Cytotoxicity of ruthenium(II) piano-stool complexes with imidazole-based PN ligands

Article abstract of DOI:10.1016/j.jorganchem.2012.06.027

A series of p-cymene ruthenium(II) complexes with imidazol-2-yl phosphines as PN ligands was prepared. Depending on the number of imidazolyl substituents in the ligands Ph_3-nP(im)_n {1-3: n = 1-3, im = imidazol-2-yl (a), 1-methylimidazol-2-yl (b)} different coordination modes were observed: κP, κ²N,N or κ³N,N,N. The complexes were tested for their cytotoxicity in different cancer cell lines. Most of the compounds were found to be non-toxic; The compounds [(p-cymene)Ru(1a)Cl₂] (4a) shows cytotoxicity towards A2780sens and Hct116 cells in the μM range but not in H4IIE cells. The cytotoxicity is decreased upon introduction of a methyl group as [(p-cymene)Ru(1b)Cl ₂] (4b) shows only modest toxicities in the cell lines investigated. The κP compound [(p-cymene)Ru(2a)Cl₂] (5a) shows selective toxicity in H4IIE cells after 72 h whereas the κ²N,N compound [(p-cymene)Ru(2a)Cl]OTf (5a′) showed no toxicity in the cell lines investigated which again.

Full text of DOI:10.1016/j.jorganchem.2012.06.027

Journal of Organometallic Chemistry 717 (2012) 187e194
Contents lists available at SciVerse ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Cytotoxicity of ruthenium(II) piano-stool complexes with imidazole-based PN
ligands
,
,
,b,
*
Wilhelm Huber b, Philip Bröhler c, Wim Wätjen c, Walter Frank b 1, Bernhard Spingler d 1, Peter C. Kunz a
a Institut für Pharmazeutische und Medizinische Chemie, Heinrich-Heine-Universität, Universitätsstr. 1, D-40225 Düsseldorf, Germany
b Institut für Anorganische Chemie und Strukturchemie, Heinrich-Heine-Universität, Universitätsstr. 1, D-40225 Düsseldorf, Germany
c Universitätsklinikum Düsseldorf, Institut für Toxikologie, Heinrich-Heine-Universität, Universitätsstr. 1, D-40225 Düsseldorf, Germany
d University of Zürich, Institute of Inorganic Chemistry, Winterthurerstrasse 190, CH-8057 Zürich, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
A
series of p-cymene ruthenium(II) complexes with imidazol-2-yl phosphines as PN ligands
Received 4 May 2012
Received in revised form
22 June 2012
was prepared. Depending on the number of imidazolyl substituents in the ligands Ph3ꢀnP(im)n {1e3:
n ¼ 1e3, im ¼ imidazol-2-yl (a), 1-methylimidazol-2-yl (b)} different coordination modes were
observed: kP, k k
2N,N or 3N,N,N. The complexes were tested for their cytotoxicity in different cancer cell
Accepted 27 June 2012
lines. Most of the compounds were found to be non-toxic; The compounds [(p-cymene)Ru(1a)Cl2] (4a)
shows cytotoxicity towards A2780sens and Hct116 cells in the M range but not in H4IIE cells. The
cytotoxicity is decreased upon introduction of a methyl group as [(p-cymene)Ru(1b)Cl2] (4b) shows only
m
Keywords:
Arene complexes
Cytotoxicity
Metal-based drugs
PN ligands
modest toxicities in the cell lines investigated. The kP compound [(p-cymene)Ru(2a)Cl2] (5a) shows
selective toxicity in H4IIE cells after 72 h whereas the k2N,N compound [(p-cymene)Ru(2a)Cl]OTf (5a0)
showed no toxicity in the cell lines investigated which again.
Ó 2012 Elsevier B.V. All rights reserved.
Ruthenium
1. Introduction
Interestingly, NAMI-A is more active against metastases than
against primary tumours [19]. Half-sandwich Ru(II) arene
Since the authorization of cisplatin in 1978, the interest in and
development of metal-based drugs prospers consistently. Still,
cisplatin, cis-[Pt(NH3)2Cl2], and its analogues, especially oxaliplatin
and carboplatin, are basic chemotherapeutics in combination
therapy [1e3]. The therapeutic effect of cisplatin is based on DNA
platination which triggers apoptosis [4]. The major draw-back for
a successful chemotherapy by platinum-based drugs is acquired
resistance towards the applied drug during the course of therapy
[5]. In order to circumvent those resistance mechanisms drugs,
which address alternative cellular targets have to be developed
[6e8].
The most promising metallodrugs besides cisplatin analogues
are ruthenium-based drugs [9e13]. Compounds of Ru(II) and Ru(III)
are able to overcome cisplatin resistance. Their cytotoxicity and
antimetastatic properties are combined with low overall
toxicity [9,14e17]. Trans-[RuCl4(DMSO)(Im)]ImH (NAMI-A, where
complexes of the type [(h
6-arene)Ru(YZ)(X)], where YZ is a biden-
tate chelating ligand and X is a good leaving group, show promising
in vitro and in vivo anticancer activity [16]. These compounds
coordinate to guanine N7 of DNA, which can be complemented by
intercalative binding of an extended arene, as well as specific
hydrogen-bonding interactions [20,21]. For example, increasing the
size of the coordinated arene is accompanied by an increase in
activity in human ovarian cancer cell lines [16] and the nature of
the chelating ligand YZ and leaving group X seems influence their
kinetics and even can change their nucleobase selectivity [22]. The
RAPTA family of organometallic Ru(II) compounds contain the
water-soluble phosphine ligand phosphaadamantane (pta) or
derivatives thereof. Usually these compounds exhibit moderate
in vitro activity, and some compounds show no activity in healthy
cells up to millimolar concentrations. The pta compounds show
little activity against primary tumours in vivo, although they exhibit
some capacity to reduce lung metastases derived from a mammary
Im
¼
imidazole) has completed phase I clinical trials [18].
carcinoma xenograft grown in mice [23]. The cytotoxicity of [Ru(h6
-
p-cymene)Cl2(pta)], in EAC cells is thought to be mediated by
mitochondrial and Jun-N (amino)-terminal kinase (JNK) e p53
pathways [24]. For all Ru(II) compounds it is believed that in vivo,
analogous to cisplatin, aquation of the chlorido complex is largely
suppressed in intracellular fluids (with chloride concentrations are
* Corresponding author. Institut für Pharmazeutische und Medizinische Chemie,
Heinrich-Heine-Universität, Universitätsstr. 1, D-40225 Düsseldorf, Germany.
Tel.: þ49 211 81 12873; fax: þ49 211 81 12287.
E-mail address: peter.kunz@uni-duesseldorf.de (P.C. Kunz).
1
X-ray structure analysis.
0022-328X/$ e see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2012.06.027

188
W. Huber et al. / Journal of Organometallic Chemistry 717 (2012) 187e194
about 100 mM), whereas in the cell nucleus with a much lower
chloride concentration (ca. 4 mM) the active aqua species forms
[25,26]. Although their mechanism of action is still largely
unknown, there is some evidence that RAPTA compounds work on
molecular targets other than DNA [27e29], implying a biochemical
mode of action profoundly different from classical platinum anti-
cancer drugs.
to a suspension of 2a (80 mg, 0.33 mmol) in dry CH3CN (10 mL). The
reaction mixture was refluxed for 1 h and stirred for 24 h at
ambient temperature. The resulting yellow solution was concen-
trated to ca. 3 mL and Et2O was added. The mixture was kept
at ꢀ18 ꢁC. The yellow precipitate was filtered off, washed with Et2O
and dried in vacuo. Yield: 154 mg (70%). 1H NMR (200 MHz,
methanol-d4):
d
¼ 1.31 (d, J ¼ 6.9 Hz, 6H), 2.04 (s, 3H), 2.91 (sept.,
We are currently examining the use of imidazole-based PN
ligands in biomedical applications [30e32] as well as in catalysis
[33,34]. Here we present coordination chemistry of these PN
J ¼ 6.9 Hz, 1H), 5.73 (m, 4H), 7.30 (d, J ¼ 1.47 Hz, 2H), 7.46 (d,
J ¼ 1.5 Hz, 2H), 7.68 (m, 5H). 31P{1H}-NMR (81 MHz, methanol-d4):
d
¼ ꢀ22 (s). ESI-MS (CH3OH): m/z (%) ¼ 493.5 (100) [M þ O]þ, 477.5
ligands towards (
h
6-cymene)Ru(II) and basic cytotoxicity studies in
(15) [M]þ, C23H25ClF3N4O3PRuS$½ H2O (671.03): calc. C 41.16, H
different cancer cell lines.
3.91, N 8.49; found C 41.15, H 3.42, N 8.43.
2. Experimental section
2.1.4. [(cym)Ru(k
2N,N-2b)Cl]Cl (5b)
[Ru(cym)Cl2]2 (100 mg, 0.16 mmol) and 2b (118.9 mg,
0.33 mmol) were dissolved in dry CH3CN (15 mL) and stirred for
24 h. The orange solution was concentrated to 5 ml and Et2O was
added. The resulting solid was filtered off, washed with Et2O and
dried in vacuo. Yield: 158 mg (84%). 1H NMR (200 MHz, CDCl3):
The ligands Ph3ꢀnP(im)n {1e3: n ¼ 1e3, im ¼ imidazol-2-yl (a),1-
methylimidazol-2-yl (b)} and [(cym)Ru(kP-1b)Cl2] (4b) were
prepared according published procedures [32,34e37]. All reactions
were carried out in Schlenk tubes under an atmosphere of dry
nitrogen using anhydrous solvents purified according to standard
procedures. All chemicals were purchased from commercial sources
and used as received. 1H and 31P NMR spectra were recorded on
a Bruker DRX 200 and Bruker DRX 500 spectrometer. The 1H spectra
were calibrated against the residual proton signal of the solvent as
d
¼ 1.25 (d, J ¼ 7.0 Hz, 6H), 1.67 (s, 3H), 2.66 (sept., J ¼ 7.0 Hz, 1H),
4.20 (s, 6H), 4.94 (m, 4H), 7.15 (d, J ¼ 1.4 Hz, 2H), 7.56 (m, 5H), 7.83
(d, J ¼ 1.4 Hz, 2H). 31P{1H}-NMR (81 MHz, CDCl3):
d
¼ ꢀ60 (s). ESI-
MS (CH3OH): m/z (%) ¼ 541.4 (76) [M]þ, 505.5 (35) [M ꢀ Cl]þ, 407.3
(100) [M
ꢀ
C10H14]þ, 371.4 (37) [M
ꢀ
C10H14
ꢀ
Cl]þ.
an internal reference (methanol-d4: dH
¼
3.31 ppm; D2O:
C24H29Cl2N4PRu$2H2O (612.50): calc. C 47.1, H 5.4, N 9.1; found C
dH ¼ 4.79 ppm, CDCl3: dH ¼ 7.26 ppm) while the 31P{1H} NMR spectra
were referenced to external 85% H3PO4. The MALDI mass spectra
were recorded on a Bruker Ultraflex MALDI-TOF mass spectrometer.
The elemental composition of the compounds was determined with
a PerkinElmer Analysator 2400 at the Institut fur Pharmazeutische
und Medizinische Chemie, Heinrich-Heine-Universitat Dusseldorf.
47.1, H 5.3, N 9.2.
2.1.5. [(cym)Ru(k
3N,N,N-3a)]Cl2 (6a)
[Ru(cym)Cl2]2 (50 mg, 0.082 mmol) and 3a (38.2 mg, 0.16 mmol)
were dissolved in dry CH3CN (30 mL) and refluxed for 2 h. The
yellow precipitate was collected by filtration, washed with a small
amount of CH3CN and dried in vacuo. Yield: 21 mg (24%). 1H NMR
2.1. Synthesis of (cym)Ru complexes
(200 MHz, methanol-d4):
d
¼ 1.20 (d, J ¼ 6.7 Hz, 6H), 2.45 (s, 3H),
3.24 (sept., J ¼ 6.7 Hz, 1H), 6.28 (m, 4H), 7.47 (dd, J ¼ 1.6 Hz,
2.1.1. [(cym)Ru(kP-1a)Cl2] (4a)
J ¼ 2.94 Hz, 3H), 8.23 (d, J ¼ 1.6 Hz, 3H). 31P{1H}-NMR (81 MHz,
Ligand 1a (83 mg, 0.33 mmol) and [Ru(cym)Cl2]2 (100 mg,
0.16 mmol) were dissolved in dry CH2Cl2 (15 mL) and stirred for
24 h. The dark red solution was concentrated to 5 mL and Et2O was
added. The precipitate was collected and dissolved in thf, filtered
and again precipitated upon addition of n-hexane. The red solid
was filtered off and dried in vacuo. Yield: 52 mg (28%). 1H NMR
methanol-d4):
d
¼ ꢀ103 (s). ESI-MS (CH3OH): m/z (%) ¼ 467.3 (100)
[M]þ, 234 (58) [M ꢀ 3a]þ. C19H23Cl2N6PRu$5/2H2O (583.38): calc. C
39.1, H 4.8, N 14.4; found C 39.3, H 4.5, N 14.1.
2.1.6. [(cym)Ru(k
3N,N,N-3b)]Cl2 (6b)
[Ru(cym)Cl2]2 (100 mg, 0.16 mmol) and 3b (91 mg, 0.33 mmol)
were dissolved in dry CH3CN (25 mL) and refluxed for 1.5 h. The
yellow precipitate was collected, washed with a small amount of
CH3CN and dried in vacuo. Yield: 26 mg (18%). 1H NMR (200 MHz,
(200 MHz, methanol-d4):
d
¼ 0.97 (d, J ¼ 7.0 Hz, 6H), 1.83 (s, 3H),
2.42 (sept., J ¼ 7.0 Hz, 1H), 5.42 (m, 4H), 7.10 (d, J ¼ 1.2 Hz, 2H), 7.72
(m, 10H). 31P{1H}-NMR (81 MHz, methanol-d4):
d
¼ 22 (s). EI-MS
(CH3OH): m/z (%) ¼ 558 (40) [M]þ, 523 (28) [M ꢀ Cl]þ, 486 (100)
[M ꢀ 2Cl]þ, 389 (27) [M ꢀ C10H14]þ, 352 (45) [M ꢀ C10H14 ꢀ Cl]þ.
C25H27Cl2N2PRu (558.45): calc. C 53.77, H 4.87, N 5.02; found C
53.44, H 4.98, N 4.77.
methanol-d4):
d
¼ 1.23 (d, J ¼ 6.7 Hz, 6H), 2.45 (s, 3H), 3.23 (sept.,
J ¼ 6.7 Hz, 1H), 4.03 (s, 9H), 6.29 (m, 4H), 7.55 (dd, J ¼ 1.6 Hz,
J ¼ 4.0 Hz, 3H), 8.21 (d, J ¼ 1.6 Hz, 3H). 31P{1H}-NMR (81 MHz,
methanol-d4):
d
¼ ꢀ116 (s). ESI-MS (CH3OH): m/z (%) ¼ 509 (100)
[M ꢀ Cl]þ, 461 (29) [M ꢀ C4N2H6 þ Cl]þ. C22H29Cl2N6PRu (580.46):
2.1.2. [(cym)Ru(kP-2a)Cl2] (5a)
calc. C 45.5, H 5.0, N 14.5; found C 45.2, H 4.9, N 14.1.
Ligand 2a (50 mg, 0.21 mmol) and [Ru(cym)Cl2]2 (63 mg,
0.1 mmol) were dissolved in dry CH3CN (25 mL) and stirred for 24 h.
The dark red solution was concentrated to 5 mL. The red precipitate
2.1.7. [(cym)Ru(k
2N,N-en)Cl]Cl (7)
[Ru(cym)Cl2]2 (100 mg, 0.16 mmol) was solved in dry CH3CN
(25 mL). A yellow solid precipitated upon addition of an excess of
ethylenediamine (en). The reaction mixture was stirred for 20 min
to complete the reaction. The solid was filtered off, washed with dry
CH3CN and dried in vacuo. Yield: 99 mg (84%). 1H NMR (200 MHz,
was filtered off, washed with Et2O and dried in vacuo. Yield: 27 mg
̈
(24%). 1H NMR (200 MHz, CDCl3):
d
¼ 0.97 (d, J ¼ 7.0 Hz, 6H), 1.76 (s,
̈
̈
3H), 2.47 (sept., J ¼ 7.0 Hz, 1H), 5.83 (m, 4H), 7.26 (d, J ¼ 1.2 Hz, 4H),
7.39 (m, 5H). 31P{1H}-NMR (81 MHz, CDCl3):
d
¼ ꢀ1 (s). ESI-MS
(CH3OH): m/z (%) ¼ 513.4 (43) [M ꢀ Cl]þ, 477.4 (100) [M ꢀ 2Cl]þ.
C22H25Cl2N4PRu$H2O (566.43): calc. C 46.65, H 4.80, N 9.89; found C
47.08, H 5.20, N 9.65.
methanol-d4):
d
¼ 1.27 (d, J ¼ 7.0 Hz, 6H), 2.43 (s, 3H), 2.70 (m, br,
4H), 2.81 (sept., J ¼ 7.0 Hz, 1H), 5.81 (m, 4H). ESI-MS (CH3OH): m/z
(%) ¼ 295 (100) [M ꢀ Cl]þ, 235 (34) [M ꢀ Cl ꢀ en]þ.
2.1.3. [(cym)Ru(k2N,N-2a)Cl]OTf (5a0)
2.2. Distribution coefficients (log D)
[Ru(cym)Cl2]2 (101 mg, 0.16 mmol) and AgOTf (85 mg,
0.33 mmol) were dissolved in dry CH3CN (15 mL) and refluxed for
1 h. Precipitated AgCl was filtered off and the red filtrate was added
The n-octanolewater distribution coefficients of the compounds
were determined using a shake-flask method. PBS buffered bi-

W. Huber et al. / Journal of Organometallic Chemistry 717 (2012) 187e194
189
distilled water (100 mL, phosphate buffer, c(PO34e) ¼ 10 mM,
graphite monochromatized Mo K
a
radiation (
l
¼ 0.71073 A). Unit
ꢀ
c(NaCl) ¼ 0.15 M, pH adjusted to 7.4 with HCl) and n-octanol
(100 mL) were shaken together using a laboratory shaker (Perki-
nElmer), for 72 h to allow saturation of both phases. 1 mg of each
compound was mixed in 1 mL of aqueous and organic phase,
respectively for 10 min using a laboratory vortexer. The resultant
emulsion was centrifuged (3000g, 5 min) to separate the phases.
The concentrations of the compounds in the organic and aqueous
phases were then determined using UV absorbance spectroscopy
(230 nm). log DpH was defined as the logarithm of the ratio of the
concentrations of the complex in the organic and aqueous phases
log D ¼ log([compound(org)]/[compound(aq)]), the value reported is
the mean of three separate determinations.
cell parameters were determined by least-squares refinements on
the positions of 8000 reflections. Space group type no. 14 was
uniquely determined for both compounds. Corrections for Lorentz
and polarization effects were applied. The structures were solved
by direct methods (SHELXS-86) [40] and subsequent DF-syntheses.
Approximate positions of all the hydrogen atoms were found in
different stages of converging refinements by full-matrix least-
squares calculations on F2 [40]. Anisotropic displacement parame-
ters were refined for all atoms heavier than hydrogen. With
idealized bonds lengths and angles assumed for all the CH, CH2 and
CH3 groups, the riding model was applied for the corresponding H
atoms and their isotropic displacement parameters were con-
strained to 120%, 120% and 150% of the equivalent isotropic
displacement parameters of the parent carbon atoms, respectively.
In addition, the H atoms of the CH3 groups were allowed to rotate
around the neighbouring CeC bonds. Crystallographic data of
5b,2CH3CN$½H2O were collected at 183(2) K on an Oxford
2.3. DNA binding studies
The UV/Vis kinetic studies and thermal denaturation tempera-
ture Tm determinations for 1:5 complex/DNA mixtures [DNA
concentration ¼ M(base pairs)] were performed in a 10 mM phos-
phate buffer at pH ¼ 7.4. Melting curves were recorded at 2 ꢁC steps
for the wavelength 260 nm with an Analytik Jena SPECORD 100
Diffraction Xcalibur system with a Ruby detector using Mo K
a
ꢀ
radiation (
l
¼ 0.7107 A) that was graphite-monochromated. Suit-
able crystals were covered with oil (Infineum V8512, formerly
known as Paratone N), mounted on top of a glass fibre and
immediately transferred to the diffractometer. The program suite
CrysAlisPro was used for data collection, multi-scan absorption
correction and data reduction [41]. The structure was solved with
direct methods using SIR97 [42] and was refined by full-matrix
least-squares methods on F2 with SHELXL-97 [40]. The structure
was checked for higher symmetry with help of the program
Platon [43].
spectrometer and a thermostat.
DTm values were calculated by
determining the midpoints of melting curves. The experimental DTm
values are estimated to be accurate within ꢂ1 ꢁC. Concentrations of
calf-thymus (ct) DNA were determined spectrophotometrically
using the molar extinction coefficient 3
¼ 13200 Mꢀ1 cmꢀ1 [38].
260
2.4. Cell culture
Hct116 human colon carcinoma and H4IIE rat hepatoma cells
were grown in Dulbecco’s modified Eagle’s medium (DMEM,
GIBCO; Germany), A2780 human ovarian carcinoma cells were
grown in RPMI cell culture medium; all media contained 10% foetal
calf serum (PAA Laboratories; Austria), penicillin (100 U/mL) and
3. Results and discussion
3.1. Syntheses and characterization
streptomycin (100
m
g/mL) at 5% CO2 and 37 ꢁC.
We prepared the (cym)Ru complexes (cym ¼ p-cymene) by
reaction of [{(cym)RuCl2}2] and the corresponding ligands
1a,be3a,b in acetonitrile (Scheme 1). The corresponding complexes
precipitated from solution or were obtained in analytically pure
form after addition of Et2O.
2.5. Determination of cytotoxicity
The effect of the compounds on cell viability was determined
using the MTT assay [39]. Cells were plated on 96-multiwell plates
(H4IIE, Hct116 cells: 15.000/well, A2780 cells: 35.000/well),
allowed to attach for 24 h and then treated with different
concentrations of the substances for indicated time points. In all
experiments compounds were dissolved in DMSO. The DMSO
concentration was equal at all compound concentrations analyzed.
The highest DMSO concentration used was 1%; no toxic effect was
detected at this concentration. After treatment medium was
changed and cells were incubated for 30 min under cell culture
conditions with 1 mg/mL MTT. Then the cells were lysed with 100%
DMSO. The concentration of reduced MTT as a marker for cell
viability was measured photometrically (560 nm) using a Wallace
Victor2 1420 multilabel counter (PerkinElmer).
All compounds exhibit sharp singlet resonances in their 31P{1H}
NMR spectra (Table 1). The analytical data of complex 4a are nearly
identical to the one of 1b which has been previously described by
Caballero et al. Complexes 6a and 6b show one set of signals for the
ligands in the 1H NMR (methanol-d4), indicating local C3v
symmetry. Additionally, their 31P{1H} NMR spectra show the typical
coordination shift to about ꢀ110 ppm which is indicative for the
k
3N,N,N-coordination mode in complexes of tris(imidazolyl)phos-
phine ligands [44e46].
The ligands 1a/b in complexes 4a and 4b show the kP coordi-
nation mode and the ligands 3a/b in complexes 6a and 6b show
the
k
3N,N,N mode. The coordination modes found in the isolated
complexes of ligands 2a and 2b are more complex. Initially, the
reaction of [{(cym)RuCl2}2] with the PN phosphine ligands
proceeds via a P coordinated species. The course of the reactions
was monitored by 31P NMR and even during formation of 6a and
2.6. Statistical analysis
All data were analyzed using one-way analysis of variance, fol-
lowed by Bonferroni or Dunnet post hoc analysis to determine
statistical significance. P values < 0.05 were considered statistically
significant. The analysis was performed with GraphpadPrism 5.0c.
6b signals for a transient species, tentatively assigned to
(see Fig. ESI1), was observed. All attempts to ‘trap’
ligands 3a/b, e.g. by reaction of these ligands with [(cym)Ru(en)Cl]
(7), finally resulted in formation of compounds 6a and 6b,
respectively.
k
k
P species
P-bound
2.7. Crystallography
When ligands 2a and 2b were used,
compete with the chelating
2N,N binding mode. Those coordina-
tion modes were found in compounds 5a0 and 5b (both 2N,N) and
5a ( P) which were unambiguously assigned by their single crystal
structures (see below). As mentioned previously, P coordination
kP coordination can
k
Crystals of compounds 5a and 5a0$CH2Cl2 suitable for X-ray
study were selected by means of a polarization microscope and
investigated with an STOE Imaging Plate Diffraction System, using
k
k
k

190
W. Huber et al. / Journal of Organometallic Chemistry 717 (2012) 187e194
Ru
Cl
Cl
Cl
Ph2P
4a (R = H)
+ 1a
N
4b (R = CH3)
or 1b
Ru
Cl
RN
CH3CN
Cl
Cl
Cl
Ru
Cl
NCCH3
Ru
Cl
+ 2a
+ 2a
+ AgOTf
– AgCl
Ru
+ 3a
or 3b
Cl
+ 2b
5a
H
N
PhP
N
2
Ru
Ru
Ru
+ AgOTf
– AgCl
2Cl
N
Cl
Cl
N
Cl
OTf
N
N
N
N
N
N
R
N
N
N
R
N
N
H
N
R
H
P
P
P
Ph
Ph
6a (R = H)
6b (R = CH3)
5b
5a'
Scheme 1. Reaction of [{(cym)RuCl2}2] in acetonitrile with ligands 1e3 yields compounds 4e6.
leads to a classical coordination shift of the 31P resonance towards
lower field whereas chelating
2N,N (or 3N,N,N in 6a,b) results in
a shift towards higher field (Table 1). The reaction of 2b with
[{(cym)RuCl2}2] yields 5b, where only
2N,N coordination of 2b is
observed. The related ligand 2a can adopt both coordination P and
2N,N modes depending on the chloride concentration in solution.
The reaction of stoichiometric amounts of [{(cym)RuCl2}2] and 2a
gives a mixture of [(cym)Ru(
2P-2a)Cl2] (5a) and [(cym)Ru( 2N,N-
solvate 5b$3CH3CN$½H2O in the monoclinic space group P 21/n.
k
k
The ruthenium atom in all structures is in octahedral coordination
sphere with the
octahedron. In 5a the other three positions are occupied by two
chlorido ligands and
P-bound 2a (Fig. 2). In 5a0 and 5b the 2N,N-
h
6-cymene ligand occupying one face of the
k
k
k
k
k
coordination mode of the PNN ligands 2a and 2b is found (Fig. 3
and Fig. 4). The ruthenium atom is coordinated by 2a/b in the
k
k
chelating k
2N,N mode and a chlorido ligand. The other chloride
2a)Cl]þ and unreacted ligand (Fig. 1). Addition of one equivalent of
Ag(O3SCF3) to that reaction mixture quantitatively yields 5a0. This
reaction is not reversible as 5a0 is persistent even in the presence of
high chloride concentration (up to >100 fold) as was shown by 31P
NMR spectroscopy. Therefore small electronic changes in the ligand
system, like introduction of electron donating methyl groups, can
favour one binding mode over the other. This might be of impor-
tance as this might alter their cytotoxicity profiles, e.g. in the
presence of other coordinating ligands as chloride.
or CF3SOꢀ3 acts as counter ion to the complex cations,
respectively. The metric parameters found in 5a and 5a0/5b are
within the range found for other compounds [(cym)Ru(PR3)Cl2]
[31,35,47e49] and complexes [(cym)Ru(NeN)Cl]þ with diamino
ligands [31,50,51].
In the solid-state structures of 5a and 5b the phenyl substitu-
ents of the corresponding ligand 2a/b and the p-cymene ligand at
the ruthenium atom adopt cis (5b) and trans positions (5a0)
(Scheme 2). The compounds 5a and 5a0, bearing ligand 2a with NH
functionalities, show hydrogen bonding in their solid-state
structures.
3.2. Solid-state structures
In the solid state of 5a0 intermolecular hydrogen bonds are
formed between N2H1/O1(triflate) and N4H2/Cl1 thus forming
one-dimensional arrays. Also in the solid state of 5a the intermo-
lecular hydrogen bonds N3H2/Cl2 result in formation of one-
dimensional arrays. Additionally a weak bifurcated intramolecular
hydrogen bond between N1H1 and Cl1 and Cl2 is found.
The solid-state structures of compounds 5a0, 5a and 5b were
determined by single crystal analysis and crystallographic data is
summarized in Table 2. Compounds 5a0 and 5a crystallized in the
monoclinic space group P 21/c. Compound 5b crystallized as
Table 1
3.3. Biological studies
31P{1H} resonances (dC) of the (cym)Ru complexes 4a,be6a,b and the corresponding
coordination shifts (Dd
¼
dC
ꢀ
dL).
Biological studies were performed on compounds 4a,b, 5a,b and
5a0 as well as 6a and 6b. The n-octanolewater distribution coeffi-
cients of compounds 4a,b, 5a,b and 5a0 were determined as log D7.4
values using phosphate buffered saline (PBS). The log D7.4 value
decreases within the series 4a > 4b > 5a0 > 5a > 5b (Table 3). The
Compound
d
(
31P)/ppm
Dd(
31P)/ppm
4a
4b
5a
5a0
5b
6a
6b
22a
þ45
þ35
þ46
þ23
ꢀ15
ꢀ43
ꢀ44
8b
1b
ꢀ22a
introduction of additional imidazolyl groups in the kP complexes
ꢀ60b
ꢀ103a
ꢀ116a
increases the water-solubility of the compounds as does the
introduction of charge in complexes 5a/b. Surprisingly, the
complexes having N-methyl groups in the PN ligands are more
water soluble than their NH-congeners.
a
In methanol-d4.
In CDCl3.
b

W. Huber et al. / Journal of Organometallic Chemistry 717 (2012) 187e194
191
Table 2
Crystallographic data for compounds 5a, 5a0 and 5b$2CH3CN$½H2O.
Compound
5a
5a0$CH2Cl2
5b$2CH3CN$½H2O
Empirical formula
Formula weight
Temperature/K
Wavelength/A
Crystal system
Space group
C
22H25Cl2N4PRu
C24H27Cl3F3N4O3PRuS
746.96
291(2)
0.71073
Monoclinic
P 21/c
C28H36Cl2N6O0.5PRu
667.57
183(2)
0.71073
Monoclinic
P 21/n
548.40
291(2)
0.71073
Monoclinic
P 21/c
ꢀ
Unit cell dimensions
ꢀ
a/A
17.2041(9)
9.7578(8)
14.1705(7)
90
103.729(6)
90
2310.9(3)
4
1.576
0.996
1112
0.3 ꢃ 0.3 ꢃ 0.3
2.42e24.99ꢁ
ꢀ20 ꢄ h ꢄ 20,
ꢀ11 ꢄ k ꢄ 11,
ꢀ16 ꢄ l ꢄ 16
29,702
14.985(2)
7.7408(8)
26.642(4)
90
98.273(18)
90
3058.2(7)
4
1.622
0.947
1504
0.2 ꢃ 0.2 ꢃ 0.2
1.91e25.92ꢁ
ꢀ18 ꢄ h ꢄ 18,
ꢀ9 ꢄ k ꢄ 9,
ꢀ32 ꢄ l ꢄ 32
35,580
10.98070(16)
12.8114(2)
23.9097(4)
90
93.5635(14)
90
3357.07(9)
4
1.321
ꢀ
b/A
ꢀ
c/A
ꢁ
ꢁ
a
/
/
b
ꢁ
g
/
3
ꢀ
Volume/A
Z
Density (calculated)/Mg/m3
Absorption coefficient/mmꢀ1
F(000)
0.701
1372
Crystal size/mm3
0.24 ꢃ 0.18 ꢃ 0.11
2.63e33.14ꢁ
ꢀ16 ꢄ h ꢄ 14,
ꢀ19 ꢄ k ꢄ 13,
ꢀ36 ꢄ l ꢄ 36
31,289
Theta range for data collection
Index ranges
Reflections collected
Independent reflections
Completeness to theta
Absorption correction
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2sigma(I)]
R indices (all data)
4013 [R(int) ¼ 0.0438]
98.3% to 24.99ꢁ
None
5920 [R(int) ¼ 0.0977]
99.4% to 25.92ꢁ
None
12,777 [R(int) ¼ 0.0355]
99.9% to 33.14ꢁ
Semi-empirical from equivalents
Full-matrix least-squares on F2
12,777/7/380
Full-matrix least-squares on F2
4013/0/279
1.045
Full-matrix least-squares on F2
5920/0/364
0.932
0.967
R1 ¼ 0.0283, wR2 ¼ 0.0665
R1 ¼ 0.0365, wR2 ¼ 0.0681
0.626 and ꢀ0.323
R1 ¼ 0.0373, wR2 ¼ 0.0603
R1 ¼ 0.1079, wR2 ¼ 0.0653
0.569 and ꢀ0.650
R1 ¼ 0.0456, wR2 ¼ 0.1232
R1 ¼ 0.0790, wR2 ¼ 0.1324
1.027 and ꢀ0.703
ꢀꢀ3
Largest diff. peak and hole/e A
A decrease in thermal denaturation temperature of 7 ꢁC in the
presence of 4a, 4b and 5a0, respectively and of 4 ꢁC in the presence
5b was recorded for calf-thymus DNA (ctDNA) at a molar ratio of
r ¼ 0.2 where [DNA] is given in M(base pairs) (Table 3 and
Supporting information). Interestingly, the melting curve in the
presence of 5a shows two stages corresponding to decreases in
thermal denaturation temperature of 8 and 28 ꢁC respectively. This
might reflect a rearrangement of the coordination mode of the
ligand in the complex as discussed before.
The cytotoxicity of the compound was determined towards
three different cell lines using the MTT assay method. For
comparison, the compound [(cym)Ru(en)Cl]Cl (7) was also intro-
duced in the cell line studies (Table 4 and Fig. 5). Cell lines used
were Hct116 human colon carcinoma, H4IIE rat hepatoma
and A2780 human ovarian carcinoma cells (cisplatin sensitive).
The cytotoxicity values for these compounds fall in the range
commonly observed for various Ru(arene)-type complexes
[31,52,53].
Fig. 1. 31P{1H}-NMR spectrum of the reaction mixture of [{(cym)RuCl2}2] and 2a in DMSO-d6. The ligand 2a and complexes [(cym)Ru( 2P-2a)Cl2] (5a) and [(cym)Ru( 2N,N-2a)Cl]þ
k k
with
k
P and
k
2N,N-coordination mode show resonances at ꢀ50, 3 and 23 ppm.

192
W. Huber et al. / Journal of Organometallic Chemistry 717 (2012) 187e194
Fig. 4. Molecular structure of 5b$2CH3CN$½H2O. Uncoordinated counter ion,
hydrogen atoms and solvent molecules are omitted for clarity. The displacement
ꢁ
ꢀ
ellipsoids are shown on a 50% level. Selected bond lengths [A] and angles [ ]: Ru1-N1
2.090(2), Ru1-N3 2.097(2), Ru1-Cl1 2.3897(7), N1-Ru1-N3 2.3897(7), N1-Ru1-Cl1
85.21(6) N3-Ru1-Cl1 84.61(6).
Fig. 2. Molecular structure of 5a. Non-acidic hydrogen atoms are omitted for clarity.
ꢀ
The displacement ellipsoids are shown on a 50% level. Selected bond lengths [A] and
angles [ꢁ]: Ru1-P1 2.3477(7), Ru1-Cl1 2.4173(7), Ru1-Cl2 2.4165(8), P1-Ru1-Cl1
87.41(3), P1-Ru1-Cl2 86.60(3), Cl1-Ru1-Cl2 87.39(3).
+
+
iPr
iPr
Ph
Ru
Ru
N
NR
N
NR
As expected, the complexes coordinatively saturated 6a and 6b
Cl
Cl
P
P
N
N
are not cytotoxic at concentrations up to 100
72 h of incubation, respectively (Table 4). Compounds 5a0 and 5b
containing
2N,N bonded ligands show no toxicity in the cell lines
mM after 24, 48 and
Ph
NR
NR
k
Scheme 2. The trans and cis isomers found in the solid-state structures of 5a0 (R ¼ H)
and 5b (R ¼ CH3).
used. The steric repulsion of the imidazolyl groups in 5b may
hinder coordination to DNA bases. Additionally, the NH groups in
5a0 point away from a possible binding site due to the ligand
geometry. In complexes [(arene)Ru(en0)X] with ethylenediamine
derivatives (en0) it has been shown that NH functionalities in the
en0 ligand are essential for efficient DNA binding [20,54,55].
The compounds [(p-cymene)Ru(1a)Cl2] (4a) shows cytotoxicity
H4IIE cells. The cytotoxicity is decreased upon introduction of
a methyl group as [(p-cymene)Ru(1b)Cl2] (4b) shows only modest
toxicities in the cell lines investigated. In general 4a is more cyto-
toxic then 4b. As mentioned above, free NH functions as in [(arene)
Ru(en0)X] should favour DNA binding. This is not a valid explana-
tion here, as the melting temperatures for 4a and 4b are essentially
the same.
towards A2780sens and Hct116 cells in the mM range but not in
Although 5a and 4a both have the kP binding mode and there-
fore resemble RAPTA-type [(arene)RuCl2(pta)] complexes, 5a is
almost non-toxic in the cell lines investigated within 24 of incu-
bation. After 72 and 96 h of incubation a selective toxicity of 5a
Table 3
Experimental n-octanol/water (PBS buffer pH 7.4) distribution coefficients (log D7.4
)
and DNA melting temperatures (DTm) at a ratio of metal complex/DNA (in base pairs)
of 1/5. The
D
Tm values are estimated to be accurate within ꢂ1 ꢁC.
Compound
log D7.4
Tm/ꢁC
D
Tm/ꢁC
4a
4b
5a
5a0
5b
1.25 ꢂ 0.03
1.14 ꢂ 0.05
0.14 ꢂ 0.03
ꢀ0.33 ꢂ 0.02
ꢀ0.68 ꢂ 0.01
66
66
45, 65
66
69
7
7
28, 8
7
4
Table 4
Cell viability tests after 24 and 72 h of incubation, IC50 values given in
determined).
mM (n.d. not
Cell line
Incubation time 4a
4b
83 >100 >100 >100 >100 n.d.
38 95 >100 >100 >100 n.d.
5a
5a0
5b
6a
6b
A2780sens 24 h
72 h
n.d.
n.d.
Fig. 3. Molecular structure of 5a0$CH2Cl2. Uncoordinated counter ion, non-acidic
Hct116
24 h
72 h
24 h
72 h
66 >100 >100 >100 >100 >100 >100
58 >100 n.d. n.d. n.d. >100 >100
>100 >100 >100 >100 >100 >100 >100
>100 >100 54 >100 >100 >100 >100
hydrogen atoms and solvent molecule are omitted for clarity. The displacement
ꢁ
ꢀ
ellipsoids are shown on a 50% level. Selected bond lengths [A] and angles [ ]: Ru1-N1
2.096(4), Ru1-N3 2.087(3), Ru1-Cl1 2.4006(15), N1-Ru1-N3 84.28(14), N1-Ru1-Cl1
85.21(12), N3-Ru1-Cl1 85.84(11).
H4IIE

W. Huber et al. / Journal of Organometallic Chemistry 717 (2012) 187e194
193
in which ligand 2a is bound in the
k
P mode, shows a selective
cytotoxicity towards H4IIE cells after 72 h. In contrast complex 5a0,
100
75
50
25
0
in which ligand 2a is bound in the
effects in the cell lines investigated.
k
2N,N mode, shows no cytotoxic
H4IIE cells
A2780 cells
Appendix A. Supplementary material
CCDC 867104, 867105 and 861637 contain the supplementary
crystallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Appendix B. Supplementary information
control
5a
5a´
5b
7
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jorganchem.2012.
06.027.
Fig. 5. Cytotoxic effects 5a, 5a0, 5b and 7 in H4IIE and A2780 cells. H4IIE rat hepatoma
and A2780 human ovarian carcinoma cells were incubated with 5a, 5a0, 5b or 7
(100
mM) for 72 h, then MTT reduction as a marker of cell viability was measured
(absorbance at 560 nm). Results are expressed as viable cells in percent of control
value ꢂ SD (n ¼ 3, *p < 0,05 vs. corresponding DMSO control).
References
[1] O. Rixe, W. Ortuzar, M. Alvarez, R. Parker, Biochem. Pharmcol. 52 (1996)
1855e1865.
125
100
75
[2] B. Spingler, D. Whittington, S.J. Lippard, Inorg. Chem. 40 (2001) 5596e5602.
[3] A.V. Klein, T.W. Hambley, Chem. Rev. 109 (2009) 4911e4920.
[4] R. Ingo Ott, Pharm. Unserer Zeit 35 (2006) 124e133.
[5] P. Abada, S.B. Howell, Met. Based Drugs (2010). Article ID 317581.
[6] E. Alessio (Ed.), Bioinorganic Medicinal Chemistry, Wiley-VCH, Weinheim,
2011.
[7] M. Gielen, E.R.T. Tiekink (Eds.), Metallotherapeutic Drugs and Metal-Based
Diagnostic Agents: The Use of Metals in Medicine, John Wiley & Sons, Chi-
chester, 2005.
[8] P.C. Bruijnincx, P.J. Sadler, Curr. Opin. Chem. Biol. 12 (2008) 197e206.
[9] A. Bergamo, G. Sava, Dalton Trans. 40 (2011) 7817e7823.
[10] E.S. Antonarakis, A. Emadi, Cancer Chemother. Pharmacol. 66 (2010) 1e9.
[11] S.H. van Rijt, P.J. Sadler, Drug Discov. Today 14 (2009) 1089e1097.
[12] G. Sava, Met. Based Drugs (2007). Article ID 16260.
[13] G. Jaouen (Ed.), Bioorganometallics. Biomolecules, Labeling, Medicine, Wiley
VCH, 2006.
incubation: 48 h
incubation: 72 h
incubation: 96 h
50
[14] C. Vock, C. Scolaro, A. Phillips, R. Scopelliti, G. Sava, P.J. Dyson, J. Med. Chem. 49
(2006) 5552e5561.
25
[15] W. Han Ang, P.J. Dyson, Eur. J. Inorg. Chem. 2006 (2006) 4003e4018.
[16] R. Aird, J. Cummings, A. Ritchie, M. Muir, R. Morris, H. Chen, P.J. Sadler,
D. Jodrell, Br. J. Cancer 86 (2002) 1652e1657.
[17] C.A. Smith, A.J. Sutherland-Smith, F. Kratz, E. Baker, B.K. Keppler, J. Biol. Inorg.
Chem. 1 (1996) 424e431.
0
0
25
50
75
100
[18] J.M. Rademaker-Lakhai, D. van den Bongard, D. Pluim, J.H. Beijnen,
J.H.M. Schellens, Clin. Cancer Res. 10 (2004) 3717e3727.
[19] S. Zorzet, A. Bergamo, M. Cocchietto, A. Sorc, B. Gava, E. Alessio, E. Iengo,
G. Sava, J. Pharmacol. Exp. Ther. 295 (2000) 927e933.
[20] H. Chen, J.A. Parkinson, S. Parsons, R.A. Coxall, R.O. Gould, P.J. Sadler, J. Am.
Chem. Soc. 124 (2002) 3064e3082.
[21] H. Chen, J.A. Parkinson, O. Nováková, J. Bella, F. Wang, A. Dawson, R. Gould,
S. Parsons, V. Brabec, P.J. Sadler, Proc. Natl. Acad. Sci. U.S.A. 100 (2003)
14623e14628.
concentration of 5a (µM)
Fig. 6. Cytotoxic effects of 5a in H4IIE. H4IIE rat hepatoma cells were incubated with
different concentrations of 5a for 48, 72 and 96 h, then MTT reduction as a marker of
cell viability was measured (absorbance at 560 nm). Results are expressed as viable
cells in percent of control value ꢂ SD (n ¼ 3, *p < 0,05 vs. corresponding DMSO
control).
[22] H. Chen, J.A. Parkinson, R.E. Morris, P.J. Sadler, J. Am. Chem. Soc. 125 (2003)
173e186.
[23] A. Bergamo, A. Masi, P.J. Dyson, G. Sava, Int. J. Oncol. 33 (2008) 1281e1289.
[24] S. Chatterjee, S. Kundu, A. Bhattacharyya, C.G. Hartinger, P.J. Dyson, J. Biol.
Inorg. Chem. 13 (2008) 1149e1155.
[25] F. Wang, H. Chen, S. Parsons, I.D.H. Oswald, J.E. Davidson, P.J. Sadler, Chem.
Eur. J. 9 (2003) 5810e5820.
[26] C. Scolaro, A. Bergamo, L. Brescacin, R. Delfino, J. Med. Chem. 48 (2005)
4161e4171.
[27] A. Casini, F. Edafe, M. Erlandsson, L. Gonsalvi, A. Ciancetta, N. Re, A. Ienco,
L. Messori, M. Peruzzini, P.J. Dyson, Dalton Trans. (2010) 1e16.
[28] X. Sun, C.-N. Tsang, H. Sun, Metallomics 1 (2009) 25e31.
[29] A. Casini, C. Gabbiani, F. Sorrentino, M.P. Rigobello, A. Bindoli, T.J. Geldbach,
A. Marrone, N. Re, C.G. Hartinger, P.J. Dyson, et al., J. Med. Chem. 51 (2008)
6773e6781.
[30] C. Wetzel, P.C. Kunz, M.U. Kassack, A. Hamacher, P. Böhler, W. Wätjen, I. Ott,
R. Rubbiani, B. Spingler, Dalton Trans. 40 (2011) 9212e9220.
[31] A.L. Noffke, undefined author, M. Bongartz, undefined author, W. Wätjen,
P. Böhler, undefined author, B. Spingler, P.C. Kunz, J. Organomet. Chem. 696
(2011) 1096e1101.
towards H4IIE cells is found (Table 4 and Fig. 6). Reason for the
decreased toxicity of compound 5a to 4a might be the hydropho-
bicity. The selective toxicity of 5a in H4IIE cells might be an effect of
5a in the cell cycle leading to delayed toxic effects. In addition it is
well known, that RAPT-type complexes not only act as DNA binders.
The poor correlation found here could also point towards
a different mechanism, where proteins are favoured targets for the
compounds. For corresponding complexes this has been shown by
the groups of Messori and Davey [29,56].
4. Conclusion
We prepared a series of ruthenium(II) arene complexes with
ambidentate PN ligands. All compounds displaying a chelating
k
2N,N or 3N,N,N-coordination mode show no cytotoxicity towards
k
[32] P.C. Kunz, M.U. Kassack, A. Hamacher, B. Spingler, Dalton Trans. (2009)
7741e7747.
[33] P.C. Kunz, I. Thiel, A.L. Noffke, G.J. Reiß, F. Mohr, B. Spingler, J. Organomet.
Chem. 697 (2012) 33e40.
the used Hct116, H4IIE and A2780 cell lines. The most lipophilic
complex 4a show the highest cytotoxicity within the series of
compounds, with IC50 values in the micromolar range. Complex 5a,
[34] C. Wetzel, P.C. Kunz, I. Thiel, B. Spingler, Inorg. Chem. 50 (2011) 7863e7870.

194
W. Huber et al. / Journal of Organometallic Chemistry 717 (2012) 187e194
[35] A. Caballero, F. Jalon, B. Manzano, G. Espino, Organometallics 23 (2004)
5694e5706.
[48] C. Allardyce, P.J. Dyson, D. Ellis, S. Heath, Chem. Commun. (2001) 1396e1397.
[49] I. Moldes, E. de la Encarnación, J. Ros,
Á Alvarez-Larena, J.F. Piniella,
[36] S. Moore, G.M. Whitesides, J. Org. Chem. 47 (1982) 1489e1493.
[37] N. Curtis, R. Brown, J. Org. Chem. 45 (1980) 4038e4040.
[38] H.-Q. Liu, T.-C. Cheung, S.-M. Peng, C.-M. Che, J. Chem. Soc. Chem. Commun.
(1995) 1787e1788.
J. Organomet. Chem. 566 (1998) 165e174.
[50] P. Kumar, A.K. Singh, R. Pandey, P.-Z. Li, S.K. Singh, Q. Xu, D.S. Pandey,
J. Organomet. Chem. 695 (2010) 2205e2212.
[51] F. Marchetti, C. Pettinari, R. Pettinari, A. Cerquetella, C. Di Nicola,
A. Macchioni, D. Zuccaccia, M. Monari, F. Piccinelli, Inorg. Chem. 47 (2008)
11593e11603.
[39] T. Mosmann, J. Immunol. Methods 65 (1983) 55e63.
[40] G.M. Sheldrick, Acta Crystallogr. Sect.
E Struct. Rep. Online A64 (2008)
112e122.
[52] D.J.M. Snelders, A. Casini, F. Edafe, G. van Koten, R.J.M.K. Gebbink, P.J. Dyson,
J. Organomet. Chem. 696 (2011) 1108e1116.
[53] A.K. Renfrew, A.D. Phillips, A.E. Egger, C.G. Hartinger, S.S. Bosquain,
A.A. Nazarov, B.K. Keppler, L. Gonsalvi, M. Peruzzini, P.J. Dyson, Organome-
tallics 28 (2009) 1165e1172.
[41] Oxford Diffraction Ltd, CrysAlisPro Software System (2007).
[42] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo,
A. Guagliardi, A.G.G. Moliterni, G. Polidori, Riccardo Spagna, J. Appl. Cryst. 32
(1999) 115e119.
[43] A.L.L. Spek, J. Appl. Cryst. 36 (2003) 7e13.
[54] S. Das, S. Sinha, R. Britto, K. Somasundaram, A.G. Samuelson, J. Inorg. Biochem.
104 (2010) 93e104.
[55] A. Habtemariam, M. Melchart, R. Fernández, S. Parsons, I.D.H. Oswald,
A. Parkin, F.P.A. Fabbiani, J.E. Davidson, A. Dawson, R.E. Aird, et al., J. Med.
Chem. 49 (2006) 6858e6868.
[44] P.C. Kunz, M. Börgardts, F. Mohr, Inorg. Chim. Acta 380 (2012) 392e398.
[45] P.C. Kunz, W. Huber, A. Rojas, U. Schatzschneider, B. Spingler, Eur. J. Inorg.
Chem. (2009) 5358e5366.
[46] P.C. Kunz, G.J. Reiß, W. Frank, W. Kläui, Eur. J. Inorg. Chem. (2003) 3945e3951.
[47] M.R.J. Elsegood, M.B. Smith, N.M. Sanchez-Ballester, Acta Crystallogr. Sect. E
Struct. Rep. Online E62 (2006) 2838e2840.
[56] B. Wu, M.S. Ong, M. Groessl, Z. Adhireksan, C.G. Hartinger, P.J. Dyson,
C.A. Davey, Chem. Eur. J. 17 (2011) 3562e3566.