Inorganica Chimica Acta 480 (2018) 83–90

Contents lists available at ScienceDirect

Inorganica Chimica Acta

journal homepage: www.elsevier.com/locate/ica

Research paper

Synthesis, characterization and antitumor activity of two new

dipyridinium ylide based lanthanide(III) complexes
Andreea Cârâc a, Rica Boscencu a, Rodica Mihaela Dinică b, Joana F. Guerreiro c, Francisco Silva c,
Fernanda Marques c, Maria Paula Cabral Campello c, Călin Moise d, Oana Brîncoveanu d,
Marius Enăchescu d, Geta Cârâc b, Aurel Tăbăcaru b,⇑
a
Faculty of Pharmacy, ‘‘Carol Davila” University of Medicine and Pharmacy, 6 Traian Vuia Street, 020956 Bucharest, Romania
b
Department of Chemistry, Physics and Environment, Faculty of Sciences and Environment, ‘‘Dunarea de Jos” University of Galati, 111 Domneasca Street, 800201 Galati, Romania
c
Centro de Ciências e Tecnologias Nucleares (C2TN), Instituto Superior Técnico, Universidade de Lisboa, Portugal
d
Center for Surface Science and Nanotechnology, Politehnica University of Bucharest (CSSNT-UPB), 313 Splaiul Independentei, 060042 Bucharest, Romania

a r t i c l e i n f o a b s t r a c t

Article history: Two new dinuclear lanthanide(III) complexes, La-DPY and Nd-DPY, with the stoichiometric formula
Received 20 November 2017 [{Ln(Et3N)(SO4)}2(l-DPY)(l4-SO4)] (Ln = La, Nd; DPY = ylide form of DPB, Et3N = triethylamine), were
Received in revised form 29 April 2018 obtained through the reaction of N-heterocylic diquaternary salt N,N’-diphenacyl-4,40 -dipyridinium
Accepted 5 May 2018
dibromide (DPB) and lanthanide(III) sulfate in methanol, in the presence of triethylamine. The obtained
complexes were characterized by Fourier transform infrared spectroscopy, proton nuclear magnetic res-
onance, elemental analysis, UV–vis spectroscopy, thermogravimetric analysis and powder X-ray diffrac-
Keywords:
tion. Scanning electron microscopy (SEM) was used to investigate the morphology and particles size of
Trivalent lanthanide complexes
Heterocyclic ligands
the complexes, confirming that their particles are quite homogenous and uniform. The antitumor activity
Ylide of the complexes was evaluated in the human cancer cells MCF7 and A2780 and compared to cisplatin,
Dipyridinium ylide the metal-based drug in cancer therapy. The complexes were found to induce apoptotic cell death and, to
Antitumor activity a lesser extent, production of ROS, although these are not the unique mechanisms of action. In conclusion,
we anticipate that these types of Ln(III) complexes have potential and could be further explored for appli-
cations as antitumor agents.
Ó 2018 Elsevier B.V. All rights reserved.

1. Introduction
Lanthanide (Ln) chemistry is currently a very active area of
research. Due to their electronic structure these elements have
unusual properties that make them suitable for applications in
catalysis, organometallic synthesis, electronic and luminescent
materials [1–11]. In the biomedical field, lanthanide complexes
have received considerable attention, particularly in bioanalysis,
imaging and radioimmunotherapy [2]. Recently, some lanthanide
complexes have shown photocytotoxic activities in tumoral cells
and hence with potential use in photodynamic therapy (PDT)
[12–16].
As far as the clinical applications are concerned, lanthanides
cannot be administered in the form of simple salts or metal ions,
due to their toxic effects, but they can be administered in the form
of thermodynamically and kinetically stable complexes. Upon
⇑ Corresponding author.
E-mail address: aurel.tabacaru@ugal.ro (A. Tăbăcaru).
coordination, the ligands play a key role in tuning the properties
of the corresponding complexes [17], which is particularly relevant
to biological, biochemical and medical applications [18].
Due to their size, lanthanide ions form stable complexes with
high geometric flexibility and high coordination number [15–19].
As strong Lewis acids, lanthanide ions coordinate with highly elec-
tronegative donor atoms such as nitrogen or oxygen. N-donor
ligands such as quaternary pyridinium derivatives present a posi-
tive charge at the nitrogen atom and therefore are expected to be
more suitable for the synthesis of very large complexes of
4f-elements than O-donor ligands [9,20]. Nevertheless, the
synthesis of lanthanide complexes with these types of ligands pre-
sent certain limitations, i.e., if anhydrous conditions are not met,
the presence of a nitrogen donor ligand such as 4,40 -bipyridine
can favor deprotonation of any coordinated water, with formation
of undesired sub-products as hydroxo- or oxo-lanthanide
derivatives [20,21].
Heterocyclic ligands present intrinsic versatility to form metal
complexes with biomedical importance and, in this direction, a

https://doi.org/10.1016/j.ica.2018.05.003
0020-1693/Ó 2018 Elsevier B.V. All rights reserved.
84 A. Cârâc et al. / Inorganica Chimica Acta 480 (2018) 83–90
wide range of complexes with biological activities such as antibac-
terial, antifungal, antiviral, and antitumor activities have been
reported [22–29]. Novel synthetic strategies with the advances
made in coordination chemistry strongly influenced the design of
prospective lanthanide-based drugs with higher selectivity and
improved toxicological and pharmacokinetic profiles [30,31]. In
this context, we describe here the synthesis and characterization
of the sulfate complexes of La3+ and Nd3+ with the diquaternary
N,N’-diphenacyl-4,40 -dipyridinium dibromide (DPB) proligand
(Scheme 1). The antitumor activity of these complexes in cancer
cell lines was performed and compared with cisplatin, the metal-
based drug in clinical use, in order to allow their evaluation as
chemotherapeutic agents. The mechanisms involved in cell death
were also explored.
2. Experimental section
2.1. Materials and methods
All the chemicals and reagents were purchased from Sigma-
Aldrich Co and used without further purification. The N-heterocylic
diquaternary salt N,N’-diphenacyl-4,40 -dipyridinium dibromide
(DPB) was prepared by the previously published method [32]. Ln2(-
SO4)3 (Ln = La, Nd) was prepared by dissolving Ln2O3 in concen-
trated H2SO4. The solvents were used as supplied or distilled
using standard methods.
Elemental analyses (C, H, N, S) were performed in-house with
Fisons Instruments 1108 CHNS-O Elemental Analyzer. Before per-
forming the analytical characterization, all samples were dried in
vacuo (50 °C, 104 bar) until a constant weight was reached.
Melting points are uncorrected and were taken on an SMP3
Stuart instrument with a capillary apparatus.
The IR spectra were recorded from 4000 to 450 cm1 with a
Spectrum Two IR spectrometer from PerkinElmer. In the following,
the IR bands are classified as very weak (vw), weak (w), medium
(m), strong (s) and very strong (vs).
The 1H NMR spectrum of La-DPY was acquired at room temper-
ature in dimethyl sulfoxide (DMSO) with a VXR-300 Varian spec-
trometer operating at 400 MHz, using tetramethylsilane (TMS) as
internal standard. In the following, the 1H NMR signals are classi-
fied as singlet (s), doublet (d), triplet (t), quartet (q) and multiplet
(m). Due to the paramagnetism of Nd(III) ion, the 1H NMR spec-
trum of Nd-DPY was not acquired.
UV–vis spectra were recorded in the 200–900 nm range for both
the ligand and the two complexes at 105 M in ethanolic solutions
using a LAMBDA 950 spectrophotometer from PerkinElmer. For the
stability tests of the complexes, the UV–vis spectra, performed in
105 M DMSO solutions, were collected in a Shimadzu UV-1800
Spectrophotometer (range: 200–800 nm).
Thermogravimetric analyses (TGA) were carried out under a N2
flow from room temperature to 1200 °C and with a heating rate of
10 °C/min using a PerkinElmer STA 6000 Simultaneous Thermal
Analyzer.
Powder X-ray diffraction (PXRD) analyses were carried out in
the 5–90° 2h range on a Rigaku SmartLab X-ray diffractometer
using Cu Ka radiation (k = 0.154060 nm) in the general (Bragg
Brentano Focusing) type of measurements, operating at room tem-
perature. The generator was set at 45 kV and 200 mA.
Scheme 1. Structure of N,N’-diphenacyl-4,40 -dipyridinium dibromide (DPB).
Scanning electron microscopy (SEM) measurements were car-
ried out using a Hitachi SU 8230 Scanning Electron Microscope
equipped with a detector able to achieve the low angle backscat-
tering electrons (LA-BSE), for both morphological and composi-
tional contrast information.
2.2. Synthesis of Ln(III) complexes: General procedure
Complexes La-DPY and Nd-DPY were synthesized according to
the following procedure: in two 50 mL round bottom flasks, 0.277
g of N,N’-diphenacyl-4,40 -dipyridinium dibromide (0.5 mmol) were
dissolved in 20 mL of methanol under heating at 50–60 °C and stir-
ring until complete dissolution. Lanthanum(III) sulfate (0.283 g,
0.5 mmol) or neodymium(III) sulfate (0.288 g, 0.5 mmol) were
then added, and after several minutes of stirring, 1 mL of triethy-
lamine was added. Then, deep violet precipitates started to appear,
which were left under reflux and continuous stirring overnight.
After cooling, the solid products were filtered off, washed three
times with 5 mL of methanol and distilled water, and were finally
dried under vacuo for 24 h.
[{La(Et3N)(SO4)}2(l-DPY)(l4-SO4)] (La-DPY). Violet powder.
Yield: 85%. M.p. 200–205 °C dec., weakly soluble in methanol,
and dimethylformamide. Anal. Calc. for C38H50La2N4O14S3 (FW =
1160.83 g mol1): C, 39.32; H, 4.34; N, 4.83; S, 8.29%. Found: C,
39.15; H, 4.45; N, 4.62; S, 7.98%. IR m(cm1): 3100–3000 (vw),
3000–2800 (vw), 1580 (m), 1542 (m), 1487 (vs), 1454 (s), 1420
(vs), 1336 (s), 1281 (w), 1223 (w), 1190 (vs), 1086 (w), 1014 (m),
943 (vw), 889 (s), 845 (w), 818 (s), 785 (w), 694 (vs), 664 (w),
503 (m). 1H NMR (DMSO d6, 400 MHz): d, 1.56q, 3.28 t (30H, (CH3-
CH2)3N), 5.60 s (2H, CHmethine), 7.14–7.30 m (10H, CHphenyl), 7.60d
(J = 8.4 Hz, 4H, CHbipyridyl), 8.65d (J = 8.4 Hz, 4H, CHbipyridyl).
[{Nd(Et3N)(SO4)}2(l-DPY)(l4-SO4)] (Nd-DPY). Violet powder.
Yield: 81%. M.p. 200–205 °C dec., weakly soluble in methanol,
and dimethylformamide. Anal. Calc. for C38H50Nd2N4O14S3 (FW =
1171.50 g mol1): C, 38.96; H, 4.30; N, 4.78; S, 8.21%. Found: C,
38.75; H, 4.48; N, 4.57; S, 7.89%. IR m(cm1): 3100–3000 (vw),
3000–2800 (vw), 1580 (m), 1542 (m), 1487 (vs), 1454 (s), 1420
(vs), 1336 (s), 1281 (w), 1223 (w), 1190 (vs), 1086 (w), 1014 (m),
943 (vw), 889 (s), 845 (w), 818 (s), 785 (w), 694 (vs), 664 (w),
503 (m).
2.3. Biological assays
2.3.1. Cellular viability by the MTT assay
Human tumor cell lines, breast MCF7 and ovarian (cisplatin sen-
sitive) A2780, were cultured in RPMI (A2780) or DMEM culture
media (Gibco) supplemented with 10% fetal bovine serum (FBS)
and 1% antibiotics at 37 °C, 5% CO2 in a humidified atmosphere
(Heraeus, Germany). The cells were adherent in monolayers and
upon confluence were harvested by digestion with trypsin-EDTA
(Gibco). Cell viability was evaluated using the tetrazolium salt
MTT ([3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
mide]), which is reduced by a mitochondrial succinate dehydroge-
nase in metabolically active cells to insoluble purple formazan
crystals [33]. For the assays, cells were seeded in 96-well plates
at a density that ensures exponential growth of controls (untreated
cells) throughout the experiments. Cells (2-3  104 cells/200 lL
medium) were allowed to settle overnight followed by the addition
of dilution series of the compounds in fresh medium in aliquots of
200 lL per well. Ligand and complexes were first solubilized in
DMSO and then in medium, and added to final concentrations from
1 mM to 50 mM. The final concentration of DMSO in cell culture
medium did not exceed 1%. After continuous exposure to the com-
pounds for 24 h and 48 h at 37 °C/5% CO2, the medium was
removed, and the cells were incubated with 200 lL of MTT solution
in phosphate buffer saline (PBS) (0.5 mg/mL). After 3 h at 37 °C/5%
 A. Cârâc et al. / Inorganica Chimica Acta 480 (2018) 83–90
CO2, the solution was discarded, and the purple formazan crystals
formed inside the cells were dissolved in 200 lL DMSO. The cellu-
lar viability was evaluated by measuring the absorbance at 570 nm
on a plate spectrophotometer (PowerWave Xs, Bio-Tek Instru-
ments, Winooski, VT, USA). The cytotoxic effects of the compounds
were quantified by calculating the compound concentration that
inhibit tumor cell growth by 50% (IC50), based on non-linear regres-
sion analysis of dose response data (GraphPad Prism software vs
5.0). All compounds were tested in at least two independent exper-
iments, each comprising six replicates per concentration.
2.3.2. Apoptosis evaluation by Hoechst staining
A2780 cells were plated in a cell culture slide at a concentration
of 2  105 cells/mL and incubated for 48 h in culture medium in
the absence or presence of the compounds at their IC50. The
Hoechst staining method (Hoechst 33342, Thermo Fisher Scien-
tific) was used to detect apoptotic nuclei as described elsewhere
[34], with minor adaptations. Briefly, the culture medium was
removed, and the cells were washed thwice with PBS before fixa-
tion with 4% (v/v) paraformaldehyde (in PBS) for 15 min at room
temperature. After fixation, cells were washed thwice with PBS
and incubated with Hoechst dye 33342 at a 1:2000 dilution,
according to the manufacturer’s instructions. Following three
washing steps with PBS, the cells were mounted in anti-fade
mounting media (Vectashield H-100, Vector Laboratories, Burlin-
game, Canada). The slides were photographed under 20 (for
counting purposes) or 63 magnification in a Zeiss Axioplan2 flu-
orescence microscope. Several random microscopic fields per sam-
ple were analyzed, corresponding to at least 300 nuclei per sample,
in two independent experiments.
2.3.3. Reactive oxygen species by the NBT assay
A2780 cells were seeded in 96-well plates with a density of 2 
104 cells/200 mL. After 24 h, the medium was replaced with fresh
medium containing the compounds at selected concentrations.
ROS generation was detected by the nitroblue tetrazolium
(NBT) assay following a previously described method [35]. Briefly,
after 1 h pretreatment with the compounds, 20 lL of a NBT solu-
tion (10 mg/mL in water) (NBT = 2,20 -bis(4-Nitrophenyl)-5,50 -
diphenyl-3,30 -(3,30 -dimethoxy-4,40 -diphenylene)ditetrazolium
chloride) was added to the medium and incubation continued for
another 1 h at 37 °C. The NBT solution was discarded and the for-
mazan deposits were solubilized with 200 lL of 90% DMSO:10%
0.01 M NaOH plus 0.1% SDS. After careful homogenization of the
wells deposits, the absorbance was measured at 550 nm. All com-
pounds were tested in at least two independent experiments, each
comprising 4 replicates per concentration. Results (mean ± SD) are
expressed as % of controls (no treatment).
3. Results and discussion
3.1. Synthesis and characterization of complexes
The synthesis of the two new lanthanide(III) complexes, La-DPY
and Nd-DPY, has been conducted in methanol solution by the reac-
tion of Ln(III) sulfates with the diquaternary salt of 4,40 -dipyri-
dinium DPB, as proligand, in the presence of triethylamine, using
a 1:1 molar ratio. Worthy of note, the absence of a base, either
inorganic or organic, did not lead to the formation of any complex.
On the other hand, attempts to prepare Ln(III) complexes in the 1:2
or 1:3 molar ratio were not successful, presumably due to the
steric factors of both the Ln(III) ions and the ligands [36].
Both complexes were isolated in good yields as violet powders
and showed to be stable at air and moisture. The violet color of the
complexes was an indicator of the transformation in basic condi-
85
tions (Et3N) of DPB proligand (Scheme 2, A) into its reactive dipyri-
dinium ylide intermediate DPY (Scheme 2, B) which was in
resonance with its enolic form (Scheme 2, C) [37]. The resulting
DPY species actually acted as the ligand towards the complexation
with the two lanthanide(III) ions, making both of the complexes
colored violet.
A preliminary PXRD analysis of the two complexes revealed
their isostructurality (Fig. 1). Attempts to obtain single crystals
suitable for the X-ray structure determination of these complexes
were so far unsuccessful, and therefore, their crystal structure
remains presently unknown. However, based on several character-
ization techniques (IR, EA, 1H NMR, UV–vis, TGA), we were able to
propose a possible structure for the obtained lanthanide(III) com-
plexes. The elemental analysis was found to be consistent with
the general stoichiometric formula [{Ln(Et3N)(SO4)}2(l-DPY)(l4-
SO4)] (Ln = La, Nd) for both complexes.
The IR spectra of the complexes show medium and strong bands
in the region 1543–1420 cm1 assigned to the stretching vibra-
tions C@C and C@N double bonds from the aromatic rings. While
the IR spectrum of the DPB proligand presents a strong band at
1694 cm1, specific to the carbonyl (C@O) group, the IR spectra
of both complexes lack such a band, which indicates the transfor-
mation of DPB proligand into the ylide form DPY (Fig. 2). This
transformation is also supported by the appearance of a medium
band located at 1010 cm1 and another weak band at 1063 cm1,
which were assigned to the bending mode of vibration of the
methine AC@CH group and to the stretching vibration of the
CAO group resulting from the carbonyl group [38].
The IR spectrum of the free groups, i.e. ionic, sulfate (having the
Td symmetry) displayed very strong bands at 1104 cm1 and at
613 cm1, which were assigned to the m3 stretching [md(SO)] and
respectively, m4 bending [dd(OSO)] modes [39]. On the other hand,
for the IR spectrum of the two complexes, the bands observed in
the range 1275–1090 cm1 were ascribed to the m1 and m3 vibra-
tions modes of the sulfate ions in low symmetry, suggesting the
occurrence within the same complex of both the tetradentate coor-
dination of bridging sulfate group in a bis-chelate mode and the
bidentate coordination of sulfate group in a mono-chelate mode
[39]. This relatively rare occurrence was also observed, e.g., with
the Sm(III) complex (H2prz)[Sm2(HIDC)2(SO4)2] (HIDC = imida-
zole-4,5-dicarboxylic monoacid, H2prz = piperazine) for the bridg-
ing bis-chelate mode [40], or the La(III) complexes (H2en)2[La2M
(SO4)6(H2O)2] (M = Co, Ni) [41]. The intense absorption band
Scheme 2. Transformation route of the DPB proligand into the DPY ligand.
86 A. Cârâc et al. / Inorganica Chimica Acta 480 (2018) 83–90

6000 and two Et3N molecules per formula unit. Each Ln(III) ion is coor-
dinated by one oxygen atom from the DPY ligand, two oxygen
5000 atoms from one bidentate mono-chelate sulfate group, two oxygen
atoms from one tetradentate bridging sulfate group and one nitro-
4000 gen atom from Et3N, therefore assuming an octahedral geometry.
Intensity (cps)

The two Ln(III) ions, interconnected by the bridging tetradentate

3000
sulfate group, are closing a cycle with the exo-bidentate DPY
DPB ligand, and the resulting complexes are dinuclear. Additionally,
2000
the positive charges of the nitrogen atoms are compensated, within
1000
Nd-DPY the same structure of the complexes, by the negative charges of the
enolate oxygen atoms, as was also observed in a palladium(II)–
La-DPY
0 quaternary 2,20 -bipyridylium ylide complex reported by Dinica
10 20 30 40 50 60 70 80 90 et al. [43]. These six-coordinate Ln(III) complexes are thus coming
2 theta (deg.) to join the very scarce family of complexes containing species with
CN = 6 [44–47]. The higher coordination requirements of Ln(III)
Fig. 1. PXRD patterns of DPB (black), La-DPY (red) and Nd-DPY (blue). (For ions are a probable cause for the occurrence of dinuclear com-
interpretation of the references to color in this figure legend, the reader is referred plexes, for which the ‘large ionic radii – low coordination numbers’
to the web version of this article.)
relationship [45] could be strategically achieved by the use of the
bulky DPY ligand.
Thermogravimetric analysis (TGA) was employed to investigate
the thermal stabilities of the obtained Ln(III) complexes compared
Nd-DPY to the DPB proligand, along with the evaluation of the main weight
losses recorded upon heating from room temperature to 1200 °C
under a nitrogen atmosphere. Moreover, this analysis was used
as a complementary tool for the confirmation of the proposed
T (%)

structure of the obtained complexes. The TGA traces of both the

La-DPY
DPB proligand and the two complexes are collectively gathered
in Fig. 4. The free DPB proligand is stable up to 225 °C, temperature
at which a rapid decomposition was registered until a black car-
bonaceous residue was found. The decomposition of DPB proligand
DPB at such a high temperature is likely due to its quaternary ammo-
nium nature [48,49]. Both Ln(III) complexes showed a slightly
lower stability with respect to the DPB proligand, their decomposi-
4000 3500 3000 2500 2000 1500 1000 500 tion onset being established at about 205 °C. At this temperature, a
-1 rapid decomposition of the complexes began with a weight loss of
Wavenumber (cm ) ca. 18.5% for La-DPB (theoretical weight loss 17.4%) and ca. 16.5%
Fig. 2. IR spectra of DPB proligand (black), La-DPY (red) and Nd-DPY (blue). (For
for Nd-DPY (theoretical weight loss 17.3%), corresponding to the
interpretation of the references to color in this figure legend, the reader is referred evaporation of the coordinated Et3N molecules in both complexes.
to the web version of this article.) A progressive decomposition is continued, until 1200 °C, with the
complete evolution of three sulfate groups and one DPY ligand,
amounting to ca. 59.5% for La-DPB (theoretical weight loss
located at 817 cm1 can be attributed to the Ln–O vibration [42], 58.6%) and ca. 54.9% for Nd-DPY (theoretical weight loss 55.7%).
whereas the weak absorption band observed at 1225 cm1 was The final residues, which have remained at the end of the heating
assigned to the C–N stretching from coordinated triethylamine
molecules [39]. The 1H NMR spectrum of La-DPY also revealed
the presence of a quartet and a triplet at 1.56 and 3.28 ppm, which
were assigned to the protons from coordinated triethylamine
molecules.
Corroboration of the analytical data with the spectroscopic 100
techniques so far described allowed to propose the structure of
complexes La-DPY and Nd-DPY (Fig. 3), which contain one DPY 80
Weight loss (%)

ligand in the enolic form, two Ln(III) ions, three sulfate groups

60

La-DPY
40

Nd-DPY
20
DPB
0
0 200 400 600 800 1000 1200
Temperature (oC)
Fig. 4. TGA traces of DPB proligand (blue), La-DPY (red) and Nd-DPY (blue). (For
Fig. 3. Proposed structure of complexes [{Ln(Et3N)(SO4)}2(l-DPY)(l4-SO4)] (Ln = interpretation of the references to color in this figure legend, the reader is referred
La, Nd). to the web version of this article.)
 A. Cârâc et al. / Inorganica Chimica Acta 480 (2018) 83–90 87

process, were associated to the high melting point metallic lan-

thanum and neodymium.
The UV–vis absorption spectra of DPB proligand and the two
lanthanide(III) complexes in ethanol are shown in Fig. 5. The
DPB proligand exhibits a single absorption band at 260 nm, which
is assigned to intraligand (p–p⁄) charge-transfer transitions [50].
The related lanthanide(III) complexes exhibit more intense
absorption bands located at the same wavelength as DPB, and a
hyperchromic effect is therefore attained as a consequence of
metal complexation. The UV–vis spectra of the complexes also
show two weak and broad absorption bands in the visible region
which are likely due to the ligand field effect and could be
assigned to the metal-to-ligand charge transfer (MLCT) phe-
nomenon [51].
For the biological assays, the lanthanide complexes were ini-
tially dissolved in DMSO. In order to assess the stability of the com-
plexes in this medium, UV–vis absorption spectra were recorded at
room temperature at different time intervals. In DMSO solution,
the single absorption band of the DPB proligand is shifted to 351
nm (see Fig. S1). As shown in Fig. 6, the pattern of the spectra of
both lanthanide complexes in DMSO is similar to that observed
in ethanolic solution and reproducible over time, although the
three absorption bands observed in DMSO are slightly shifted rel-
atively to those observed in ethanolic solution. These findings led Fig. 6. Room temperature UV–vis spectra of lanthanide complexes, recorded on
us to conclude that the compounds are stable in DMSO. 105 M DMSO solutions at different time intervals: Nd-DPY, t = 0 (blue); Nd-DPY, t
= 48 h (pink), La-DPY, t = 0 (black); La-DPY, t = 48 h (red). (For interpretation of the
Scanning electron microscopy (SEM) was used to investigate
references to color in this figure legend, the reader is referred to the web version of
the morphology and particles size of the synthesized Ln(III) com- this article.)
plexes. Both complexes possess multiform crystal structures and
morphologies (elongated micro- and nano- particles, aggregated
micro- and nano- particles. Fig. 7 shows the SEM images of DPB
ligand, La-DPY and Nd-DPY complexes at different magnifications. texture. Moreover, quite a few spheroids, with diameter between
A comparison of the SEM images of covalently bonded complexes 10 nm and 40 nm, have emerged at the fracture surface, and we
and ligand shows that the former appears to be quite homogenous suggest that the ligand triggered the formation of this attractive
and uniform. The morphology of the La and Nd complexes reveals morphology. The first microstructure of microrods, with a length
aggregates of nanoparticles with dimensions varying from 10 to from 2 to 6 mm and width from 0.1 to 0.5 mm, can be distinguished
40 nm, and from 30 to 250 nm, respectively. in La-DPY and Nd-DPY, and it is associated with the crystal struc-
These micro- and nanoparticles might be formed by the coordi- ture of Ln2(SO4)3 (Ln = La and Nd), whereas the second one corre-
nation of N-heterocyclic bipyridine to the lanthanide(III) ions [52]. sponds to the secondary phase, agglomerates of the smaller size
Complexes with well-defined shape and dimensionality were particles with no well-defined shape (Fig. 8). These particles might
achieved and their morphology is determined mainly by the inter- be formed by the coordination of the DPY ligand to the lanthanide
action between lanthanide ions and the ligand [53]. The SEM ions under the experimental conditions. The chelating ligand was
images of these complexes also reveal information regarding their unable to coordinate to the same Ln3+ because of the limits of spa-
tial geometry. The coordination interaction therefore occurs inter-
molecularly, but not intramolecularly. The results highlight that
the confinement did not disturb the first coordination sphere of
1.0 the lanthanide ions.
La-DPY
Nd-DPY 3.2. Biological activity
0.8 DPB

3.2.1. Cellular viability

Absorbance

0.6 The cytotoxicity of the complexes was assessed in the human

tumor cells A2780 (ovarian cisplatin sensitive) and MCF7 (breast)
0.4 within the concentration range 1 mM–50 mM. The effects of the
compounds on the cellular viability were evaluated by the MTT
assay after continuous exposure for 24 h and 48 h. As can be
0.2 observed in Fig. 9 (Table S1 and Fig. S2), the cytotoxic activity of
DPB increased upon coordination to the two Ln(III) ions, especially
0.0 in A2780 cells. The lanthanide sulfate salts were also included in
this study and showed considerably lower activity than the corre-
200 400 600 800 sponding complexes in both cell lines (IC50 > 100 mM). The cyto-
toxic effect observed was higher in ovarian cells compared to
Wavelength (nm) breast cells, increasing drastically for longer incubation times
(48 h). The antitumor activity displayed by cisplatin in A2780 cells
Fig. 5. Room temperature UV–vis spectra of DPB (black), La-DPY (red) and Nd-DPY
(blue), recorded on 105 M ethanolic solutions. (For interpretation of the references
after 24 h incubation was of the same order of magnitude than the
to color in this figure legend, the reader is referred to the web version of this Nd-DPY complex (26.0 ± 5.0 vs 30 ± 11). In MCF7 cells, both com-
article.) plexes were even more active than cisplatin, in particular the
88 A. Cârâc et al. / Inorganica Chimica Acta 480 (2018) 83–90

Fig. 7. SEM images of DPB ligand (a), La-DPY (b), and Nd-DPY (c), at different magnifications (top 5kx, and bottom 25kx).

Fig. 8. SEM images of La-DPY (a) and Nd-DPY (b), showing the particle and rod dimensions.

3.2.2. Apoptosis evaluation

In order to elucidate the mechanisms that might underlie the
cytotoxicity of the complexes, the Hoechst nuclei staining assay
was used to estimate the level of apoptosis occurring in a cell pop-
100 ulation of A2780 cells exposed, or not, to the complexes. This assay
80
is based on the ability of Hoechst 33342 to permeate the cells and
stain the DNA, allowing the detection of nuclei exhibiting apoptotic
IC 50 (μM)

60 markers, such as condensation or fragmentation of chromatin and

Nd2(SO4)3
40
formation of apoptotic bodies [54]. The results obtained indicate
La2(SO4)3
DPB that both complexes can induce apoptosis in A2780 cells, as evi-
20 NdDPY denced by the slight increase in the number of cells displaying typ-
0
LaDPY ical apoptotic markers (Fig. 10 and Fig. S3). However, this effect is
not as marked as expected based on the fact that the IC50 concen-
tration for the complexes was used. This suggests that apoptosis is
Fig. 9. Cytotoxic activity measured as the half-inhibitory concentration (IC50) for
not the predominant pathway of cell death at play, and raises the
the complexes, the ligand and the lanthanide salts against MCF7 and A2780 cells question whether the cytotoxic effects induced by the complexes
after 24 h and 48 h incubation. are also dependent on other types of programmed cell death.

Nd-DPY complex. Compared with cisplatin, the widely used 3.2.3. ROS production
chemotherapy drug, and using the same experimental conditions, The NBT assay was used to evaluate the induction of ROS
both complexes presented important cytotoxic properties, in par- (superoxide anion) in A2780 cells exposed to the compounds. This
ticular in ovarian cancer cells. colorimetric assay is based on the reduction of the membrane
 A. Cârâc et al. / Inorganica Chimica Acta 480 (2018) 83–90 89

Control

10 μm

10 μm 10 μm

La-DPY Nd-DPY
Fig. 10. Staining of A2780 cells with Hoechst 33342 in the absence (control) or upon exposure to the complexes for visualization of apoptotic nuclei. The images shown are
representative of two independent experiments where at least 300 nuclei were counted per sample. The white arrows indicate the presence of chromatin condensation,
nuclear fragmentation, or apoptotic bodies, typical of apoptosis.

permeable yellow-colored nitroblue tetrazolium to an insoluble relation to controls (no treatment), slightly increasing for La-DPY
blue-product formazan by O–2. NBT conversion occurs intracellu- and Nd-DPY only at 50 mM. In these experimental conditions, the
larly through mitochondrial NAD(P)H oxidase [55]. As can be seen oxidative burst i.e., the rapid release of ROS (O–2), may not be the
from Fig. 11, the production of ROS is not considerably different in exclusive mechanism responsible for cellular cytotoxicity.

4. Conclusions

Two new lanthanide(III) complexes La-DPY and Nd-DPY were

Fig. 11. NBT reduction in A2780 cells in response to the compounds at selected
concentrations: 10, 20 and 50 mM after 2 h incubation.
synthesized through the reaction between Ln2(SO4)3 (Ln = La, Nd)
and the diquaternary bis(pyridinium) salt DPB, acting as proligand,
in alcoholic medium. The analytical and spectroscopic data
revealed the dinuclear structure of both complexes, also highlight-
ing the conversion of the DPB proligand into its ylide reactive inter-
mediate DPY ligand in basic conditions. The antitumor activity of
the complexes was evaluated in both ovarian and breast cancer
cells and was found to be superior to cisplatin, in particular for
Nd-DPY. However, induction of apoptotic cell death and produc-
tion of ROS do not seem to be the exclusive mechanisms responsi-
ble for cellular cytotoxicity, so these will be worthy of further
clarification in the future. In brief, we anticipate that these types
of Ln(III) complexes have potential and could be further explored
as antitumor agents.
90 A. Cârâc et al. / Inorganica Chimica Acta 480 (2018) 83–90

Acknowledgements [22] A. Srishailam, Y.P. Kumar, P.V. Reddy, N. Nambigari, U. Vuruputuri, S.S. Singh,
S. Satyanarayana, J. Photochem. Photobiol. B Biol. 132 (2014) 111.
[23] S. Rubino, P. Portanova, F. Giammalva, M.A. Girasolo, S. Orecchio, G. Calvaruso,
Andreea Cârâc gratefully acknowledge the support through G.C. Stocco, Inorg. Chim. Acta 370 (2011) 207.
POSDRU/159/1.5/S/138907-EXCELIS project. C2TN/IST authors [24] P. Martins, J. Jesus, S. Santos, L.R. Raposo, C. Roma-Rodrigues, P.V. Baptista, A.R.
Fernandes, Molecules 20 (2015) 16852.
gratefully acknowledge the FCT support through the UID/
[25] G. Liao, Z. Ye, Y. Liu, B. Fu, C. Fu, PeerJ. 5 (2017) e3252.
Multi/04349/2013 project. The authors also thank the helpful [26] Q.Y. Zhu, J. Dai, Coord. Chem. Rev. 330 (2017) 95.
assistance of Geanina Mihai, Cosmin Gheorghe, Raluca Mesterca, [27] C. Saturnino, I. Barone, D. Iacopetta, A. Mariconda, M.S. Sinicropi, C. Rosano, A.
Campana, S. Catalano, P. Longo, S. Andò, Futu. Med. Chem. 8 (2016) 2213.
Oana Lazar and Aida Pantazi, PhD students from CSSNT-UPB.
[28] E. Sirignano, A. Pisano, A. Caruso, C. Saturnino, M.S. Sinicropi, R. Lappano, A.
Professors Claudio Pettinari and Fabio Marchetti from the Botta, D. Iacopetta, M. Maggiolini, P. Longo, Anti-Cancer Agent. Med. Chem. 15
University of Camerino are also acknowledged for helpful (2015) 468.
experimental assistance. [29] A. Caporale, G. Palma, A. Mariconda, V. Del Vecchio, D. Iacopetta, O.I. Parisi, M.
S. Sinicropi, F. Puoci, C. Arra, P. Longo, C. Saturnino, Molecules 22 (2017) 526.
[30] K. Wang, Y. Cheng, X. Yang, R. Li, Met. Ions Biol. Syst. 40 (2003) 709.
[31] D.S. Kopchuk, A.P. Krinochkin, D.N. Kozhevnikov, P.A. Slepukhin, Polyhedron
Appendix A. Supplementary data 118 (2016) 30.
[32] R.M. Dinica, I.I. Druta, C. Pettinari, Synlett 7 (2000) 1013.
Supplementary data associated with this article can be found, in [33] M.V. Berridge, P.M. Herst, A.S. Tan, Biotechnol. Annu. Rev. 11 (2005) 127.
[34] O.A. Lenis-Rojas, C. Roma-Rodrigues, A.R. Fernandes, F. Marques, D. Pérez-
the online version, at https://doi.org/10.1016/j.ica.2018.05.003. Fernández, J. Guerra-Varela, L. Sánchez, D. Vázquez-García, M. López-Torres, A.
J.J. Fernández, Inorg. Chem. 56 (2017) 7127.
[35] J. Lopes, D. Alves, T.S. Morais, P.J. Costa, M.F. Piedade, F. Marques, M.J. Villa de
References Brito, M.H. Garcia, J. Inorg. Biochem. 169 (2017) 68.
[36] S.A. Cotton, Lanthanide and Actinide Chemistry, John Wiley & Sons, West
[1] Jean-Claude G. Bünzli, J. Coord. Chem. 67 (2014) 3706. Sussex, 2006.
[2] S.A. Cotton, P.R. Raithby, Coord. Chem. Rev. 340 (2017) 220. [37] I.I. Druta, R.M. Dinica, E. Bacu, I. Humelnicu, Tetrahedron 54 (1998) 10811.
[3] M. Komiyama, Chem. Lett. 45 (2016) 1347. [38] R.M. Silverstein, F.X. Webster, D.J. Kiemle, D.L. Bryce, Spectrometric
[4] J. Vuojola, T. Soukka, Methods Appl. Fluoresc. 2 (2014) 012001. Identification of Organic Compounds, 8th Ed., John Wiley & Sons, Hoboken,
[5] R.W. Mewis, S.J. Archibald, Coord. Chem. Rev. 254 (2010) 1686. New Jersey, 2014.
[6] E.S. Andreiadis, D. Imbert, J. Pécaut, R. Demadrille, M. Mazzanti, Dalton Trans. [39] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
41 (2012) 1268. Compounds, 6th Ed., John Wiley & Sons, Hoboken, New Jersey, 2009.
[7] Y. Chen, R. Guan, C. Zhang, J. Huang, L. Ji, H. Chao, Coord. Chem. Rev. 310 (2016) [40] W.-G. Lu, D.-C. Zhong, L. Jiang, T.-B. Lu, Cryst. Growth Des. 12 (2012) 3675.
16. [41] Y.-P. Yuan, R.-Y. Wang, D.-Y. Kong, J.-G. Mao, A. Clearfield, J. Solid State Chem.
[8] S. Shinoda, Chem. Soc. Rev. 42 (2013) 1825. 178 (2005) 2030.
[9] J.C. Wei, W. Ye, R. Ning, G. Li-Na, Z. Jian-Jun, J. Chem. Thermodynam. 103 [42] Y. Xing, Z. Shi, G. Li, W. Pang, Dalton Trans. 5 (2003) 940.
(2016) 181. [43] R.M. Dinica, F. Marchetti, C. Pettinari, B.W. Skelton, A.H. White, Inorg. Chim.
[10] C. Zhao, Y. Sun, J. Ren, X. Qu, Inorg. Chim. Acta 452 (2016) 50. Acta 360 (2007) 2609.
[11] A. Tăbăcaru, B. Furdui, I.O. Ghinea, G. Cârâc, R.M. Dinică, Inorg. Chim. Acta 455 [44] S. Liu, L.-W. Yang, S.J. Rettig, C. Orvig, Inorg. Chem. 32 (1993) 2773.
(2017) 329. [45] K. Dehnicke, A. Greiner, Angew. Chem. Int. Ed. 42 (2003) 1340.
[12] J. Zhang, Y. Li, X. Hao, Q. Zhang, K. Yang, L. Li, L. Ma, S. Wang, X. Li, Mini-Rev. [46] S.A. Cotton, Compt. Rend. Chim. 8 (2005) 129.
Med. Chem. 11 (2011) 678. [47] B. Na, X.-J. Zhang, W. Shi, Y.-Q. Zhang, B.-W. Wang, C. Gao, S. Gao, P. Cheng,
[13] H.L. Wu, G.L. Pan, H. Wang, X.L. Wang, Y.C. Bai, Y.H. Zhang, J. Photochem. Chem. Eur. J. 20 (2014) 15975.
Photobiol. B Biol. 135 (2014) 33. [48] B. Furdui, R.M. Dinica, A. Tabacaru, C. Pettinari, Tetrahedron 68 (2012) 6164.
[14] B.R. Karreddula, S. Akkili, R.K. Hussain, J. Iran. Chem. Soc. 12 (2015) 1473. [49] B. Furdui, O. Constantin, A. Tabacaru, R.M. Dinica, Rev. Chim. 63 (2012) 667.
[15] N.C. Martinez-Gomez, H.N. Vu, E. Skovran, Inorg. Chem. 55 (2016) 10083. [50] H.H. Perkampus, UV-Vis Spectroscopy and Its Applications, Springer-Verlag,
[16] A. Hussain, S. Gadadhar, T.K. Goswami, A.A. Karande, A.R. Chakravarty, Eur. J. Berlin, 1992.
Med. Chem. 50 (2012) 319. [51] M.K. Nazeeruddin, S.M. Zakeeruddin, R. Humphry-Baker, M. Jirousek, P. Liska,
[17] A.P. Krinochkin, D.S. Kopchuk, D.N. Kozhevnikov, Polyhedron 102 (2015) 556. N. Vlachopoulos, V. Shklover, C.-H. Fischer, M. Grätzel, Inorg. Chem. 38 (1999)
[18] J.C.G. Bünzli, S.V. Eliseeva, Chem. Sci. 4 (2013) 1939. 6298.
[19] A.R. Ramya, D. Sharma, S. Natarajan, M.L.P. Reddy, Inorg. Chem. 51 (2012) [52] Y. Zhang, D. Fang, R. Liu, S. Zhao, X. Xiao, C. Wang, Y. Cao, W. Xu, Dye Pigment.
8818. 130 (2016) 129.
[20] A. Datta, K. Das, S. Mendiratta, C. Massera, E. Garribba, J.-H. Huang, S.B. Mane, [53] K. Mohanan, N. Subhadrambika, R. Selwin Joseyphus, S.S. Swathy, V.P. Nisha, J.
C. Sinha, M.S.E. Fallah, Inorg. Chim. Acta 453 (2016) 128. Saudi Chem. Soc. 29 (2016) 379.
[21] L. Armelao, D.B.D. Amico, L. Bellucci, G. Bottaro, L. Labella, F. Marchetti, S. [54] N. Atale, S. Gupta, U.C.S. Yadav, V. Rani, J. Microsc. 255 (2014) 7.
Samaritani, Polyhedron 119 (2016) 371. [55] A. Agarwal, A. Majzoub, Indian, J. Urol. 33 (2017) 199.