Study of the anticancer properties of methyl- and phenyl-substituted carbon- and silicon-bridged ansa-titanocene complexes

Article abstract of DOI:10.1016/j.jorganchem.2013.07.059

The previously known complexes [Ti{(Me₂CMe₂C) (η⁵-C₅H₄)₂}Cl₂] (1), [Ti{Me₂C(η⁵-C₅H₄) ₂}Cl₂] (2), [Ti {Me₂

Full text of DOI:10.1016/j.jorganchem.2013.07.059

Journal of Organometallic Chemistry xxx (2013) 1e7
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Study of the anticancer properties of methyl- and phenyl-substituted
carbon- and silicon-bridged ansa-titanocene complexes
ꢀ a
ꢀ a
ꢀ a
ꢀ a
ꢁ ꢀ
ꢁ ꢀ a
Sanja Mijatovic , Mirna Bulatovic , Marija Mojic , Stanislava Stosic-Grujicic ,
b,
c
ꢀ a
ꢀ
ꢁ
*
Djordje Miljkovic , Danijela Maksimovic-Ivanic , Santiago Gómez-Ruiz , Jirí Pinkas ,
c
,
ꢀ d
**
ꢁ
Michal Horácek , Goran N. KaluCerovic
a Institute for Biological Research “Sinisa Stankovic”, University of Belgrade, Bulevar despota Stefana 142, 11060 Belgrade, Serbia
b Departamento de Química Inorgánica y Analítica, E.S.C.E.T., Universidad Rey Juan Carlos, 28933 Móstoles, Madrid, Spain
c
ꢁ
J. Heyrovsky Institute of Physical Chemistry of Academy of Sciences of the Czech Republic, v.v.i., Dolejskova 3, 182 23 Prague 8, Czech Republic
d Institut für Chemie, Martin-Luther-Universität Halle-Wittenberg, Kurt-Mothes-Straße 2, 06120 Halle, Germany
a r t i c l e i n f o
a b s t r a c t
5-C5H4)2}Cl2] (1), [Ti{Me2C(
5-C5Me4)2}Cl2] (6) have been
5-C5H4)2}Cl2] (3) has been
h
5-C5H4)2}Cl2] (2), [Ti
Article history:
The previously known complexes [Ti{(Me2CMe2C)(
{Me2Si(
prepared following reported procedures. The novel complex [Ti{MePhC(
h
h h h
5-C5H4)2}Cl2] (4), [Ti{MePhSi( 5-C5H4)2}Cl2] (5) and [Ti{MePhSi(
Received 27 May 2013
Received in revised form
5 July 2013
h
prepared and characterized. The cytotoxic activity of 1e6 has been tested after 72 h on melanoma A375
Accepted 22 July 2013
and B16, prostate cancer DU145 and LNCaP and colon cancer HCT116, SW620 and CT26CL25 cell lines
observing a high cytotoxic activity of complexes 1 and 6 compared to the reference compound ([Ti(h5
-
Keywords:
Titanocene derivatives
Cytotoxicity
C5H5)2}Cl2]). 1 and 6 have also been tested against primary normal mouse keratinocytes and lung fi-
broblasts. While viability of both type of primary cells was significantly less affected by 1 in comparison
to the reference compound [Ti(h
5-C5H5)2Cl2], compound 6 was completely nontoxic for nonmalignant
Primary cells
cells, indicating a potential selectivity of this compound towards cancer cell lines. In addition CFSE
staining, cell cycle analysis, AnnexinV-FITC/PI staining, detection of caspase activity and mitochondrial
potential showed that 1 and 6 were acting through inhibition of proliferation and subsequent induction
of mitochondrial dependent apoptosis in colon cancer cell lines, HCT116 and SW620, which express low
sensitivity to cisplatin. Compound 6 was found to be the leading drug in this group since it shows the
fastest and most selective anticancer profile.
Cell cycle analysis
Caspase detection
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
patients with breast metastatic carcinoma [7] and advanced renal
cell carcinoma [8], which showed a low activity discouraging
Titanocene derivatives are together with ferrocene complexes
the most studied metallocenes in preclinical trials in the treatment
of cancer [1]. The starting point of this research field was the work
of Köpf and Köpf-Maier which showed the cytotoxic properties of
further studies.
However, after the recent work of many different groups
worldwide, the interest in this topic has been renewed [9e13].
Thus, aminoacid-functionalized titanocene derivatives [14,15],
benzyl-substituted titanocene or ansa-titanocene complexes [13],
ionic titanocene compounds [16,17], alkylammonium-substituted
titanocene derivatives [18e20], steroid-functionalized titanocenes
[21], and alkyl-, aryl- and/or alkenyl-substituted titanocene or
ansa-titanocene derivatives [22e25] have been synthesized, char-
acterized and studied in preclinical trials against different types of
human cancer cell lines and primary cells.
titanocene dichloride, [Ti(h
5-C5H5)2Cl2] and some of its analogs in
the early 1980’s [2]. Titanocene dichloride was studied in phase I
clinical trials in 1993 [3e5] and some water soluble formulations
were subsequently developed by Medac GmbH (Germany) [6].
Phase I clinical trials were not as satisfactory as expected, but the
research in the topic continued with some phase II clinical trials in
In general, the cytotoxic activity of titanocene complexes has
been correlated to their structure, and the study of the anticancer
mechanism of titanocene derivatives is currently a very active
research field. Several studies have led to the proposal that after
hydrolysis of titanocene complexes, titanium ions reach the cell
* Corresponding author. Tel.: þ34 914888507; fax: þ34 914888143.
** Corresponding author. Fax: þ49 345 5527028.
E-mail
addresses:
santiago.gomez@urjc.es,
santigomezruiz@gmail.com
ꢀ
(S. Gómez-Ruiz), goran.kaluderovic@chemie.uni-halle.de (G.N. KaluCerovic).
0022-328X/$ e see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2013.07.059
ꢀ
Please cite this article in press as: S. Mijatovic, et al., Journal of Organometallic Chemistry (2013), http://dx.doi.org/10.1016/
j.jorganchem.2013.07.059

ꢀ
S. Mijatovic et al. / Journal of Organometallic Chemistry xxx (2013) 1e7
2
cytoplasm assisted by the major iron transport protein “transferrin”
[26e29]. Inside the cell, titanium ions are transported to the nu-
cleus (probably assisted by ATP) and then bind to DNA and lead to
cell death [30e32]. However, recent experiments have also re-
ported on the potential interaction of a ligand-bound Ti(IV) com-
plex to other proteins [33e35], which may also be implicated in the
induction of cell death.
There are several reports on the cytotoxicity of titanocene de-
rivatives with different substituents attached to Cp rings [1,9,13],
but not many studies on the mechanistic properties of alkyl- or
alkenyl-substituted silicon- or carbon-bridged ansa-titanocene
derivatives [22,23]. With the aim to contribute to the understand-
ing of the mechanism of cancer cell death promoted by silicon- or
carbon-bridged ansa-titanocene derivatives with simple alkyl or
aryl substituents, a complete study which consists of the synthesis,
characterization, cell cycle analysis, AnnexinV-FITC/PI staining and
caspase detection has been carried out and is reported here.
was added dropwise during 30 min. The resulting redeorange
mixture was stirred for 14 h and then refluxed for 5 h. The dark
redeorange solution was separated and residual NaOH suspension
was washed twice with 120 mL of diethyl ether. Collected organic
phases were combined and the resulting solution was washed
subsequently with an NH4Cl solution, brine and finally with water
to neutral reaction and dried over Na2SO4. The solution was filtered,
volatiles were evaporated on rotary evaporator and the redeorange
oily residue was fractionally distilled at reduced pressure (0.01 mm
Hg). The first fraction up to 80 ꢀC contained almost exclusively 6-
methyl-6-phenylfulvene as was determined by GCeMS and 1H
NMR spectrometry. The fraction collected at 118e125 ꢀC contained
mainly the desired product in ca 90% purity besides ca. 10% of 6-
methyl-6-phenylfulvene as was determined by GCeMS. The re-
sidual 6-methyl-6-phenylfulvene was distilled off from the mixture
at reduced pressure (0.01 mm Hg) using a boiling water bath,
leaving product as an orange oily liquid. Yield 39.1 g (24%). For this
synthesis, the temperature of the heating bath should be kept
below 130 ꢀC during distillation, otherwise the polymerization of
2. Experimental section
the crude product could be initiated. Elemental analysis C18H18
,
calculated C, 92.26; H, 7.74%, found C, 92.35; H 7.80%. GCeMS, m/z
(relative abundance): 235 (19), 234 (Mþꢁ; 100), 219 ([M ꢂ Me]þ;
77), 205 (23), 204 (45), 203 (36), 202 (25), 191 (35), 178 (33), 169
(25), 168 (21), 165 (22), 154 (24), 153 (30), 152 (26), 142 (19), 141
(30), 128 (21), 115 (30), 91 (25), 77 (11). After the preparation of
2.1. General manipulations
All reactions were performed using standard Schlenk tube
techniques in an atmosphere of dry argon. Solvents were distilled
from the appropriate drying agents and degassed before use. [Ti
MePhC(C5H5)2 the synthesis of [Ti{MePhC(
carried out according to Scheme 2.
h
5-C5H4)2}Cl2] (3) was
{(Me2CMe2C)(
[37], [Ti{Me2Si(
(5) [39] and [Ti{MePhSi(
according to previously described synthetic methods. Reference
h h
5-C5H4)2}Cl2] (1) [36], [Ti{Me2C(
5-C5H4)2}Cl2] (2)
5-C5H4)2}Cl2]
5-C5Me4)2}Cl2] (6) [39] were prepared
h
5-C5H4)2}Cl2] (4) [38], [Ti{MePhSi(
h
LiBun (28.8 mL, 2.5 M, 72 mmol) was dropped to a solution of the
ligand MePhC(C5H5)2 (8.42 g, 36 mmol) in THF (100 mL). The
resulting redeorange mixture was stirred for 3 h and then slowly
transferred to another Schlenk vessel containing cold (ꢂ75 ꢀC)
suspension of TiCl4(THF)2 (prepared in situ by reaction of TiCl4
(4.0 mL, 36 mmol) with 6 mL of THF) in 100 mL of toluene. The
mixture was allowed to warm to room temperature and then
heated to 65 ꢀC for 16 h. The resulting dark red mixture was
evaporated almost to dryness and treated with 200 mL of hexane.
The formed brown precipitate was isolated, washed with hexane
(3 ꢃ 60 mL) and dried on air. The solid was extracted in boiling
dichloromethane in a Soxhlet extractor and then recrystallized
from hexane. The title complex was obtained as a dark brown
crystalline solid. Yield: 0.8 g (6%). Elemental analysis C18H16Cl2Ti,
calculated C, 61.58; H, 4.59%, found C, 61.39; H 4.51%. M.p.168 ꢀC. 1H
NMR (300 MHz, CDCl3): 1.99 (s, 3H, CMe); 5.52e5.57 (m, 2H, C(5)H,
C5H4); 5.74e5.78 (m, 2H, C(2)H, C5H4); 6.98e7.05 (m, 4H, C(3)H and
h
compound titanocene dichloride ([Ti(h
5-C5H5)2}Cl2] (Ref)) was
purchased from SigmaeAldrich (Spain).
Acetophenone, NaOH, NH4Cl, Na2SO4, LiBun (solution 2.5 M in
hexanes) and TiCl4 were purchased from Aldrich. All the commercial
reagents were used directly without further purification. IR spectra
were recorded on a Thermo Nicolet Avatar 330 FT-IR spectropho-
tometer. 1H and 13C{1H} NMR spectra were recorded on a Varian
Mercury 300 spectrometer in CDCl3 solutions at 298 K. Chemical
shifts (d/ppm) are given relative to solvent signal (dH: 7.26 ppm; dC:
77.16 ppm). 2D NMR experiments such as 1He1H COSY and 13Ce1H
HSQC have been carried out for the correct assignation of the signals.
Microanalyses were carried out with a PerkineElmer 2400 or LECO
CHNS-932 microanalyzer. Mass spectroscopic analyses were per-
formed on a VG7070E spectrometer at 70 eV.
3
4
2.2. Synthesis of [Ti{MePhC(h
5-C5H4)2}Cl2] (3)
C(4)H, C5H4); 7.38 (tt, JHH ¼ 7.5 Hz, JHH ¼ 1.2 Hz, 1H, CHpara, Ph);
7.44e7.52 (m, 2H, CHmeta, Ph); 7.61e7.67 (m, 2H, CHortho, Ph). 13C
{1H}(CDCl3): 27.24 (CMe); 48.49 (CMe); 112.94 (C(2), C5H4); 113.14
(C(5), C5H4); 114.98 (C(1), C5H4); 128.07 (CHortho, Ph); 128.11
(CHpara, Ph); 129.56 (CHmeta, Ph); 129.73 (C(3), C5H4); 133.48 (C(4),
C5H4). EI-MS, m/z (relative abundance): 354 (16), 353 (19), 352 (73),
The first step for the preparation of compound 3 is the synthesis
of MePhC(C5H5)2 according to Scheme 1.
The ligand was prepared from cyclopentadiene and acetophe-
none in the presence of NaOH as a base, according to a reported
method [40].
The freshly cracked cyclopentadiene (93.0 g, 1.4 mol) was added
to a suspension of NaOH (112.0 g, 2.8 mol) in 400 mL of THF. The
mixture was stirred for 1 h and then acetophenone (84.0 g, 0.7 mol)
ꢁ
351 (32), 350 (Mþ , 100), 316 (40), 315 (46), 314 ([M ꢂ HCl]þ, 88),
313 (18), 299 ([M ꢂ HCl ꢂ Me]þ, 12), 298 (10), 280 (12), 279 (23),
278 ([M ꢂ 2HCl]þ, 83), 277 (18), 276 (20), 263 ([M ꢂ 2HCl ꢂ Me]þ,
15), 216 (11), 215 (30), 165 (18), 153 (15), 152 (22), 147 (30), 139 (20),
Scheme 1. Synthesis of the carbon-bridged bis-cyclopentadienyl ligand.
ꢀ
Please cite this article in press as: S. Mijatovic, et al., Journal of Organometallic Chemistry (2013), http://dx.doi.org/10.1016/
j.jorganchem.2013.07.059

ꢀ
S. Mijatovic et al. / Journal of Organometallic Chemistry xxx (2013) 1e7
3
2.5. Preparation of adult mouse lung fibroblasts
Lungs were aseptically removed, rinsed in PBS and cut into small
pieces. Tissue was then digested in collagenase IV from C. Hys-
toliticum (0.7 mg/mL) þ DNase (30 mg/mL) with slow stirring for
1 h at 37 ꢀC. Cells were washed three times and resuspended in
RPMI-15% FCS. After 24 h non-adherent cell were discarded and
medium was replaced. Cells that reached 80e90% of confluence
were used for experiments.
2.6. Determination of cell viability by MTT and crystal violet assay
Fig. 1. Titanocene complexes used in this study.
The viability of adherent viable cells was evaluated with crystal
violet (CV) assay, while mitochondrial dehydrogenase activity, as
indicator of cell viability, was used for non-adherent cells. Cells
(5 ꢃ 103/well) were treated with the wide range of doses of the
drugs for 72 h and viability was measured as described [41]. In brief,
at the end of cultivation period adherent cells were fixed with
methanol 10 min at RT and subsequently stained 15 min with 1% CV
solution. Then, the cells were washed, dried and the dye was dis-
solved in 33% acetic acid. The absorbance of dissolved dye was
measured at 540 nm with the reference wavelength at 640 nm. For
determination of viability of non-adherent cells, MTT was added to
cell cultures in a final concentration of 0.5 mg/mL. Cells were
collected after 2 h, centrifuged and pellets were dissolved in DMSO.
The absorbance that corresponds to the level of MTT reduction to
formazan, was measured at 540 nm with the reference wavelength
at 640 nm. Results are calculated as percentage of control that was
arbitrarily set to 100%. All tests were carried out in triplicate cul-
tures and repeated three times. IC50 value was calculated using the
following formula:
115 (18). IR (KBr): 3103 (w); 3088 (m); 3056 (vw); 3027 (vw); 2976
(m); 2933 (m); 2873 (vw); 1558 (vw); 1541 (vw); 1494 (m); 1474
(w); 1464 (w); 1448 (m); 1417 (m); 1378 (w); 1276 (m); 1243 (w);
1105 (m); 1075 (w); 1040 (w); 1025 (m); 918 (w); 815 (vs); 767 (s);
727 (s); 703 (vs); 633 (vw); 594 (vw); 574 (m); 463 (w); 427 (w).
2.3. Reagents, cells and animals
Fetal calf serum (FCS), RPMI-1640, phosphate-buffered saline
(PBS), dimethyl sulfoxide (DMSO), 3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide (MTT), collagenase IV from Clos-
tridium Hystoliticum, and propidium iodide (PI) were obtained from
Sigma (St. Louis, MO). AnnexinV-FITC (AnnV) was from Biotium
(Hayward, CA). Acridin orange (AO) was from Labo-Moderna (Paris,
France). The B16 murine melanoma was kind provided by Dr. Sinisa
Radulovic (Institute for Oncology and Radiology of Serbia, Belgrade,
Serbia) while human melanoma A375, colon cancer HCT116 and
ð50% ꢂ Low Inh%Þ
IC50
¼
þ Low Conc
ðHigh Inh% ꢂ Low Inh%Þ ꢃ ðHigh Conc ꢂ Low ConcÞ
SW620, prostate LNCaP and DU145 and mouse colon CT26CL25
were kindly provided by Prof. Ferdinando Nicoletti (Department of
Biomedical Sciences, University of Catania, Italy).
where Low Inh%/High Inh% is % inhibition directly below/above 50%
inhibition and Low Conc/High Conc is the corresponding concen-
trations of test compound, presented as mean ꢄ SD from three
independent experiments.
Cells are routinely maintained in HEPES-buffered RPMI-1640
medium supplemented with 10% FCS, 2 mM L-glutamine, 0.01%
sodium pyruvate, and antibiotics (culture medium) at 37 ꢀC in a
humidified atmosphere with 5% CO2. After standard trypsinization,
cells were seeded at 5 ꢃ 103/well in 96-well plates for viability
determination and 2.5 ꢃ 105/well in 6-well plate for flow
cytometry.
For isolation of fibroblasts and keratinocytes, 2.5 month old
C57BL6 female mices were used. Animals were bred and kept in our
own facility of the Institute for Biological Research “Sinisa Stan-
kovic” (Belgrade, Serbia) under standard laboratory conditions
(nonspecific pathogen free) with free access to food and water. All
procedures were approved by the local Institutional Animal Care
and Use Committee.
2.7. Cell cycle analysis
Cells (2.5 ꢃ 105/well) were treated with IC50 dose (87 and 98
mM
of 1, 75 and 68 mM of 6 and 142 and 200 mM of Ref for SW620 and
HCT116 cells, respectively) for 24 h, then trypsinized and fixed in
70% ethanol at 4 ꢀC for 30 min. After washing in PBS, cells were
incubated with PI (20 mg/mL) and RNase (0.1 mg/mL) for 30 min at
37 ꢀC in dark. Red fluorescence was analyzed with FACS Calibur
2.4. Preparation of keratinocytes
Mouse ears were split into two pieces and exposed to 0.5%
trypsin-PBS solution for 1 h at 37 ꢀC then epidermis was taken and
cut into small pieces. Cells were resuspended and filtered through
nylon meshes. Cells were finally resuspended into RPMI-15% FCS.
Scheme 2. Synthesis of titanocene complex 3.
ꢀ
Please cite this article in press as: S. Mijatovic, et al., Journal of Organometallic Chemistry (2013), http://dx.doi.org/10.1016/
j.jorganchem.2013.07.059

ꢀ
S. Mijatovic et al. / Journal of Organometallic Chemistry xxx (2013) 1e7
4
Table 1
IC50
[
mM] values of titanium(IV) complexes 1e6 and reference complex, Ref ¼ [Ti(
h
5-C5H5)2Cl2] (72 h of action against selected tumor cell lines). IC50 values were calculated as
mean ꢄ SD from three independent experiments.
Cell line/Compound
Ref
1
2
3
4
5
6
A375
B16
161 ꢄ 1
>200
>200
141 ꢄ 1
154 ꢄ 1
163 ꢄ 36
>200
124 ꢄ 36
86 ꢄ 7
152 ꢄ 7
182 ꢄ 1
148 ꢄ 1
>200
127 ꢄ 15
178 ꢄ 19
144 ꢄ 25
132 ꢄ 18
148 ꢄ 24
117 ꢄ 43
163 ꢄ 11
170 ꢄ 17
>200
>200
158 ꢄ 1
163 ꢄ 1
142 ꢄ 1
197 ꢄ 1
181 ꢄ 9
>200
160 ꢄ 1
199 ꢄ 2
>200
105 ꢄ 29
43 ꢄ 4
HCT116
SW620
CT26CL25
DU145
LnCap
98 ꢄ 19
87 ꢄ 2
68 ꢄ 6
75 ꢄ 1
119 ꢄ 3
93 ꢄ 34
100 ꢄ 20
>200
62 ꢄ 13
83 ꢄ 24
66 ꢄ 19
175 ꢄ 1
156 ꢄ 1
>200
>200
flow cytometer (BD, Heildelberg, Germany). The distribution of
cells in different cell cycle phases was determined with Cell Quest
Pro software (BD).
ligand MePhC(C5H5)2 was carried out according to literature pro-
cedures (Scheme 1). This ligand was di-lithiated by treatment with
two equivalents of LiBun and subsequently reacted with one
equivalent of [TiCl4(THF)2] (Scheme 2).
The NMR, mass and IR spectra and elemental analysis showed
that all the complexes (1e6), isolated as crystalline solids, which
were of high purity.
2.8. AnnexinV-FITC/PI staining and caspase detection
Cells (2.5 ꢃ 105/well) were treated with IC50 dose of each
For the new complex [Ti{MePhC(h
5-C5H4)2}Cl2] (3) the 1H NMR
compound (87 and 98 mM of 1, 75 and 68 mM of 6 and 142 and
spectrum shows three multiplets between 5.52 and 7.05 ppm with
1:1:2 intensity ratio which correspond to the four magnetically
unequivalent protons of the cyclopentadienyl rings. In addition, a
singlet at 1.99 ppm (corresponding to the protons of the methyl
group of the bridge) and three multiplets between 7.38 and
7.67 ppm (assigned to the protons of the phenyl group) were also
observed.
200 mM of Ref for SW620 and HCT116 cells, respectively) for 48 or
72 h, then trypsinized and stained with AnnV-FITC/PI (Biotium,
Hayward, CA) or apostat (R&D Systems, Minneapolis, MN USA)
according to the manufacturer’s instructions. Cells were analyzed
with FACS Calibur flow cytometer (BD, Heidelberg, Germany) using
Cell Quest Pro software (BD).
The 13C{1H} NMR spectrum shows the expected signals for the
different carbon atoms of the complex (see Experimental section
2.2). Characterization of 3 by mass spectrometry shows the mo-
lecular ion peak as the base peak of the spectrum.
2.9. CFSE staining
Cells were stained with 1
ester (CFSE) for 10 min at 37 ꢀC, then washed and treated with IC50
dose of each compound (87 and 98 M of 1, 75 and 68 M of 6 and
142 and 200 M of Ref for SW620 and HCT116 cells, respectively)
mM of carboxyfluorescein succinimidyl
m
m
m
3.2. Biological studies
for 72 h. At the end of cultivation cells were trypsinazed, washed
and analyzed with FACS Calibur flow cytometer.
3.2.1. Effect of titanium complexes on tumor cell viability
Antitumor potential of titanium complexes (1e6) and the
reference compound [Ti(h
5-C5H5)2Cl2] (Ref) on melanoma A375
2.10. JC-1 staining
and B16, prostate cancer DU145 and LNCaP and colon cancer
SW620, HCT116 and CT26CL25 cell lines was evaluated by the
Cells (2.5 ꢃ 105/well) were treated with IC50 dose of each
treatment with [Ti(h
5-C5H5)2Cl2] and its analogs (1e6). Cell
compound (87 and 98 mM of 1, 75 and 68 mM of 6 and 142 and
200 mM of Ref for SW620 and HCT116 cells, respectively) for 72 h,
viability was determined by CV after 72 h of treatment.
The obtained results (Table 1) clearly indicate that two of the
studied compounds, namely 1 and 6, showed improved antitumor
potential in comparison to the reference compound titanocene
dichloride (Ref) except in the case of A375 cells.
then trypsinized and stained with JC-1 dye (5 mg/mL) (Biotium) for
20 min at 37 ꢀC. Finally, cells were washed, resuspended in PBS and
analyzed with FACS Calibur flow cytometer.
2.11. Statistical analysis
The results of viability assays are presented as mean ꢄ SD of
triplicate observations from one representative of at least three
experiments with similar results. The significance of the differences
between various treatments was determined by ANOVA followed
by Student NewmaneKeuls test. P < 0.05 was considered to be
significant.
3. Results and discussion
3.1. Synthesis and characterization
[Ti{(Me2CMe2C)(
Cl2] (2) [37], [Ti{Me2Si(
C5H4)2}Cl2] (5) [39] and [Ti{MePhSi(
were prepared according to previously described synthetic
methods.
h
5-C5H4)2}Cl2] (1) [36], [Ti{Me2C(
5-C5H4)2}Cl2] (4) [38], [Ti{MePhSi(h5
5-C5Me4)2}Cl2] (6) [39] (Fig. 1)
h
5-C5H4)2}
h
-
h
Fig. 2. The effect of titanocene complexes on viability of normal cells. (A) Keratino-
cytes (8 ꢃ 104 cells/well) and (B) fibroblasts (3 ꢃ 104 cells/well) were treated with 90
and 70
mM of 1 and 6 and 200 mM of Ref for 72 h, after which cell viability was
For the preparation of the carbon-bridged ansa-titanocene [Ti
determined by MTT and CV assay, respectively. The data are presented as mean ꢄ SD
{MePhC(
h
5-C5H4)2}Cl2] (3), the synthesis of the corresponding
from representative of three independent experiments.
ꢀ
Please cite this article in press as: S. Mijatovic, et al., Journal of Organometallic Chemistry (2013), http://dx.doi.org/10.1016/
j.jorganchem.2013.07.059

ꢀ
S. Mijatovic et al. / Journal of Organometallic Chemistry xxx (2013) 1e7
5
From these results one can see that incorporation of an ethylene
ansa-bridge increases the cytotoxicity as it is confirmed by the
much lower IC50 values associated to 1 compared to its analog 2
(with just a single carbon atom in the ansa-bridge).
results reported for highly active titanocene compounds like
titanocene-Y or some other analogs [13]. However, we decided to
study some mechanistic aspects of these compounds which were
subsequently analyzed.
In addition, incorporation of a phenyl ring attached directly to
the bridging atom of the carbon-bridged titanocene derivatives, 3,
increases the cytotoxicity, indicated by the IC50 values which are
lower than those of 2 in all the studied cancer cell lines. However,
this behavior seems to induce opposite effect regarding in vitro
activity in the silicon-bridged compounds, because 4 demonstrated
lower IC50 values than titanocene compound 5 (except in the
HTC116 cell line).
Finally, as reported in our recent work [23], the incorporation of
tetramethyl-substituted silicon-bridged ansa-biscyclopentadienyl
ligands as in the case of complex 6, increases the cytotoxic activity
in comparison with their non-substituted analogs (4 and 5).
Complexes 2e5 are less active compared to 1 and 6 due to the
non-substitution of cyclopentadienyl rings, as it was previously
observed by our group [22,23]. This substitution decreases the
hydrophobic nature of the complexes, and thus the possibility of
formation of micellar species decreasing the stability of the Ti-
cyclopentadienyl bonds, leading to a higher degree of hydrolysis
which may induce a lower cytotoxic activity. This phenomenon is
counterbalanced in complex 1 due to the incorporation of an
ethylene ansa-bridge which increases the cytotoxicity.
3.2.2. Effect of complexes 1 and 6 on primary cells
In view of the interesting results obtained with complexes 1 and
6, an evaluation of the selectivity of these compounds by the study
of their action against primary mouse keratinocytes and lung fi-
broblasts was carried out. Primary cells were treated with 90 and
70
mM solutions of 1 and 6, respectively. The chosen doses are
average IC50 calculated from the values obtained on relevant cell
lines (B16, HCT116 and SW620). Cell viability was determined by CV
(for fibroblasts) and MTT assays (for keratinocytes) after 72 h of
treatment. In the same time frame, viability of keratinocytes was
significantly less affected by 1 than by the reference compound
[Ti(h
5-C5H5)2Cl2] (Ref). On the other hand, IC50 concentration of 6
was completely nontoxic to both keratinocytes and fibroblasts,
indicating a potential selectivity of this compound towards cancer
cell lines (See Fig. 2).
3.2.3. Antitumor action mechanism of 1 and 6
HCT116 and SW620 cells (IC50 for cisplatin >120
mM) were
selected for further analysis To determine the cause of decreased
viability, colon cancer HCT116 and SW620 cell lines were treated
The highest activity of complexes 1 and 6 relative to the refer-
ence compound (Ref) has been observed against B16, HCT116,
SW620 and CT26CL25 cancer cells. Importantly, the compounds
were efficient against cell lines low responsive to cisplatin such as
HCT116 and SW620. These results indicate that treatment with 1
and 6 may overcome the problems associated to the development
of resistance of the cancer cells to cisplatin.
with IC50 doses of 1 and 6 and the reference compound [Ti(h5
-
C5H5)2Cl2] (Ref) determined for each cell line, and after 72 h cell
cycle distribution was analyzed.
As presented on Fig. 3A and B (left panel) treatment with 1 and 6
resulted in accumulation of hypodiploid cells in subG compartment
in both cell lines indicated proapoptotic pattern of their anticancer
activity. The effect of 6 on SW620 cells was the most pronounced in
comparison to all the other studied compounds, while reference
Compounds 1 and 6 have higher cytotoxicity than titanocene
dichloride but only modest cytotoxicity in comparison to the
compound [Ti(h
5-C5H5)2Cl2] (Ref) promoted DNA fragmentation
Fig. 3. The effect of titanocene complexes on cell cycle progression. (A) SW620 cells and (B) HCT116 cells (2 ꢃ 105/well) were exposed to IC50 doses of each compound (87 and
98 mM of 1, 75 and 68 mM of 6 and 142 and 200 mM of Ref for SW620 and HCT116 cells, respectively) for 72 h and cell cycle distribution (left panel) was evaluated. For determination
of cellular proliferation (right panel) cells were pre-stained with CFSE and then exposed to treatment as indicated above.
ꢀ
Please cite this article in press as: S. Mijatovic, et al., Journal of Organometallic Chemistry (2013), http://dx.doi.org/10.1016/
j.jorganchem.2013.07.059

ꢀ
S. Mijatovic et al. / Journal of Organometallic Chemistry xxx (2013) 1e7
6
Fig. 4. The effect of Ti complexes on apoptotic process. (A) SW620 cells and (B) HCT116 cells (2 ꢃ 105/well) were exposed to IC50 doses of each compound (87 and 98
mM of
compound 1, 75 and 68 mM for compound 6 for SW620 and HCT116 cells, 142 and 200 mM Ref respectively) and Ann/PI double staining after 48 h, ApoStat staining after 48 and 72 h
and mitochondrial depolarization after 72 h were determined.
only in HCT116 cells. At the same time point, analysis of cellular
proliferation by CFSE staining, revealed significant inhibition of cell
division although the kinetic was different in the studied cell lines.
While on SW620 cells proliferation upon the treatment with both
compounds was blocked after the second division, number of
divided HCT116 cells significantly decreased already in the first
division. These results indicated that titanocene derivatives worked
through inhibition of proliferation and further induction of
apoptosis (Fig. 3A and B, right panel).
To confirm apoptosis as major mechanism responsible for
decreased viability of SW620 and HCT116 cells, Ann/PI double
staining was performed after 48 h of incubation with tested com-
pounds. Obtained results showed intensive accumulation of early
apoptotic cells in HCT116 and SW620 cells exposed to both agents.
The effect of 6 was dominant and faster in comparison to 1 as
judged by the elevated presence of double positive late apoptotic
cells (Fig. 4). On the other hand, negative results of LDH release
assay for determination of necrotic cell death in the first 24 h of
incubation (not shown), eliminates the possibility of necrosis being
the pivotal mode of drug action. The apoptotic process confirmed
by cell cycle analysis as well as Ann/PI staining was accompanied
with caspase activation. This phenomenon was time-dependent
and dominant in cultures exposed to 6. To precise the intracel-
lular pathway involved in the induction of the apoptotic process,
mitochondrial membrane potential was estimated by JC-1 staining.
A 72 h long treatment with 1, triggered hyperpolarization of
mitochondrial membrane in both cell lines. However, same
experimental conditions using 6 led to mitochondrial membrane
collapse in HCT116. These results showed again a faster and more
potent activity of 6 relative to compounds 1 and Ref.
salvaging pathway in response to stress. However, depending on the
conditions, this process managed cell death. Unchanged presence of
autophagosomes, determined by AO supravital staining, indicated
that autophagic process is irrelevant for drug action, even in cyto-
protective or destructive manner (not shown).
4. Conclusions
The novel complex [Ti{MePhC(
synthesized and characterized by traditional methods. The cyto-
toxicity of 3 and the known complexes [Ti{(Me2CMe2C)(
5-C5H4)2}
Cl2] (1), [Ti{Me2C(
5-C5H4)2}Cl2] (2), [Ti{Me2Si( 5-C5H4)2}Cl2] (4),
[Ti{MePhSi(
5-C5H4)2}Cl2] (5) and [Ti{MePhSi( 5-C5Me4)2}Cl2] (6)
and the reference compound [Ti(
5-C5H5)2Cl2] (Ref) has been
h
5-C5H4)2}Cl2] (3) has been
h
h
h
h
h
h
tested against seven different cancer cell lines.
An increase in the cytotoxic activity has been observed by the
incorporation of an ethylene ansa-bridge in the titanocene deriv-
ative (complex 1), compared to its analogous with just a single
carbon atom in the ansa-bridge (complex 2). In addition, it seems
that the incorporation of a phenyl ring attached directly to the
bridging atom decreases the viability of the cancer cells in carbon-
bridged compounds, while increases the viability of the cancer cells
in silicon-bridged systems. Furthermore, one can envisage that the
incorporation of tetramethyl-substituted silicon-bridged ansa-bis-
cyclopentadienyl ligands significantly decreases the viability of the
cancer cells, in comparison with their non-substituted silicon-
bridged ansa-biscyclopentadienyl analogs. However, all these pro-
posed structureeactivity relationships are subjected to further tests
that should be carried out using more complexes of this type.
Complexes 1 and 6 are the most cytotoxic of all the analyzed
complexes against all the studied cancer cell lines with IC50 values
It is well known that apoptotic process is most commonly asso-
ciated with the collapse of mitochondrial membrane potential.
However, hyperpolarization often precedes the total mitochondrial
collapse and therefore can be considered as an earlysign of increased
apoptosis [42]. Apoptosis is often a trigger for an autophagic process,
of up to 43 ꢄ 4
compounds are generally more cytotoxic than the reference com-
pound [Ti(
5-C5H5)2Cl2] (Ref). Of particular importance is activity
of complexes 1 and 6 against cisplatin non-sensitive clones of colon
mM (for 6 against B16 cancer cell). All the studied
h
ꢀ
Please cite this article in press as: S. Mijatovic, et al., Journal of Organometallic Chemistry (2013), http://dx.doi.org/10.1016/
j.jorganchem.2013.07.059

ꢀ
S. Mijatovic et al. / Journal of Organometallic Chemistry xxx (2013) 1e7
7
cancer cells HCT116 and SW620, indicating that these compounds
References
may overcome the problems associated to the development of
resistance of the cancer cells to cisplatin.
[1] See for example: S. Gómez-Ruiz, New Directions in the Fight Against Cancer:
From Metal Complexes to Nanostructured Materials, in: B. Pignataro (Ed.),
Molecules at Work. Selfassembly, Nanomaterials, Molecular Machinery,
Wiley, Weinheim, 2012, pp. 207e242.
[2] P. Köpf-Maier, H. Köpf, Chem. Rev. 87 (1987) 1137e1152.
[3] A. Korfel, M.E. Scheulen, H.J. Schmoll, O. Gründel, A. Harstrick, M.F. Knoche,
L.M. Fels, M. Skorzec, F. Bach, J. Baumgart, G. Sass, S. Seeber, E. Thiel,
W.E. Berdel, Clin. Cancer Res. 4 (1988) 2701e2708.
[4] C.V. Christodoulou, D.R. Ferry, D.W. Fyfe, A. Young, J. Doran, T.M.T. Sheehan,
A. Eliopoulos, K. Hale, J. Baumgart, G. Sab, D.J. Kerr, J. Clin. Oncol. 16 (1998)
2761e2769.
[5] K. Mross, P. Robben-Bathe, L. Edler, J. Baumgart, W.E. Berdel, H. Fiebig,
C. Unger, Oncologie 23 (2000) 576e579.
[6] B.W Müller, R. Müller, S. Lucks, W. Mohr, Medac Gesellschaft fur Klinische
Spzeilpräparate GmbH, U.S. Patent 5,296,237, 1994.
[7] N. Kröger, U.R. Kleeberg, K. Mross, L. Edler, G. Saß, D.K. Hossfeld, Oncologie 23
(2000) 60e62.
[8] G. Lummen, S. Sperling, H. Luboldt, T. Otto, H. Rübben, Cancer Chemother.
Pharmacol. 42 (1998) 415e417.
[9] E. Meléndez, Crit. Rev. Oncol. Hematol. 42 (2002) 309e315.
[10] F. Caruso, M. Rossi, in: A. Sigel, H. Sigel (Eds.), Metal Ions in Biological System,
Metal Complexes in Tumor Diagnostics and as Anticancer Agents, vol. 42,
Marcel Dekker, Inc., New York, 2004.
The most cytotoxic compounds, 1 and 6 have also been
tested against primary mouse keratinocytes and lung fibroblasts
observing that viability of both types of primary cells were signif-
icantly less affected by 1 than by the reference compound titano-
cene dichloride. In addition, 6 was completely nontoxic to primary
cells, indicating that possesses a potential selectivity.
Finally, the mechanism of action of 1 and 6 on cisplatin resistant
colon cancer HCT116 and SW620 cell lines has been tested by using
cell cycle analyses. The results show that treatment with 1 and 6
resulted in accumulation of hypodiploid cells in subG compartment
in both cell lines, carrying out their action through inhibition of
proliferation and further induction of apoptosis. Treatment with 1
triggered hyperpolarization of mitochondrial membrane in both cell
lines, while 6 led to mitochondrial membrane collapse in HCT116.
While SW620 cell proliferation when treating them with 1 and 6
showed a blocking after the second division, number of divided
HCT116 treated cells significantly decreased already in the first
division. These results showed that titanocene derivatives worked
through inhibition of proliferation and further induction of
apoptosis.
[11] J.C. Dabrowiak, in: Metals in Medicine, Wiley, West Sussex, United Kingdom,
2009.
[12] U. Olszewski, G. Hamilton, Anti-Cancer Agents Med. Chem. 10 (2010) 302e
311.
[13] K. Strohfeldt, M. Tacke, Chem. Soc. Rev. 37 (2008) 1174e1187.
[14] R. Hernandez, J. Lamboy, L.M. Gao, J. Matta, F.R. Roman, E. Melendez, J. Biol.
Inorg. Chem. 13 (2008) 685e692.
[15] R. Hernandez, J. Mendez, J. Lamboy, M. Torres, F.R. Roman, E. Melendez,
Toxicol. In Vitro 24 (2010) 178e183.
Further work will now be focused on the possibility of inhibition
of topoisomerases by using different titanocene derivatives, which
would be essential for several cellular events such as DNA repli-
cation, transcription, chromosome condensation and segregations.
[16] L.M. Gao, J. Matta, A.L. Rheingold, E. Melendez, J. Organomet. Chem. 694
(2009) 4134e4139.
[17] A. Gansäuer, I. Winkler, D. Worgull, T. Lauterbach, D. Franke, A. Selig,
L. Wagner, A. Prokop, Chem. Eur. J. 14 (2008) 4160e4163.
[18] O.R. Allen, L. Croll, A.L. Gott, R.J. Knox, P.C. McGowan, Organometallics 23
(2004) 288e292.
[19] O.R. Allen, A.L. Gott, J.A. Hartley, R.J. Knox, P.C. McGowan, Dalton Trans. (2007)
5082e5090.
Acknowledgments
The authors would like to acknowledge financial support from
the Ministerio de Educación y Ciencia, Spain (Grant nos. CTQ-2011-
24346 and CTQ-2012-30762) and from the Ministry of Science and
Technological Development of the Republic of Serbia (Grant No.
173013). J.P. and M.H. thank to Grant Agency of the Czech Republic
(Project No. P207/12/2368). We would also like to thank Karel Mach
for being the starting point of this collaborative work.
[20] G.D. Potter, M.C. Baird, S.P.C. Cole, Inorg. Chim. Acta 364 (2010) 16e22.
[21] L.M. Gao, J.L. Vera, J. Matta, E. Melendez, J. Biol. Inorg. Chem. 15 (2010) 851e
859.
ꢀ
[22] S. Gómez-Ruiz, G.N. KaluCerovic, D. Polo-Cerón, S. Prashar, M. Fajardo,
ꢁ
ꢁ
ꢁ
ꢀ
Z. Zizak, Z.D. Juranic, T.J. Sabo, Inorg. Chem. Commun. 10 (2007) 748e752.
ꢀ
[23] S. Gómez-Ruiz, G.N. KaluCerovic, S. Prashar, D. Polo-Cerón, M. Fajardo,
ꢁ
ꢁ
ꢁ
ꢀ
Z. Zizak, T.J. Sabo, Z.D. Juranic, J. Inorg. Biochem. 102 (2008) 1558e1570.
ꢁ
ꢁ
ꢁ
Abbreviations
ꢀ
ꢀ
[24] S. Gómez-Ruiz, G.N. KaluCerovic, Z. Zizak, I. Besu, Z.D. Juranic, S. Prashar,
M. Fajardo, J. Organomet. Chem. 694 (2009) 1981e1987.
[25] G.N. KaluCerovic, V. Tayurskaya, R. Paschke, S. Prashar, M. Fajardo, S. Gómez-
ꢀ
Ruiz, Appl. Organomet. Chem. 24 (2010) 656e662.
[26] H. Sun, H. Li, R.A. Weir, P.J. Sadler, Angew. Chem. Int. Ed. 37 (1998) 1577e
1579.
[27] M. Guo, P.J. Sadler, J. Chem. Soc. Dalton Trans. (2000) 7e9.
[28] M. Guo, H. Sun, S. Bihari, J.A. Parkinson, R.O. Gould, S. Parsons, P.J. Sadler,
Inorg. Chem. 39 (2000) 206e215.
AO
acridin orange
CFSE
COSY
CV
carboxyfluorescein succinimidyl ester
correlation spectroscopy
crystal violet
[29] M. Guo, H. Sun, H.J. McArdle, L. Gambling, P.J. Sadler, Biochemistry 39 (2000)
10023e10033.
DMSO dimethyl sulfoxide
EI-MS
FACS
FCS
electronic impact mass spectrometry
fluorescence-activated cell sorting
fetal calf serum
[30] P. Köpf-Maier, D. Krahl, Chem.-Biol. Interact. 44 (1983) 317e328.
[31] P. Köpf-Maier, D. Krahl, Naturwissenschaften 68 (1981) 273e274.
[32] P. Köpf-Maier, J. Struct. Biol. 105 (1990) 35e45.
[33] A.D. Tinoco, C.D. Incarvito, A.M. Valentine, J. Am. Chem. Soc. 129 (2007) 3444e
3454.
FITC
FT-IR
fluorescein isothiocyanate (FITC)
Fourier transformed infrared spectroscopy
[34] A.D. Tinoco, E.V. Eames, A.M. Valentine, J. Am. Chem. Soc. 130 (2007) 2262e
2270.
HEPES (4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid)
[35] M. Pavlaki, K. Debeli, I.-E. Triantaphyllidou, N. Klouras, E. Giannopoulou,
A.J. Aletras, J. Biol. Inorg. Chem. 14 (2009) 947e957.
[36] J.J. Eisch, X. Shi, F.A. Owuor, Organometallics 17 (1998) 5219e5221.
[37] I.E. Nifant’ev, K.A. Butakov, Z.G. Aliev, I.F. Urazowski, Metalloorg. Khim. 4
(1991) 1265e1268.
HSQC
LDH
MTT
heteronuclear single quantum coherence
lactate dehydrogenase
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide
phosphate-buffered saline
propidium iodide
[38] C.S. Bajgur, W. Tikkanen, J.L. Petersen, Inorg. Chem. 24 (1995) 2539e2546.
PBS
PI
ꢁ
ꢁ
[39] M. Horácek, R. Gyepes, J. Merna, K. Kubista, K. Mach, J. Pinkas, J. Organomet.
Chem. 695 (2010) 1425e1433.
[40] I.E. Nifant’ev, P.V. Ivchenko, M.V. Borzov, J. Chem. Res. Synop. 23 (1992)
RPMI
THF
Roswell Park Memorial Institute
tetrahydrofuran
162.
ꢀ
ꢀ
ꢀ
ꢀ
ꢁ
ꢀ
ꢀ
ꢀ
[41] L.E. Mihajlovic, A. Savic, J. Poljarevic, I. Vuckovic, M. Mojic, M. Bulatovic,
ꢀ
ꢀ
ꢀ
ꢁ ꢀ ꢁ ꢀ
D. Maksimovic-Ivanic, S. Mijatovic, G.N. KaluCerovic, S. Stosic-Grujicic,
D. Miljkovic, S. Grguric-Sipka, T.J. Sabo, J. Inorg. Biochem. 109 (2012)
ꢁ
ꢀ
ꢀ
Appendix A. Supplementary data
40e48.
[42] S. Paglin, Y.N. Lee, C. Nakar, M. Fitzgerald, J. Plotkin, B. Deuel, N. Hackett,
M. McMahill, E. Sphicas, N. Lampen, J. Yahalom, Cancer Res. 23 (2005) 11061e
11070.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2013.07.059.
ꢀ
Please cite this article in press as: S. Mijatovic, et al., Journal of Organometallic Chemistry (2013), http://dx.doi.org/10.1016/
j.jorganchem.2013.07.059