New iron cyclopentadienyl complexes bearing different phosphane co-ligands: Structural factors vs. cytotoxicity

Article abstract of DOI:10.1016/j.jorganchem.2017.10.004

A new family of piano stool iron-cyclopentadienyl compounds bearing different phosphane co-ligands has been synthesized. All the compounds, with the general structure [Fe(Cp)(CO)(PR₃)(L)]ⁿ (PR₃ = triphenylphosphane, 4-(diphenylphosphino) benzoic acid or tris(4-fluorophenyl)phosphane; when L = I, n = 0; when L = 4-aminobenzonitrile, n = +1) were fully characterized by the usual analytical and spectroscopic techniques. Interestingly, compound [Fe(Cp)(CO)(PPh₃)I] 1 crystalizes in the orthorhombic space group P2₁2₁2₁ and its crystal packing only contains one enantiomer, while compound [Fe(η⁵-Cp)(CO)(P(Ph-p-F)₃)I] 3 crystalizes in the centrosymmetric space group Pbca presenting an important disorder in the structure, probably due to the presence of the two enatiomers in the crystal packing. All the compounds presented adequate stability in aqueous solution and they were tested against cervical HeLa human cancer cells. The cationic complexes bearing triphenylphosphane (4) or tris(4-fluorophenyl)phosphane (6) were found to be highly cytotoxic, causing cell death by apoptosis. The results point out that the electronic features of the new compounds might be related to their cytotoxic activity.

Full text of DOI:10.1016/j.jorganchem.2017.10.004

Journal of Organometallic Chemistry 852 (2017) 34e42
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
New iron cyclopentadienyl complexes bearing different phosphane
co-ligands: Structural factors vs. cytotoxicity
,
,
Adhan Pilon a 1, Patrícia Gírio a 1, Guilherme Nogueira a, Fernando Avecilla b,
Harry Adams c, Julia Lorenzo d, M. Helena Garcia a **, Andreia Valente a
Centro de Química Estrutural, Faculdade de Ciencias da Universidade de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
,
, *
a
^
b
ꢀ
~
Grupo Xenomar, Centro de Investigacions Científicas Avanzadas (CICA), Departamento de Química, Facultade de Ciencias, Universidade da Coruna,
~
~
Campus de A Coruna, 15071 A Coruna, Spain
c Department of Chemistry, University of Sheffield, Sheffield S3 7HF, UK
Institut de Biotecnologia i de Biomedicina, Departament de Bioquímica i Biologia Molecular, Universitat Autonoma de Barcelona, 08193 Bellaterra,
Barcelona, Spain
d
ꢁ
a r t i c l e i n f o
a b s t r a c t
Article history:
A new family of piano stool iron-cyclopentadienyl compounds bearing different phosphane co-ligands
has been synthesized. All the compounds, with the general structure [Fe(Cp)(CO)(PR3)(L)]n
(PR3 ¼ triphenylphosphane, 4-(diphenylphosphino) benzoic acid or tris(4-fluorophenyl)phosphane;
when L ¼ I, n ¼ 0; when L ¼ 4-aminobenzonitrile, n ¼ þ1) were fully characterized by the usual
analytical and spectroscopic techniques. Interestingly, compound [Fe(Cp)(CO)(PPh3)I] 1 crystalizes in the
orthorhombic space group P212121 and its crystal packing only contains one enantiomer, while com-
Received 18 August 2017
Received in revised form
3 October 2017
Accepted 5 October 2017
Available online 7 October 2017
pound [Fe(h
5-Cp)(CO)(P(Ph-p-F)3)I] 3 crystalizes in the centrosymmetric space group Pbca presenting an
Keywords:
important disorder in the structure, probably due to the presence of the two enatiomers in the crystal
packing. All the compounds presented adequate stability in aqueous solution and they were tested
against cervical HeLa human cancer cells. The cationic complexes bearing triphenylphosphane (4) or
tris(4-fluorophenyl)phosphane (6) were found to be highly cytotoxic, causing cell death by apoptosis. The
results point out that the electronic features of the new compounds might be related to their cytotoxic
activity.
Iron cyclopentadienyl
Electronic flow
Anticancer
Apoptosis
© 2017 Elsevier B.V. All rights reserved.
1. Introduction
used in cancer therapy. In this frame, the search for new iron
compounds with anticancer properties has been essentially
Iron is a biometal essential for human life being a building
element in several important proteins [1]. Its presence in many
biological functions, importance in cell replication, metabolism and
growth, along with the anticancer properties already shown for
some iron complexes, makes it an appealing candidate to be used in
anticancer drugs. Indeed, the discovery [2] and structural charac-
terization [3] of the “sandwich” compound ferrocene, in the early
focused on ferrocene derivatives [6]. Ferrocene by itself is not a
particularly cytotoxic compound [7]. However, ferrocifens, de-
rivatives of tamoxifen (a chemotherapeutic agent for patients with
hormone-dependent breast cancer), revealed anticancer activity
against hormone-dependent (MCF7) and independent (MDA-MB-
231) breast cancer cell lines [8e10]. Ferrocene derivatives suffer,
however, from bioavailability problems, restricting them from
entering into clinical studies [7].
€
€
1950s and the pioneer work of Kopf and Kopf-Maier discovering
the cytotoxic properties [4,5] of ferrocenium, the oxidized form of
ferrocene, paved the way for the search of iron compounds to be
The encouraging results obtained by our group with piano stool
structured compounds based on the “RuCp” fragment (Cp ¼ h5
-
C5H5) [11e18] together with the success found for ferrocene de-
rivatives lead us to enlarge our work to the compounds with “FeCp”
bearing a half-sandwich geometry. We have recently published our
results concerning two new families of organometallic iron com-
pounds based on the general cationic structure [Fe(Cp)(dppe)(L)]þ,
where dppe ¼ ethylenebis(diphenylphosphane) and L ¼ imidazole
* Corresponding author.
** Corresponding author.
E-mail addresses: mhgarcia@fc.ul.pt (M.H. Garcia), amvalente@fc.ul.pt
(A. Valente).
1
These authors contributed equally to this work.
https://doi.org/10.1016/j.jorganchem.2017.10.004
0022-328X/© 2017 Elsevier B.V. All rights reserved.

A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
35
based ligands [19] or nitrile ligands [20]. These two new families
showed cytotoxicity against breast MCF7 [19], cervical HeLa [19],
ovarian A278019 and leukemia HL-6020 human cancer cells lower
than those found for cisplatin in the same experimental conditions.
Other complexes bearing the “FeCp” scaffold and bearing nitrile
carbohydrate derivative ligands also showed good cytotoxicities
against a colon cancer cell line (HCT116) [21].
Microcrystalline green powder. Single crystals for X-ray diffraction
studies were obtained by crystallization from THF/n-hexane solu-
tion. IR (KBr, cmꢀ1):
n
(C-H aromatics) 3047,
n
(C≡O) 1936,
n(C-C ar-
omatics) 1473, 1427. 1H NMR (DMSO-d6, Me4Si,
d/ppm): 7.47 (comp,
15, H2þH3þH4); 4.59 (s, 5, Cp). 13C NMR (DMSO-d6, Me4Si,
d/ppm):
1
221.40 (d, JCP
¼
31.2, C≡O); 135.34 (d, JCP
¼
43.5, C1);
133.15 þ 128.28 (d, JCP ¼ 9.4; d, JCP ¼ 9.6, C3 þ C2); 130.21 (C4);
Complexes [FeCp(CO)2X] (X
[FeCp(CO)2]2 showed cytotoxicity towards MDA-MB-231 breast and
M, 24 h incubation),
¼
halide, NCS, BFꢀ4 ) and
82.93 (Cp). 31P NMR (DMSO-d6,
d/ppm): 67.04 (s). UVeVis in DMSO,
lmax/nm (ε/Mꢀ1 cmꢀ1): 275 (Sh); 325 (2470); 387 (Sh); 448 (760);
HeLa cervical cancer cells (IC50 ¼ 3.0e17.3
m
626 (155). Elemental analysis (%) Found: C 53.5, H 3.5. Calc. for
while being non-cytotoxic towards normal mammary epithelial
cells MCF-10A [22].
C
24H20FeIOP: C 53.5, H 3.7.
[Fe(
5-Cp)(CO)(PPh2(C6H4COOH))I] 2
Yield: 84% (489 mg; 0.84 mmol). Irradiation time: 3 h. Dark
green crystalline powder. IR (KBr, cmꢀ1):
(OH) 3530, (C-H aro-
(C-C ar-
/ppm): 8.06 (d, 2,
h
In this study, we have decided to explore the simultaneous ef-
fect of different
character of the carbonyl co-ligand, that might tune the complex
cytotoxicity [16]. The competitive acceptor effect of the benzo-
s
donor phosphane ligands with the
p
acceptor
n
n
matics) 3055,
n
(C≡O) 1944,
n
(C¼O carboxylic acid) 1689,
n
p
omatics) 1674, 1442. 1H NMR (DMSO-d6, Me4Si,
d
H3 ); 7.62e7.43 (comp, 10, H2’þH2þH3þH4); 4.61 (s, 5, Cp). 13C NMR
0
nitrile derivative ligand is expected to impart strong electronic ef-
fects on the iron complexes due to its involvement in strong metal-
(DMSO-d6, Me4Si,
d
/ppm): 220.90 (d, JCP ¼ 31.2, C≡O); 166.99
1
ligand
p
-backdonation via the d metal-
p
* NC orbitals [23].
(COOH); 141.10 (d, JCP ¼ 41, C10); 134.69 (dd, JCP ¼ 16, 44, C1);
2
133.37 (dd, JCP ¼ 9, 18, C2); 133.12 (d, JCP ¼ 9, C20); 132.05 (C40);
2. Experimental section
130.52 (C4); 128.90 (d, 3JCP ¼ 9, C30); 128.48 (d, 3JCP ¼ 9, C3); 83.03
(Cp). 31P NMR (DMSO-d6,
d/ppm): 68.35 (s). UVevis DMSO, lmax/nm
2.1. General procedures
(ε/Mꢀ1 cmꢀ1): 274 (14555); 330 (Sh); 447 (830); 627 (175).
Elemental analysis (%) Found: C 51.3, H 3.6. Calc. for C25H20FeIO3P: C
51.5, H 3.6.
All reactions and manipulations were performed under nitrogen
atmosphere using Schlenk techniques. All solvents used were dried
and freshly distilled under nitrogen prior to use, using standard
methods [24]. 1H, 13C and 31P NMR spectra were recorded on a
Bruker Avance 400 spectrometer at probe temperature using
commercially available deuterated solvents. 1H and 13C chemical
shifts (s ¼ singlet; d ¼ duplet; t ¼ triplet; m ¼ multiplet;
comp ¼ complex) are reported in parts per million (ppm) down-
field from internal standard Me4Si and the 31P NMR spectra are
reported in ppm downfield from external standard, 85% H3PO4.
Coupling constants are reported in Hz. All assignments were
attributed using 13C APT or DEPT-135, COSY, HMBC, HSQC and
HMQC NMR techniques. Infrared spectra were recorded on KBr
pellets using a Mattson Satellite FT-IR spectrophotometer. Only
considered relevant bands were cited in the text. Electronic spectra
were obtained at room temperature on a Jasco V-560 spectrometer
from solutions of 10ꢀ3-10ꢀ5 M in quartz cuvettes (1 cm optical
[Fe(h
5-Cp)(CO)(P(Ph-p-F)3)I] 3
Yield: 64% (379 mg; 0.64 mmol). Irradiation time: 5 h. Micro-
crystalline green powder. Single crystals for X-ray diffraction
studies were obtained by crystallization from dichloromethane/n-
hexane solution. IR (KBr, cmꢀ1):
1944,
n
(C-H aromatics) 3062,
n(C≡O)
n
(C-C aromatics) 1581, 1496. 1H NMR (DMSO-d6, Me4Si,
d/
ppm): 7.52 (s, 6, H3); 7.33 (s, 6, H2); 4.65 (s, 5, Cp). 13C NMR (DMSO-
d6, Me4Si,
d
/ppm): 220.20 (d, JCP ¼ 31.2, C≡O); 161.54 (d, 1JCF ¼ 248,
C4); 135.54 (t, J ¼ 10, C3); 131.30 (d, 1JCP ¼ 45, C1); 115.60 (dd, J ¼ 11,
21, C2); 82.81 (Cp). 31P NMR (DMSO-d6,
d/ppm): 65.97 (s). UVevis in
DMSO, lmax/nm (ε/Mꢀ1 cmꢀ1): 281 (Sh), 329 (Sh), 437 (830), 626
(159). Elemental analysis (%) Found: C 47.0, H 2.6. Calc. for
C24H17F3FeIOP: C 46.2, H 2.7.
2.4. General procedure for the synthesis of [Fe(h5
Cp)(CO)(PR3)(C7H6N2)]þ complexes 4-6
-
ꢀ
path). Elemental analyses were performed at Laboratorio de
ꢀ
ꢀ
Analises, at Instituto Superior Tecnico, using a Fisons Instruments
EA1 108 system. Data acquisition, integration and handling were
performed using a PC with the software package EAGER-200 (Carlo
Erba Instruments).
To a stirred and degassed solution of complexes 1e3 (0.17 mmol
for 5, 6; 0.30 mmol for 4) in dry acetone (30 ml) was added AgPF6
(0.25 mmol for 5, 6; 0.45 mmol for 4). After
1 h 4-
aminobenzonitrile (0.17 mmol for 5, 6; 0.30 mmol for 4) was
added and the reaction followed for 24 h at room temperature. The
precipitates were separated by cannula-filtration and the solvent
was evaporated under vacuum. The residue was twice recrystal-
lized from dry acetone/n-hexane and dry THF/n-hexane.
2.2. Synthesis
The starting material [Fe(h
5-Cp)(CO)2I] was prepared from the
commercially available dimer [Fe(
literature procedure [25].
h
5-Cp)(CO)2]2 following the
[Fe(
Yield: 85% (172 mg; 0.26 mmol). Dark red crystalline. IR (KBr,
cmꢀ1):
(C-H aromatics) 3080, (N≡C) 2245, (C≡O) 1982, (C-C
aromatics) 1620, 1512, 1435,
(P-F) 840. 1H NMR (DMSO-d6, Me4Si,
h
5-Cp)(CO)(PPh3)(C7H6N2)][PF6] 4
n
n
n
n
2.3. General procedure for the synthesis of [Fe(
h
5-Cp)(CO)(PR3)I]
n
complexes 1-3
d
/ppm): 7.61e7.53 (comp, 9, H3þH4); 7.40e7.32 (m, 6, H2); 6.88 (d,
2, JHH ¼ 8.5, H7); 6.45 (d, 2, JHH ¼ 8.5, H8); 5.08 (s, 5, Cp). 13C NMR
To a stirred and degassed solution of [Fe(
h
5-Cp)(CO)2I] (1 mmol)
(DMSO-d6, Me4Si,
d
/ppm): 217.11 (d, JCP ¼ 45.4, C≡O); 153.99 (C5);
2
in dry acetone (30 mL) PR3 (1 mmol) was added. The reaction
mixture was then irradiated under UV light (125 W) for 3e7 h (see
below). The precipitate was separated by cannula-filtration and the
solvent was evaporated under vacuum. The residue was twice
recrystallized from dry dichloromethane/n-hexane and dark green
products are obtained.
137.61 (C6); 134.03 (C7); 132.76 (d, JCP ¼ 10, C2); 131.93 (d,
1JCP ¼ 45, C1); 131.25 (C4); 129.23 (d, JCP ¼ 10, C3); 112.92 (C8);
3
93.58 (C9); 85.10 (Cp). 31P NMR (DMSO-d6,
d/ppm): 66.84
(s); ꢀ144.19 (setp, JPF ¼ 712, PF6). UVeVis in DMSO, lmax/nm
(ε/Mꢀ1 cmꢀ1): 284 (21555); 321 (23250); 409 (900); 509 (Sh).
Elemental analysis (%) Found: C 52.9, H 3.5, N 3.8. Calc. for
[Fe(h
5-Cp)(CO)(PPh3)I] 1
C
31H26F6FeN2OP2: C 52.4, H 3.7, N 3.9.
Yield: 69% (371 mg; 0.69 mmol). Irradiation time:
4
h.
[Fe(h
5-Cp)(CO)(PPh2(C6H4COOH))(C7H6N2)][PF6] 5

36
A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
Yield: 88% (107 mg; 0.15 mmol). Dark red crystalline. IR (KBr,
cmꢀ1):
(OH) 3530, (C-H aromatics) 2970, (N≡C) 2245, (C≡O)
(C-C aromatics) 1697, 1604,
collection and refinement for the new compound were collected in
Table 1.
n
n
n
n
1990,
1411,
n
(C¼O carboxylic acid) 1705,
n
n
(P-F) 840. 1H NMR (DMSO-d6, Me4Si,
d
/ppm): 13.39 (broad, 1,
2.6. Stability studies in DMSO and DMSO/DMEM
0
COOH); 8.07 (m, 2, H3 ); 7.75e7.34 (comp, 12, H2’þH2þH3); 6.90 (d,
2, H7); 6.45 (comp, 4, H8þNH2); 5.13 (s, 5, Cp). 13C NMR (DMSO-d6,
For the stability studies, all the complexes were dissolved in
DMSO or 5% DMSO/95% DMEM at ca. 1 ꢂ 10ꢀ4 M and their elec-
tronic spectra were recorded in the range allowed by the solvents at
set time intervals.
Me4Si, d/ppm): 222.28 (C≡O); 166.66 (COOH); 154.06 (C5); 137.82
(C6); 134.12 (C7); 133.7 (d, 1JCP ¼ 18, C1); 133.45 (C10); 133.23 (C20);
2
132.97 (d, JCP ¼ 10, C2); 131.52 (C40); 131.33 (C4); 129.82 (C30);
3
129.44 (t, JCP ¼ 10, C3); 112.95 (C8); 93.59 (C9); 85.26 (Cp). 31P
2
NMR (DMSO-d6,
d
/ppm): 68.87 (s); ꢀ144.20 (sept, JCP ¼ 10, C2);
2.7. Biological studies
1
3
131.93 (d, JCP ¼ 45, C1); 131.25 (C4); 129.23 (d, JCP ¼ 10, C3);
112.92 (C8); 93.58 (C9); 85.10 (Cp). 31P NMR (DMSO-d6,
d/ppm):
2.7.1. Cell lines and culture conditions
66.84 (s); ꢀ144.19 (setp, JPF ¼ 712, PF6). UVeVis in DMSO, lmax/nm
(ε/Mꢀ1 cmꢀ1): 282 (38980); 319 (Sh); 410 (Sh); 510 (Sh). Elemental
analysis (%) Found: C 46.9, H 3.5, N 3.3. Calc. for C32H26F6Fe-
N2O3P2$25AgI: C 47.3, H 3.2, N 3.4.
Human cervix adenocarcinoma HeLa cell line (ATC CCL-2) was
maintained in a Minimum Essential Medium (MEM) alpha medium
supplemented with 10% (v/v) heat inactivated fetal bovine serum
(FBS) in a highly humidified atmosphere of 95% air with 5% CO2 at
37 ꢁC.
[Fe(
Yield: 88% (109 mg; 0.15 mmol). Dark red crystalline. IR (KBr,
cmꢀ1):
(C-H aromatics) 3078, (N≡C) 2252, (C≡O) 1990, (C-C
aromatics) 1589, 1496,
(P-F) 848. 1H NMR (DMSO-d6, Me4Si,
h
5-Cp)(CO)(P(Ph-p-F)3)(C7H6N2)][PF6] 6
n
n
n
n
2.7.2. Cytotoxicity assays
n
d/
Cells were seeded into a 96-well plate at a cell density of
3.0 ꢂ 103 cell/well and incubated for 24 h before the addition of the
ligands or iron compounds a concentrations ranging from 0 to
ppm): 7.45e7.37 (comp, 12, H2þH3); 6.98 (d, 2, JHH ¼ 8.3, H7); 6.48
(m, 4, H8þNH2); 5.13 (s, 5, Cp). 13C NMR (DMSO-d6, Me4Si,
d/ppm):
216.84 (d, JCP ¼ 47.0, C≡O); 163.69 (d, 1JCF ¼ 250, C4); 154.12 (C5);
137.84 (C6); 135.45 þ 116.69 (C2þC3); 134.12 (C7); 128.03 (d,
200
mM. The growth inhibitory effect was measured after 24 and
72 h treatment by the XTT assay [28]. Aliquots of 20
mL of XTT so-
1JCP ¼ 51, C1); 112.98 (C8); 93.46 (C9); 85.26 (Cp). 31P NMR (DMSO-
lution [2,3-bis-(2-methoxy-4-nitro- 5-sulfophenyl)-2H-tetrazo-
lium-5-carboxanilide] were added to each well. After 3 h, the
colour formed was quantified by a spectrophotometric plate Reader
(PerkinElmerVictor3 V) at 490 nm wavelength. Cell cytotoxicity
was evaluated in terms of cell-growth inhibition in treated cultures
and expressed as a % of the control conditions. Each experiment
was repeated at least three times, and each concentration tested in
at least six replicates.
2
d6,
d
/ppm): 68.87 (s); ꢀ144.20 (sept, JCP ¼ 10, C2); 131.93 (d,
3
1JCP ¼ 45, C1); 131.25 (C4); 129.23 (d, JCP ¼ 10, C3); 112.92 (C8);
93.58 (C9); 85.10 (Cp). 31P NMR (DMSO-d6,
d/ppm): 66.84
(s); ꢀ144.19 (setp, JPF ¼ 712, PF6). UVevis in DMSO, lmax/nm
(ε/Mꢀ1 cmꢀ1): 284 (17248); 311 (18339); 406 (600). Elemental
analysis (%) Found: C 49.8, H 3.4, N 3.0. Calc. for C31H23F9Fe-
N2OP2$½CH2Cl2: C 49.1, H 3.1, N 3.6.
2.5. X-ray crystal structure determination
2.7.3. In vitro apoptosis assay
Induction of apoptosis in vitro by iron compounds was deter-
mined by a flow cytometric assay with Annexin V-FITC by using an
Annexin V-FITC Apoptosis Detection Kit (Roche). Exponentially
growing HeLa cells in 6-well plates (4 ꢂ 105 cells per well) were
exposed to concentrations equal to the IC50 of the iron compounds
for 24 h or cisplatin as a reference. Cells were collected, washed
Three-dimensional X-ray data for [Fe(
[Fe(
5-Cp)(CO)(P(Ph-p-F)3)I] 3 were collected on a Bruker SMART
Apex CCD diffractometer at 100(2) K, using a graphite mono-
chromator and Mo-Ka radiation ( scan
¼ 0.71073 Å) by the
method. Reflections were measured from a hemisphere of data
collected of frames each covering 0.3ꢁ in
. A total of 2892 re-
flections were measured, all of which were corrected for Lorentz
and polarization effects and for absorption by semi-empirical
methods based on symmetry-equivalent and repeated reflections.
Of the total, 2687 independent reflections exceeded the signifi-
h
5-Cp)(CO)(PPh3)I] 1 and
h
l
f-u
u
with PBS, and resuspended in 100
staining was accomplished following the product instruction
(Roche). In brief, 2 L of Annexin V-fluorescein isothiocyanate
(FITC) and 2 L of propidium iodide (PI) were added to the samples
mL of binding buffer. Annexin V
m
m
for 15 min at room temperature in the dark. The amount of
apoptotic cells was analysed by flow cytometry (FACSCalibur, Bec-
ton Dickinson).
cance level jFj/ (jFj) > 4.0. After data collection, in each case a
s
multi-scan absorption correction (SADABS) [26] was applied, and
the structure was solved by direct methods and refined by full
matrix least-squares on F [2] data using SHELX suite of programs
[27]. The structure was solved by direct methods and refined by
full-matrix least-squares methods on F2. The non-hydrogen atoms
were refined with anisotropic thermal parameters in all cases.
Hydrogen atoms were included in calculation positions and refined
in the riding mode. A final difference Fourier map showed a re-
sidual density outside due to disorder, which could not be refined:
3. Results and discussion
3.1. Synthesis
New iron(II) organometallic compounds have been synthesized.
The
neutral
compounds
[FeCp(CO)(PR3)I]
(where
PR3 ¼ triphenylphosphane 1, 4-(diphenylphosphino)benzoic acid 2
or tris(4-fluorophenyl)phosphane 3) have been synthesized by UV
irradiation of a solution of [Fe(Cp)(CO)2I] and the corresponding
phosphane ligand in acetone (Scheme 1). The cationic compounds
2.303 and ꢀ1.542 e.Åꢀ3
.
A
weighting scheme
w
¼
1/
[
s
2(Fo2) þ (0.088000 P) [2] þ 48.334400 P] for 1, where P ¼
(jFoj2 þ 2jFcj2)/3, was used in the latter stages of refinement. CCDC
No. 1536687 and 1565463 contains the supplementary crystallo-
graphic data for 1 and 3, respectively. These data can be obtained
free of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.
html, or from the Cambridge Crystallographic Data Centre, 12
Union Road, Cambridge CB2 1EZ, UK; fax: (þ44) 1223-336-033; or
e-mail: deposit@ccdc.cam.ac.uk. Crystal data and details of the data
[Fe(Cp)(CO)(PR3)(4-aminobenzonitrile)]þ
(where
PR3 ¼ triphenylphosphane 4, 4-(diphenylphosphino)benzoic acid 5
or tris(4-fluorophenyl)phosphane 6) were obtained by iodide
abstraction using AgPF6 and reaction with 4-aminobenzonitrile in
acetone at room temperature for 24 h (Scheme 1). All the com-
pounds were obtained in good yields (64e88%) as dark green

A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
37
Table 1
Crystal data and structure refinement for [Fe(h h
5-Cp)(CO)(PPh3)I] 1 and for [Fe( 5-Cp)(CO)(P(Ph-p-F)3)I] 3.
1
3
Formula
Formula weight
C24H20FeIOP
538.12
C24H17F3FeIOP
592.10
T, K
100(2)
100(2)
Wavelength, Å
Crystal system
Space group
0.71073
Orthorhombic
P212121
0.71073
Orthorhombic
Pbca
a/Å
b/Å
8.6776(6)
13.6856(9)
17.1867(11)
2041.1(2)
4
15.4421(5)
16.5176(6)
16.7701(6)
4277.5(3)
8
c/Å
V/Å3
Z
F000
1064
1.751
2.342
1.90 to 28.34
0.0535
2320
1.839
2.264
2.18 to 26.42
0.0705
Dcalc/g cmꢀ3
m
/mmꢀ1
q
/(ꢁ)
Rint
Absolute structure parameter (Flack parameter)
Crystal size/mm3
0.08(11)
0.14 ✕ 0.13 ✕ 0.13
1.132
0.0799
0.2222
0.124 ✕ 0.090 ✕ 0.085
1.329
0.1031
0.2080
2.213 and ꢀ1.977
Goodness-of-fit on F2
a
R1
b
wR2 (all data)
Largest differences peak and hole (eÅꢀ3
)
2.303 and ꢀ1.542
a
R1 ¼ SrrFor - rFcrr/SrFor.
b
wR2 ¼ {S[w(rrFor2 -rFcr2r)2]r/S[w(F2o)2]}1/2
.
Scheme 1. Reaction scheme for compounds 1e6, numbered for NMR purposes.
(complexes 1e3) or dark red compounds (complexes 4e6). The
formulation and purity of all the new compounds has been ascer-
tained by elemental analysis and FT-IR and multinuclear (1H, 13C
and 31P) NMR spectroscopy. Compound 1 has been previously re-
ported [29]. In this paper from 1966, the reaction was carried out in
benzene at reflux temperature for 18 h. During the reaction two
products were formed, being [FeCp(CO)(PPh3)I] isolated in ~34%
yield and only characterized by FT-IR and elemental analysis. Later,
a different synthesis approach was suggested [30], based on a paper
describing the synthesis of iron cyclopentadienyl complexes with
triphenylphosphite ligands [31] using UV irradiation in cyclo-
hexane, benzene or THF. However, neither synthesis details nor
characterization is provided. Here, we present an improved syn-
thesis (69% yield) based on those papers, together with a complete
spectroscopic characterization and the X-ray structure of 1 (see
below).
with the aromatic rings of the phosphane ligands (3200-3000 cmꢀ1
and 1600-1400 cmꢀ1). Additional bands at ~850 cmꢀ1 confirm the
presence of the counter-ion PFꢀ6 for compounds 4e6. The presence
of carbon monoxide can be confirmed by the infrared stretching
vibration occurring at ca. 1940 cmꢀ1 in compounds 1e3 and at ca.
1990 cmꢀ1 in compounds 4e6. These vibrations in compounds 1e3
are present at lower values than those of the parent dicarbonyl
compound [FeCp(CO)2I] (
n
¼ 2037 and 1975 cmꢀ1). These differ-
ences are consistent with the enhancement of bonding between
the metal and the remaining terminal carbonyl group, due to the
replacement of one CO in [FeCp(CO)2I] by the poorer
p-acceptor
phosphane. The replacement of the iodide by the nitrile ligand in
compounds 4e6 leads to a strengthening of the CO bond and ac-
counts for the competition effect of the metaleligand
p-back-
donation. The coordination of the 4-aminobenzonitrile in
compounds 4e6 can be also confirmed by the characteristic
stretching vibration of the nitrile functional group appearing at
2214 cmꢀ1 in the free ligand and at ~2247 cmꢀ1 after coordination.
The solid state FT-IR spectra (KBr pellets) of all the complexes
present the characteristic band for the cyclopentadienyl ring along

38
A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
Table 2
NMR assignments for compounds 1e6, phosphanes and 4-aminobenzonitrile in DMSO-d6.
1H NMR
Cp
31P NMR
PR3
H7
H8
Triphenylphosphane
PPh3
4-(diphenylphosphino)benzoic acid
PPh2(C6H4COOH)
tris(4-fluorophenyl)phosphane
P(Ph-p-F)3
e
e
e
e
e
ꢀ7.0
e
e
e
e
ꢀ6.5
ꢀ10.6
4-aminobenzonitrile
e
7.35
e
6.60
e
e
1
2
3
4
5
6
4.59
4.61
4.65
5.10
5.17
5.13
67.04
68.35
65.97
66.84
67.48
65.87
e
e
e
e
6.88
6.90
6.98
6.45
6.45
6.47
This variation of ~30 cmꢀ1 for nN≡C accounts for the
s
coordination
recorded in 1 ꢂ 10ꢀ3 to 1 ꢂ 10ꢀ5 M solutions in dimethyl sulfoxide
in the wavelength range of 268e900 nm. The electronic data is
summarized in the experimental section and in Fig. 1.
and the effect of metaleligand
p-backdonation through dmetale-
p*
interaction which was as also observed in other related
ligand
compounds [30].
The spectra are characterized by a strong absorption band below
285 nm and a shoulder at around 320e330 nm. The first is attrib-
The NMR spectra in DMSO-d6 shows the expected signals of the
Cp ligand at ~4.6 ppm for compounds 1e3 and at ~5.1 ppm for
compounds 4e6, more deshielded in this later case due to their
uted to
p/p* transitions on the chromophores (phosphanes and
also nitrile for 4e6). The latter is attributed to electronic transitions
occurring in the organometallic fragment {FeCp}þ (by analogy with
the iron parental complex [Fe(Cp)(CO)2I], Fig. S1). These bands are
much more intense in the cationic compounds. In addition, one or
two weak bands in the visible region were found for all complexes.
cationic nature (Table 2). The replacement of one CO strong
p-
acceptor ligand in the parent complex [Fe(Cp)(CO)2I], by a more
electron donating phosphane ligand is accompanied by a shielding
on the Cp (5.28 vs. ~4.6 ppm), owing to the increased electron
density at the metal centre. This donation effect of the phosphane is
The first one (600 Mꢀ1cmꢀ1
< 3
< 850 Mꢀ1cmꢀ1) is probably due to
in agreement with the phosphane deshielding observed on the 31
P
metal-to-ligand charge transfer bands (MLCT), since 1H NMR and
FT-IR studies showed the presence of -backdonation from the
NMR sharp singlet resonance upon coordination (Table 2). The 1H
NMR signals of the 4-aminobenzonitrile are shielded in all the
cationic complexes 4e6, being this effect more pronounced in the
ortho protons (Dd ¼ ꢀ0.37 to ꢀ0.47 ppm), in agreement with an
electronic flow from the metal centre to the nitrile functional group
p
metal centre to the nitrile and to the carbonyl, as also observed for
other related compounds [30]. The second band in the visible range,
with molar absorptivity values below 200 Mꢀ1cmꢀ1 might be due
to DMSO-d6 iron transitions.
due to the effect of
p-backdonation. Accordingly, an increased
electronic density is observed on the 4-aminobenzonitrile aromatic
ring. The 13C NMR spectra shows the same general effect observed
for the protons in all complexes.
3.2.2. Complexes stability
Envisaging the use of these new compounds as cytotoxic agents
and their study in human cancer cell lines, their stability and
behaviour was studied in DMSO (co-solvent used in the biological
assays) and in culture cellular media (DMEM), using 5% DMSO. All
the compounds are fairly stable in both solutions for a period of at
least 24 h as one can observe by UVeVis (Figs. S2 and S3 in Sup-
porting Information) or NMR spectroscopy (in DMSO-d6, Fig. S4),
3.2. UVevisible (UVeVis) studies
3.2.1. Compounds characterization
The electronic absorption spectra of all compounds was
Fig. 1. Electronic spectra of the new iron compounds in DMSO solutions.

A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
39
Fig. 2. ORTEP plot for the complex A) [Fe(h h
5-Cp)(CO)(PPh3)I] 1 and B) [Fe( 5-Cp)(CO)(P(Ph-p-F)3)I] 3. All the non-hydrogen atoms are presented by their 50% probability ellipsoids.
Hydrogen atoms are omitted for clarity.
since no changes in band format or the appearance of new bands
was noticed.
interactions are observed between Cp and Ph rings in these dimeric
moieties. The rings are not parallel. The distance between cen-
troids, dc1-c2, is 3.958(16) Ǻ [c1, C(12A)-C(13A)-C(14A)-C(15A)-
C(16A)-C(17A), c2, C(1K)-C(2K)-C(3K)-C(4K)-C(5K)]. The obtained
Flack parameter (see Table 1) corresponds with an enantiomerically
pure compound [32].
3.3. Single crystal structure of [Fe(h -
5-Cp)(CO)(PPh3)I] 1 and [Fe(h5
Cp)(CO)(P(Ph-p-F)3)I] 3
[Fe(
tion as red blocks (crystal dimensions 0.14 ✕ 0.13 ✕ 0.13 mm).
Fig. 2A shows an ORTEP representation of [Fe(
5-Cp)(CO)(PPh3)I] of
h
5-Cp)(CO)(PPh3)I] 1 crystallizes from THF/n-hexane solu-
[Fe(h 3 crystallizes from dichloro-
5-Cp)(CO)(P(Ph-p-F)3)I]
methane/n-hexane solution as black blocks (crystal dimensions
0.124 ✕ 0.090 ✕ 0.085 mm). Fig. 2B shows an ORTEP representation
of 3. The compound crystallizes in a centrosymmetric space group,
h
1. The compound crystallizes in a non-centrosymmetric space
group, P212121. Crystal packing only contains one enantiomer of
Pbca. Short contacts and
p-p interactions are not present in the
iron complex due to short contacts and
p
-
p
angular interactions,
crystal packing of this structure (see Fig. 4). The compound presents
an important disorder in the structure, which can be justified for
the presence of the two enatiomers in the crystal packing.
In the molecular structures, the ruthenium centres adopt a
“piano stool” distribution formed by the iron-Cp unit bound to one
phosphane, one CO molecule and one iodide anion. The C-O bond
length, 1.071(17) Å in 1 and 1.07(2) Å in 3, is shorter than in other
which prevent the free rotoinversion around Fe-I edge. Short con-
tacts are observed between phenyl rings and oxygen atoms of CO
molecules in compound 1 (see Fig. 3). The distance between O(1)-
C(10) is 2.862(13) Å and between O(1)-C(11) is 2.495(14) Å (through
symmetry operation, 1-x,-1/2 þ y,1.5-z). The result is the presence
of dimeric structures in the crystal packing. Weak
p-p angular
Fig. 3. Dimeric structures present in the crystal packing of [Fe(
h
5-Cp)(CO)(PPh3)I] 1.
p-
p
angular interactions between phenyl and Cp groups are showed by thick dashed black lines.
Short contacts between CO molecules and phenyl groups are showed by fine dashed black lines.

40
A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
Fig. 4. Partial section in the crystal packing of [Fe(h
5-Cp)(CO)(P(Ph-p-F)3)I] 3.
compounds, where the Fe back-bonding into the CO
p
* orbital is
Supporting Information (Fig. S5). Among the ligands, only tris(4-
more important [33]. The Fe-C bond length is determined pre-
dominantly by the steric constraints of the phosphane ligand rather
than being a consequence of the bond. The distance between Fe and
fluorophenyl)phosphane showed some cytotoxicity at 72 h with
an IC50 of 60.81 3.43 mM (ligands IC50 is presented in Supporting
Information). None of the neutral compounds 1e3 was cytotoxic in
this cell line. Relatively to the cationic compounds 4e6, one can
observe the effect of the phosphane ligand on the overall cytotox-
icity. Compound 5 bearing 4-(diphenylphosphino)benzoic acid is
non-cytotoxic, while the compounds bearing tris(4-fluorophenyl)
phosphane 6 and triphenylphosphane 4 were found to be quite
cytotoxic against HeLa cells (in the micromolar range) being the
cytotoxicity dependent on the exposition time. These values are in
the same range for those obtained for [Fe(Cp)(dppe)(L)]þ com-
pounds (dppe ¼ ethylenebis(diphenylphosphane); L ¼ imidazole
based ligands) [19] and for [Fe(Cp)(CO)2X] (X ¼ halide, NCS, BFꢀ4 )
[22]. Noteworthy, comparing with the latter example, the
replacement of one CO ligand in [Fe(Cp)(CO)2I]22 by a phosphane
the centroid of the p-bonded cyclopentadienyl moiety is 1.669(15)
Å to Fe centre in 1 (ring slippage 0.026 Å) and 1.7403(16) Å for unit
A and 1.6793(17) Å for unit B in 3 (ring slippage, 0.077 Å and
0.088 Å, respectively). The mean value of the Fe-C bond distance is
2.072(14) Å in 1 and 2.123(18) Å for unit A and 2.110(3) Å for unit B,
in 3. Table 3 contains selected bond lengths and angles for the two
compounds.
3.4. Cytotoxicity of the iron complexes against HeLa cells
The cytotoxic effect of the iron complexes (1e6) was examined
on human cervical cancer cells (HeLa) using the XTT assay, a
colorimetric determination of cell viability during in vitro treatment
with a drug. Cells were exposed to each compound (Fe(II) com-
plexes, and all the ligands) continuously for a 24 or 72 h period and
then assayed for growth using the XTT end point assay. The IC50
values of iron complexes for the growth inhibition of Hela cells are
summarized in Table 4 and the IC50 curves can be found in
[Fe(Cp)(CO)(PR3)I] (1e3) led to a sharp decrease (6.7
vs. > 200 M at 24 h incubation) on the compounds cytotoxicity
emphasizing the importance of the compounds electronic features.
Substitution of iodine by a second acceptor ligand such as 4-
1.1
m
p
aminobenzonitrile, [Fe(Cp)(CO)(PR3)(C7H6N2)] (4 and 6) leads to
the reestablishment of the cytotoxicity with IC50 of 3.76 and
4.92 mM, for 4 and 6, respectively, at 72 incubation. These values are
better than those of cisplatin (7.0 2.70) in the same conditions
[19]. The relation between the electronic flow throughout the
compounds creating an asymmetric charge density in the molecule
and their cytotoxicity has been already established for some
ruthenium-cyclopentadienyl families of compounds [16,17]. The
lack of activity in the case of compound 5 bearing the 4-(diphe-
nylphosphino)benzoic acid seems to be related to the substituent
group on the phosphane ligand. One could postulate that the car-
boxylic acid group might be negatively charged in the cellular
media, which may well hinder its cellular uptake [34].
Table 3
Bond lengths [Å] and angles [ꢁ] for [Fe(
h
5-Cp)(CO)(PPh3)I]
1 -
and [Fe(h5
Cp)(CO)(P(Ph-p-F)3)I] 3.
Bond lengths
1
3
Fe(1)-I(1)
Fe(1)-C(2)
Fe(1)-P(1)
2.6218(16)
2.043(14)
2.227(3)
2.6226(17)
2.108(12)
2.209(3)
A atoms B atoms
2.123(17) 2.06(3)
2.075(18) 2.05(3)
2.137(18) 2.20(4)
2.173(17) 2.13(4)
1.775(19) 1.78(3)
1.07(2) 1.07(3)
Fe(1)-C(1)
Fe(1)-C(3)
Fe(1)-C(4)
Fe(1)-C(5)
Fe(1)-C(24)
O(1)-C(24)
2.068(15)
2.104(12)
2.083(14)
2.064(15)
1.806(14)
1.071(17)
3.5. Determination of cell death mechanism by Annexin V/PI assay
The cell death mechanism was evaluated by Annexin V/PI flow
cytometry for the two cytotoxic compounds 4 and 6. HeLa cells
were incubated at equitoxic concentrations (IC50 values) during
24 h and cisplatin was used as a reference. Annexin V binds
phosphatidyl serine residues, which are asymmetrically distributed
toward the inner plasma membrane but migrate to the outer
plasma membrane during apoptosis. As can be seen in Table 5, both
complexes, 4 and 6, induce apoptosis in great extent, comparable to
Bond angles
1
3
C(2)-Fe(1)-P(1)
98.8(4)
137.4(3)
A atoms B atoms
92.9(6) 83.7(13)
118.0(4) 111.2(8)
155.8(4) 151.5(8)
92.1(5) 83.9(10)
100.7(4) 97.4(7)
C(24)-Fe(1)-P(1)
C(5)-Fe(1)-P(1)
C(1)-Fe(1)-P(1)
C(4)-Fe(1)-P(1)
C(3)-Fe(1)-P(1)
93.2(4)
114.6(5)
88.8(4)
156.3(5)
136.9(4)

A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
41
Table 4
IC50 values (
m
M) of iron compounds (1e6), [Fe(Cp)(dppe)(L)]þ (dppe ¼ ethylenebis(diphenylphosphane); L ¼ imidazole based ligands),
[Fe(Cp)(CO)2X] (X ¼ halide, NCS, BFꢀ4 ), [Ru(Cp)(PPh3)2(clotrinazole)]þ and cisplatin against HeLa cells.
Complex
IC
(m
M)
IC50 (mM)
50
24 h
72 h
1
2
3
>200
>200
>200
>200
>100
>200
4
5
18.97 1.95
>200
3.76 1.07
>100
6
18.44 2.14
e
6.7 1.1e18.3 1.222
e
4.92 1.25
1.4 0.35e6.3 1.1019
e
[Fe(Cp)(dppe)(L)]þ
[Fe(Cp)(CO)2X]
[Ru(Cp)(PPh3)2(clotrinazole)]þ
Cisplatin
3.6 1.135
e
7.0 2.7019
Table 5
Percentage of HeLa cells in each state after treatment with complexes 4, 6 and Cisplatin at IC50 concentration for 24 h of incubation.
% vital cells
(R1)
% apoptotic cells
(R2)
% late apoptosis or dead cells
(R3)
% damaged cells
(R4)
Control
Cisplatin
4
6
90.52
62.19
57.60
67.28
4.67
3.25
1.57
0.53
1.12
1.42
23.85
24.63
18.25
13.43
16.64
13.05
the values found for cisplatin and correlating well with the cyto-
toxicity assays. This is a highly desirable feature for a drug candi-
date since apoptosis is a controlled mechanism of cell death.
funding from the Spanish Ministry of Economy Industry and
Competitiveness contract BIO2016-78057-R.
Appendix A. Supplementary data
4. Conclusion
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.jorganchem.2017.10.004.
A new family of Fe(II) compounds based on the general structure
[Fe(Cp)(CO)(PR3)(L)]n (PR3 ¼ triphenylphosphane, 4-(diphenyl-
phosphino)benzoic acid or tris(4-fluorophenyl)phosphane; when
L ¼ I, n ¼ 0; when L ¼ 4-aminobenzonitrile, n ¼ þ1) has been
successfully synthesized and characterized. Compound [Fe(Cp)(-
CO)(PPh3)I] 1 crystalizes in the orthorhombic space group P212121,
References
[1] S.V. Torti, F.M. Torti, Nat. Rev. Cancer 13 (2013) 342e355.
[2] T.J. Kealy, P.L. Pauson, Nature 168 (1951) 1039e1040.
[3] G. Wilkinson, M. Rosenblum, M.C. Whiting, R.B. Woodward, J. Am. Chem. Soc.
74 (1952) 2125e2126.
while compound [Fe(h
5-Cp)(CO)(P(Ph-p-F)3)I] 3 crystalizes in the
centrosymmetric space group Pbca. In both cases the iron centre
adopts the expected “piano stool” structure.
€
€
[4] P. Kopf-Maier, H. Kopf, E.W. Neuse, Angew. Chem. Int. Ed. Engl. 23 (1984)
456e457.
€
€
[5] P. Kopf-Maier, H. Kopf, E.W. Neuse, J. Cancer Res. Clin. Oncol. 108 (1984)
336e340.
The cationic compounds 4 and 6 were found to be cytotoxic
against cervical human cancer cells (HeLa) in the micromolar range,
inducing cell death by apoptosis within the same range as cisplatin.
Structural features observed on our spectroscopic data point out
that the asymmetric charge distribution created by the electronic
flow through the complexes might be related to their cytotoxic
activity. This effect can be explained by an enhancement of the
interaction of the compounds with the negatively charged cancer
cell membranes, disrupting the negative potential of cell mem-
branes. Importantly, compounds 4 and 6 present similar activity to
others Ru-Cp related compounds for HeLa cell line at the same time
of incubation [35]. This pinpoint that FeCp compounds, based on
bio essential and less expensive metal (relatively to others
commonly used in the search of metallodrugs), deserve more
detailed studies to unravel their mechanism of action.
[6] S.S. Braga, A.M.S. Silva, Organometallics 32 (2013) 5626e5639.
[7] G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem. 54 (2011) 3e25.
[8] M.J. Scandlyn, E.C. Stuart, T.J. Somers-Edgar, A.R. Menzies, R.J. Rosengren, Br. J.
Cancer 99 (2008) 1056e1063.
[9] H. Seeger, J. Huober, D. Wallwiener, A.O. Mueck, Horm. Metab. Res. 36 (2004)
277e280.
[10] M. Murty, Can. J. Physiol. Pharmacol. 85 (2007) 952e955.
^
[11] T.S. Morais, F.C. Santos, T.F. Jorge, L. Corte-Real, P.J.A. Madeira, F. Marques,
M.P. Robalo, A. Matos, I. Santos, M.H. Garcia, J. Inorg. Biochem. 130 (2014)
1e14.
ꢀ
[12] V. Moreno, M. Font-Bardia, T. Calvet, J. Lorenzo, F.X. Aviles, M.H. Garcia,
T.S. Morais, A. Valente, M.P. Robalo, J. Inorg. Biochem. 105 (2011) 241e249.
[13] M. Helena Garcia, T.S. Morais, P. Florindo, M.F.M. Piedade, V. Moreno,
C. Ciudad, V. Noe, J. Inorg. Biochem. 103 (2009) 354e361.
[14] A.I. Tomaz, T. Jakusch, T.S. Morais, F. Marques, R.F.M. de Almeida, F. Mendes,
E.A. Enyedy, I. Santos, J.C. Pessoa, T. Kiss, M.H. Garcia, J. Inorg. Biochem. 117
(2012) 261e269.
[15] A. Valente, M.H. Garcia, F. Marques, Y. Miao, C. Rousseau, P. Zinck, J. Inorg.
Biochem. 127 (2013) 79e81.
^
[16] L. Corte-Real, M. Paula Robalo, F. Marques, G. Nogueira, F. Avecilla, T.J.L. Silva,
Acknowledgements
F.C. Santos, A. Isabel Tomaz, M. Helena Garcia, A. Valente, J. Inorg. Biochem.
150 (2015) 148e159.
[17] T.S. Morais, A. Valente, A.I. Tomaz, F. Marques, M.H. Garcia, Future Med. Chem.
8 (2016) 527e544.
This work was financed by the Portuguese Foundation for Sci-
~
^
ence and Technology (Fundaçao para a Ciencia e Tecnologia, FCT)
within the scope of project UID/QUI/00100/2013. Andreia Valente
and Adhan Pilon acknowledge the Investigator FCT2013 Initiative
for the project IF/01302/2013 (acknowledging FCT, as well as POPH
and FSE - European Social Fund). Julia Lorenzo acknowledge
[18] N. Mendes, F. Tortosa, A. Valente, F. Marques, A. Matos, T.S. Morais, A.I. Tomaz,
€
F. Gartner, M.H. Garcia, Anticancer. Agents Med. Chem. 17 (2017) 126e136.
[19] A.C. Gonçalves, T.S. Morais, M.P. Robalo, F. Marques, F. Avecilla, C.P. Matos,
I. Santos, A.I. Tomaz, M.H. Garcia, J. Inorg. Biochem. 129 (2013) 1e8.
^
[20] A. Valente, A.M. Santos, L. Corte-Real, M.P. Robalo, V. Moreno, M. Font-Bardia,
T. Calvet, J. Lorenzo, M.H. Garcia, J. Organomet. Chem. 756 (2014) 52e60.

42
A. Pilon et al. / Journal of Organometallic Chemistry 852 (2017) 34e42
[21] P.R. Florindo, D.M. Pereira, P.M. Borralho, C.M.P. Rodrigues, M.F.M. Piedade,
A.C. Fernandes, J. Med. Chem. 58 (2015) 4339e4347.
[22] H.T. Poh, P.C. Ho, W.Y. Fan, RSC Adv. 6 (2016) 18814e18823.
[23] M.H. Garcia, M.P. Robalo, a R. Dias, M. Fa, J. Organomet. Chem. 619 (2001)
252e264.
1177e1181.
[30] M.H. Garcia, M.P. Robalo, A.P.S. Teixeira, A.R. Dias, M.F.M. Piedade,
M.T. Duarte, J. Organomet. Chem. 632 (2001) 145e156.
[31] A.N. Nesmeyanov, Y.A. Chapovskey, Y.A. Ustynyuk, J. Organomet. Chem. 9
(1967) 345e353.
[24] W.L.F. Armarego, C.L.L. Chai, Purif. Lab. Chem. (2009) 61e79.
[25] T.S. Piper, G. Wilkinson, J. Inorg. Nucl. Chem. 2 (1956) 38e45.
[26] G.M. Sheldrick, SADABS, Version 2.10, University of Gottingen, Germany,
[32] H.D. Flack, G. Bernardinelli, Chirality 20 (2008) 681e690.
[33] L. Mercs, G. Labat, A. Neels, A. Ehlers, M. Albrecht, Organometallics 25 (2006)
5648e5656.
€
€
2004.
[34] E. Frohlich, Int. J. Nanomedicine 7 (2012) 5577e5591.
[27] G.M. Sheldrick, Acta Crystallogr. Sect. A A64 (2008) 112e122.
[28] A. Cory, T. Owen, J. Barltrop, J. Cory, Cancer Commun. 3 (1991) 207e212.
[29] P.M. Treichel, R.L. Shubkin, K.W. Barnett, D. Reichard, Inorg. Chem. 5 (1966)
[35] E. Rodríguez Arce, C. Sarniguet, T.S. Moraes, M. Vieites, A.I. Tomaz,
ꢀ
A. Medeiros, M.A. Comini, J. Varela, H. Cerecetto, M. Gonzalez, F. Marques,
M.H. García, L. Otero, D. Gambino, J. Coord. Chem. 68 (2015) 2923e2937.