New ferrocene-based 2-thio-imidazol-4-ones and their copper complexes. Synthesis and cytotoxicity

Article abstract of DOI:10.1039/C8DT03164A

Synthesis, characterization (HRMS, NMR, EPR, XANES, UV-Vis spectroscopy, and electrochemistry), DNA and BSA binding and in vitro biological screening of two new ferrocene-incorporated thiohydantoin derivatives (5 and 6) and their copper coordination compounds are reported. The ferrocene-based thiohydantoin derivatives were prepared by copper-catalyzed azide alkyne cycloaddition reactions between alkynyl ferrocenes and 5-(Z)-3-(2-azidoethyl)-2-(methylthio)-5-(pyridin-2-ylmethylene)-1H-imidazol-4H-one. Alkynyl ferrocenes necessary for these syntheses were prepared by new procedures. Intermolecular redox reactions between the ferrocene fragment and copper(+2) coordinated ions were studied by different methods to determine the mechanism and kinetic constants of redox processes. Ferrocene-containing imidazolones (5 and 6) and their copper complexes were also tested for their in vitro cytotoxic activity against MCF-7 and A-549 carcinoma cells, and also against the noncancerous cell line Hek-293. The results showed modest cytotoxicity against the subjected cancer cell line compared with cisplatin. The ability of the obtained compounds to cause DNA degradation and cell apoptosis was investigated, and the distribution of cytosol/pellets was studied by AAS.

Full text of DOI:10.1039/C8DT03164A

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
New ferrocene-based 2-thio-imidazol-4-ones
and their copper complexes. Synthesis and
cytotoxicity†
Cite this: Dalton Trans., 2018, 47,
17357
a,b
D. A. Guk,a O. O. Krasnovskaya,a N. S. Dashkova,a D. A. Skvortsov,
M. P. Rubtsova,a,b V. P. Dyadchenko,a E. S. Yudina,a M. A. Kosarev,a A. V. Soldatov,c
V. V. Shapovalov,c A. S. Semkina,d K. Y. Vlasova,a V. I. Pergushov,a R. R. Shafikov,
a
A. A. Andreeva,f M. Ya. Melnikov,a N. V. Zyk,a A. G. Majougaa,e,f and
a
E. K. Beloglazkina
*
Synthesis, characterization (HRMS, NMR, EPR, XANES, UV-Vis spectroscopy, and electrochemistry),
DNA and BSA binding and in vitro biological screening of two new ferrocene-incorporated thiohydan-
toin derivatives (5 and 6) and their copper coordination compounds are reported. The ferrocene-
based thiohydantoin derivatives were prepared by copper-catalyzed azide alkyne cycloaddition reac-
tions between alkynyl ferrocenes and 5-(Z)-3-(2-azidoethyl)-2-(methylthio)-5-(pyridin-2-ylmethyl-
ene)-1H-imidazol-4H-one. Alkynyl ferrocenes necessary for these syntheses were prepared by new
procedures. Intermolecular redox reactions between the ferrocene fragment and copper(+2) co-
ordinated ions were studied by different methods to determine the mechanism and kinetic constants
of redox processes. Ferrocene-containing imidazolones (5 and 6) and their copper complexes were
also tested for their in vitro cytotoxic activity against MCF-7 and A-549 carcinoma cells, and also
against the noncancerous cell line Hek-293. The results showed modest cytotoxicity against the sub-
jected cancer cell line compared with cisplatin. The ability of the obtained compounds to cause DNA
degradation and cell apoptosis was investigated, and the distribution of cytosol/pellets was studied
by AAS.
Received 3rd August 2018,
Accepted 13th November 2018
DOI: 10.1039/c8dt03164a
rsc.li/dalton
reduce their cytotoxic effect are often ineffective.1,2 Therefore,
it is important to search for new cytotoxic coordination com-
1. Introduction
Despite the fact that platinum-containing coordination com- pounds based on endogenous metals, which have lower
pounds are successfully used in clinical practice, attempts to general toxicity. The fact that transition metals such as iron,
cobalt, zinc and copper are natural participants of intracellular
metabolism makes the syntheses of metal complexes based on
them and investigation of their biological activity expedient. At
present, several coordination compounds of iron, cobalt3 and
aChemistry Department, Lomonosov Moscow State University, Leninskie Gory 1/3,
copper4 are at the first and second stages of preclinical trials
Moscow 119991, Russia
bCenter of Life Sciences, Skolkovo Institute of Science and Technology, Skolkovo,
as antitumor drugs.
Moscow, 143026, Russia
According to the Web of Science, a number of publications
cPhysics Department, Southern Federal University, Ulitsa Zorge, 5, Rostov 344000,
devoted to the development of antineoplastic drugs based on
Russia
copper continue to grow and exceed a number of similar publi-
cations devoted to other biogenic metals. Based on the early
preclinical studies of copper(II) coordination compounds with
N,N-bidentate heterocyclic ligands, such complexes are able to
intercalate DNA with subsequent cleavage5 and inhibit the
growth of tumor cells.6 Also, the concentration of copper ions
in cancer cells is higher than in healthy cells, which is necess-
dPirogov Russian National Research Medical University (RNRMU)
Ostrovitianov str. 1, Moscow, 117997, Russia
eNational University of Science and Technology MISIS, Leninskiy prospect 4,
Moscow 101000, Russia
fD. Mendeleev University of Chemical Technology of Russia, Miusskaya Ploshchad’ 9,
Moscow 125047, Russia
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8dt03164a
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17357–17366 | 17357

View Article Online
Paper
Dalton Transactions
ary for angiogenesis in malignant breast,7,8 prostate,9,10 large active ferrocenyl fragment would allow the synthesis of new
intestine,10 and lung11 tumors. Copper-containing complexes
classes of Cu(I) or Cu(II/I) coordination compounds with an
increased bioavailability, in comparison with non-modified
analogs. Thus, the purpose of this work was to synthesize
are inexpensive and have lower overall toxicity in comparison
with platinum analogues.12,13
ferrocenyl-containing ligands
5
and
and
8
6
and their copper
Most of the biologically active coordination compounds of
copper described in the literature are Cu(II) complexes, due to
the simplicity of the synthesis and the stability of such deriva-
tives. At the same time, it is well known that the penetration of
copper-containing therapeutic agents into a cell occurs when
copper is in valence Cu(I), but not in the case of Cu(II),14 which
requires the presence of reducing agents. For example,
ascorbic acid or glutathione (GSH) is able to reduce Cu(II) ions
to Cu(I) before they penetrate into the cell.15 Taking into
account the fact that the effective penetration of the drug
into the tumor cell leads to an increase in cytotoxicity and,
as a result, the effectiveness of the drug, the principal task
is the development of stable Cu(I)-containing cytotoxic
compounds.
coordination compounds
shown below:
7
whose structures are
Previous studies of our research group were devoted to the
derivatives of 2-thioimidazolones and 2-alkylthioimidazolin-4-
ones, and coordination compounds of Cu(II,I) based on them,
which showed an effective cytotoxic effect on tumor cells.16–18
In this paper, we first propose a new approach for the stabiliz-
ation of the +1 oxidation state of copper in such coordination
compounds. In order to stabilize the copper ion in the mono-
valent state, we have proposed to introduce a redox-active frag-
ment into the thiohydantoin ligand, which would make possi-
ble the intramolecular reduction of the copper ion from Cu(II)
to Cu(I). Ferrocene was chosen as a redox-active fragment due
to its biological activity and suitable electrochemical reduction
potential (E0Cp2Fe(II)/Cp2Fe(III) = + 0.4 V), which is close to the
reduction potential of copper(II) ions (E0Cu(II)\Cu(I) = + 0.153
0.39 V19). Thus, we have assumed the possibility of an intra-
molecular redox reaction Fe2+ + Cu2+ → Fe3+ + Cu1+. In
addition, ferrocene itself is a pharmacophore fragment; ferro-
cene-containing compounds exhibit the antitumor activity
against lymphocytic leukemia (P-388 cell line),20 lung carci-
noma (A549 cell line),21 mammary adenocarcinoma (Ca-75522
and the MCF-723 cell lines), and solid tumors.24 Also, literature
data suggest that the introduction of a ferrocene fragment into
the antitumor drug structure can lead to an increase of the
cytotoxic activity and, in some cases, the selectivity of the
therapeutic agents.25 So, one of the ferrocene-containing
derivatives demonstrated the activity exceeding that of cispla-
tin.26 In addition, ferrocene, due to its lipophilic nature, is
able to facilitate further penetration of the coordination
compound through the cell membrane. According to the data
from ref. 27, the introduction of a redox-active ferrocene-
containing derivative of natural amino acids into a structure
of the curcumin complex with divalent copper leads not
only to an increase of the cytotoxic activity of the conjugate
compared to individual starting materials, but also to the
accumulation of the ferrocene-containing conjugate in the
cytosol.
2. Results and discussion
Materials and methods and the details of synthetic procedures
for all mentioned substances are given in the ESI (ESI section
1†).
2.1. Synthesis
The synthesis of the target ligands 5 and 6 (Scheme 1) was
carried out using a copper-catalyzed azide–alkyne cyclo-
addition reaction (CuAAC) between the azide-containing imid-
azol-4-one 2 and the alkynyl-substituted ferrocene derivatives
3 and 4. 3-Azidoethyl-substituted 2-methylthio-imidazole-4-one
was chosen for the insertion of the copper ion chelating frag-
ment into the target ligand molecules. Ethynylferrocene 3 and
4-ferrocenylbutyn-1 4 were used as sources of the ferrocene
moiety; these molecules differ by the electronic characteristics
of the ferrocene nucleus: for the first compound, cyclopenta-
dienyl, the fragment is linked to the acceptor alkyne substitu-
ent, and for the second compound, it is linked to the donor
alkyl fragment. It should be noted that very few click reactions
with ferrocene derivatives are described in the literature.28
When the CuAAC reaction conditions were optimized, it was
found that in the organic solvents (dichloromethane, aceto-
nitrile, tetrahydrofuran, dimethylformamide) the reactions do
not proceed or proceed with the yield not exceeding 5%, and
also that the main reaction by-products for the reactions
carried out in an aqueous medium are the products of partial
and complete hydrolysis of the initial substances, intermedi-
ates and target molecules. Thus, in order to achieve optimal
yields, it is necessary to minimize the amount of water in the
reaction mixture. The optimal conditions for the azide–alkyne
cycloaddition reaction were the use of copper(II) sulfate
in the presence of a twofold excess of sodium ascorbate in
methylene chloride, with the addition of a few drops of
We have assumed that the modification of 3-alkyl-2- water to dissolve inorganic salts (see the experimental part in
thioxo-4H-tetrahydroimidazolin-4-one ligands with a redox- the ESI†).
17358 | Dalton Trans., 2018, 47, 17357–17366
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
solution of copper(II) perchlorate in acetonitrile in a molar
ratio of 1 : 1. When mixing the reagents, rapid dissolution of
the suspended ligand was observed and the formation of a
dark green solution of coordination compound 7 or 8, the
proposed structure of which is shown in Scheme 2.
Complexes 7 and 8 were characterized by MALDI and HRMS
methods, UV-vis, IR and XANES spectroscopy, as well as using
cyclic voltammetry data to determine the oxidation states and
local environment of metal ions. Broad bands near
1100 cm−1 and a sharp band near 625 cm−1 are observed in
the IR spectra of the complexes 7 and 8 (Fig. S18 and S20†),
which could be due to the antisymmetric stretching and
bending of perchlorate ions, respectively. No splitting pattern
of the perchlorate peak was observed. This is in favor of the
fact that the perchlorate ions are most likely not coordinated
with the metal ions and are present in the crystal lattice as
counterions.30,31
The data of reaction mixture monitoring by CV, UV-vis and
EPR spectroscopy show that the process of complexation pre-
cedes the reduction of copper and the end of the complexation
and the beginning of the oxidation–reduction reaction occur
within a time of ∼1 min from the mixing of the starting com-
pounds (section 2.2). The kinetics of the oxidation–reduction
process in the bimetallic “Fe2+–Cu2+” redox system was studied
in more detail by EPR (section 2.4) with the example of the
ligand 6 and CuCl2·2H2O interaction.
Scheme 1 Synthesis of ferrocene-based ligands 5 and 6.
Thus, bimetallic coordination compounds Fe3+/Cu+ of a
previously unknown structural type have been synthesized. At
the moment, as far as we know, only coordination compounds
Since ligands 5 and 6 contain redox-active ferrocene frag- of ferrocene derivatives with Cu(II) or Cu(I) in which intra-
ments in their structure, we assumed that they can reduce molecular redox processes do not occur are described in the
copper(II) to copper(I) during complexation with copper(II) salts literature.32,33 In contrast, the ferrocene-containing organic
to form ferrocenyl cations. Therefore, it was necessary to avoid ligands 5 and 6 described in this work in the complexation
carrying out the reaction in the presence of nucleophiles reactions efficiently reduce the coordinated copper(II) ions and
capable of interacting with ferrocenyl cations.
stabilize them in the valence state (+1).
+
It is known that ferrocenylium-cation FeCp2 is stable in
acetonitrile, acetone and nitromethane in the absence of
strong nucleophiles, but it is decomposed by Cl− and Br− (to
FeCp2 and FeHal4−, and in some cases to other products),
and also by solvents such as DMF, DMSO, and HMPTA, and
nitrogen-containing ligands PHEN and BIPY (to FeCp2 and
Fe2+ complex, octahedrally coordinated by the molecules of
the added ligand).29 It is believed that in the first stage, the
ligand exchange occurs at the Fe3+ atom, as a result of which
2.2 Cyclic voltammetry
Ligands 5 and 6 were investigated by cyclic voltammetry (CV)
and rotating disk electrode (RDE) techniques in DMF and
CH3CN solutions on a glassy carbon electrode in the presence
of 0.1 M Bu4NClO4 as an indifferent electrolyte. The obtained
results are summarized in Table 1.
−
the C5H5 anion is liberated, being the strongest reducing
+
agent and freeing up the formed FeCp2 and Fe3+ complex to
the final products. Since the ligand exchange and reduction
reactions occur almost simultaneously, and the transform-
ation of the C5H5 radicals is irreversible, in reality the
+
destruction of FeCp2 is also possible with nucleophiles that
are much weaker than C5H5−. Taking this into account, com-
plexation reactions of ligands 5 and 6 with Cu(II) was carried
out in acetonitrile, and Cu(ClO4)2 containing a non-nucleo-
philic perchlorate anion was chosen as the initial copper salt.
The best results of the complexation reactions were obtained
by mixing a slurry of an organic ligand in acetonitrile with a Scheme 2 Synthesis of complexes 7 and 8.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17357–17366 | 17359

View Article Online
Paper
Dalton Transactions
Table 1 Electrochemical potentials (vs. Ag/AgCl/KCl (aq. sat.)) of in-
organic copper salts, ligands 5 and 6 and their copper complexes in
CH3CN or DMF (0.1 M Bu4NClO4, glass-carbon electrode; potential scan
coordination compounds in the solution and is consistent
with the EPR data (see section 2.4). The second peak dis-
appears after a few minutes; the intensity of the peak at −0.48
V, related to the copper(I)-containing complex, in contrast
increases. In addition, the reduction of the ligand fragment
(E = −1.13 V) is facilitated by 270 mV compared with the free
ligand (Fig. 1). These changes indicate the formation of
complex 8, with the assumed structure shown in Scheme 2.
Thus, ligand 6 with a linker between the ferrocene fragment
and the triazole ring, according to the CV, forms an ultimate
copper complex more slowly than ligand 5, in which ferrocene
and triazole are directly linked. In addition, in the case of
ligand 6, it is possible to affirm the existence in CH3CN solu-
tion of an intermediate copper-containing compound that
turns into complex 8.
rate 100 mV s−1
)
Compound
Solvent
CH3CN
Epc, B
Epa, B
Cu(CH3CN)4ClO4
Cu(ClO4)2·6 H2O
−0.85/−0.42, −0.13
(desorption)
1.03/1.11
1.16/1.06
CH3CN
—
−0.68/−0.42, −0.19
(desorption)
0.11/1.23
CuCl2·2 H2O
5
DMF
0.95
−0.21
CH3CN
−1.28
0.52/0.47
1.24
1.68
−1.88
−2.01
1.85
6
CH3CN
1.30
−1.78
0.44/0.39
0.87
1.44
When the ligand 6 reacts with CuIICl2·2H2O in DMF poorly
soluble copper(I) salt (possibly CuICl) apparently forms and
precipitates on the surface of the electrode, since only low
intense peaks are observed in the reduction region in the CV
besides the Fc+ → Fc reduction peak (Fig. S4†). Apparently, in
the presence of chloride anions, the formed Fc+ is unstable
and decays, with the loss of iron, as described in ref. 37. Thus,
in this case, the ligand 6 and copper salt interaction seems to
proceed according to the sequence:
1.70
0.54/0.47
1.56
7
8
9
CH3CN
CH3CN
DMF
−0.47
−0.85/−0.17
(desorption)
−0.48
0.46/0.39
1.34
−1.13/−0.04
(desorption)
−0.76
0.54/0.46
−1.26/−1.23
6 þ CuIICl2 ! CuICl þ 6þ ! products of 6þ decomposition:
The oxidation of ligand 6 containing the donor alkyl substi-
tuent in the cyclopentadienyl ring proceeds reversibly at E1/2
The interaction of ligand 5 with CuIICl2·2H2O in DMF takes
place slowly; in contrast to all the other reactions studied, this
process is the only reaction with Cu(II) salts, where the color
change of the solution to blue-green occurs in 10–15 minutes
at room temperature. The complete disappearance of the Cu2+
→ Cu+ peak occurs only after ∼1.5 hours after the mixing of
the ligand and the salt. Thus, the reaction
=
0.42 V (CH3CN) over the ferrocene (Fc) fragment. Ligand 5 oxi-
dizes at a higher potential (E1/2 = 0.50 V), which is apparently
due to the accepting nature of the triazole moiety conjugated
to the cyclopentadienyl ring. The first stage of the reduction of
ligands 5 and 6 at E = 1.28–1.30 V occurs irreversibly, appar-
ently, over the conjugated pyridylmethyleneimidazolinone
fragment, as was previously observed for ligands, derivatives of
2-thiohidantoin.34,35
5 þ CuIICl2 ! ½5+ꢀCuICl:
The addition of Cu(II) salts (CuCl2·2H2O or Cu(ClO4)2·6H2O)
to an electrochemical cell, containing a solution of ligand 5 or
6, leads to significant changes in CV. In all cases, oxidation of
the ferrocene fragment to the ferrocenyl cation and the
reduction of Cu2+ to Cu1+ occur. The formation of the ferroce-
nyl cation is indicated by a change in the color of the solution
from light yellow (the color of the initial ligand) to blue-green
(the characteristic color of the ferrocenyl cation,36 Fig. S1†)
instantaneously or for several minutes, as well as by the
characteristic changes in RDE: the current of Fc+/Fc becomes
cathodic (Fig. S2†), which corresponds to the reduction
process.
in this case is slow. Note also that based on the results of
the electrochemical study the copper(II) reduction steps follow
its coordination with the ligand (Scheme 3); this is evidenced
For complex 7, forming from the ligand 5, according to the
CV data, Cu+ → Cu0 reduction occurs at −0.85 V, and the sub-
sequent reduction peaks of the ligand fragment (E = −1.13 V)
are shifted to the anode potential region (Fig. S3†). But when
the ligand 6 was mixed with CuII(ClO4)2 in CH3CN, on the first
scans of the CV curves in the reduction region, there are two
peaks corresponding to the Cu+ → Cu0 transitions at −0.48 V
Fig. 1 CV curves of a ligand 6 mixture with CuII(ClO4)2·6H2O (solid)
and the first CV curve after 6 and CuII(ClO4)2·6H2O were mixed (dotted
and −0.85 V, which indicates the formation of two different line) in CH3CN (10−4 M, GC electrode, 0.1 M Bu4NClO4).
17360 | Dalton Trans., 2018, 47, 17357–17366
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Scheme 3 The sequence of reduction and coordination of the copper
(II) ion by the ligand 5 based on the CV data.
by the absence of any peaks of the Cu(I) derivative reduction in
the voltammogram, except for the peaks of the final complex.
Note that for all the processes described in this section, a
catalytic increase in the oxidation current Fc → Fc+ was
observed on the reverse CV scans (Fig. 1 and S2–S4†), which
can be explained by the participation of the regenerated Cu2+
in the oxidation of ferrocene fragments (Fc → Fc+), which are
at these potentials in an electrochemically reduced state.
Thus, similar values of ferrocene fragment oxidation poten-
tials and CuII reduction potentials make it possible for elec-
tronic transition from the HOMO Fc to the LUMO Cu(II),
which confirms the hypothesis that the organic ligands 5 and
6 are able to effectively stabilize Cu(I) ions. Furthermore, the
stage of formation of the copper(II) complex with the ferro-
cene-based ligand precedes the stage of Cu(II)
reduction.
→ Cu(I)
2.3 XANES spectroscopy
For the coordination compound 7, XANES absorption spectra
for iron and copper were recorded. Comparing the XANES
spectrum of the K-edge of copper for this sample with the
spectra of the reference samples (Fig. 2, top), we can conclude
that there is a similarity in the local copper environment of
the investigated complex and copper(II) nitrate. However, in
the absorption edge region, the spectrum of complex 7 has an
appreciable feature, which is caused by the presence of
Cu1+ ions in the sample. It shoud be noted, that compound 7
gradually oxidized by air oxygen forming the copper(II) deriva-
tive if it is stored without an argon atmosphere.
By analyzing the XANES spectrum behind the K-edge of the
iron for this sample (Fig. 2, bottom), it can be seen that the
experimental curve lies between the spectra of the standard
reference samples. Despite the fact that Fe2+ ions are present
in the sample with high probability, the presence of Fe3+ is
also evident.
Thus, in view of the presence of Fe(III) ions and Cu(II) ions
in the sample of the coordination compound 7, it can be con-
cluded that even in a solid sample an oxidation–reduction
process Fe2+ + Cu2+ ⇔ Fe3+ + Cu1+ discussed in sections 2.1
and 2.2 takes place. Also, it can be assumed that the oxidation
of Cu(I) ions by the oxygen of the air leads to the observed
result. However, we assumed that under the reducing con-
ditions of the cellular environment this process will not occur.
Fig. 2 XANES spectra and their first derivatives behind the K-edge of
copper for the sample of complex
7 and reference compounds
Cu(NO3)2, Cu(CH3COO)2 and Cu2O (top) and α-Fe2O3, Fe3O4, and FeO
(bottom).
copper ions, we investigated the kinetics of redox reactions. As
soon as paramagnetic copper(II) ions are converted to diamag-
netic copper(I), electron-paramagnetic resonance spectroscopy
(EPR) was used.
EPR has several advantages that are significant in the
framework of this study, it gives additional structural infor-
mation on the symmetry of copper(II) ions’ coordination
environment.
Based on the cyclic voltammetry data (section 2.2), the reac-
tion of ligand 6 with CuIICl2 in DMF was chosen to study the
kinetics of the process due to the long enough reaction time
span.
A comparison of the EPR spectra of a mixture of ligand 6
with copper chloride dehydrate (Fig. 3) shows that the integral
intensity of paramagnetic Cu2+ ions expectedly gradually
decreases.
2.4 Kinetic studies of complexation and redox
transformations by EPR spectroscopy
At the same time, the shape of the EPR spectrum does not
change with time, meaning that the complexation occurs
To confirm the presence of the redox-processes with the par- immediately after mixing the ligand with copper chloride solu-
ticipation of ferrocene electronic switches and coordinated tion, before registration of the first EPR spectrum (Scheme 3).
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17357–17366 | 17361

View Article Online
Paper
Dalton Transactions
Scheme 4 The proposed paths of the complex 8 formation (based on
the EPR data).
Fig. 3 6 with an equivalent amount of copper(II) chloride dihydrate in
DMF over time. The arrows indicate the position of the components of
the hyperfine structure of the diamagnetically diluted solution of Mn2+
ions in MgO (the internal standard for the value of the magnetic field
induction).
Fig. 5 The kinetic curve corresponding to the consumption of para-
magnetic Cu2+ ions during the redox process in the solution of com-
pound 6 and its processing in coordinates C–t1/2
.
trations of 8a and 8b are associated with large errors intro-
duced in the simulated EPR spectra.
Fig. 4 The experimental EPR spectrum of the solution of coordination
compound 8, and the simulated EPR spectra of isomeric coordination
compounds 8a (pyramidal copper surroundings) and 8b (octahedral
copper surroundings).
EPR spectra do not undergo fundamental changes during
the redox process, meaning that there is an equilibrium
between isomers 8a and 8b in the reaction mixture.
Based on the EPR data, we can propose a hypothesis that
two isomeric intermediate copper(II) complexes 8a and 8b with
different coordination environments exist in the solution and
then both are converted into complex 8.
The experimental EPR spectra (Fig. 4) correlate with the
sum of the simulated spectra for the octahedrally coordinated
isomer 8a (∼40%) and the square pyramidally coordinated
isomer 8b (∼60%). The proposed structure of the isomeric
complexes is shown in Scheme 4.
2.5 In vitro biological studies
2.5.1 MTT assay. The MTT assay was conducted on MCF-7
The kinetic curves of Cu2+ ion concentration (C) decrease (human breast adenocarcinoma cell line), A549 (human lung
and its evolution (concentration vs. square root of reaction carcinoma), HEK293 (human embryonic kidney), and VA13
time) are shown in Fig. 5. It should be noted that in some (fissile lung fetal cells) cell lines in accordance with the stan-
cases this dependence is characteristic of diffusion-controlled dard MTT method with three repeats for averaging the results.
processes.
CC50 values of compounds 7 and 8 and the corresponding
The redox reactivity of 8a and 8b cannot be correctly esti- ligands 5 and 6 were compared to determine the role of copper
mated in the presence of compound 6, as individual concen- ions in toxicity. The cytotoxicity data for compounds 5–8 are
17362 | Dalton Trans., 2018, 47, 17357–17366
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Table 2 Cytotoxicity of compounds 3–6 compared to that of doxo-
rubicin and cisplatin
Table 3 Concentration stability constants of conjugates of coordi-
nation compounds 5 and 6 with bovine serum albumin, literature data
for doxorubicin are given for comparison
Compound
CC50 HEK293 CC50 MCF7 CC50 A549 CC50 Va-13
Coordination
compound
Number of binding
sites
5, µM
6, µM
7, µM
8, µM
>
∼3
16
>
>
>
11
Kapp (BSA)
∼11
17.7
∼33
5.5 0.4
28.1 2.4 21.1 1.8
54.7 42.7 2.7
2
(1.47 0.04) × 104
(4.5 0.3) × 105
(4.85 0.05) × 105
(9.2 0.5) × 101
(7.8 0.7) × 103
0.98
1.15
1.16
0.54
—
2
2
5
6
7
8
28.7 3.5
7
Doxorubicin, nM 11.5 3.2
Cisplatin, µM
(ref. 17)
55.4 11.8 47.9 7.9 159.9 27.4
64.1 3.9 >30 2.9 0.3
12.4 3.9
Doxorubicin
given in Table 2. All the tested derivatives have a cytotoxicity
comparable to or higher than cisplatin and doxorubicin.
2.5.3 Inhibition of telomerase activity. Since the first years
of telomerase research, the enzyme has been considered a uni-
Ferrocene derivative 6a exhibits a noticeable cytotoxicity. versal target for anticancer therapy. Recent studies have also
The introduction of a (CH2)2 linker between the ferrocene and considered the telomerase activity as an important potential
triazole moieties of the ligand leads to a cytotoxicity increase, target in drug development and anti-cancer therapy.43
as seen from the cytotoxicity of compounds 5 and 6. Typically,
Earlier, we have described16 the coordination compounds
the cytotoxicity of metal–thiohydantoin complexes is defined of copper(II,I) based on 2-alkylthioimidazolin-4-ones with
by a metal.38 Surprisingly, the introduction of the copper ion promising cytotoxicity, wherein the mechanism of cytotoxic
into the ferrocene ligand 6 results in cytotoxicity reduction, action was associated with the inhibition of telomerase
which we associate with the planar molecular geometry activity. In this research work, we have designed new Cu(I)/
change and, consequently, its inability to bind to the target in Fe(III) coordination compounds with a pronounced cytotoxic
the cell.
activity against tumor cells (section 2.5.1). It was assumed that
2.5.2 BSA binding. The interaction between transport pro- the cytotoxic activity of the ferrocene-containing ligands 5 and
teins and various drugs is crucial for a wide range of pharma- 6 and binuclear Cu(I)/Fe(III) coordination compounds 7 and 8
cological, biological, and clinical applications.39 The under- could also be due to the inhibition of telomerase activity. The
standing of the binding of drugs to serum albumin in vitro can effects of compounds 5–8 (20/200 µM) and BIBR (50 µM) on
help in providing guidelines for rational drug design.40 Serum the telomerase activity from HEK293 T cells were evaluated via
albumin is the most abundant of the proteins in blood RQ-TRAP.
plasma, and it serves as a transport protein for drugs.
In the case of the classical inhibitor of telomerase activity,
For the investigation of the BSA binding ability of 7 and 8 BIBR (50 μm), the telomerase activity is only reduced when the
and predicting their stability under physiological conditions of inhibitor is added to the telomerase reaction, and in the case
the blood, a standard experiment using fluorescence quench- of the inhibitor addition after the reaction, the telomerase
ing was used.
activity persists before PCR. The data obtained for ligands 5
Fig. S6 and S7† show the plots of fluorescence quenching and 6 and coordination compounds 7 and 8 show that only
of bovine serum albumin in response to an increasing concen- ligand 6 exhibits a significant inhibition of telomerase activity
tration of compounds 7 and 8. The steady decrease in the (35 19% telomerase activity and 89 51% PCR control in the
intensity of fluorescence in response to an increase in the con- presence of 20 µM of 6; see Fig. S8 in the ESI†). It is important
centration of coordination compounds allows us to conclude to note that the ability to inhibit the telomerase enzyme is cor-
that all of the tested compounds bind to bovine serum related with cytotoxicity; reducing the length of the linker and
albumin. To quantify the binding by the Sketchard method, introducing a copper ion into the ferrocene ligand 6 reduce
the obtained dependences of the decrease in fluorescence the ability to inhibit the telomerase enzyme and cytotoxicity,
intensity on the concentration of coordination compounds respectively.
were linearized in double inverse logarithmic coordinates. The
linear graphs of dependencies by Sketchard are presented in agent, which is
Hereby, we have designed a ferrocene-containing cytotoxic
ferrocene-containing conjugate of
a
Fig. S7 in the ESI.†
2-alkylthioimidazolin-4-one.
The obtained values are shown in Table 3. The number of
2.5.4 DNA cleavage assay. We have studied the ability of
binding sites and constants, measured in the experiment, is compounds 7 and 8 to intercalate DNA with fluorescence emis-
well consistent with the values obtained earlier for analogous sion using ethidium bromide and have shown that both the
copper coordination compounds and the literature data41 that complexes are not DNA intercalators (see the ESI†).
one copper binding site is present in the molecules of bovine
The half-life of DNA by spontaneous hydrolysis will be thou-
and human serum albumin. Thus, compounds 5–7 have sands to billions of years.44 However, it could be hydrolyzed
sufficiently good affinities for BSA and can be transported with the help of artificial nucleases or be damaged by reactive
through the bloodstream due to the interaction with blood oxygen species (ROS). Many transition metal complexes were
transport proteins.39,42
used as DNA cleavage agents. The coordination compounds of
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17357–17366 | 17363

View Article Online
Paper
Dalton Transactions
Table 4 The results of pUC18 DNA cleavage catalyzed by 0.250 mM of
different cleavage agents for 4 h
Table 5 Cytosol/pellet distribution of coordination compounds 7 and 8
investigated by the AAS method
Cleavage agent
LC DNA (%)
NC DNA (%)
SC DNA (%)
Cytosol
Pellet
CuCl2
Control (without
treatment)
5
6
7
8
0
7
25
1
75
92
7
8
% of the dose added to the cells
% of the amount of the drug
penetrated into the cell
% of the dose added to the cells
% of the amount of the drug
penetrated into the cell
11.18 2.65
75.92 7.53 24.08 7.53
3.45 0.61
18
5
9
18
31
75
60
64
64
16
32
11.10 0.23 15.68 5.79
42.44 9.67 57.56 9.67
8
Table 6 The results of apoptosis test catalyzed by ligand 6 and coordi-
nation compound 8
copper with nuclease activity are widely described.45 Due to
the biologically accessible redox potential, Cu complexes belong
to the most frequently studied artificial metallonucleases. We
have investigated the nuclease activity of compounds 5–8 on the
substrate, supercoiled pUC18 plasmid DNA, by gel electrophor-
esis on agarose. The DNA cleavage ability was obtained from
the conversion amount of supercoiled DNA (SC) to nicked circu-
lar form (NC) and linearized DNA (LC).
C, μM
%, apoptotic cells
8
8
6
6
50
100
5
15.9
19.4
33.1
43.4
16.6
20
DMSO
To be sure of the nuclease active point in ferrocene-contain-
ing ligands 5 and 6 or coordination compounds 7 and 8, the
cleavage reaction catalyzed by compounds 5–8 was compared
with those reactions catalyzed by an equimolar amount of CuCl2.
The results are presented in Table 4 and in Fig. S8.† The
nuclease activity of complexes 7 and 8 was not the simple sum-
mation of the activities of free ligands 5 and 6 and copper(II) salt.
Fig. 6 Flow cytometry analysis in A549 cells after ligand 6 and complex 8 treatment. A549 cells treated with 50 μM of complex 8 (A), 100 μM of
complex 8 (B), 5 μM of ligand 6 (C) and 20 μM of ligand 6 (D).
17364 | Dalton Trans., 2018, 47, 17357–17366
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
The addition of increasing amounts of ligands 5 and 6 to studying their interaction with copper salts. The possibility of
pUC18 DNA does not lead to a significant change in the DNA the intramolecular redox reaction of the obtained ferrocene-
pattern. In the case of the attachment to the equimolar amount substituted imidazolones with copper(II) chloride and per-
of DNA of the coordination compounds 7 and 8, multiple chlorate has been shown. The preliminary in vitro biological
double-stranded DNA breaks are observed. Thus, Cu(I),(I) ions studies of the obtained ferrocene-substituted ligands and their
play an important role in the scission ability of the complexes.
2.5.5 Cytosol/pellet distribution investigated by AAS. This
work is devoted to the development of high-permeability drugs
that can effectively overcome the cell membrane. We have investi-
gated the cytosol/pellet distribution of coordination compounds 7
and 8. Mammary carcinoma cells 4T1 were incubated with coordi-
nation compounds 7 and 8, and the accumulation of copper in
the pellet and cytosol was determined by atomic adsorption spec-
troscopy (AAS). The results are presented in Table 5.
Cellular accumulation alone is insufficient to enable any
activity; any potential therapeutic agent must also be able to
reach its target within the cell. As can be seen from the data
obtained, the length of the linker significantly affects the
ability of the coordination compound to accumulate in pellet.
The coordination compound 8 has a lower cytotoxicity than its
analogue 7, however, it is predominantly localized in the
pellet, which provides the possibility of its effect on DNA-oper-
ating enzymes, DNA cleavage and mitochondria.
Fe, Cu-containing coordination compounds were carried out.
Author contributions
D. A. Guk, N. S. Dashkova, O. Krasnovskaya, V. P. Dyadchenko,
E. S. Yudina, M. A. Kosarev – synthesis of ligands and
complexes. D. A. Skvortsov, M. P. Rubcova, A. S. Semkina,
K. Y. Vlasova, R. R. Shafikov, A. A. Andreeva – biological
research. V. I. Pergushov, M. Ya. Melnikov
–
EPR
spectroscopy. A. V. Soldatov, V. V. Shapovalov – XANES
spectroscopy. N. V. Zyk, A. G. Majouga, E. K. Beloglazkina –
spectral studies, preparation of an article for publication.
Conflicts of interest
There are no conflicts of interest to declare.
2.5.6 Flow cytometry-based apoptosis detection. Apoptosis
is a form of programmed cell death that results in the orderly
and efficient removal of damaged cells, such as those resulting
from DNA damage or during development. Apoptosis can be
triggered by signals from within the cell, such as genotoxic
stress, or by extrinsic signals, such as the binding of ligands to
cell surface death receptors.46 The cytotoxic activity of a large
number of antitumor drugs occurs due to cellular apopto-
sis.47,48 In order to test whether the cytotoxic activity of ligand
6 and copper-containing complex 8 is based on the induction
of cell apoptosis, flow cytometry using the FITC Annexin V
Apoptosis Detection Kit protocol was conducted. A549 cells
were incubated with the investigated compounds 6 and 8 for
24 h. The results are presented in Table 6. The populations P3
and P4 show the percentage of apoptotic cells (double positive
cells). The treatment of cells with ligand 6 (Fig. 6C and D)
results in the accumulation of two populations of apoptotic
cells with different efficiencies of PI-staining. The intensity of
PI-staining demonstrates the phase of apoptosis: more inten-
sive staining corresponds to the late phase.
Acknowledgements
The authors are thankful to Thermo Fisher Scientific Inc.,
Textronica AG group (Moscow, Russia), and personally to Prof.
Alexander Makarov for providing the Orbitrap Elite mass
spectrometer for this work. This work in part of NMR study
was supported by the M.V. Lomonosov Moscow State
University Program of Development. We are grateful to the
Russian Foundation of Basic Research (Projects 16-03-00921
and 18-29-08060) for financial support.
Notes and references
1 G. J. Dougbartey, L. J. Peppone and I. A. de Graaf, Toxicol.,
2016, 371, 58–66.
2 J. Kuduk-Jaworska and J. J. Jański, J. Inorg. Biochem., 2017,
170, 148–159.
3 C. R. Munteanu and K. Suntharalingam, Dalton Trans.,
2015, 44(31), 13796–13808.
4 J.-X. Liang and H.-J. Zhong, J. Inorg. Biochem., 2017, 177,
276–286.
As can be seen from the data obtained, ligand 6 induces
cell death along the apoptotic pathway. Unlikethis, the coordi-
nation compound 8, even at concentrations an order of magni-
tude greater than the ligand, does not show traces of early
apoptosis. Since the ability to cause apoptosis correlates with
the cytotoxicity data, it can be assumed that the antitumor
activity of ligand 6 is associated with apoptotic cell death.
5 M. Chikira, C. Hee Ng and M. Palaniandavar, Int. J. Mol.
Sci., 2015, 16, 22754–22780.
6 C. Santini, M. Pellei, V. Gandin, M. Porchia, F. Tisato and
C. Marzano, Chem. Rev., 2014, 114, 815–862.
7 H. W. Kuo, S. F. Chen, C. C. Wu, D. R. Chen and J. H. Lee,
Biol. Trace Elem. Res., 2002, 89(1), 1–11.
8 S. L. Rizk and H. H. Sky-Peck, Cancer Res., 1984, 44(11),
5390–5394.
3. Conclusions
In conclusion, the ferrocene-containing derivatives of 2-thio-
hydantoins were firstly synthesized as redox active ligands for
9 F. K. Habib, T. C. Dembinski and S. R. Stitch, Clin. Chim.
Acta, 1980, 104(3), 329–335.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 17357–17366 | 17365

View Article Online
Paper
Dalton Transactions
10 S. B. Nayak, V. R. Bhat, D. Upadhyay and S. L. Udupa, 28 R. Prins, A. R. Korswagen and A. G. T. G. Kortbeek,
Indian J. Physiol. Pharmacol., 2003, 47(1), 108–110. J. Organomet. Chem., 1972, 39, 335–344.
11 M. Díez, M. Arroyo, F. J. Cerdàn, M. Muñoz, M. A. Martin 29 M. Zábranský, L. Kolářová, I. Císařová and P. Štěpnička,
and J. L. Balibrea, Oncology, 1989, 46(4), 230–234. J. Organomet. Chem., 2017, 846, 217–222.
12 D. C. Brady, M. S. Crowe, M. L. Turski, G. A. Hobbs, X. Yao, 30 S. Anbu, S. Kamalraj, B. Varghese, J. Muthumary and
A. Chaikuad, S. Knapp, K. Xiao, S. L. Campbell, D. J. Thiele
and C. M. Counter, Nature, 2014, 509, 492–496.
13 P. Sathyadevi, P. Krishnamoorthy, R. R. Butorac,
A. H. Cowley and N. Dharmaraj, Metallomics, 2012, 4, 498–
511.
M. Kandaswamy, Inorg. Chem., 2012, 51, 5580–5592.
31 S. Anbu, M. Kandaswamy, P. Sathya Moorthy,
M. Balasubramanian and M. N. Ponnuswamy, Polyhedron,
2009, 28, 49–56.
32 M. Shabbir, Z. Akhter, I. Ahmad, S. Ahmed, M. Bolte,
H. Ismail and B. Mirza, Inorg. Chim. Acta, 2017, 463, 102–
111.
14 M. Chikira, C. Hee Ng and M. Palaniandavar, Int. J. Mol.
Sci., 2015, 16, 22754–22780.
15 C. E. Paulsen and K. S. Carroll, Chem. Rev., 2013, 113, 33 E. K. Beloglazkina, A. G. Majouga, A. A. Moiseeva,
4633–4679.
M. G. Tsepkov and N. V. Zyk, Russ. Chem. Bull., 2007, 56(2),
351–355.
16 A. G. Majouga, M. I. Zvereva, M. P. Rubtsova,
D. A. Skvortsov, A. V. Mironov, D. M. Azhibek, 34 A. G. Majouga, E. K. Beloglazkina, A. A. Moiseeva,
O. O. Krasnovskaya, V. M. Gerasimov, A. V. Udina,
N. I. Vorozhtsov, E. K. Beloglazkina, A. Leonid,
L. V. Mikhina, A. V. Tretyakova, N. V. Zyk, N. S. Zefirov,
O. V. Shilova, E. A. Manzheliy, M. A. Lebedeva, E. S. Davies,
A. N. Khlobystov and N. V. Zyk, Dalton Trans., 2013, 42,
6290–6293.
A. V. Kabanov and O. A. Dontsova, J. Med. Chem., 2014, 57, 35 M. Cuartero, R. G. Acres, J. Bradley, Z. Jarolimova, L. Wang,
6252–6258.
E. Bakker, G. A. Crespo and R. De Marco, Electrochim. Acta,
2017, 238, 357–367.
17 K. Tishchenko, E. Beloglazkina, M. Proskurnin,
V. Malinnikov, D. Guk, M. Muratova, O. Krasnovskaya, 36 P. Zanello, G. Opromolla, L. Pardi, K. H. Pannell and
A. Udina, D. Skvortsov, R. R. Shafikov, Y. Ivanenkov, H. K. Sharma, J. Organomet. Chem., 1993, 450, 193–196.
V. Aladinskiy, I. Sorokin, O. Gromov, A. Majouga and 37 O. O. Krasnovskaya, Y. V. Fedorov, V. M. Gerasimov,
N. Zyk, J. Inorg. Biochem., 2017, 175, 190–197.
18 E. K. Beloglazkina, O. O. Krasnovskaya, D. A. Guk,
V. A. Tafeenko, A. A. Moiseeva, N. V. Zyk and
A. G. Majouga, Polyhedron, 2018, 148, 129–137.
19 D. X. West and A. E. Liberta, Coord. Chem. Rev., 1993, 123,
49–71.
D. A. Skvortsov, A. A. Moiseeva, A. V. Mironov,
E. K. Beloglazkina, N. V. Zyk and A. G. Majouga, Arabian
J. Chem., 2016, DOI: 10.1016/j.arabjc.2016.04.013.
38 A. G. Majouga, E. K. Beloglazkina, A. A. Moiseeva,
O. V. Shilova, E. A. Manzheliy, M. A. Lebedeva, E. S. Davies,
A. N. Khlobystov and N. V. Zyk, Dalton Trans., 2013, 42,
6290–6293.
20 Á Mooney, A. J. Corry, C. N. Ruairc, T. Mahgoub,
D. O’Sullivan, N. O’Donovan, J. Crown, S. Varughese, 39 S. Sangeetha and M. Murali, Int. J. Biol. Macromol., 2018,
S. M. Draper and D. K. Rai, Dalton Trans., 2010, 39, 8228–
8239.
21 B. Long, S. Liang, D. Xin, Y. Yang and J. Xiang, Eur. J. Med.
Chem., 2009, 44, 2572–2576.
22 L. V. Snegur, Y. S. Nekrasov, N. S. Sergeeva, Z. V. Zhilina,
V. V. Gumenyuk, Z. A. Starikova, A. A. Simenel,
107, 2501–2511.
40 Y. Kawasaki and E. Friere, Drug Discovery Today, 2011,
16(21–22), 985–990.
41 M. Anjomshoa, M. Torkzadeh-Mahani, J. Janczak,
C. Rizzoli, M. Sahihi, F. Ataei and M. Dehkhodaei,
Polyhedron, 2016, 119, 23–38.
N. B. Morozova, I. K. Sviridova and V. N. Babin, Appl. 42 D. Agudelo, P. Bourassa, J. Bruneau, G. Bérubé, É Asselin,
Organomet. Chem., 2008, 22, 139–147. H.-A. Tajmir-Riahi and D. S. Sem, PLoS One, 2012, 7, 8.
23 A. P. Ferreira, J. L. F. da Silva, M. T. Duarte, M. F. M. da 43 Q. Qi-Pin, M. Ting, T. Ming-Xiong, L. Yan-Cheng and
Piedade, M. P. Robalo, S. G. Harjivan, C. Marzano, L. Hong, Eur. J. Med. Chem., 2018, 143, 1597–1603.
V. Gandin and M. M. Marques, Organometallics, 2009, 28, 44 G. K. Schroeder, C. Lad, P. Wyman, N. H. Williams and
5412–5423.
R. Wolfenden, Proc. Natl. Acad. Sci. U. S. A., 2006, 103,
4052–4055.
24 R. A. Hussain, A. Badshah, J. M. Pezzuto, N. Ahmed,
T. P. Kondratyuk and E.-J. Park, J. Photochem. Photobiol., B, 45 H. Yu, L. Qiao-Sen, Z. Ji, Z. Zhong-Wei and Y. Xiao-Qi,
2015, 148, 197–208. Bioorg. Med. Chem., 2008, 16(3), 1103–1110.
25 R. Schobert, S. Knauer, S. Seibt and B. Biersack, Curr. Med. 46 G. Pistritto, D. Trisciuoglio, C. Ceci, A. Garufi and
Chem., 2011, 18, 790–807. G. D’Orazi, Aging, 2016, 8(4), 603–619.
26 A. Nguyen, S. Top, A. Vessieres, P. Pigeon, M. Huche, 47 R. S. Y. Wong, J. Exp. Clin. Cancer Res., 2011, 30(87), DOI:
E. A. Hillard and G. Jaouen, J. Organomet. Chem., 2007, 692, 10.1186/1756-9966-30-87.
1219–1225; D. R. K. Weerasuriya, W. P. S. L. Wijesinghe 48 J. B. Foo, M. L. Low, J. H. Lim, Y. Z. Lor, A. R. Zainol, V. Eh
and R. M. G. Rajapakse, Mater. Sci. Eng., 2017, 71, 206–213.
27 T. K. Goswami, S. Gadadhar, B. Gole, A. A. Karande and
A. R. Chakravarty, Eur. J. Med. Chem., 2013, 63, 800–810.
Dam, N. Abdul Rahman, C. Y. Beh, L. C. Chan, C. W. How,
Y. S. Tor and L. Saiful Yazan, BioMetals, 2018, 31(4),
505–515.
17366 | Dalton Trans., 2018, 47, 17357–17366
This journal is © The Royal Society of Chemistry 2018