Synthesis, chemical characterization and cancer cell growth-inhibitory activities of Cu(ii) and Ru(iii) aliphatic and aromatic dithiocarbamato complexes

Article abstract of DOI:10.1039/c8dt02965b

In this paper, we focused on the analysis of the effects mediated by different cyclic dithiocarbamic ligands (DTC) on three classes of antiproliferative coordination compounds, namely, Ru(iii) complexes with the general formulae [Ru(DTC)₃] and [Ru₂(DTC)₅]Cl, and the neutral Cu(ii) derivatives of the type [Cu(DTC)₂]. In particular, we present the synthesis and characterization of a library of total 23 coordination compounds containing Ru(iii) or Cu(ii) as the biologically-active metal center and two or more dithiocarbamato (DTC) ligands derived from cyclic amines (aliphatic or aromatic). Several techniques including elemental analysis, X-ray crystallography, ESI-MS, ¹H-NMR spectroscopy, FT-IR and UV-Vis spectrophotometry were used to characterize the compounds, which highlighted the different electronic behaviors generated by the substituents within the DTC moiety. Moreover, the synthesized compounds were tested for their stability in order to investigate their antiproliferative activity against 3 human cancer cell lines, namely, HeLa, HepG2 and HepG2/SB3. In particular, HepG2/SB3 was chosen for its aggressiveness due to upregulation of the anti-apoptotic protein SerpinB3. Finally, the most promising compounds are studied in terms of log?P. Overall, the results reveal the drug-likeness of some of the derivatives of copper(ii) and ruthenium(iii).

Full text of DOI:10.1039/c8dt02965b

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Synthesis, chemical characterization and cancer
cell growth-inhibitory activities of Cu(II) and
Ru(III) aliphatic and aromatic dithiocarbamato
complexes†
Cite this: Dalton Trans., 2018, 47,
15477
a,b
a
a
a
c
d
L. Brustolin,‡ C. Nardon,‡ N. Pettenuzzo, N. Zuin Fantoni, S. Quarta, F. Chiara,
d
d
e
c
a
A. Gambalunga, A. Trevisan, L. Marchiò, P. Pontisso and D. Fregona
*
In this paper, we focused on the analysis of the effects mediated by different cyclic dithiocarbamic
ligands (DTC) on three classes of antiproliferative coordination compounds, namely, Ru(III) complexes
3 2 5
with the general formulae [Ru(DTC) ] and [Ru (DTC) ]Cl, and the neutral Cu(II) derivatives of the type
[
Cu(DTC) ]. In particular, we present the synthesis and characterization of a library of total 23 coordination
2
compounds containing Ru(III) or Cu(II) as the biologically-active metal center and two or more dithio-
carbamato (DTC) ligands derived from cyclic amines (aliphatic or aromatic). Several techniques including
1
elemental analysis, X-ray crystallography, ESI-MS, H-NMR spectroscopy, FT-IR and UV-Vis spectrophoto-
metry were used to characterize the compounds, which highlighted the different electronic behaviors
generated by the substituents within the DTC moiety. Moreover, the synthesized compounds were
tested for their stability in order to investigate their antiproliferative activity against 3 human cancer cell
lines, namely, HeLa, HepG2 and HepG2/SB3. In particular, HepG2/SB3 was chosen for its aggressiveness
due to upregulation of the anti-apoptotic protein SerpinB3. Finally, the most promising compounds are
studied in terms of log P. Overall, the results reveal the drug-likeness of some of the derivatives of
copper(II) and ruthenium(III).
Received 20th July 2018,
Accepted 29th September 2018
DOI: 10.1039/c8dt02965b
rsc.li/dalton
gress has resulted in a significant number of clinically-estab-
lished entities (or in clinical trials), which are compounds
Since the beginning of the twentieth century, a number of bio- derived from both essential and non-essential metal elements.
Introduction
5,6
1
–3
logical processes involving metal ions have been identified,
Undoubtedly, the milestone of this class of therapeutics is cispla-
th
and this year is the 56 anniversary of the first “Metals in tin (CisDDP, cis-diamminodichloridoplatinum(II)), a chemothera-
7
Biology” Gordon Research Conference (1962), which recog- peutic agent widely used in the treatment of solid tumors.
4
nizes bioinorganic chemistry as a distinct discipline. This pro- Moreover, particular attention has been paid to the affinity of the
platinum(II) center towards soft bases, such as sulfur-contain-
ing biomolecules, which is associated with both toxicity and
a
Department of Chemical Sciences, University of Padova, Via F. Marzolo 1,
8–10
resistance to the treatment.
In this context, Borch et al.
3
5131 Padova, Italy. E-mail: dolores.fregona@unipd.it
demonstrated that the administration of a sulfur-based chelat-
ing agent (i.e., a dithiocarbamate, DTC) favors the selective
removal of the platinum center from the thiol groups of
b
Department of Surgical, Oncologic and Gastroenterological Sciences,
University of Padova, Via Giustiniani 2, 35128 Padova, Italy
c
Department of Medicine, University of Padova, Via Giustiniani 2, 35128 Padova,
Italy
different proteins (thus acting as a chemoprotectant) without
d
Department of Cardiac, Thoracic, and Vascular Sciences, University of Padova,
11,12
reversing the cytotoxic Pt-DNA adducts.
Subsequently, the
Via Giustiniani 2, 35128 Padova, Italy
next step was the exploration of the antiblastic features of
almost all transition elements to take advantage of their
e
Department of Chemical, Life and Sustainability Sciences, Parco Area delle Scienze
11/a, 43124 Parma, Italy
†
Electronic supplementary information (ESI) available: The synthetic and instru- different coordination geometries, oxidation states and redox
1
3
mental procedures, along with the biological techniques for the in vitro screen- properties.
1
ings. Moreover, the tables related to the diagnostic peaks in the IR, H-NMR,
Among the inorganic derivatives studied worldwide as anti-
cancer agents in the last few decades, metal dithiocarbamates
stand out due to their promising activity/toxicity ratio.
and UV-Vis spectra of all the synthesized complexes are provided, along with the
experimental spectra. CCDC 1845778 and 1845779. For ESI and crystallographic
data in CIF or other electronic format see DOI: 10.1039/c8dt02965b
1
4–16
‡
These authors contributed equally.
These coordination compounds, which were introduced in
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15477–15486 | 15477

View Article Online
Paper
Dalton Transactions
Fig. 1 Two limit forms of the DTC ligand (A), where on the left the nitrogen electron density is shifted to the N–CSS moiety, generating a
π-basic ligand (thioureidic form); on the other side, in form II the nitrogen lone pair is hampered from delocalization, thus allowing a metal
back-donation on the π* orbitals located on the sulfur atoms. Aliphatic ligands a–f (B) theoretically prefer form I, whereas aromatics g–i (C) belong
to form II.
1
7
oncological research by Fregona et al. in 2001, combine a meric effect), thus endowing the N–CSS bond with single-bond
32
biologically-reactive metal center with the chemoprotective character (Fig. 1A, form II). Therefore, the ligand shows
action of one or more dithiocarbamato ligands in a single π-acceptor properties due to its π* orbitals arising from the
small molecule. The most relevant results in terms of in vitro low-energy d-orbitals of the sulfur atoms, characterized by lower
cytotoxicity have been obtained with Au(III)–, Cu(II) and Ru(III)– electron density when compared with the aliphatic derivatives.
dithiocarbamato derivatives, which also exhibit excellent
Based on these considerations, the aim of this study was
1
8–20
in vivo results against orphan tumors.
Moreover, besides the synthesis and physico-chemical characterization of a
the metal center, the biological activity is modulated by library of total 23 Cu(II)– and Ru(III)–dithiocarbamato com-
ligands, which are crucial for stabilizing the metal oxidation plexes with ligands derived from both aliphatic amines
state and for the hydrophobic/hydrophilic features of the with different degrees of inductive effect (i.e., pyrrolidine (a),
2
1
coordination compound. In this regard, complexes involving piperidine (b), morpholine (c), indoline (d), L-proline methyl-
the pyrrolidine-dithiocarbamato (PDT) ligand, for instance (e) and tert-butyl ester (f), Fig. 1B) and aromatic amines (i.e.,
III
22
II
23
[
Au X
2
(PDT)] (X = Cl and Br), [Cu (PDT)
2
], and the ionic carbazole (g), indole (h) and pyrrole (i), Fig. 1C). Subsequently,
III
20,24
dinuclear β-[Ru (PDT)
5
]Cl,
exhibit greater anticancer the synthetized metal–DTC derivatives were tested for their
2
activity than the corresponding non-cyclic ligands, such as sar- antiproliferative activity against three different human tumor
cosine- or diethyl dithiocarbamate. Moreover, a number of cell lines, namely HeLa (human cervix carcinoma), HepG2
studies performed with cyclic DTCs coordinating different (human hepatocellular carcinoma), and the apoptosis-resistant
transition metal centers have focused on the relationship HepG2/SB3 (human hepatocellular carcinoma cell line stably
between the nature of the N-substituents and the electronic pro- transfected to overexpress Serpin-B3). These screenings were
25–28
perties of the resulting complexes.
The presence of an elec- followed by the evaluation of the lipophilic nature of the com-
tron-donating group attached to the dithiocarbamic N atom plexes via log P determination.
enhances the electron density around the nitrogen atom itself,
and hence in the –CSS moiety; conversely, electron-withdrawing
groups have the opposite effect. Accordingly, dithiocarbamato
Results and discussion
Considerations on the synthetic routes
salts exhibit the ability to simultaneously be strong- and weak-
29
field ligands. As an example, in DTCs of aliphatic amines, the
lone pair on the nitrogen atom is mostly shifted towards the The syntheses of the dithiocarbamato salts derived from ali-
CSS moiety due to the inductive effect mediated by the substitu- phatic amines (ligands b–f, Fig. 1) followed a general route,
ents, resulting in an increase in electron density on the sulfur taking into account the nucleophilic attack of the nitrogen
–
30,31
33
atoms (Fig. 1A, form I).
In this case, the d-orbitals of the two lone pair to the electron poor carbon of CS2. Moreover, their
S atoms are partially filled, making the ligand a potential aromatic counterparts (ligands g–i, Fig. 1) were obtained in dry
π-donor species. Moreover, in aromatic-DTC analogues, the nitro- THF under an inert atmosphere using NaH to force the equili-
3
4
gen lone pair is involved in the aromaticity of the ring (meso- brium towards dithiocarbamic salt formation.
15478 | Dalton Trans., 2018, 47, 15477–15486
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
Concerning the Cu(II) derivatives, the precursor CuCl
2
reacts in a 2 : 1 ligand-to-metal molar ratio, forming the
Cu(DTC) ] complexes (Scheme 1A), wherein the DTC ligands
[
2
coordinate the metal center through both sulfur atoms. Moreover,
focusing on the Ru(III)–DTC derivatives, many studies have
reported that the chelating dithiocarbamato ligand can totally
replace the chloride ligands of the starting material RuCl
resulting in a mixture of neutral mononuclear [Ru(DTC)3]
Scheme 1B) and ionic dinuclear [Ru (PDT) ]Cl, as a mixture
3
,
(
2
5
of α and β isomers (Scheme 1C and D), respectively. After silica
gel purification, the thermal conversion to the most thermo-
dynamically stable β-[Ru
2 5
(DTC) ]Cl was performed by breaking
the Ru–S bond (bridge) in the α isomer, followed by rearrange-
3
5,36
ment in methanol or dichloromethane.
To date, we
obtained and well characterized via different techniques all
the mono and dinuclear Ru(III) complexes of PDT, PipeDTC,
MorphDTC, IndolineDTC, ProOMeDTC, ProOtBuDTC, and
CDT. Conversely, we were not able to isolate the IndDTC and
PyrrDTC derivatives yet due to the instability of these ligands
in the presence of the Ru(III) center. It is worth noting that
some of the Cu(II) coordination compounds (i.e., the com- Fig. 2 Molecular structure of [Cu(ProOMeDTC)
2
]
(top) and
2
3,37
[Cu(ProOtBuDTC) ] (bottom) with thermal ellipsoids drawn at a 30%
2
pounds with a, d and g–i in Fig. 1),
and Ru(III) (with the
2
4
probability
level.
Selected
geometric
parameters
(Å)
for
PDT and CDT ligands) have been already reported in litera-
ture, but they were re-synthetized herein to compare their
chemical and biological properties.
[
[
Cu(ProOMeDTC)
Cu(ProOtBuDTC)
2
] are Cu–S(1) = 2.305(2) and Cu–S(2) = 2.285(2). The
] has Cu–S(11) = 2.306(1); Cu–S(12) = 2.283(1); Cu–S
2
(21) = 2.856(1) and Cu–S(22) = 2.313(2).
X-ray structural analysis
2
−
in comparison to the geometry of the CuCl
Moreover, the complex [Cu(ProOtBuDTC)
4
complex anion.
2
] shows a more
Dark-green crystals of the Cu(II)-prolineDTC derivatives were
obtained by the slow evaporation of CH Cl solutions of the
2 2
complexes, and their molecular structure were solved. Two
ligands complete the metal coordination sphere. In
regular square planar geometry for the metal center. The struc-
tures are chiral according to the presence of the chiral dithio-
carbamato ligand, which has an S configuration.
[
Cu(ProOMeDTC)
oriented on the same side of the coordination plane, whereas
in [Cu(ProOtBuDTC) ] (Fig. 2, down) they are arranged on
opposite sides. The geometry of the metal center in
Cu(ProOMeDTC) ] can be described as a flattened tetrahedron
2
] (Fig. 2, top), the methyl-ester groups are
FT-IR characterization and comparative analysis
2
−
1
FTIR spectrophotometry (both medium 4000–500 cm and far
1
5
00–200 cm− ) is the main tool to detect the molecular signa-
[
2
tures for the two limit forms presented in Fig. 1A (see Tables
SI 2–4 in the ESI†). Indeed, the ν(N–CSS) band for both the
salts and metal–DTC complexes should be found at
640–1690 cm− when form I is prevalent and at lower wave-
1
1
−
1
numbers (1250–1350 cm ) if the DTC ligand contains a single
C–N bond (form II). It is worth highlighting that the actual be-
havior is intermediate between the two limit forms depicted in
Fig. 1A, which is shifted toward form I or II based on the
extent of lone pair sharing in the –NCSS moiety.
Comparing the ν(N–CSS) values shown in Table 1 for the
free ligands a–i (Fig. 1) and the corresponding complexes, the
substituents with a significant inductive effect, such as those
in pyrrolidine (a) and piperidine (b), confer a preference for
the form I on the corresponding DTC ligand. In contrast, the
leads to aromatic DTC carbazole (g), indole (h) and pyrrole (i) are
Scheme 1 The addition of two equivalents of DTC to CuCl
2
II
the neutral [Cu (DTC)
2
] (A). Conversely, when RuCl
3
and DTC react in a
] (B) and α + β
5
]Cl (C and D, respectively) are generated,
associated with the form II. In fact, their nitrogen lone pair is
engaged in the aromaticity of the ring (mesomeric effect), thus
making the electron shift to the –NCSS group impossible. The
DTC salts of the morpholine (c), indoline (d), and L-proline
ester (e and f) derivatives contain either electron-withdrawing
III
1
: 3 metal-to-ligand molar ratio, a mixture of [Ru (DTC)
3
III
2
isomers of the ionic [Ru (DTC)
and, after purification via chromatography, the dinuclear complex is
completely isomerized to the thermodynamically more stable β-[Ru 2I II
(
5
DTC) ]Cl.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15477–15486 | 15479

View Article Online
Paper
Dalton Transactions
Table 1 Collection of the IR-vibrations related to the ν(N–CSS) band of the selected dithiocarbamato ligands a–i and their Cu(II) and Ru(III) coordi-
nation compounds
Ligand salt
[Cu(DTC)
2
]
[Ru(DTC)
3
]
2 5
β-[Ru (DTC) ]Cl
1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
a
b
c
d
e
f
g
h
i
1422 cm−
1498 cm
1502 cm
1485 cm
1422 cm
1472 cm
1470 cm
1333 cm
1334 cm
1334 cm
1487, 1470, 1444 cm
1488, 1455, 1439 cm
1486, 1431 cm
1506, 1473, 1448 cm
1417 cm−
1
1503, 1441 cm
−1
−
1
1
1
1
1
1
1
−1
−1
1410 cm
1503, 1435 cm
1371 cm−
1387, 1348, 1322 cm
−1
−1
1431, 1324 cm
1480, 1448 cm
−1
1392 cm−
1440 cm
−1
−1
−
−1
−1
1388 cm
1440 cm
1481, 1450 cm
1281 cm−
1290 cm−
1259 cm−
1363, 1324, 1295 cm
—
—
1364, 1326, 1299 cm
−1
—
—
groups (i.e., ester moiety for the two prolines or an oxygen either electron-withdrawing groups or resonance groups, they
atom for morpholine) or resonance groups with a mesomeric exhibit an inductive effect as well, which is mediated by the
effect (i.e., the aromatic ring of indoline), which lead them to piperidine-like ring for morpholine and the pyrrolidine ring
3
8
be halfway between the two cases.
for the prolines and indoline. The steric influence of different
In light of these observations, the plot of N–CSS stretch- cyclic substituents may account for small deviations from
4
4
ing frequency versus the pK of the conjugate acid of the linearity with specific effects being difficult to foresee.
parent amines
a
3
9–43
highlights a linear correlation (Fig. 3). Focusing on the impact of the metal center and its oxidation
Thus, the weaker the conjugate acid of the parent amine, state on the ν(N–CSS) mode, upon coordination the nitrogen
the stronger the N–CSS bond, which is supported by the larger lone pair experiences a shift toward the –CSS moiety due to
propensity of the nitrogen atom to share its lone pair. Ligands the electron-withdrawing properties of the transition metal,
a
(
and
ca. 1417 cm ) and i the lowest (1259 cm ). In other words,
the inductive and mesomeric effects underlying the base ascribable to the asymmetric stretching ν
strength of the investigated amines are the same that affect possible to detect the symmetric counterpart ν
b
showed the highest ν(N–CSS) wavenumbers thus increasing the energy of the corresponding vibration.
−
1
−1
The second diagnostic region spans from 800 to 1100 cm−1,
(CSS). It is also
(CSS) in the
Contrary to the N–CSS stretching
a
s
the shift of the nitrogen lone pair towards the –CSS moiety. The range of 410–700 cm−
1 29,45
.
DTC salts of the morpholine (c), indoline (d), and L-proline mode, ν(CSS) shows no correlation with the base strength of
ester (e and f) derivatives have intermediate behavior (about the parent amine in the free ligand, likely due to the steric
1
44
1
380 cm− ) between the forms I and II. In fact, besides bearing hindrance of the N-substituents. Moving to the investigated
complexes, the general trend follows the paradigm the higher
the ν(N–CSS) frequency, the lower the ν(–CSS) based on the
main influence of the thioureido or dithiocarbamato form.
Indeed, the thioureido form I (Fig. 1A) reflects the presence of
electron density in the low energy d-orbitals of the sulfur
atoms due to the shift of the nitrogen lone pair towards
the –CSS moiety. As a result, the tendency of the metal center
to back-donate to the sulfur atoms is reduced. On the
contrary, the dithiocarbamato form II (Fig. 1A) provides
empty d-orbitals in the sulfur atoms, which can accept metal
2
5,27–29
π-retrodonation.
tron density in the
This effect is related to the higher elec-
ring and, consequently, to the more
covalent character of the M–S bonds. Finally, according to the
4
6
Bonati–Ugo method, the presence of only one band in the
diagnostic region of the ν(CSS) mode is assumed to indicate a
symmetric bidentate ligand, as confirmed by the reported
crystal structures (Fig. 2).
1
H-NMR characterization
1
Fig. 3 Plot of the pK
for the synthetized dithiocarbamato salts against the wavenumbers of
a
values of the conjugate acid of the parent amines The H-NMR analysis was useful to follow the progress of the
reaction for the formation of the dithiocarbamato salts, as well
as that of metal complexes, focusing on the C–H signals
closest to the nitrogen atom (see Tables SI 5–7 in the ESI†).
Moreover, it was suitable to assess the presence of some labile
functional groups such as esters. The assignments were con-
the N–CSS stretching mode. The plot follows an almost linear corre-
2
lation (ρ = 0.97). The pK
a
values are: pyrrolidine, 11.27; piperidine, 11.22;
morpholine, 8.36; L-proline methyl ester, 8.50; L-proline tert-butyl ester,
8
−
.47; indoline, 4.90; carbazole, −6.00; indole −6.50; and pyrrole
5.90.
39–43
15480 | Dalton Trans., 2018, 47, 15477–15486
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
ducted by comparing the spectra of the obtained complexes to
those of the ligands and the free amines.
1
The H-NMR spectra of Cu(II) and Ru(III) compounds are
9
5
affected by paramagnetic effects associated with the d and d
low spin) electronic configurations of the metal centers,
respectively. Consequently, hyperfine shifts (δ) in NMR signals
and shortening of both the nuclear longitudinal (T ) and trans-
The evidence of
the first phenomenon is the shift and lower intensity observed
for the methylene protons H(2) and H(5) in [Ru(PDT) ] (δ =
5.83 and 44.29 ppm). Further, for the Cu(II) derivatives
(
1
4
7,48
verse (T ) relaxation times are observed.
2
3
3
involving PipeDTC and L-proline ester DTCs as ligands, the
signals of the α-protons completely disappear. This behavior
is related to the relaxation times, and in particular, strictly
1
Fig. 4 H-NMR spectrum of [Ru(ProOtBuDTC)
CDCl
observed, which correspond to three magnetically different α-protons
3
] (300.13 MHz, 298 K,
3
) in the range of 50–20 ppm. 12 different singlets are
depends on the electronic relaxation time of the paramagnetic (positions 2 and 5).
4
8–50
metal center (τ
s
).
The paramagnetism of Cu(II) affects to a
lesser extent the ester protons of both proline derivatives since
the –OR group (R = methyl, tert-butyl) is spatially more distant cates the diamagnetic nature of these derivatives, which
3
6
from the metal center. The corresponding peaks are less possess two antiferromagnetically-coupled Ru(III) ions. The
broadened with respect to the other resonances within the complexity of the spectra might be ascribed to the different
same coordination compounds and their presence confirms coordination modes adopted by the five ligands, which cause
the absence of hydrolysis side-reactions during the synthesis the proton signals to be different even when placed in similar
5
7
of the [Cu(DTC) ] complexes.
chemical environments. However, despite the high number
2
1
The case of the mononuclear Ru(III) derivatives is quite of multiplets, the H-NMR analysis of this type of complexes is
peculiar and deserves a detailed discussion. Indeed, besides a fundamental to verify the completeness of the isomerization
paramagnetic shift, an additional effect influences the reaction between the α and β isomers by observing the dis-
1
H-NMR spectra of some Ru(DTC) derivatives, namely their appearance of specific signals and the increase in intensity of
3
5
1–54
“
dynamic stereochemistry”.
In fact, at the NMR timescale, others.
this type of complex undergoes a metal-centered inversion
between the two optical isomers Λ and Δ, thus generating a
signal for each species at equilibrium. Such intramolecular
inversion may occur according to four different mecha-
UV-Vis characterization
The electronic spectra of the Cu(II) and Ru(III)–DTC derivatives
2 2
were acquired in CH Cl (800–240 nm) at 298 K (see Tables SI
5
5,56
nisms,
the most favored (rotation around the C
transitional state D3h). To better explain this phenomenon, the
spectra of the [Ru(PDT) ] and [Ru(ProOtBuDTC) ] complexes
but in our case the trigonal twist (Bailar twist) is
8
and 9†). The copper complexes containing an aromatic
3
symmetry axis, with a
moiety (ligands d and h–g, Fig. 1) were not analyzed due to
their very low solubility in all solvents.
3
3
For the Cu(II) aliphatic dithiocarbamato complexes
are compared since they contain symmetric and asymmetric
(
Fig. 5A), the band I at about 260 nm and band II at around
90 nm are ascribed to the π* ← π absorption in the –NCSS
group and the intraligand π* ← π transition in the –CSS
1
ligands, respectively. In the first case, the H-NMR analysis
2
shows the presence of two different broad singlets at 35.83
and 44.29 ppm, related to the α-CH
isomers, whose unambiguous attribution is not possible.
Conversely, the spectrum of the [Ru(ProOtBuDTC) ] derivative
2
protons of the Λ and Δ
5
8,59
moiety, respectively.
In particular, by analyzing the band II
maximum wavelength, a red shift is observed on going
from the PDT and PipeDTC derivatives (ca. 285 nm) to the
ProlineDTC (ca. 290 nm) counterparts, which is likely due to
3
is more complex as 12 singlets are detected in three four-mem-
bered sets. Considering ProOtBuDTC as an asymmetric ligand,
the –CH group in position 2 and the two protons in position 5
6
0
the inductive effects by the ester groups.
(diastereotopic) are magnetically different, thus leading to
three distinct resonances. Moreover, from the spectrum ana-
lysis, one can infer that for each of them there is a resonance
split, likely due to the presence of the Λ and Δ optical isomers
and cis or trans geometric isomers (obtained by hindered
5
1
rotation around the C–N bond, Fig. 4). Based on this hypoth-
esis, the resonances could come from the coupling of
3
(protons) × 2 (optical isomers Λ and Δ) × 2 (geometric
isomers cis and trans) = 12 total signals.
1
Concerning the H-NMR spectra of the Ru(III) dinuclear
Fig. 5 Representative UV-Visible spectra (CH
2
Cl
2
,
298 K):
II
III
III
species, the absence of signals out of the 0–12 ppm range indi- [Cu (ProOMeDTC)
2
] (A), [Ru (PipeDTC)
3
] (B), and β-[Ru (PipeDTC) ]Cl (C).
5
2
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15477–15486 | 15481

View Article Online
Paper
Dalton Transactions
In the 350–600 nm region, taking into account that the Table 2 IC50 values (μM) calculated after 72 h-treatment. Data rep-
9
resent the mean ± SD of at least three independent experiments
Cu(II) ion (d ) experiences the Jahn–Teller effect (favoring dis-
(vehicle: DMSO, except for saline solution used for testing cisplatin)
torted structures), Choi et al. were able to calculate the energy
sequence of the d-orbitals of a Cu–DTC complex as dxy > d
z
2
>
IC50 (μM)
d
x2−y2
> dxz,yz. Accordingly, the next transition (band III) is
ascribable to dxz,yz ← dxy, combined with either a metal-to- Compound
HeLa
HepG2/CTR
HepG2/SB3
ligand charge transfer (MLCT) or a ligand-to-metal charge
β-[Ru
transfer (LMCT), whereas band IV and band V are assigned to β-[Ru
β-[Ru
2
2
2
2
(PDT)
(PipeDTC)
(ProOMeDTC)
(ProOtBuDTC)
5
]Cl
0.45 ± 0.02
0.46 ± 0.04
3.95 ± 0.07
0.27 ± 0.01
0.62 ± 0.08
0.66 ± 0.01
9.2 ± 0.2
0.59 ± 0.01
0.75 ± 0.02
>10
5
]Cl
5
]Cl
]Cl
the lowest energy d–d transitions d
x
2
−y
2
← dxy and d
z
2 ← dxy,
β-[Ru
5
1.28 ± 0.05
0.90 ± 0.02
6
1
respectively.
Similarly to the previously discussed cases, in the Ru(III)– [Cu(PDT)
DTC complexes (both mono- and dinuclear, Fig. 5B and C,
respectively) the band I (at 240–250 nm for the aliphatic com-
plexes) is assigned either to an intraligand π* ← π transition [Cu(ProOtBuDTC)
2
]
0.32 ± 0.04
0.6 ± 0.3
6.7 ± 0.2
1.03 ± 0.03
1.13 ± 0.07
1.99 ± 0.03
>10
>10
>10
2.97 ± 0.07
5.0 ± 0.1
>10
>10
>10
>10
2.8 ± 0.1
6.6 ± 0.1
>10
[
[
[
Cu(PipeDTC)
Cu(MorfDTC)
Cu(ProOMeDTC)
2
]
2
]
2
]
2
]
[
Cu(PyrroleDTC) ]
2
located in the –NCSS moiety or to an intraligand p ← d tran-
6
2
sition between levels originating from the sulfur atoms. In
addition, the band II (250–260 nm) and band III (280–290 nm) are
attributed to other intraligand π* ← π transitions in the N–C–S
Cisplatin
6.6 ± 0.3
>10
>10
6
3
and the S–C–S moieties of the dithiocarbamate, respectively.
In the 300–800 nm range, the absorption band IV 10 μM, 5 μM, 2 μM, 1 μM, 0.5 μM and in some cases lower) for
330–360 nm) can be assigned to a ligand-to-metal charge- 72 hours. Quadruplicate conditions were established for each
(
transfer (d ← π, LMCT) rather than to d–d transitions (ε > 100 treatment and at least three independent experiments were
−
1
−1 64,65
M
cm ).
ca. 600 nm may be assigned to d-d transitions.
The absorption spectra of the β-[Ru (DTC)
Moreover, band V (440–460 nm) and band VI at performed for each compound. HeLa cells were chosen since
they are generally used as a kind of standard for preliminary
]Cl derivatives screening studies of new cytotoxic agents. Concerning that
2
5
show in the 300–800 nm region an intense band IV at hepatocellular carcinoma is one of the most spread malignan-
3
30–360 nm and a broad weak band V at 420–470 nm. No cor- cies worldwide, the studies were focused on the HepG2 line
relation can be established between the spectra acquired for and its more aggressive HepG2/SB3 counterpart. The latter is
the mono- and di-nuclear complexes due to their quite dissim- associated with resistance to oncological treatments due to the
ilar structures and the presence of two ruthenium centers anti- overexpression of the anti-apoptotic protein SerpinB3 (high
3
6
ferromagnetically coupled. Therefore, since a strong metal– SerpinB3 levels have been detected in patients with the most
6
6
metal interaction may take place between the two units, the aggressive cancer subtypes). Indeed, this serine-protease
absorption band IV can be ascribed to a metal–metal to ligand inhibitor, which is typically found in cancer cells of epithelial
charge transfer (MMLCT), whereas the very broad band V may origin and of the liver, significantly attenuates apoptosis
be due to a metal-to-metal charge transfer (MMCT) since the triggered by anti-cancer drugs or TNF-α, thus allowing tumor
molar extinction coefficient values are higher than expected growth, dictating hepatocellular carcinoma progression and
6
5
67,68
for a d–d transition.
invasion.
Moreover, this protein is located in the inner
mitochondrial compartments, where it binds to the respiratory
Complex I and protects cells from the pro-oxidant action
UV-Vis stability studies
Due to their poor water solubility, the synthetized metal com- of some chemotherapeutic agents, such as doxorubicin and
6
9
pounds derived from cyclic dithiocarbamates were pre-dis- cisplatin. Among the 23 metal–dithiocarbamato compounds
solved in DMSO to carry out the biological in vitro studies. The synthetized, only 20 complexes were subjected to biological
Ru(III) compounds, both mono- and dinuclear, remained screening. Specifically, the Cu(II) complexes derived from
stable in this solvent over 72 hours, as well as the Cu(II) deriva- ligands d, h and g (Fig. 1) were excluded due to DMSO-solubi-
tives. Moreover, the most interesting compounds in terms of lity issues.
antiproliferative activity (see Table 2) were analyzed in the pres-
The obtained cell viabilities were plotted against the corres-
ence of cell culture medium and human serum (both diluted ponding compound concentration to determine the IC50 (con-
in phosphate buffered saline). No change in the related UV-Vis centration of the test agent inducing a 50% reduction in cell
spectra was observed, thus highlighting their general stability number compared with the control cell cultures). For compari-
under these biological conditions (spectra reported in the son, the cytotoxicity of cisplatin (dissolved in saline solution)
ESI†).
was evaluated under the same experimental conditions.
In this regard, it should be noted that 72 h experiments were
chosen taking into account the time necessary for cisplatin to
Antiproliferative assays and log P investigations
For testing the antiproliferative activity of the investigated com- exert its activity. For clarity, Table 2 only presents the com-
pounds, HeLa, HepG2/CTR and HepG2/SB3 cells were exposed pounds with IC50 values lower than 10 μM, which was set as
to different concentrations of the metal–DTC complexes (i.e., our experimental cut-off for future development activities.
15482 | Dalton Trans., 2018, 47, 15477–15486
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
1
0 complexes demonstrated remarkable cytotoxic activity Table 3 Log P values of the most biologically-active metal–DTC
derivatives, determined via UV-Vis analysis. Data represent the mean ±
(
IC50 < 10 μM) against HeLa cells, whereas 6 and 5 inhibited
SD of at least three independent measurements
the proliferation of HepG2 and HepG2/SB3 cells, respectively.
This first result highlights the resistance of hepatocellular
carcinoma (HCC) cells to chemotherapy compared to cervical
cancer. Moreover, the screening of the free dithiocarbamato
Compound
log P
β-[Ru
β-[Ru
β-[Ru
2
2
2
2
(PDT)
(PipeDTC)
(ProOMeDTC)
(ProOtBuDTC)
5
]Cl
0.92 ± 0.04
0.9 ± 0.2
0.78 ± 0.07
1.4 ± 0.2
5
]Cl
2 2
salts, as well as the metal precursors CuCl ·2H O and
5
]Cl
]Cl
RuCl ·3H O, highlighted the absence of biological activity at β-[Ru
5
3
2
the concentration of 10 μM. Among the most promising com-
plexes, the dinuclear Ru(III)–DTC derivatives stand out, with
[
Cu(PDT)
2
]
1.54 ± 0.04
1.5 ± 0.1
[
Cu(PipeDTC)
2
]
IC50 values at the micromolar and sub-micromolar level. On [Cu(MorphDTC)
the contrary, the mononuclear derivatives of [Ru(DTC) ] (not
3
2
]
1.37 ± 0.04
1.3 ± 0.2
1.6 ± 0.3
[
[
[
Cu(ProOMeDTC)
Cu(ProOtBuDTC)
Cu(PyrrDTC)
2
]
2
]
presented in Table 2) did not show any antiproliferative activity
against all the tested human cell cultures. This behavior was
2
]
2.01 ± 0.04
2
0
already observed for other Ru(III)–DTC derivatives,
thus
7
2–74
supporting the hypothesis that the particular shape and/or the metabolism and excretion.
The general parameter defin-
electronic interaction (anti-ferromagnetism) between the two ing the lipophilicity of a molecule is log P, which is the
Ru(III) metal centers, mediated by the S atoms of the bridged decimal logarithm of the n-octanol/water partition coefficient
DTC ligands, may be the key element for the great activity of P. In particular, P identifies the ratio between the concen-
our dinuclear complexes.
tration of a non-ionized drug in octanol and that in water at
The Cu(II)–dithiocarbamato complexes similarly showed equilibrium at 25 °C. Interestingly, the analysis of 592 oral
high antiproliferative activity against HeLa cells but were less drugs approved worldwide between 1983 and 2007 shows that
7
5,76
active against the liver cancer cultures. However, except for the the mean log P value ranges from 2.3 to 2.6.
two proline-based complexes, these copper derivatives are
Based on these considerations, in this study, to better eluci-
poorly soluble in DMSO as vehicle, and thus it is possible that date the reported cytotoxicity data, we focused on the log P
the final observed activity was underestimated.
analysis of the most active compounds. This was performed
Concerning the ligand nature, the most active compounds using the bi-phasic solvent system n-octanol/water, which is
contain the aliphatic dithiocarbamates associated with the commonly accepted in the scientific community since it well
7
7
thioureidic form of the ligand, and hence a stronger N–C bond mimics the interface of the water/phospholipid membrane.
and weaker metal–sulfur bond. In other words, the complexes The lipophilic character parameter of the most promising
containing aromatic moieties showed no cytotoxic activity metal–DTC compounds was evaluated via UV-Vis spectrophoto-
except for [Cu(PyrroleDTC) ], which exhibited an antiprolifera- metry. We identified the concentration of the complex in the
2
2
4,78
tive activity at about 2 μM. The aromatic derivatives involve two distinct phases after equilibration,
ligands that, besides preferring the dithiocarbamic form reported as mean ± SD of at least three independent measure-
endowed with a strong M–S bond), have pronounced hydro- ments, are presented in Table 3.
phobic character, which hampers their solubility in the cell
Overall, the synthetized derivatives are associated with posi-
and the results,
(
culture medium. Remarkably, none of the ligand salts resulted tive log P values, as expected, due to the presence of hydro-
in cytotoxicity towards the tested cell cultures, thus confirming phobic ligands. Intriguingly, despite their ionic nature, the
that the antiproliferative activity belongs in a peculiar way to dinuclear Ru(III) compounds showed higher affinity for the
the whole complex.
organic phase. In other words, the aliphatic surface made up
Finally, comparison of the IC50 values obtained for our of five dithiocarbamato ligands predominates on the +1
metal–DTC complexes with those of clinically-established drugs charge of the complex.
is of paramount importance. In this regard, most of the tested
Concerning the copper(II) derivatives, the data well correlate
derivatives exhibit greater cytotoxicity than the reference drug with the neutral form of the complexes, with the great hydro-
cisplatin (Table 2), which is used in standard regimens for the phobic character of the aromatic pyrrole resulting in the
7
0
treatment of terminal-stage cervical cancer. In the case of highest-recorded log P value.
HCC and considering sorafenib (Nexavar®), the only che-
motherapeutic agent able to reduce the spread of this cancer,
7
1
Experimental
its reported IC50 value is 8.99 μM, which is about 15-fold
higher than that recorded for our dinuclear Ru(III)–DTC com-
pounds. Taken together, these findings highlight the great
anticancer potential of some of the synthetized metal–DTC
compounds.
This material is available in the ESI.†
Conclusions
It is worth reminding that in the process of drug discovery,
lipophilicity is very important to optimize the pharmacological The biological activity of a coordination compound depends
response, toxicological profile, absorption, distribution, not only on the nature of its metal center, but also from the
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15477–15486 | 15483

View Article Online
Paper
Dalton Transactions
79,80
synergic effect of the metal–ligand system.
In this study, the
Acknowledgements
selected metal centers were Cu(II) and Ru(III), which have inter-
esting chemical features exploitable at the pharmaceutical level. This study was financially supported by the University of
The first goal of this study was the synthesis and compara- Padova (PRAT2015 - CPDA158302), the L’Oréal Foundation –
tive characterization of a library of compounds, wherein each UNESCO (C. N.’s post-doc Fellowship, http://www.forwome-
metal ion coordinates cyclic N,N-disubstituted dithiocarba- ninscience.com/en/fellowships), T.R.N. IMBALLAGGI – logistic
mato (DTC) ligands. In particular, the chosen DTC chelating services (N. P.’s PhD Fellowship, http://www.trnimballaggi.it/
agents were derived from cyclic amines, both aliphatic (i.e., en/) and the A.R.TE.M.O. Association ONLUS (http://www.asso-
pyrrolidine, piperidine, morpholine, indoline, and L-proline ciazioneartemo.it). The authors gratefully acknowledge
methyl- and tert-butyl ester) and aromatic (carbazole, indole, Dr Loris Calore for elemental analyses.
and pyrrole). DTCs can indeed confer great stability on the
metal ion both at the redox and coordination level, thus redu-
cing side-interactions with sulfur-containing bio-molecules
and enzymes under physiological conditions.
Notes and references
The dithiocarbamato ligands and their complexes were
synthetized using different strategies aimed at improving the
reaction yields and purity of the compounds, which were
characterized via elemental analysis, X-ray crystallography,
1 N. P. E. Barry and P. J. Sadler, Chem. Commun., 2013, 49,
5106.
2 P. Chellan and P. J. Sadler, Philos. Trans. R. Soc., A, 2015,
373, 20140182.
3 W. Maret, Int. J. Mol. Sci., 2016, 17, 1.
4 Metals in Chemical Biology, Nat. Chem. Biol., 2008, 4, 143.
5 K. D. Mjos and C. Orvig, Chem. Rev., 2014, 114, 4540.
6 S. Medici, M. Peana, V. M. Nurchi, J. I. Lachowicz,
G. Crisponi and M. A. Zoroddu, Coord. Chem. Rev., 2015,
284, 329.
7 N. J. Wheate, S. Walker, G. E. Craig and R. Oun, Dalton
Trans., 2010, 39, 8113.
8 P. Mistry, L. R. Kelland, G. Abel, S. Sidhar and K. R. Harrap,
Br. J. Cancer, 1991, 64, 215.
9 R. P. Miller, R. K. Tadagavadi, G. Ramesh and W. B. Reeves,
Toxins, 2010, 2, 2490.
1
ESI-MS, H-NMR spectroscopy, FT-IR and UV-Vis spectropho-
tometry. The obtained complexes are endowed with the follow-
II
ing stoichiometry: [Cu (DTC)
2
] for the copper derivatives and
III
III
the two forms [Ru (DTC)
nium derivatives.
3
] and β-[Ru (DTC)
5
]Cl for the ruthe-
2
Concerning the structural and electronic properties of the
synthetized metal–DTC derivatives, FT-IR spectrophotometry
undoubtedly highlighted the difference between the dithiocar-
bamic (CDT, IndDTC, and PyrrDTC (and generally
IndolineDTC)) and thioureidic (PDT and PipeDTC) forms of
the coordinated ligands based on the nature of the cyclic
amine. Moreover, the MorphDTC and L-proline ester dithiocar-
bamato ligands displayed intermediate properties between the 10 N. D. Eljack, H. M. Ma, J. Drucker, C. Shen, T. W. Hambley,
two limiting forms, with a preference for the thioureidic form.
The underlying rationale of this study was to elucidate how
E. J. New, T. Friedrich and R. J. Clarke, Metallomics, 2014,
6, 2126.
the nature of the ligand (aromatic/aliphatic, and hence the 11 R. F. Borch and M. E. Pleasants, Proc. Natl. Acad. Sci. U. S.
strength of the metal–sulfur bond and lipophilicity of the final A., 1979, 76, 6611.
complex) with equal compound stoichiometry affects the bio- 12 R. F. Borch, J. C. Katz, P. H. Lieder and M. E. Pleasants,
logical activity. Thus, the synthetized metal–DTC derivatives Proc. Natl. Acad. Sci. U. S. A., 1980, 77, 5441.
were tested in vitro for their cytotoxicity against three human 13 N. P. E. Barry and P. J. Sadler, Pure Appl. Chem., 2014, 86,
tumor cell lines (HeLa, HepG2/CTR and HepG2/SB3). In 1897.
general, the compounds containing PDT, PipeDTC, 14 G. Hogarht, Mini-Rev. Med. Chem., 2012, 12, 1202.
ProOMeDTC and ProOtBuDTC generated promising results, 15 C. Nardon and D. Fregona, Curr. Top. Med. Chem., 2016,
also against the apoptosis-resistant cell line HepG2/SB3.
16, 360.
Based on the collected results, the IC50 values of the most 16 A. Casini, W.-Y. Sun and I. Ott, in Metallo-Drugs:
active compounds were flanked with the corresponding lipo-
philic index (log P). Among the aliphatic complexes, the
proline derivatives of all the tested metal complexes were
Development and Action of Anticancer Agents, ed. A. Siegel,
H. Siegel, E. Freisinger and R. K. O. Siegel, Walter de
Gruyter GmbH & Co, 2018, ch. 7, pp. 200–213.
associated with a very good solubility profile (assuming DMSO 17 G. Faraglia, D. Fregona, S. Sitran, L. Giovagnini,
as the vehicle). Focusing on the metals, the dinuclear Ru(III)
species and the Cu(II) compounds showed the lowest IC50
C. Marzano, F. Baccichetti, U. Casellato and R. Graziani,
J. Inorg. Biochem., 2001, 83, 31.
values combined with exceptional stability in both DMSO and 18 C. Nardon, G. Boscutti and D. Fregona, Anticancer Res.,
aqueous media (i.e., cell culture medium and human serum).
2014, 34, 487.
1
9 L. Giovagnini, S. Sitran, M. Montopoli, L. Caparrotta,
M. Corsini, C. Rosani, P. Zanello, Q. P. Dou and
D. Fregona, Inorg. Chem., 2008, 47, 6336.
0 C. Nardon, L. Brustolin and D. Fregona, Future Med. Chem.,
2016, 8, 211.
Conflicts of interest
2
There are no conflicts to declare.
15484 | Dalton Trans., 2018, 47, 15477–15486
This journal is © The Royal Society of Chemistry 2018

View Article Online
Dalton Transactions
Paper
2
1 T. Storr, in Ligand Design in Medicinal Inorganic Chemistry, 50 I. Bertini, A. Dei and A. Scozzafava, Inorg. Chem., 1975, 14,
014, pp. 1–449. 1526.
2 C. Nardon, F. Chiara, L. Brustolin, A. Gambalunga, 51 M. C. Palazzotto, D. J. Duffy, B. L. Edgar, L. Que Jr. and
2
2
F. Ciscato, A. Rasola, A. Trevisan and D. Fregona,
ChemistryOpen, 2015, 4, 183.
L. H. Pignolet, J. Am. Chem. Soc., 1973, 95, 4537.
52 L. Que Jr. and L. H. Pignolet, Inorg. Chem., 1974, 13,
2
2
3 V. Milacic, D. Chen, L. Giovagnini, A. Diez, D. Fregona and
Q. P. Dou, Toxicol. Appl. Pharmacol., 2008, 231, 24.
4 S. Scintilla, L. Brustolin, A. Gambalunga, F. Chiara,
351.
53 D. J. Duffy and L. H. Pignolet, Inorg. Chem., 1974, 13,
2045.
A. Trevisan, C. Nardon and D. Fregona, J. Inorg. Biochem., 54 L. H. Pignolet, Inorg. Chem., 1974, 13, 2051.
2
017, 166, 76.
5 F. A. Cotton and J. A. McCleverty, Inorg. Chem., 1964, 3,
398.
6 E. J. Kupchik and P. J. Calabretta, Inorg. Chem., 1965, 4,
73.
7 R. R. Eley, R. R. Myers and N. V. Duffy, Inorg. Chem., 1972,
1, 1128.
55 A. Rodger and B. F. G. Johnson, Inorg. Chem., 1988, 27,
3061.
56 H. S. Rzepa and M. E. Cass, Inorg. Chem., 2007, 46,
8024.
57 E. M. Nagy, A. Pettenuzzo, G. Boscutti, L. Marchiò, L. Dalla
Via and D. Fregona, Chem. – Eur. J., 2012, 18, 14464.
58 S. K. Verma and V. K. Singh, RSC Adv., 2015, 5, 53036.
59 F. Takami, S. Wakahara and T. Maeda, Tetrahedron Lett.,
1971, 12, 2645.
60 C. Oktavec, J. Stefanec, B. Siles, E. Beinrohr, V. Konecny
and J. Garay, Collect. Czech. Chem. Commun., 1982, 47,
2867.
2
2
2
1
9
1
2
2
8 B. Leon and D. K. Straub, Inorg. Chim. Acta, 1989, 156, 13.
9 K. Nakamoto, J. Fujita, R. A. Condrate and Y. Morimoto,
J. Chem. Phys., 1963, 39, 423.
0 G. J. M. van der Kerk and H. L. Klöpping, Recl. Trav. Chim.
Pays-Bas, 1952, 71, 1179.
3
3
3
1 H. K. Hall Jr., J. Am. Chem. Soc., 1956, 78, 2570.
2 J. P. Cartier and C. Sandorfy, Can. J. Chem., 1963, 41,
61 S. Choi, E. R. Menzel and J. R. Wasson, J. Inorg. Nucl.
Chem., 1977, 39, 417.
2
759.
62 L. Ronconi, L. Giovagnini, C. Marzano, F. Bettìo,
R. Graziani, G. Pilloni and D. Fregona, Inorg. Chem., 2005,
44, 1867.
3
3
3 D. M. Miller and R. A. Latimer, Can. J. Chem., 1962, 40, 246.
4 R. D. Bereman and D. Nalewajek, J. Inorg. Nucl. Chem.,
1
978, 40, 1313.
63 G. E. Manoussakis and C. A. Bolos, Inorg. Chim. Acta, 1985,
108, 215.
64 E. M. Nagy, C. Nardon, L. Giovagnini, L. Marchiò,
A. Trevisan and D. Fregona, Dalton Trans., 2011, 40,
11885.
3
5 A. R. Hendrickson, J. M. Hope and R. L. Martin, J. Chem.
Soc., Dalton Trans., 1976, 2032.
6 L. Giovagnini, S. Sitran, I. Castagliuolo, P. Brun,
M. Corsini, P. Zanello, A. Zoleo, A. Maniero, B. Biondi and
D. Fregona, Dalton Trans., 2008, 6699.
3
65 J. F. Endicott and M. Jamal Uddin, Coord. Chem. Rev., 2001,
219, 687.
3
7 R. D. Bereman and D. Nalewajek, Inorg. Chem., 1978, 17,
1
085.
8 O. Esenturk and R. A. Walker, Phys. Chem. Chem. Phys.,
003, 5, 2020.
66 C. Turato, A. Vitale, S. Fasolato, M. Ruvoletto, L. Terrin,
S. Quarta, R. Ramirez-Morales, A. Biasiolo, G. Zanus,
N. Zali, P. S. Tan, Y. Hoshida, A. Gatta, U. Cillo and
P. Pontisso, Br. J. Cancer, 2014, 110, 2708.
67 P. Pontisso, Ann. Hepatol., 2014, 13, 722.
3
2
3
4
4
9 H. K. Hall Jr., J. Am. Chem. Soc., 1957, 79, 5441.
0 M. A. Slifkin and S. M. Ali, J. Mol. Liq., 1984, 29, 75.
1 H. J. Chen, L. E. Hakka, R. L. Hinman, A. J. Kresge and 68 L. Vidalino, A. Doria, S. Quarta, M. Zen, A. Gatta and
E. B. Whipple, J. Am. Chem. Soc., 1971, 93, 5102. P. Pontisso, Autoimmun. Rev., 2009, 9, 108.
2 R. L. Hinman and J. Lang, J. Am. Chem. Soc., 1964, 86, 69 F. Ciscato, M. Sciacovelli, G. Villano, C. Turato, P. Bernardi,
796–3806. A. Rasola and P. Pontisso, Oncotarget, 2014, 5, 2418.
3 Y. Chiang and E. B. Whipple, J. Am. Chem. Soc., 1963, 85, 70 M. Candelaria, A. Garcia-Arias, L. Cetina and A. Dueñas-
4
4
3
2
763.
Gonzalez, Radiat. Oncol., 2006, 1.
4
4
4 H. K. Hall Jr., J. Am. Chem. Soc., 1957, 79, 5444.
5 C. N. Rao and R. Venkataraghavan, Spectrochim. Acta Part A:
Mol. Spectrosc., 1989, 45, 299.
6 F. Bonati and R. Ugo, J. Organomet. Chem., 1967, 10, 257.
7 F. Rastrelli and A. Bagno, Magn. Reson. Chem., 2010, 48,
S132.
8 N. N. Murthy, K. D. Karlin, I. Bertini and C. Luchinat,
J. Am. Chem. Soc., 1997, 119, 2156.
9 K. A. Hötzer, A. Klingert, T. Klumpp, E. Krissinel,
71 F. Pascale, L. Bedouet, M. Baylatry, J. Namur and
A. Laurent, Anticancer Res., 2015, 35, 6497.
72 M. J. Waring, Expert Opin. Drug. Discov, 2010, 5, 235.
73 J. A. Arnott and S. L. Planey, Expert Opin. Drug. Discovery,
2012, 7, 863.
74 C. A. S. Bergström and M. Yazdanian, AAPS J., 2016, 18,
1095.
75 P. D. Leeson and B. Springthorpe, Nat. Rev. Drug Discovery,
2007, 6, 881.
4
4
4
4
D. Bürssner and U. E. Steiner, J. Phys. Chem. A, 2002, 106, 76 A. L. Hopkins, G. M. Keserü, P. D. Leeson, D. C. Rees and
2
207.
C. H. Reynolds, Nat. Rev. Drug Discovery, 2014, 13, 105.
This journal is © The Royal Society of Chemistry 2018
Dalton Trans., 2018, 47, 15477–15486 | 15485

View Article Online
Paper
Dalton Transactions
7
7
7 G. Klopman and H. Zhu, Mini-Rev. Med. Chem., 2005, 5, 127. 79 M. Marloye, G. Berger, M. Gelbcke and F. A. Dufrasne,
8 A. V. Rudnev, L. S. Foteeva, C. Kowol, R. Berger, Future Med. Chem., 2016, 8, 2263.
M. A. Jakupec, V. B. Arion, A. R. Timerbaev and 80 C. Nardon and D. Fregona, Curr. Med. Chem., 2018, 25,
B. K. Keppler, J. Inorg. Biochem., 2006, 100, 1819. 434.
15486 | Dalton Trans., 2018, 47, 15477–15486
This journal is © The Royal Society of Chemistry 2018