compounds
showed a decrease in absorption with a slight blue shift (hypochro-
mism), which confirms an electrostatic interaction between N3 and
DNA (Fig. 5a). The hypochromic effect is thought to be due to the in-
teraction between the electronic states of the binding chromophore and
1.45 × 10
3.54 × 10
1.99 × 10
2.01 × 10
4.47 × 10
9.83 × 103
potentials of N3 shifted cathodically with the decrease in peak current.
The substantial decay in peak current is attributed to the formation of a
slowly diffusing N3-DNA supramolecular complex resulting in a low-
ering of the concentration of the free drug (that is mainly responsible
for the transfer of current) (Fig. 4b). The DNA binding mode of a drug
can be judged from the variation in the formal potential. The obvious
negative peak potential shift (cathodic shift) in the CV behavior of N3 is
related to the electrostatic interaction of its positively charged ferro-
cenium state with the negatively charged oxygen of DNA. The diffusion
those of the DNA bases [54].
A
DNA binding constant of
4
−1
6.908 × 10 M
for compound N3 is in close agreement with the
value obtained from CV (Fig. 5b).
3.4. DFT study
Computational calculations were performed for these electroactive
compounds in order to complement the experimental outcomes of
cyclic voltammetry and vibrational spectroscopy. The reduction and
oxidation potentials of the compounds determined by cyclic voltam-
metry revealed an analogous trend to that predicted from the DFT
work. Fig. 6a and b provide graphical demonstrations of the HOMO and
LUMO orbitals of N3. The HOMO orbitals are localized on the ferrocene
moiety (Fe metal), whereas, the LUMO orbitals are extended over the
thiourea functionality (containing electron withdrawing N and O
atoms), and the chloro-substituted phenyl ring. Generally, the energy
values of the HOMO/LUMO orbitals and the energy gaps reflect the
chemical activities of the molecules. The HOMO orbitals indicate the
preferred site for the oxidation of the electroactive species, whereas the
LUMO orbitals correspond to the probable site for the reduction of the
optimized compound. The HOMO orbital as an electron donor re-
presents the ability to donate an electron, while the LUMO orbital as an
electron acceptor indicates the ability to receive an electron [55,56].
The HOMO-LUMO energy gap reveals the intramolecular charge
transfer (ICT) interaction that occurs within the molecule. The presence
of a para-linked ferrocenyl unit in the thioureas N1–N3 results in better
π-conjugation compared to that in the meta-substituted ferrocenyl
fragment in N4–N6, which leads to low HOMO-LUMO band gap values
−
7
2
−1
−
coefficient of N3-DNA is 4.18 × 10 cm
s
, which is less than the
7 2 −1
diffusion coefficient of free N3 (6.54 × 10 cm s ). The difference
may be attributed to the formation of a slowly diffusing heavy drug-
DNA adduct (Fig. 4c). The decrease in the peak current as a con-
sequence of varying DNA concentrations provided the N3-DNA binding
4
−1
constant of 4.027 × 10 M according to Eq. (2) (Fig. 4d). The binding
site size of 0.321 bp was calculated with the help of Eq. (4) (Fig. 4e),
which also illustrates the electrostatic mode of interaction.
The essential DNA binding parameters of the compounds studied are
listed in Table 4. The DNA binding constant results reveal that those
compounds having ferrocene at the para position (N1–N3) have a
slightly greater binding capability than do the corresponding meta de-
rivatives (N4–N6). This situation may be due to more delocalization of
electrons when the cyclopentadienyl ring of ferrocene is at the para
position, thus rendering the nitrogen atom to become more polarized,
stable and favorable to bind electrostatically with a negatively charged
DNA [20]. The lower K (binding constant) values of N3–N6 may also be
related to the steric factor of the methyl groups on the phenyl rings
attached to ferrocene, which do not permit uniform binding of the
compound with DNA.
[
57–59]. The HOMO-LUMO energy gaps are in close agreement with
Fig. 5. (a) Representative plots of absorbance vs. wavelength of 25 µM N3 in ethanol with increasing concentration of DNA (4–20 µM). (b) Plot of A
DNA] for determination of DNA binding constant of N3.
o
/A-A vs. 1/
o
[
14

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Fig. 6. (a) Representative graphical demonstration of HOMO of N3. (b)
Representative graphical representation of LUMO orbitals of N3.
Table 5
E
HOMO and ELUMO values obtained through DFT calculations.
Sample
E
HOMO (eV)
E
LUMO (eV) HOMO-LUMO
Electrochemical Band
Code
Band Gap (eV) Gap (V)
Fig. 7. Comparison of oxidation and reduction potentials of 1 mM ferrocenyl
thioureas N1-N6 in ethanol recorded at a scan rate of 200 mV/s.
N1
N2
N3
N4
N5
N6
−0.3171
−0.3515
−0.3350
−0.4546
−0.4211
−0.4902
−0.2256
−0.2402
−0.2341
−0.3304
−0.2823
−0.3708
0.0915
0.1113
0.1009
0.1242
0.1388
0.1194
0.1203
0.1344
0.1268
0.1429
0.1496
0.1372
(
correlation coefficient = 0.997). The largest difference between the
−1
calculated and observed frequencies is 21 cm
.
3.5. Protein kinase inhibition assays
the electrochemical band gaps (correlation coefficient = 0.986), com-
puted from the difference between the oxidation and reduction poten-
tial of the species (Eredox = Eoxi − Ered) (Fig. S1 and Table 5).
Protein kinase inhibitors comprise a distinct class of oncogenic ki-
nase inhibitors. The protein kinase inhibition data for the ferrocenyl
thioureas are listed in Table 7. These complexes display a varying de-
gree of inhibition with the zones of inhibition ranging from
The ease of reduction of the compounds varied in the following
sequence: N6 > N4 > N5 > N2 > N3 > N1. A similar trend was
obtained from the ELUMO values, with the reduction potential of N6
being the highest owing to the easiest reduction and hence the most
negative ELUMO value [60,61]. A more negative ELUMO favors addition
of electrons as the energies of the orbitals are reduced. The EHOMO
values obtained from DFT were compared with the oxidation potentials
obtained from the CV measurements. The oxidation potentials observed
experimentally for the compounds varied as N1 > N3 > N2 >
N5 > N4 > N6. This observation is supported by the DFT calcula-
tions on comparing the EHOMO values; the less negative value for N1
indicates its ease of oxidation as compared to other three compounds
1
4.3 ± 1.15 to 21.8 ± 0.85 mm. Surfactin was used as the control.
The compounds N1–N6 were found to be effective inhibitors as they
produced the maximum zone of inhibition on the culture plates. This
may be attributed to the better hydrogen bond donor ability of
thioureas and also since the sulfur is a superior donor for π–π* inter-
actions; such bonds are considered to be very important in the ligand-
kinase interaction [63]. The higher activity of these complexes may
perhaps be related to the ability of chlorine to form hydrogen bonds
with the hydrophobic pocket of the enzyme. Generally, the compounds
N1–N3, which have ferrocene at the para position, display a somewhat
greater scavenging potential as compared to the meta-ferrocenyl deri-
vatives N4–N6. This difference may be due to extended delocalization
of electrons and the fact that the steric hindrance is at a minimum in the
case of para-substituted ferrocene complexes. The results of the in-
hibition assays reveal that the synthesized compounds can be con-
sidered as potential candidates to inhibit tumor initiation.
(
Fig. 7 and Table 5). The ferrocene moiety at the meta position of
thiourea complexes N4–N6 is harder to oxidize in contrast to that in the
para-substituted compounds N1–N3. The meta branching hinders the
extended delocalization, hence it is easier to oxidize the ferrocene
fragment in N1–N3 [58,59].
The DFT-based calculations also enabled determination of the
Mulliken charges on these molecular structures, which help in identi-
fying the sites that are prone to electrophilic or nucleophilic attack
3.6. Solubility and stability
[
62]. The graphical description of the Mulliken charge distribution on
the representative compound N3 is illustrated by Fig. S2
Supplementary Information File). The molecule N3 has several pos-
The ferrocenyl thioureas (N1–N6) showed high solubility in di-
methyl sulfoxide (DMSO), dimethylformamide (DMF) and dioxane.
They were also fairly miscible with water, ethanol, acetone, chloroform
and dichloromethane, but they were insoluble in hydrogenated solvents
such as: toluene, n-hexane, etc. The stability of a complex in solution
media under physiological conditions is vital for its in vivo application
as drug. Thus, the stabilities of the ferrocene-containing thioureas
(N1–N6) were explored by UV–Vis spectroscopy [13–15,64]. The
changes occurring in absorbance were noted after certain time inter-
vals. The complexes retain their stability in 10% Dulbecco’s Modified
Eagle’s medium (DMEM) (Wisent Inc., St-Bruno, Canada), as validated
from the absorption spectra, which displayed no substantial changes
even after 72 h (Fig. S3).
(
sible sites for electrophilic (the electrophilic sites are most electro-
negative and are represented as red color) and nucleophilic attack (the
nucleophilic sites are most positive and are represented as green color).
The IR vibrational frequencies determined experimentally are
compared with those obtained theoretically from the DFT calculations
upon employing the B3LYP/6-31G(d,p) method for the optimized
geometries of compounds N1–N6. Some of the important calculated
and experimental vibrational frequencies together with their vibra-
tional assignments are listed in Table 6. The vibrational frequencies
were scaled by the typical factor 0.96. The theoretical and observed
vibrational frequencies follow a linear relationship, as shown in Fig. 8
15

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Table 6
Vibrational frequencies (for selected vibrations only) of ferrocenyl thioureas (N1-N6) calculated at the B3LYP/6-31G(d,p) level compared with the experimental
data.
Calculated and Experimental IR Vibrational Frequencies (cm−1)
Assignment
υ NeH
N1
N2
N3
N4
N5
N6
3384–3196 (Cal)
3363–3211 (Cal)
3385–3230 (Exp)
3106 (Cal)
3092 (Exp)
–
3379–3194 (Cal)
3363–3180 (Exp)
3049 (Cal)
3065 (Exp)
–
3356–3225 (Cal)
3379–3210 (Exp)
3061 (Cal)
3073 (Exp)
2943 (Cal)
2928 (Exp)
1673 (Cal)
1665 (Exp)
1600 (Cal)
1592 (Exp)
1551 (Cal)
1568 (Exp)
1436 (Cal)
1454 (Exp)
1246––1127 (Cal)
1261–1145 (Exp)
1062 (Cal)
1045 (Exp)
1033 (Cal)
1018 (Exp)
993 (Cal)
3384–3222 (Cal)
3365–3205 (Exp)
3070 (Cal)
3089 (Exp)
2941 (Cal)
2952 (Exp)
1658 (Cal)
1671 (Exp)
1592 (Cal)
1598 (Exp)
1543 (Cal)
1562 (Exp)
1431 (Cal)
1445 (Exp)
1253–1145 (Cal)
1258–1132 (Exp)
1044 (Cal)
1057 (Exp)
1012 (Cal)
1030 (Exp)
990 (Cal)
3367–3206 (Cal)
3382–3220 (Exp)
3072 (Cal)
3054 (Exp)
2935 (Cal)
2948 (Exp)
1656 (Cal)
1667 (Exp)
1602 (Cal)
1589 (Exp)
1576 (Cal)
1559 (Exp)
1446 (Cal)
1457 (Exp)
1254–1141 (Cal)
1267–1129 (Exp)
1069 (Cal)
1051 (Exp)
1014 (Cal)
1027 (Exp)
977 (Cal)
3
368–3175 (Exp)
υ CeHaromatic
υ CeHaliphatic
υ C]O
3058 (Cal)
3
–
–
077 (Exp)
–
–
1669 (Cal)
1654 (Cal)
1670 (Exp)
1590 (Cal)
1596 (Exp)
1547 (Cal)
1564 (Exp)
–
1660 (Cal)
1669 (Exp)
1605 (Cal)
1602 (Exp)
1560 (Cal)
1545 (Exp)
–
1
678 (Exp)
1607 (Cal)
598 (Exp)
υ C]C
1
β NeH
1566 (Cal)
1
–
–
551 (Exp)
δ CeHaliphatic
υ C]S
–
–
1242–1130 (Cal)
1240–1118 (Cal)
1256–1130 (Exp)
–
1263–1150 (Cal)
1247–1133 (Exp)
–
1
–
–
–
–
262–1149 (Exp)
ω CeHaliphatic
ρ CeHaliphatic
β CeHaromatic
γ CeHaromatic
β C]O
–
–
–
–
–
–
981 (Cal)
986 (Cal)
973 (Exp)
842 (Cal)
856 (Exp)
774 (Cal)
766 (Exp)
702 (Cal)
714 (Exp)
632 (Cal)
640 (Exp)
480 (Cal)
485 (Exp)
972 (Cal)
980 (Exp)
835 (Cal)
851 (Exp)
788 (Cal)
772 (Exp)
711 (Cal)
725 (Exp)
647 (Cal)
633 (Exp)
477 (Cal)
471 (Exp)
9
65 (Exp)
853 (Cal)
40 (Exp)
791 (Cal)
76 (Exp)
715 (Cal)
01 (Exp)
635 (Cal)
48 (Exp)
472 (Cal)
81 (Exp)
979 (Exp)
974 (Exp)
989 (Exp)
866 (Cal)
823 (Cal)
829 (Cal)
8
850 (Exp)
836 (Exp)
841 (Exp)
797 (Cal)
770 (Cal)
796 (Cal)
7
785 (Exp)
787 (Exp)
783 (Exp)
γ NeH
706 (Cal)
723 (Cal)
726 (Cal)
7
719 (Exp)
710 (Exp)
712 (Exp)
γ C]O
659 (Cal)
650 (Cal)
631 (Cal)
6
645 (Exp)
638 (Exp)
646 (Exp)
υ FeeCp
486 (Cal)
475 (Cal)
470 (Cal)
4
479 (Exp)
484 (Exp)
480 (Exp)
Cal = calculated frequency; Exp = experimental frequency; υ = stretching; δ = deformation; β = in-plane bending; γ = out of plane bending; ω = wagging;
ρ = rocking
3
.7. Cytotoxicity assays
(IC50) data listed in Table 8, it can be seen that the screened compounds
reveal substantial action against cancerous cells, though much less than
cisplatin. However, these ferrocenyl derivatives exhibit fewer toxic ef-
fects in normal cells. One plausible cause for the variance in action as
compared to that for cisplatin may lie in the difference in their inter-
action with DNA [65]. Cisplatin interacts covalently with the ni-
trogenous bases in the DNA, while the synthesized complexes are as-
sumed to experience electrostatic interactions, as indicated by the
electrochemical and spectroscopic studies. It is obvious from Fig. 9 that
the tested compounds restrict the growth of tumor cells in a dose-de-
pendent fashion. The time-dependent response of the ferrocenyl
thioureas was also evaluated after regular intervals (12–72 h). The in-
hibitory responses against the cancerous cells was found to be max-
imum after 72 h incubation (Fig. S4).
The anticancer activity of the synthesized complexes along with that
for cisplatin (used as a control) were assessed against MCF-7 carcinoma
cells and also towards the non-cancerous cell line MCF-10A, using the
MTT reduction assay method. From the 50% inhibitory concentration
The results of the cytotoxicity assays indicate that the para-ferro-
cene substituted thiourea complexes N1–N3 show enhanced cell growth
inhibitory effects as compared to their meta analogues N4–N6, which
might be ascribed to their better π-conjugation ability [20]. This result
indicates that the presence of ferrocene at the para position augments
the lipophilicity, which in turn helps the molecules to interact or pe-
netrate more through the cell membrane, thereby enhancing their cy-
totoxic activity. Furthermore, replacing the H atom of the phenyl rings
(
attached to Fc moiety) with a methyl group (N4–N6) leads to a de-
creased activity, which can be ascribed to their minimum lipophilic
character owing to steric hindrance. However, no particular cytotoxi-
city trend was observed by changing the position of the substituents
from ortho to para on the benzoyl group.
Fig. 8. Linear correlation graph between the experimental and calculated vi-
brational frequencies of a representative compound N3. (Correlation coeffi-
cient = 0.9975).
The cytotoxic effect of many drugs is due to the interaction with
DNA that ultimately inhibits the transcription and replication resulting
16

F. Asghar et al.
Table 7
Inorganica Chimica Acta 488 (2019) 8–18
related to the oxidation state of the central iron atom of the ferrocenyl
Protein kinase inhibition assay of synthesized ferrocenyl deri-
vatives.
group. The low oxidation potentials of N1–N6 (0.483 V, 0.453 V, 0.479,
0
.504 V, 0.490 V, and 0.497 V) as compared to ferrocene (0.518 V)
supports the idea that the ferrocenyl moiety may be easily oxidized to
Sample Code
Zone of Inhibition (mm)
3
+
3+
the Fe state. It is believed that it is mainly the Fe -ferrocenyl cation
that undergoes electrostatic interaction with the phosphate backbone of
N1
N2
N3
N4
N5
N6
PC
21.8 ± 0.85
18.4 ± 0.57
20.0 ± 1.24
15.5 ± 1.03
16.2 ± 0.76
14.3 ± 1.15
25
2+
3+
the DNA. Electron transfer reactions involving Fe - and Fe -ferro-
cenyl species may generate reactive oxygen species (ROS), such as
•
hydroxyl radicals ( OH) in vivo that cause damage to DNA, and may also
be responsible for the anticancer activity through the formation of ra-
dical metabolites that cause biological damage in the cancer cell
[
71–73].
PC = Surfactin was used as standard drug (positive control),
while DMSO was used as negative control.
4
. Conclusions
Table 8
Cytotoxicity against human tumor and normal cells, after 72 h of incubation.
Ferrocene-substituted thioureas have been synthesized in good
Sample Code
MCF-7
MCF-10A
yields and characterized by several physiochemical and spectroscopic
techniques. The crystallographic analysis of one of the new complexes
established the existence of secondary non-covalent interactions, which
led us to study their prospects for various biological targets. High
binding constant values and protein kinase inhibition activity reveal
that the inclusion of ferrocene and electronegative chlorine atoms can
enhance the therapeutic properties of thioureas by making the com-
plexes stable, polarized, and more lipophilic. The significant binding
strength and good protein kinase inhibition potential are parameters
that are pre-requisite for the design of new antitumor drugs.
Cisplatin
N1
1.55 ± 1.82
1.85 ± 1.01
3.22 ± 0.64
2.54 ± 1.07
4.43 ± 0.76
4.01 ± 0.96
4.97 ± 1.25
17.3 ± 0.59
16.8 ± 0.74
15.1 ± 1.62
16.0 ± 1.27
13.4 ± 0.95
14.2 ± 1.33
12.3 ± 0.88
N2
N3
N4
N5
N6
in cell death [66–69]. Helleday et al. have stated that “The control over
DNA can control the growth of cancerous cells” [70]. The antitumor
mechanism of the “drug” action of the ferrocenyl thioureas (N1–N6) is
Fig. 9. Cell viability (%) of cancerous and non-cancerous cells at various concentrations (0–100 µM) of (a) N1, (b) N3, and (c) N5 in ethanol, after 72 h of incubation.
17

F. Asghar et al.
Inorganica Chimica Acta 488 (2019) 8–18
Conflicts of interest
[26] V. Alagarsamy, S. Meena, K.V. Ramseshu, V.R. Solomon, K. Thirumurugan,
K. Dhanabal, M. Muruga, Eur. J. Med. Chem. 41 (2006) 1293.
[
[
27] T.K. Venkatachalam, F.M. Uckun, Synth. Commun. 35 (2005) 2039.
28] W.L. Armarego, C. Chai, Purification of Laboratory Chemicals, fifth ed.,
Butterworth-Heinemann, 2012.
There are no conflicts of interest to declare.
Acknowledgements
[
[
29] F. Asghar, A. Badshah, I.S. Butler, S. Tabassum, B. Lal, M.N. Tahir, Inorg. Chim.
Acta. 442 (2016) 46.
30] F. Asghar, A. Badshah, S. Fatima, S. Zubair, I.S. Butler, M.N. Tahir, J. Organomet.
Chem. 843 (2017) 48.
Faiza Asghar is highly obliged to the Higher Education Commission
(
HEC) of Pakistan for providing a scholarship under the program 5000
[31] F. Asghar, S. Fatima, S. Rana, A. Badshah, I.S. Butler, M.N. Tahir, Dalton Trans.
2
018 (1868) 47.
Indigenous PhD Scholarships at Quaid-i-Azam University, Islamabad
and for a 6-month scholarship under the International Research Support
Initiative Program (IRSIP) as a Graduate Research Trainee in the
Department of Chemistry, McGill University, Montreal, Quebec,
Canada.
[
32] S. Fatima, A. Badshah, B. Lal, F. Asghar, M.N. Tahir, A. Shah, I. Ullah, J. Organomet.
Chem. 819 (2016) 194.
[33] B. Lal, A. Badshah, A.A. Altaf, M.N. Tahir, S. Ullah, F. Huq, Dalton Trans. 41 (2012)
1
4643.
[
[
34] F. Asghar, A. Badshah, B. Lal, S. Zubair, S. Fatima, I.S. Butler, Bioorg. Chem. 72
(
2017) 215.
35] F. Asghar, A. Badshah, B. Lal, I.S. Butler, S. Tabassum, M.N. Tahir, Inorg. Chim.
Acta. 439 (2016) 82.
Appendix A. Supplementary data
[
[
36] G.M. Sheldrick, Acta Cryst. A64 (2008) 112.
37] F. Asghar, A. Badshah, A. Shah, M.K. Rauf, M.I. Ali, M.N. Tahir, E. Nosheen,
Z.U. Rehman, R. Qureshi, J. Organomet. Chem. 717 (2012) 1.
38] S. Zubair, F. Asghar, A. Badshah, R.A. Hussain, S. Tabassum, M.N. Tahir, J.
Organomet. Chem. 879 (2019) 60.
Supplementary data (Single crystal X-ray diffraction data for the
structural analysis has been deposited with the Cambridge
Crystallographic Data Center, CCDC No. 1527498. The copy of this
information may be obtained free of charge from deposit@ccdc.cam.ac.
uk or http://www.ccdc.cam.ac.uk.) to this article can be found online
at https://doi.org/10.1016/j.ica.2018.12.052.
[
[
39] M.K. Rauf, U. Shaheen, F. Asghar, A. Badshah, A. Nadhman, S. Azam, M.I. Ali,
G. Shahnaz, M. Yasinzai, Arch. Pharm. Chem. Life Sci. 348 (2015) 1.
40] R.L. Scott, Recl. Trav. Chim. Pays-Bas. 75 (1956) 787.
41] S.J. Smith, B.T. Sutcliffe, Rev. Comput. Chem. 70 (1997) 271.
42] F. Asghar, A. Badshah, R.A. Hussain, M. Sohail, K. Akbar, I.S. Butler, J. Organomet.
Chem. 797 (2015) 131.
[
[
[
References
[43] F. Asghar, S. Rana, S. Fatima, A. Badshah, B. Lal, I.S. Butler, New J. Chem. 42
(
2018) 7154.
[
44] G. Yao, F.M. Sebisubi, L.Y.C. Voo, C.C. Ho, G.T. Tan, L.C. Chang, J. Braz. Chem. Soc.
[
[
[
[
1] G. Jaouen, A. Vessieres, I.S. Butler, Acc. Chem. Res. 26 (1993) 36.
2
2 (2011) 1125.
2] U. Schatzschneider, N.M. Nolte, Angew. Chem. Int. Ed. 45 (2006) 1504.
3] M. Laskoski, W. Steffen, M.D. Smith, U.H. Bunz, Chem. Commun. 2001 (2001) 691.
4] N. Chavain, H. Vezin, D. Dive, N. Touati, J.F. Paul, E. Buisine, C. Biot, Mol. Pharm. 5
[45] T. Mosmann, J. Immunol. Methods. 65 (1983) 55.
[46] F. Javed, A.A. Altaf, A. Badshah, M.N. Tahir, M. Siddiq, Z.U. Rehman, A. Shah,
S. Ullah, B. Lal, J. Coord. Chem. 65 (2012) 969.
(
2008) 710.
[47] C. Colovos, T.O. Yeates, Protein Sci. 2 (1993) 1511.
[
[
5] J. Manosroi, K. Rueanto, K. Boonpisuttinant, W. Manosroi, C. Biot, J. Med. Chem.
[
[
48] M. Sirajuddin, S. Ali, A. Haider, N.A. Shah, A. Shah, M.R. Khan, Polyhedron 40
(2012) 19.
5
3 (2010) 3937.
6] C.L. Ferreira, C.B. Ewart, C.A. Barta, S. Little, V. Yardley, C. Martins, E. Polishchuk,
P.J. Smith, J.R. Moss, M. Merkel, M.J. Adam, C. Orvig, Inorg. Chem. 45 (2006) 659.
7] B. Long, S. Liang, D. Xin, Y. Yang, J. Xiang, Eur. J. Med. Chem. 44 (2009) 2572.
8] M.F.R. Fouda, M.M. Abd-elzaher, R.A. Abdelsamaia, A.A. Labib, Appl. Organomet.
Chem. 21 (2007) 613.
49] A. Shah, A.M. Khan, R. Qureshi, F.L. Ansari, M.F. Nazar, S.S. Shah, Int. J. Mol. Sci. 9
(
2008) 1424.
[
[
[50] Q. Feng, N.Q. Li, Y.Y. Jiang, Anal. Chim. Acta 344 (1997) 97.
51] M. Aslanoglu, G. Ayne, Anal. Bioanal. Chem. 380 (2004) 658.
[52] A.M.O. Brett, J. Piedade, L. Silva, V. Diculescu, Anal. Biochem. 332 (2004) 321.
[
[
9] N. Chavain, E.D. Charvet, X. Trivelli, L. Mbeki, M. Rottmann, R. Brun, C. Biot,
Bioorg. Med. Chem. 17 (2009) 8048.
[
53] A.M.O. Brett, V. Diculescu, J.A.P. Piedade, Bioelectrochemistry 55 (2002) 61.
[54] R. Fukuda, S. Takenaka, M. Takagi, J. Chem. Soc. Chem. Commun. (1990) 1028.
[55] C. Lee, W. Yang, R.G. Parr, Phys. Rev. B. 37 (1988) 785.
[
10] M. Soares, J. Oliveira, M. Santos, R. Ribeiro, P. Pigeon, G. Jaouen, A.E. Santana,
M. Goulart, E. Hillard, Drug Develop. Res. 71 (2010) 69.
[
56] Y. Ataly, D. Avci, A. Basoglu, J. Struct. Chem. 19 (2008) 239.
[
[
11] B. Maryanoff, S. Keeley, F. Persico, J. Med. Chem. 26 (1983) 226.
12] A. Mahajan, L. Kremer, S. Louw, Y. Gueradel, K. Chibale, C. Biot, Bioorg. Med.
Chem. Lett. 21 (2011) 2866.
[57] B. Dhokale, T. Jadhav, S.M. Mobin, R. Misra, RSC Adv. 5 (2015) 57692.
[
58] R. Misra, R. Maragani, T. Jadhav, S.M. Mobin, New J. Chem. 38 (2014) 1446.
[59] R. Misra, P. Gautam, J. Organomet. Chem. 776 (2015) 83.
[
[
13] B. Deka, T. Sarkar, S. Banerjee, A. Kumar, S. Mukherjee, S. Deka, K.K. Saikia,
A. Hussain, Dalton Trans. 46 (2017) 396.
[60] M. Marinov, S. Minchev, N. Stoyanov, G. Ivanova, M. Spassova, Croat. Chem. Acta
7
8 (2005) 9.
14] B. Deka, A. Bhattacharyya, S. Mukherjee, T. Sarkar, K. Soni, S. Banerjee, K.K. Saikia,
S. Deka, A. Hussain, Appl. Organometal. Chem. 32 (2018) e4287.
15] T. Sarkar, S. Banerjee, S. Mukherjee, A. Hussain, Dalton Trans. 45 (2016) 6424.
16] R. Misra, T. Jadhav, S.M. Mobin, Dalton Trans. 2014 (2013) 43.
17] T.S. Reddy, R. Maragani, B. Dhokale, S.M. Mobin, R. Misra, RSC Adv. 6 (2016)
[61] B.W.D. Andrade, S. Datta, S.R. Forrest, P. Djurovich, E. Polikarpov, M.E. Thompson,
Org. Electron. 6 (2005) 11.
[
[
[
[62] S. Chandrasekar, V. Balachandran, H.S. Evans, A. Latha, Spectrochim. Acta A. 143
(
2015) 136.
[63] A. Choudhary, R.T. Raines, Chembiochem 2011 (1801) 12.
7
746.
[64] S.A. Elsayed, B.J. Claude, I.S. Butler, S.I. Mostafa, J. Mol. Struct. 1028 (2012) 208.
[
[
[
18] B. Dhokale, T. Jadhav, Y. Patil, R. Misra, Dyes Pigments. 134 (2016) 164.
19] R. Maragani, P. Gautam, S.M. Mobin, R. Misra, Dalton Trans. 45 (2016) 4802.
20] R. Gul, A. Badshah, A. Khan, A. Junaid, M. Khawar Rauf, Spectrochim. Acta A. 117
[
65] H. Tayyem, F. Huq, J.Q. Yu, P. Beale, K. Fisher, ChemMedChem 3 (2008) 145.
[66] E.R. Jamieson, S.J. Lippard, Chem. Rev. 99 (1999) 2467.
[
67] R.A. Alderden, M.D. Hall, T.W. Hambley, J. Chem. Ed. 83 (2006) 728.
(
2014) 264.
[68] T.W. Hambley, J. Chem. Soc. Dalton Trans. (2001) 2711.
[
21] T. Stringer, D. Taylor, C. Kock, H. Guzgay, A. Au, S.H. An, B. Sanchez, R. O’Connor,
N. Patel, K.M. Land, P.J. Smith, D.T. Hendricks, T.J. Egan, G.S. Smith, Eur. J. Med.
Chem. 69 (2013) 90.
[69] R.J. Knox, F. Friedlos, D.A. Lydall, J.J. Roberts, Cancer Res. 1986 (1972) 46.
[
70] T. Helleday, J. Lo, D.C.V. Gent, B.P. Engelward, DNA Repair 6 (2007) 923.
[71] D. Osella, M. Ferrali, P. Zanello, F. Laschi, M. Fontani, C. Nervi, G. Cavigiolio, Inorg.
Chim. Acta 306 (2000) 42.
[
[
[
[
22] X. Du, E. Hansell, J.C. Engel, C.R. Caffrey, F.E. Cohen, J.H. McKerrow, Chem. Biol. 7
(
2000) 733.
[72] D. Osella, H. Mahboobi, D. Colangelo, G. Cavigiolio, A. Vessieres, G. Jaouen, Inorg.
Chim. Acta 2005 (1993) 358.
23] C. Sun, H. Huang, M. Feng, X. Shi, X. Zhang, P. Zhoub, Bioorg. Med. Chem. Lett. 16
2006) 162.
24] A. Saeed, U. Shaheen, A. Hameed, S.Z.H. Naqvi, J. Fluorine Chem. 130 (2009)
028.
25] J.S. Upadlgaya, P.K. Srivastava, J. Ind. Chem. Soc. 59 (1982) 767.
(
[73] S. Saeed, N. Rashid, P.G. Jones, M. Ali, R. Hussain, Eur. J. Med. Chem. 45 (2010)
1323.
1
18