Cytotoxic gold compounds: Synthesis, biological characterization and investigation of their inhibition properties of the zinc finger protein PARP-1

Article abstract of DOI:10.1039/c2dt11913g

The new gold(iii) complexes: [Au{2-(2′-pyridyl)imidazolate}Cl ₂] and [Au{2,6-bis(2′-benzimidazolate)pyridine}(OCOCH ₃)] and the mono- and binuclear gold(i) complexes: [Au{2-(2′-pyridyl)imidazole}(PPh₃)](PF₆), [Au(2-phenylimidazolate)(DAPTA)] (DAPTA = 3,7-diacetyl-1,3,7-triaza-5- phosphabicyclo[3.3.1]nonane), [(PPh₃Au)₂(2-R-imidazolate)] (PF₆) (R = 2-C₅H₄N, Ph) have been synthesized and characterized. The structure of the [(PPh₃Au)₂{2- (2′-pyridyl)imidazolate)](PF₆) complex was also characterized by X-ray crystallography. The antiproliferative properties of the complexes were assayed against human ovarian carcinoma cell lines, either sensitive (A2780) or resistant to cisplatin (A2780cisR), human mammary carcinoma cells (MCF7) and non-tumorigenic human kidney (HEK293) cells. Most of the studied compounds showed important cytotoxic effects. Interestingly, the compounds containing the 2-(2′-pyridyl)imidazolate ligand showed selectivity towards cancer cells with respect to the non-tumorigenic ones, with the dinuclear compound [(PPh ₃Au)₂{2-(2′-pyridyl)imidazolate)](PF₆) being the most active. Some compounds were also screened for their inhibitory effect of the zinc-finger protein PARP-1, essential for DNA repair and relevant to the mechanisms of cancer cell resistance to cisplatin. Interaction studies of the compounds with the model protein ubiquitin were undertaken by electrospray ionization mass spectrometry (ESI MS). The results are discussed in relation to the putative mechanisms of action of the cytotoxic gold compounds. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2dt11913g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 3287
www.rsc.org/dalton
PAPER
Cytotoxic gold compounds: synthesis, biological characterization and
investigation of their inhibition properties of the zinc finger protein PARP-1†
Maria Serratrice,a Fabio Edafe,b Filipa Mendes,c Rosario Scopelliti,b Shaik Mohammed Zakeeruddin,b
Michael Grätzel,b Isabel Santos,c Maria Agostina Cinellua and Angela Casini*b,d
Received 12th October 2011, Accepted 8th December 2011
DOI: 10.1039/c2dt11913g
The new gold(III) complexes: [Au{2-(2′-pyridyl)imidazolate}Cl2] and [Au{2,6-bis(2′-benzimidazolate)
pyridine}(OCOCH3)] and the mono- and binuclear gold(I) complexes: [Au{2-(2′-pyridyl)imidazole}
(PPh3)](PF6), [Au(2-phenylimidazolate)(DAPTA)] (DAPTA = 3,7-diacetyl-1,3,7-triaza-5-phosphabicyclo
[3.3.1]nonane), [(PPh3Au)2(2-R-imidazolate)](PF6) (R = 2-C5H4N, Ph) have been synthesized and
characterized. The structure of the [(PPh3Au)2{2-(2′-pyridyl)imidazolate)](PF6) complex was also
characterized by X-ray crystallography. The antiproliferative properties of the complexes were assayed
against human ovarian carcinoma cell lines, either sensitive (A2780) or resistant to cisplatin (A2780cisR),
human mammary carcinoma cells (MCF7) and non-tumorigenic human kidney (HEK293) cells. Most of
the studied compounds showed important cytotoxic effects. Interestingly, the compounds containing the
2-(2′-pyridyl)imidazolate ligand showed selectivity towards cancer cells with respect to the non-
tumorigenic ones, with the dinuclear compound [(PPh3Au)2{2-(2′-pyridyl)imidazolate)](PF6) being the
most active. Some compounds were also screened for their inhibitory effect of the zinc-finger protein
PARP-1, essential for DNA repair and relevant to the mechanisms of cancer cell resistance to cisplatin.
Interaction studies of the compounds with the model protein ubiquitin were undertaken by electrospray
ionization mass spectrometry (ESI MS). The results are discussed in relation to the putative mechanisms
of action of the cytotoxic gold compounds.
gold compounds were widely used for the treatment of several
Introduction
diseases, particularly as anti-infective and anti-tubercular
agents.11 Despite extensive clinical studies carried out during
Transition metal complexes have proved to possess outstanding
anticancer activities and several reviews summarized recent pro-
gress in the field.1–3 In this research area, gold compounds
occupy a relevant position constituting a promising class of
experimental anticancer metallodrugs.4–10 It is worth mentioning
that gold compounds have a long and important tradition in
medicine, the so-called Chrysotherapy that derives its name
from Chryses, a golden-haired heroine in Greek mythology.
Such a tradition dates back to ancient China (2500 B.C.) and
flourished especially in Europe during the late Middle Ages and
the Renaissance. In the early years of modern pharmacology,
that pioneering time, gold compounds have found rather limited
medical applications and are presently used only for the treat-
ment of severe rheumatoid arthritis.12 This is probably the result
of their relevant systemic toxicity (e.g. nephrotoxicity) and the
poor chemical stability of some of the tested compounds. For
example, auranofin, a linear bidentate gold(I) complex with
triethylphosphine and a thiosugar as ligands, which is in clinical
use as an antiarthritic drug, is also a potent cytotoxic agent in
vitro,13 but proved to be active in vivo only in mice inoculated
with lymphocytic P388 leukaemia.14
To date, several Au(I) and Au(III) compounds with profoundly
different molecular structures have been developed and tested as
anticancer agents, and initial structure-function relationships
have been outlined. Au(I) complexes with phosphine and
carbene ligands,15,16 Au(III) dithiocarbamate compounds,17
Au(III) porphyrinates18 and organogold(III) compounds19 have
been among the most widely investigated in vitro and in vivo.
Within this frame, in the last few years, we have particularly
focused on Au(III) complexes in which the metal centre is stabil-
ised by the presence of at least two nitrogen ligands, as in the
case of the substituted 2,2′-bipyridine derivatives showing prom-
ising antiproliferative effects in vitro.20–22
aDepartment of Chemistry, University of Sassari, Via Vienna 2, 07100
Sassari, Italy
bInstitut des Sciences et Ingénierie Chimiques, Ecole Polytechnique
Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland.
Fax: +41 21 6939865; Tel: +41 21 6939860
cUnidade de Ciências Químicas e Radiofarmacêuticas, Instituto
Tecnológico e Nuclear, Estrada Nacional 10, 2686-953 Sacavém,
Portugal
dResearch Institute of Pharmacy, University of Groningen, Antonius
Deusinglaan 1, 9713 AV Groningen, The Netherlands.
E-mail: a.casini@rug.nl; Fax: +31 5036 3247; Tel: +31 5036 3008
†CCDC reference number 848506. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/c2dt11913g
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3287–3293 | 3287

View Article Online
For the various families of gold compounds markedly distinct
molecular mechanisms were recently invoked, such as a direct
mitochondrial mechanism involving thioredoxin reductase inhi-
bition in the case of the gold(I) complexes,23–25 as well as inter-
actions with various thiolate/selenolate containing proteins/
enzymes.26 The influence on some apoptotic proteins—i.e.
MAPKs and Bcl-2—was also reported for gold(III) porphyrins,27
as well as proteasome inhibition for gold(III) dithiocarbamates.28
Analysis of cancer cell growth inhibition data of a series of
gold(III) compounds in comparison to 110 standard agents with
known mechanisms of action suggest that numerous molecular
mechanisms are involved, including the inhibition of cyclin-
dependent kinases or histone deacetylases.29 Recently, we also
reported on the potent inhibition of the zinc finger protein poly-
(adenosine diphosphate (ADP)-ribose) polymerase 1 (PARP-1)
by Au(I) and Au(III) complexes.30 PARPs are essential proteins
involved in cancer resistance to chemotherapies, and play a key
role in DNA repair by detecting DNA strand breaks and catalys-
ing poly(ADP-ribosylation).31,32 Information on the reactivity of
the gold complexes with the PARP-1 zinc-finger domain was
also obtained by high-resolution mass spectrometry, and an
excellent correlation between PARP-1 inhibition in protein
extracts and the ability of the complexes to bind to the zinc-
finger motif (in competition with zinc) was established.
Therefore, relying on the promising results obtained for
both Au(III) complexes and Au(I) phosphine compounds, we
have synthesized new Au(III) and Au(I) derivatives bearing
monodentate, bidentate or tridentate nitrogen donor ligands
such as 2-phenylimidazole, 2-(2′-pyridyl)imidazole and, 2,6-
bis(benzimidazol-2-yl)pyridine, as well as phosphane groups,
namely triphenylphosphine and 3,7-diacetyl-1,3,7-triaza-5-phos-
phabicyclo[3.3.1]nonane (DAPTA) (Chart 1). Interestingly, the
DAPTA ligand is known to increase the water solubility of the
resulting metal compounds, and recently Au(I)-DAPTA com-
plexes with thionate or alkynyl ligands showed promising results
as anticancer agents.33,34 The new gold complexes were screened
in vitro for their antiproliferative effects on human ovarian and
breast cancer cell lines and on non-tumorigenic cells in order to
evaluate their selectivity profiles. The most promising compound
out of the cytotoxicity screening was further tested for PARP-1
inhibition on protein extracts from cancer cells. Information on
the possible protein-bound metallo-fragments was obtained by
electrospray ionization mass spectrometry (ESI MS).
Results and discussion
The Au(III) complexes 1aAu(III) and cAu(III) with bidentate 2-
(2′-pyridyl)imidazole (a) and tridentate 2,6-bis(benzimidazol-2-
yl)pyridine (c) nitrogen donor ligands, respectively, which are
able to stabilize the Au(III) centre in a typical square-planar geo-
metry, were synthesized according to previously reported pro-
cedures for Au(III) bipyridine derivatives.21 Mono- and dinuclear
Au(I) complexes 2aAu(I), 1bAu(I), 3aAu(I)2 and 2bAu(I)2 were
obtained by reaction of the ligands 2-(2′-pyridyl)imidazole (a)
and 2-phenylimidazole (b) with one or two equivalents of the
appropriate Au(I)–phosphine precursor, as reported in the exper-
imental section. The obtained compounds are shown in Chart 1.
Interestingly, the Au(I) derivatives, with the exception of
2aAu(I), were soluble in water, while the Au(III) complexes were
poorly soluble. Single crystals of the dinuclear complex
3aAu(I)2 were obtained by slow diffusion of diethyl ether into a
dichloromethane solution and were analysed by X-ray diffrac-
tion. An ORTEP diagram of compound 3aAu(I)2 is shown in
Fig. 1, with the most relevant bond distances and angles. Crystal
data and other crystallographic parameters are reported in the
experimental section (Table 2) and in the available supplemen-
tary material†. In 3aAu(I)2 the cation [(PPh3Au)2{2-(2′-pyridyl)
imidazolate)]+ lies on a crystallographic 2-fold axis (.2.) passing
through the middle of the 2-(2′-pyridyl)imidazolate which
bridges two Au(I) metal centres. The geometry around the metal
is mostly linear, as shown by the P–Au–N angle 177.7(5)°. The
Au–P and Au–N bond lengths, 2.242(5) and 2.05(2) Å respect-
ively, can be easily compared with values found for similar
compounds.35,36
The antiproliferative properties of the new gold complexes
were assayed by monitoring their ability to inhibit cell growth
using the MTT assay (see Experimental section). Cytotoxic
activity of the compounds was determined after exposing for
72 h the human ovarian cancer A2780 cell line, and its cisplatin-
resistant variant (A2780cisR), as well as the human breast
cancer cell line MCF7, in comparison to cisplatin and auranofin.
The results are summarized in Table 1. Most of the compounds
turned out to be good cytotoxic agents with the only exceptions
of the mononuclear Au(III) and Au(I) complexes 1aAu(III) and
1bAu(I), with IC50 values > 50 μM. In general the cationic
derivatives are the most effective among the tested compounds.
In particular, the dinuclear Au(I) compounds 3aAu(I)2 and
Fig. 1 ORTEP diagram of the cation of 3aAu(I)2. Ellipsoids are drawn
at 50% probability level. Selected bond lengths (Å) and angles (°):
Au1–P1 2.242(5), Au1–N1 2.05(2), N1–Au1–P1 177.7(5).
Chart 1 Ligands and gold complexes presented in this study.
3288 | Dalton Trans., 2012, 41, 3287–3293
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 IC50 values of the Au complexes described in this study
against human ovarian carcinoma cell lines cisplatin sensitive (A2780)
and resistant (A2780cisR), human breast cancer cells (MCF7), and the
non-tumorigenic cell line HEK293, compared to cisplatin and auranofin
IC50 (μM)a
Compound A2780
A2780cisR
MCF7
HEK293
1aAu(III)
2aAu(I)
3aAu(I)2
1bAu(I)
2bAu(I)2
cAu(III)
61.9 17.3 52.9 16.2 80.3 7.3
217.9 67.8
9.6 0.9
6.4 2.4
7.2 0.6
1.1 0.01
93.7 3.7
2.1 0.7
9.7 1.3
1.25 0.5
1.9 0.6
4.1 0.2
0.5 0.2
12 4.2
5.13 0.5
69.4 12.3 149.8 0.4 53.7 13.2
1.9 0.7
6.8 1.0
1.5 0.3
35 7.0
3.3 0.9
15 3.1
1.7 0.4
20 3.0
2.6 1.0
5.8 1.5
4.62 1.1
—
Auranofin
Cisplatin
Fig. 2 PARP-1 activity levels in A2780 and A2780cisR cellular
extracts. PARP-1 activity was measured in homogenates (50 μg of
protein) treated with the compounds (50 μM) over 1 h at room tempera-
ture. Data are the mean SD of at least three experiments each per-
formed in triplicate.
a Mean SE of at least three determinations.
2bAu(I)2, bearing two AuPPh3 units, are even more effective
antiproliferative agents than cisplatin. These two compounds are
comparable to auranofin, being active in the low micromolar
range on both A2780 and A2780cisR cell lines, as well as on the
MCF7 cells. Among the derivatives with the 2-phenylimidazole
ligand, the mononuclear one (1bAu(I)), in which Au(I) is bound
to DAPTA, is ca. 40-fold less effective than the dinuclear ana-
logue with triphenylphosphines (2bAu(I)2). Finally, 2aAu(I),
and cAu(III) have IC50 values comparable to cisplatin for the
A2780 cell line, but are at least 5–7-fold more effective towards
the A2780cisR variant. These results, attesting the compounds
ability to overcome cisplatin resistance, strongly support the
hypothesis of their different mechanism of action with respect to
cisplatin.
In order to assess the compounds’ selectivity for cancerous
cells with respect to normal cell lines, the gold complexes were
also screened for their antiproliferative effects on the non-tumori-
genic human embryonic kidney cells HEK293. Interestingly,
while auranofin is still highly cytotoxic, the new gold complexes
show different activity profiles. In most cases the cytotoxicity is
comparable for the cancerous and normal cell lines. However, a
general trend was found for the complexes containing the 2-(2′-
pyridyl)imidazole/imidazolato ligand (a): the latter are less cyto-
toxic towards the HEK293 (and MCF7), and more active on the
ovarian cancer cell lines. In particular 3aAu(I)2 is ca. 12-times
more toxic on the A2780cisR (IC50 ca. 0.5 μM) than on the
HEK293 cells (IC50 ca. 6.4 μM).
with PARP-1 for gold binding. In other words, the specific
protein content/composition of each cancer cell line orients the
reactivity of the compound towards different targets. In the
specific case of the A2780 cell lines, previously reported proteo-
mic studies evidenced a few proteins to be differentially
expressed in the cisplatin resistant variant compared to the sensi-
tive one, and five of them were considered responsible for plati-
num resistance.37 Interestingly, glutathione-S-transferase (GST),
a known target for metallodrugs, was one of the five protein
candidates.
In order to shed light on the reactivity of the Au complexes
we investigated their interactions with ubiquitin (Ub), used as a
model protein, by ESI MS following established proto-
cols.38,39,40 Thus, three molar equivalents of 3aAu(I)2 were
added to an aqueous solution of Ub buffered at pH 7.4 (see
Experimental for details). Fig. 3 shows the mass spectrum of the
Ub–3aAu(I)2 sample recorded after 1 h incubation. In the spec-
trum, Ub was identified as one of the main peaks at 952.49 m/z
(ca. 8565 Da), and gold-containing species were observed at
1003.54 m/z (ca. 9023 Da) corresponding to an Ub–AuPPh3
adduct in which the original 2-(2′-pyridyl)imidazolato ligand is
absent. The observed reactivity is similar to the one described
Following our recent results that indicate some cytotoxic gold
compounds are efficient inhibitors of the zinc-finger protein
PARP-1, we tested two compounds, namely the most active and
selective on cancer cells 3aAu(I)2, and the least active 1bAu(I),
for PARP-1 inhibition on A2780 and A2780cisR cell extracts.
The PARP-1 activity was evaluated after incubation of protein
cell extracts with the compounds for 1 h at room temperature.
Fig. 2 shows the residual PARP-1 activity in protein extracts
treated with the complexes at a fixed concentration. Interestingly,
while the poorly cytotoxic compound 1bAu(I) is practically inef-
fective as a PARP-1 inhibitor in all cases, the active dinuclear
complex 3aAu(I)2 is able to efficiently inhibit PARP-1 in A2780
cell extracts reducing its activity to ca. 24%. However, 3aAu(I)2
has no effect on PARP-1 activity in cell extracts from the
A2780cisR variant, indicating that other proteins are competing
Fig. 3 ESI mass spectrum (+10, +9 and +8 ions) of Ub incubated with
3aAu(I)2 (metal complex : Ub ratio = 3 : 1) in (NH4)2CO3 buffer (pH
7.4) for 1 h at 37 °C.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3287–3293 | 3289

View Article Online
for the reference compound auranofin, which has also been
reported to release the thioglucose ligand upon protein binding,
while maintaining the phosphine moiety.41,42
31P NMR (CDCl3) δ 30.6 (PPh3), −143.6 (PF6). 1H NMR
(CDCl3); δ 8.30 (d, 1H, J = 7.8 Hz; H3′), 8.19 (d, 1H, J = 5.9
Hz; H6′), 7.96 (t, 1H, J = 7.5 Hz; H4′), 7.55–7.68 (m, 15H,
PPh3), 7.50 (s, 1H; H5), 7.38 (dd, 1H, J = 7.8, 5.9 Hz; H5′), 7.27
(s, 1H; H4). Anal. Calcd. for C26H22AuF6N3P2 (749.38 g
mol−1): C 41.67; H 2.96; N 5.61. Found: C 41.70; H 3.11; N
5.70. ESI-MS (H2O/acetone/CH3CN, positive mode) exact mass
for C26H22AuF6N3P2 (749.09): found m/z 604.12 [AuLPPh3]+
(calc. m/z 604.12).
Experimental section
Materials and methods
Chemicals were obtained from commercial suppliers and used as
received. The ligand 3,7-diacetyl-1,3,7-triaza-5-phosphabicyclo
[3.3.1]nonane (DAPTA) was prepared as previously reported.43
Cisplatin and NaAuCl4 were purchased from Sigma-Aldrich,
Auranofin, Au(III)acetate and 2-phenylimidazole (2) from
AlfaAesar. Solvents used in synthesis were dried using standard
procedures. The identity and purity (≥95%) of the gold com-
plexes were unambiguously established using elemental analysis,
mass spectrometry and NMR. Elemental analyses were per-
formed on a Carlo Erba EA 1110 CHN instrument. ESI-MS
spectra were obtained in acetonitrile on a Thermo Finnigan LCQ
Deca XP Plus quadrupole ion-trap instrument operated in posi-
tive ion mode over a mass range of m/z 150–2000. The ioniz-
ation energy was set at 30 V and the capillary temperature at
200 °C. 1H, and 31P NMR spectra were recorded at 25 °C with a
Bruker Avance 400 spectrometer operating at 400.13 and
161.98 MHz, respectively. Chemical shifts are given in ppm rela-
tively to internal TMS (1H), and external H3PO4 (31P). Number-
ing schemes of the protons are as in the drawings below.
[Au2{2-(2-pyridyl)imidazolate}(PPh3)2]PF6
(3aAu(I)2):
AgPF6 (252 mg, 1 mmol) was added to an acetone solution of
(PPh3)AuCl (604 mg, 1 mmol, 20 mL) and the resulting mixture
stirred for 10 min until the precipitation of AgCl was completed.
Then, the filtered solution was added dropwise to a solution of a
(72.8 mg, 0.5 mmol) and KOH (28.1 mg 0.5 mmol) in aceto-
nitrile (5 mL)/water (20 mL). The resulting mixture was stirred
for 24 h at room temperature in the darkness. Then, the solvent
was removed under vacuum and the white product crystallized
from dichloromethane/diethyl ether and dried under vacuum
(332 mg, 55%). Solubility: chloroform, dichloromethane,
acetone, poorly soluble in water and insoluble in ether and
hexane.
31P NMR (CDCl3) δ 31.7 (PPh3), −144.1 (PF6). 1H NMR
(CDCl3) δ: 8.53 (d, 1H, J = 8.0 Hz; H3′), 8.21 (d, 1H, J = 4.8
Hz, H6′), 7.70 (t, 1H, J = 7.8 Hz; H4′), 7.54–7,62 (m, 30 H,
2PPh3), 7.40 (br m, 1H; H5′), 7.36 (s, 2H, H4,H5). Anal. Calcd.
for C44H36Au2F6N3P3 (1207.6 g mol−1): C 43.76; H 3.00; N
3.48. Found: C 43.69; H 3.00; N 3.59. ESI-MS (H2O/acetone/
CH3CN, positive mode) exact mass for C44H36Au2F6N3P3
(1207.14): found m/z 1602.17 [AuL(PPh3)2]+ (calc. m/z
1602.17).
Synthesis
[Au{2-(2-pyridyl)imidazolate}Cl2] (1aAu(III)): an aqueous sol-
ution of NaAuCl4 (199 mg, 0.5 mmol, 20 mL) was added to a
solution of a (72.6 mg, 0.5 mmol, 20 mL) and KOH (28 mg,
0.5 mmol) in acetonitrile (5 mL)/water (10 mL). The resulting
suspension was stirred for 24 h at room temperature in the dark-
ness. Afterwards, the precipitate was filtered under vacuum and
the crude product recrystallized from acetone/diethyl ether to
give the analytical sample as a brown solid (180 mg, 87%).
Solubility: acetone, DMSO and insoluble in chloroform,
dichloromethane, water, ether and hexane.
[Au(2-phenylimidazole)(DAPTA)] (1bAu(I)): an aqueous sol-
ution of (DAPTA)AuCl (83.2 mg, 0.17 mmol, 20 mL), obtained
from [(THT)AuCl]44 and DAPTA, was added to a solution of b
(24.7 mg, 0.17 mmol) and KOH (9.54 mg, 0.17 mmol) in aceto-
nitrile (5 mL)/water (20 mL). The resulting mixture was stirred
for 24 h at room temperature in the darkness. Afterwards, the
solvent was removed under vacuum and the beige product crys-
tallized from chloroform/diethyl ether and dried under vacuum
(40.2 mg, 40%). Solubility: water, chloroform, dichloromethane,
acetone and insoluble in ether and hexane.
1H NMR (DMSO) δ: 9.04 (d, J = 6.0 Hz, 1H, H6′), 8.33 (t,
1H, J = 7.8 Hz; H4′), 8.03 (d, 1H, J = 8.0 Hz; H3′), 7.66 (t, 1H, J
= 7.0 Hz (av); H5′), 7.41 (s, 1H; H5), 7.33 (s, 1H; H4). Anal.
Calcd. for C8H6AuCl2N3 (412.03 g mol−1): C 23.32; H 1.47; N
10.20. Found: C 23.40; H 1.36; N 10.13. ESI-MS (H2O/acetone/
CH3CN, positive mode) exact mass for C8H6N3AuCl2 (410.96):
found m/z 411.97 [AuLH]+ (calc. m/z 411.97).
[Au{2-(2-pyridyl)imidazole}(PPh3)]PF6 (2aAu(I)): AgPF6
(126 mg, 0.5 mmol) was added to a dichloromethane solution of
(PPh3)AuCl (373 mg, 0.5 mmol) and the resulting mixture
stirred for 10 min until the precipitation of AgCl was completed.
Then the filtered solution was added dropwise to a dichloro-
methane solution of a (78 mg, 0.5 mmol, 20 mL), and stirred for
2 h at room temperature, in the darkness. The solution was
filtered through Celite and concentrated to a small volume;
addition of diethyl ether afforded a white solid. This was filtered
off, washed with diethyl ether (3 × 10 mL) and dried under
vacuum (253.1 mg, 59.3%). Solubility: chloroform, dichloro-
methane, acetone and insoluble in water, ether and hexane.
1
31P NMR (DMSO) δ: −28.6 (P DAPTA); H NMR (DMSO)
δ: 8.17 (d, 2H, J = 7.2 Hz; H°), 7.38 (t, 2H, J = 7.6 Hz (av);
Hm), 7.30 (t, 1H, J = 7.9 Hz; Hp), 6.99 (s, 2H, H4,H5), 5.54 (d,
1H, J = 14.9 Hz; NCH2N), 5.46 (dd, 1H, J = 14.9, 7.9 Hz,
NCH2P), 4.95 (d, 2H, J = 14 Hz, NCH2N), 4.66 (d, 1H, J = 15.0
Hz; NCH2P), 4.34 (d, 1H, J = 16.8 Hz; NCH2P), 4.13 (d, 2H, J
= 14.0 Hz; NCH2N), 4.07 (s, 2H; NCH2P), 3.81 (d, 1H, J = 15
Hz; NCH2P), 2.05 (s, 3H; CH3), 2.04 (s, 3H; Me). Anal. Calcd.
for C18H23AuN5O2P (569.13 g mol−1): C 37.97; H 4.07; N
12.30. Found: C 37.88; H 4.04; N 12.15. ESI-MS (H2O/acetone/
CH3CN, positive mode) exact mass for C18H23AuN5O2P
(569.13): found m/z 570.13 [AuLDAPTAH]+ (calc. m/z 570.13).
[Au2(2-phenylimidazolate)(PPh3)2]PF6 (2bAu(I)2): an acetone
solution of [(PPh3)Au]PF6 (604 mg, 1 mmol, 20 mL), prepared
as described above, was added dropwise to a solution of b
(72.0 mg, 0.5 mmol) and KOH (28.1 mg, 0.5 mmol) in aceto-
nitrile (5 mL)/water (20 mL). The resulting mixture was stirred
for 24 h at room temperature in the darkness. Afterwards, the
3290 | Dalton Trans., 2012, 41, 3287–3293
This journal is © The Royal Society of Chemistry 2012

View Article Online
solvent was removed under vacuum and the white product crys-
tallized from dichloromethane/diethyl ether and dried under
vacuum (216 mg, 40%). Solubility: water, chloroform, dichloro-
methane, acetone and insoluble in ether and hexane.
10% FBS and 1% penicillin/streptomycin (all from Invitrogen),
at 37 °C in a humidified atmosphere of 95% of air and 5% CO2
(Heraeus, Germany). Non-tumoral human embryonic kidney
cells HEK293 were kindly provided by Dr Maria Pia Rigobello
(CNRS, Padova, Italy), and were cultivated in DMEM medium,
added with GlutaMaxI (containing 10% FBS and 1% penicillin/
streptomycin (all from Invitrogen) and incubated at 37 °C and
5% CO2. For evaluation of growth inhibition, cells were seeded
in 96-well plates (Costar, Integra Biosciences, Cambridge, MA)
and grown for 24 h in complete medium. Solutions of the com-
pounds were prepared by diluting a freshly prepared stock sol-
ution (in DMSO) of the corresponding compound in aqueous
media (RPMI or DMEM for the A2780 and A2780cisR or
MCF7 and HEK293, respectively). Afterwards, the intermediate
dilutions of the compounds were added to the wells (100 μL) to
obtain a final concentration ranging from 0 to 150 μM, and the
cells were incubated for 72 h. DMSO at comparable concen-
trations did not show any effects on cell cytotoxicity. Following
72 h drug exposure, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenylte-
trazolium bromide (MTT) was added to the cells at a final con-
centration of 0.25 mg mL−1 incubated for 2 h, then the culture
medium was removed and the violet formazan (artificial chromo-
genic precipitate of the reduction of tetrazolium salts by dehy-
drogenases and reductases) was dissolved in DMSO. The optical
density of each well (96-well plates) was quantified three times
in tetraplicates at 540 nm using a multiwell plate reader (iEMS
Reader MF, Labsystems, US), and the percentage of surviving
cells was calculated from the ratio of absorbance of treated to
untreated cells. The IC50 value was calculated as the concen-
tration reducing the proliferation of the cells by 50% and is pre-
sented as a mean ( SE) of at least three independent
experiments.
31P NMR (CDCl3) δ 31.6 (PPh3), −143.9 (PF6). 1H NMR
(CDCl3) δ: 8.01 (d, 2H, J = 8.0 Hz; H°), 7.65–7.35 (m, 33H; Hm
+ Hp + H PPh3), 7.33 (s, 2 H; H4, H5). Anal. Calcd. for
C45H37Au2F6N2P3 (1206.6 g mol−1): C 44.79; H 3.09; N 2.32.
Found: C 44.65; H 3.18; N 2.19. ESI-MS (H2O/acetone/CH3CN,
positive mode) exact mass for C45H37Au2F6N2P3 (1206.14):
found m/z 1061.18 [AuL(PPh3)2]+ (calc. m/z 1061.17).
[Au{2,6-bis(2-benzimidazolate)pyridine}(O2CCH3)]
(cAu(III)): a solution of Au(O2CCH3)3, (187 mg, 0.5 mmol) in
acetic acid (20 mL) was added to a solution of c (155 mg,
0.5 mmol) in the same solvent (20 mL). The mixture was stirred
for 5 h at reflux (180 °C) in the darkness. Afterwards, the sol-
ution was filtered, dried under vacuum to afford and orange
solid. This was crystallized from dichloromethane/diethyl and
dried under vacuum (80.3 mg, 29%). Solubility: DMSO, chloro-
form, dichloromethane and insoluble in water, ether and hexane.
1H NMR (CDCl3) δ: 8.21 (t, 1H, J = 7.8 Hz; H4), 7.88 (d, 2H,
J = 7.8 Hz; H3,H5), 7.71 (d, 2H, J = 8.0 Hz; H7′,H7′′), 7.42 (d,
2H, J = 8.0 Hz; H4′,H4′), 7.23 (m, 4H, H5′,H6′,H5′′,H6′′), 2.48 (s,
3H, CH3). Anal. Calcd. for C21H14AuN5O2 (565.33 g mol−1): C
44.62; H 2.50; N 12.39. Found: C 44.57; H 2.52; N 12.16.
ESI-MS (acetone/CH3CN, positive mode) exact mass for
C21H14AuN5O2 (565.08): found m/z 566.08 [AuL(OOCCH3)H]+
(calc. m/z 566.08).
X-Ray crystallography
Crystals of 3aAu(I)2 were obtained by slow diffusion of diethyl
ether into a saturated solution of the complex in CH2Cl2 and
were resolved with X-ray diffraction. The data collection was
measured at low temperature [100(2) K] using Mo Kα radiation
on a Bruker APEX II CCD diffractometer having a kappa geo-
metry goniometer. Data reduction was carried out by EvalCCD45
and then corrected for absorption.46 The solution and refinement
was performed by SHELX.47 The structure was refined using
full-matrix least-squares based on F2 with all non hydrogen
atoms anisotropically defined. Hydrogen atoms were placed in
calculated positions by means of the “riding” model. Disorder
problems due to the imposed crystallographic symmetry and
dealing with the pyridine moiety were found during the last
stages of refinement. Some restraints were applied (SHELX
cards: DFIX and EADP) in order to get reasonable parameters.
An additional problem was discovered by TWINROTMAT/
PLATON,48 and was due to pseudo-merohedral twinning; a new
HKL was then generated and used for refinement with MERG 0,
HKLF 5 and two BASF parameters [0.294(6); 0.208(6)].
PARP-1 activity determinations
PARP-1 activity was determined using Trevigen’s HT Universal
Colorimetric PARP Assay. This assay measures the incorporation
of biotinylated poly(ADP-ribose) onto histone proteins in a
96 microtiter strip well format. An aliquot of protein cell extracts
(50 μg) from A2780 or A2780cisR cells was used as the enzyme
source and 3-aminobenzamide (3-AB), provided in the kit, was
used as a control inhibitor. The cells extracts were incubated
with the compounds (10 or 50 μM) for 1 h at room temperature
followed by determination of PARP-1 activity. Two cell extract
controls were always performed in parallel: a positive activity
control for PARP-1 without inhibitors, that provided the 100%
activity reference point, and a negative control, without PARP-1
to determine background absorbance. The final reaction mixture
(50 μL) was treated with TACS-Sapphire™, a horseradish per-
oxidase colorimetric substrate, and incubated in the dark for
30 min. Absorbance was read at 630 nm after 30 min. The data
corresponds to means of at least three experiments performed in
triplicate S.D.
Cell culture and inhibition of cell growth
The human breast cancer cell line MCF7 and human ovarian
cancer cell lines A2780 and A2780cisR (obtained from the Euro-
pean Centre of Cell Cultures ECACC, Salisbury, UK) were cul-
tured respectively in DMEM (Dulbecco’s Modified Eagle
Medium) and RPMI containing GlutaMaxI supplemented with
ESI-MS studies
Samples were prepared by mixing 100 μM Ub (Sigma, U6253)
with an excess of gold compound (3 : 1, metal : protein ratio) in
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3287–3293 | 3291

View Article Online
20 mM (NH4)2CO3 buffer (pH 7.4) and incubated for 1 h at
37 °C. Prior to analysis samples were extensively ultrafiltered
using a Centricon YM-3 filter (Amicon Bioseparations, Millipore
Corporation) in order to remove the unbound complex. ESI-MS
data were acquired on a Q-Tof Ultima mass spectrometer
(Waters) fitted with a standard Z-spray ion source and operated
in the positive ionization mode. Experimental parameters were
set as follows: capillary voltage 3.5 kV, source temperature
80 °C, desolvation temperature 120 °C, sample cone voltage 100
V, desolvation gas flow 400 L h−1, acquisition window
300–2000 m/z in 1 s. The samples were diluted 1 : 20 in water
and 5 μL was introduced into the mass spectrometer by infusion
at a flow rate of 20 μL min−1 with a solution of CH3CN/H2O/
HCOOH 50 : 49.8 : 0.2 (v : v : v). External calibration was
carried out with a solution of phosphoric acid at 0.01%. Data
were processed using the MassLynx 4.1 software.
when establishing the selectivity of a compound for a possible
biological target it is essential to consider the overall specific cel-
lular context under investigation. In other words, different cell
lines may express different proteins at different relative concen-
trations, as well as presenting variegate intracellular environ-
ments where the reactivity and selectivity of a metallodrug can
be importantly modulated.
Acknowledgements
The authors thank EU COST D39 action for stimulating
discussions. MS thanks the Institute of Chemical Sciences
and Engineering (EPFL) for providing her with a Master
fellowship. AC thanks the Swiss National Science Foundation
(AMBIZIONE project n° PZ00P2-136908/1) and the University
of Groningen (Rosalind Franklin fellowship) for financial
support. Authors thank Dr Maria Rigobello for providing the
HEK cells.
Conclusions
Notes and references
Gold compounds have clearly emerged as an attractive new class
of cytotoxic agents with potential application in cancer treat-
ment. So far, conspicuous experimental evidence has been gath-
ered suggesting that the pronounced antiproliferative effects
caused by gold compounds most likely arise from innovative
mechanisms of action in comparison to established anticancer
metallodrugs. Here we report on the synthesis and antiprolifera-
tive properties of novel gold(III) complexes, [Au{2-(2-pyridyl)
imidazolato}Cl2] and [Au{2,6-bis(2-benzimidazolato)pyridine}
(OCOCH3)], and mono- and dinuclear gold(I) complexes bearing
imidazolate and phosphane ligands. Interestingly, the mono and
dinuclear gold(I) compounds 2aAu(I) and 3aAu(I)2 bearing the
2-(2-pyridyl)imidazole and PPh3 ligands were the most cytotoxic
of the series, and more selective towards cancer cell lines com-
pared to the non-tumorigenic HEK293 cells. Most of the studied
compounds were also more active than cisplatin on the
A2780cisR cell line, a feature that is common to several cyto-
toxic gold complexes and that indicates different mechanisms of
biological action with respect to classical Pt(II) drugs.
Notably, not many gold complexes have been reported
showing greater selectivity for tumour cells versus normal cells.
Among them a bis-chelated Au(I) bidentate phosphine complex
of the water soluble ligand 1,3-bis(di-2-pyridylphosphino)
propane (d2pypp), namely [Au(d2pypp)2]Cl, was shown to
selectively induce apoptosis in breast cancer cells but not in
normal cells.49 Similarly, Au(I) complexes with modified dipho-
sphine ligands (e.g. bidentate pyridylphosphines) were designed
and tested on both cancer cells and isolated rat hepatocytes, the
latter taken as non-tumorigenic cell models,50,51 and showed
selectivity between different cell types.
1 J. Reedijk, Eur. J. Inorg. Chem., 2009, 1303–1312.
2 M. A. Jakupec, M. Galanski, V. B. Arion, C. G. Hartinger and B.
K. Keppler, Dalton Trans., 2008, 183–194.
3 S. M. Aris and N. P. Farrell, Eur. J. Inorg. Chem., 2009, 1293–1302.
4 S. J. Berners-Price and A. Filipovska, Metallomics, 2011, 3, 863–873.
5 E. R. Tiekink, Inflammopharmacology, 2008, 16, 138–142.
6 S. Nobili, E. Mini, I. Landini, C. Gabbiani, A. Casini and L. Messori,
Medicinal Research Reviews, 2009.
7 C. F. Shaw, Chem. Rev., 1999, 99, 2589–2600.
8 C. Gabbiani, A. Casani, G. Kelter, F. Cocco, M. A. Cinellu, H. H. Fiebig
and L. Messori, Metallomics, 2011, 3, 1318–1323.
9 X. Y. Wang and Z. J. Guo, Dalton Trans., 2008, 1521–1532.
10 A. Casini and L. Messori, Curr. Top. Med. Chem., 2011, 11, 2647–2660.
11 S. P. Fricker, Transition Met. Chem., 1996, 21, 377–383.
12 L. Messori and G. Marcon, Metal Ions in Biological Systems, Vol 41:
Metal Ions and Their Complexes in Medication, 2004, 41, 279–304.
13 T. M. Simon, D. H. Kunishima, G. J. Vibert and A. Lorber, Journal of
Rheumatology, 1979, 6, 91–97.
14 T. M. Simon, D. H. Kunishima, G. J. Vibert and A. Lorber, Cancer
Research, 1981, 41, 94–97.
15 I. Ott, Coord. Chem. Rev., 2009, 253, 1670–1681.
16 J. L. Hickey, R. A. Ruhayel, P. J. Barnard, M. V. Baker,
S. J. Berners-Price and A. Filipovska, J. Am. Chem. Soc., 2008, 130,
12570–12571.
17 D. Fregona, L. Ronconi and D. Aldinucci, Drug Discovery Today, 2009,
14, 1075–1076.
18 C. M. Che, R. W. Y. Sun, W. Y. Yu, C. B. Ko, N. Y. Zhu and H. Z. Sun,
Chem. Commun., 2003, 1718–1719.
19 M. A. Cinellu, G. Minghetti, M. V. Pinna, S. Stoccoro, A. Zucca and
M. Manassero, J. Chem. Soc., Dalton Trans., 2000, 1261–1265.
20 G. Marcon, S. Carotti, M. Coronnello, L. Messori, E. Mini, P. Orioli,
T. Mazzei, M. A. Cinellu and G. Minghetti, J. Med. Chem., 2002, 45,
1672–1677.
21 C. Gabbiani, A. Casini, L. Messori, A. Guerri, M. A. Cinellu,
G. Minghetti, M. Corsini, C. Rosani, P. Zanello and M. Arca, Inorg.
Chem., 2008, 47, 2368–2379.
22 A. Casini, M. C. Diawara, R. Scopelliti, S. M. Zakeeruddin, M. Gratzel
and P. J. Dyson, Dalton Trans., 2010, 39, 2239–2245.
23 A. Bindoli, M. P. Rigobello, G. Scutari, C. Gabbiani, A. Casini and
L. Messori, Coord. Chem. Rev., 2009, 253, 1692–1707.
24 C. Gabbiani, G. Mastrobuoni, F. Sorrentino, B. Dani, M.P. Rigobello,
A. Bindoli, M. A. Cinellu, G. Pieraccini, L. Messori and A. Casini, Med.
Chem. Comm., 2011, 2, 50–54.
25 C. Wetzel, P. C. Kunz, M. U. Kassack, A. Hamacher, P. Bohler,
W. Watjen, I. Ott, R. Rubbiani and B. Spingler, Dalton Trans., 2011, 40,
9212–9220.
In order to identify possible biological targets for the highly
cytotoxic and selective dinuclear compound 3aAu(I)2 we evalu-
ated its PARP-1 inhibition properties on cell extracts from
A2780 and A2780cisR cell lines, in comparison to the non-cyto-
toxic compound 1bAu(I). Compound 3aAu(I)2 was able to
inhibit PARP-1 activity only in extracts from A2780 cells, while
1bAu(I) was ineffective in both cases. These results confirm the
relevance of zinc finger proteins such as PARP-1 as possible
targets for cytotoxic gold compounds, but also indicate that
26 G. Mugesh, K. P. Bhabak and B. J. Bhuyan, Dalton Trans., 2011, 40,
2099–2111.
3292 | Dalton Trans., 2012, 41, 3287–3293
This journal is © The Royal Society of Chemistry 2012

View Article Online
27 C. M. Che and R. W. Y. Sun, Coord. Chem. Rev., 2009, 253, 1682–
1691.
28 L. Ronconi and D. Fregona, Dalton Trans., 2009, 10670–10680.
29 A. Casini, G. Kelter, C. Gabbiani, M. A. Cinellu, G. Minghetti,
D. Fregona, H. H. Fiebig and L. Messori, J. Biol. Inorg. Chem., 2009,
14, 1139–1149.
30 F. Mendes, G. Groessl, A. A. Nazarov, Y. O. Tsybin, G. Sava, I. Santos,
P. J. Dyson and A. Casini, J. Med. Chem., 2011, 54, 2196–2206.
31 V. Schreiber, F. Dantzer, J. C. Ame and G. de Murcia, Nat. Rev. Mol. Cell
Biol., 2006, 7, 517–528.
39 M. Picquet, F. Pelletier, V. Comte, A. Massard, M. Wenzel, S. Toulot,
P. Richard, P. Le Gendre, O. Zava, F. Edafe, A. Casini and P. J. Dyson,
J. Med. Chem., 2010, 53, 6923–6933.
40 M. Wenzel, B. Bertrand, M. J. Eymin, V. Comte, J. A. Harvey,
P. Richard, M. Groessl, O. Zava, H. Amrouche, P. D. Harvey, P. Le
Gendre, M. Picquet and A. Casini, Inorg. Chem., 2011, 50, 9472–9480.
41 C. F. Shaw, Comments Inorg. Chem., 1989, 8, 233–267.
42 J. R. Roberts, J. Xiao, B. Schleisman, D. J. Parsons and C. F. Shaw,
Inorg. Chem., 1996, 35, 424–433.
43 D. J. Daigle, Inorg. Synth., 1998, 32, 40–45.
32 P. A. Jeggo, Curr. Biol., 1997, 8, R49–R51.
33 S. Miranda, E. Vergara, F. Mohr, D. de Vos, E. Cerrada, A. Mendia and
M. Laguna, Inorg. Chem., 2008, 47, 5641–5648.
34 E. Vergara, A. Casini, F. Sorrentino, O. Zava, E. Cerrada,
M. P. Rigobello, A. Bindoli, M. Laguna and P. J. Dyson, ChemMedChem,
2010, 5, 96–102.
44 R. Uson, A. Laguna, M. Laguna, D. A. Briggs, H. H. Murray and
J. J. P. Fackler, Inorg. Synth., 1989, 26, 85–91.
45 A. J. M. Duisenberg, L. M. J. Kroon-Batenburg and A. M. M. Schreurs,
J. Appl. Cryst., 2003, 36, 220–229.
46 R. H. Blessing, Acta Cryst. A, 1995, 51, 33–38.
47 G. M. Sheldrick, Acta Cryst. A, 2008, 64, 112–122.
35 M. Maekawa, A. Nabei, T. Tominaga, K. Sugimoto, T. Minematsu,
T. Okubo, T. Kuroda-Sowa, M. Munakata and S. Kitagawa, Dalton
Trans., 2009, 415–417.
36 K. Nomiya, R. Noguchi, K. Ohsawa, K. Tsuda and M. Oda, J. Inorg.
Biochem., 2000, 78, 363–370.
37 X. D. Yan, L. Y. Pan, Y. Yuan, J. H. Lang and N. Mao, J. Proteome Res.,
2007, 6, 772–780.
38 A. Casini, C. Gabbiani, G. Mastrobuoni, L. Messori, G. Moneti and
G. Pieraccini, ChemMedChem, 2006, 1, 413–417.
48 A. L. Spek, Acta Cryst. D, 2009, 65, 148–155.
49 O. Rackham, S. J. Nichols, P. J. Leedman, S. J. Berners-Price and
A. Filipovska, Biochem. Pharmacol., 2007, 74, 992–1002.
50 J. J. Liu, P. Galettis, A. Farr, L. Maharaj, H. Samarasinha, A.
C. McGechan, B. C. Baguley, R. J. Bowen, S. J. Berners-Price and M.
J. McKeage, J. Inorg. Biochem., 2008, 102, 303–310.
51 M. J. McKeage, S. J. Berners-Price, P. Galettis, R. J. Bowen,
W. Brouwer, L. Ding, L. Zhuang and B. C. Baguley, Cancer Chemother.
Pharmacol., 2000, 46, 343–350.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3287–3293 | 3293