Cytotoxic triosmium carbonyl clusters: A structure-activity relationship study

Article abstract of DOI:10.1002/cmdc.201300394

A structure-activity relationship (SAR) study of the triosmium carbonyl cluster Os₃(CO)₁₀(NCCH₃)₂ was carried out with a series of clusters of the general formula Os ₃(CO)_12-nL_n, cationic osmium clusters and a hemi-labile maltolato-Os cluster. The SAR results showed that good solubility in DMSO and at least one vacant site are required for cytotoxicity. In vitro evaluation of these new compounds showed that some are selectively active against estrogen receptor (ER)-independent MDA-MB-231 breast cancer cell lines relative to ER-dependent MCF-7 breast cancer cells, suggesting that the compounds have a different biological target specific to MDA-MB-231 cells. In particular, the maltolato cluster exhibits strong antiproliferative activity, with an IC₅₀ value of 3 μM after only 24 h incubation. Additionally, biochemical assays conducted with the cationic cluster show that it induces apoptosis, although a biological target has not yet been identified. Further research to establish the molecular targets of these compounds and to develop improved organometallic clusters as potential breast cancer therapeutics is underway. The land of Os(mium): A structure-activity relationship study of a series of triosmium carbonyl clusters revealed that good solubility and a vacant site are necessary for activity. The newly synthesized compounds were more selective toward estrogen receptor (ER)-negative breast cancer cells versus ER-positive cells, indicating a different biological target than for tamoxifen.

Full text of DOI:10.1002/cmdc.201300394

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201300394
Cytotoxic Triosmium Carbonyl Clusters: A Structure–
Activity Relationship Study
Hui Zhi Shirley Lee,[a] Weng Kee Leong,*[a] Siden Top,*[b] and Anne Vessiꢀres[b]
A structure–activity relationship (SAR) study of the triosmium
carbonyl cluster Os3(CO)10(NCCH3)2 was carried out with
a series of clusters of the general formula Os3(CO)12ÀnLn, cation-
ic osmium clusters and a hemi-labile maltolato-Os cluster. The
SAR results showed that good solubility in DMSO and at least
one vacant site are required for cytotoxicity. In vitro evaluation
of these new compounds showed that some are selectively
active against estrogen receptor (ER)-independent MDA-MB-
231 breast cancer cell lines relative to ER-dependent MCF-7
breast cancer cells, suggesting that the compounds have a dif-
ferent biological target specific to MDA-MB-231 cells. In partic-
ular, the maltolato cluster exhibits strong antiproliferative activ-
ity, with an IC50 value of 3 mm after only 24 h incubation. Addi-
tionally, biochemical assays conducted with the cationic cluster
show that it induces apoptosis, although a biological target
has not yet been identified. Further research to establish the
molecular targets of these compounds and to develop im-
proved organometallic clusters as potential breast cancer ther-
apeutics is underway.
Introduction
The emergence of organometallic compounds within the bio-
logical sciences has been gaining momentum, especially in the
area of drug development. Much interest has been focused on
organometallic compounds that possess anticancer activity,
which can be generally classified into two groups: one group
that involves modification of organic analogues, such as the
ferroquines[1] and ferrocifens,[2] and another group exhibiting
metal-based activity, such as the arenes[3] and titanocenes.[4]
The organometallic clusters are a class of organometallic
compounds that was recently discovered to have potential as
drug candidates.[5] We were the first to report a series of trios-
mium carbonyl clusters (including clusters 1 and 3a below)
that showed antiproliferative activity against both ER+ and ERÀ
breast cancers. Mode of action studies for the most active
compounds in the series, cluster species Os3(CO)10(NCCH3)2 (1),
showed that these compounds hyperstabilized microtubules
and induced apoptosis.[6] In this previous study, we hypothe-
sized that the molecular requirement for cytotoxicity was the
availability of two coordination sites. The study was, however,
hampered by the low solubility of the compounds in DMSO,
the vehicle used in the cytotoxicity assays. This problem was
particularly acute with the monosubstituted derivatives, such
as 3a, which was not very soluble in DMSO and was found to
be inactive. The effect of solubility on the anticancer activity of
these compounds is therefore of interest.
In the present study, we decided to examine the structure–
activity relationships (SAR) of a series of similar derivatives to
assess the effect of solubility (e.g., protonation of 3a to form
4a increases its solubility in DMSO), number of vacant sites
(e.g., comparison of activity for positive control 1 with 3b and
3c), and type of ligands (e.g., g-pyrone of compound 5 in
place of the nitrile ligand of the parent compound) on activity
(Figure 1).
Results and Discussion
Syntheses of triosmium carbonyl clusters
Syntheses of osmium carbonyl clusters 1 and 3a were carried
out as previously reported[7] by heating Os3(CO)12 in excess
acetonitrile in the presence of trimethylamine-N-oxide (TMNO).
[a] H. Z. S. Lee, Prof. W. K. Leong
Division of Chemistry & Biological Chemistry
Nanyang Technological University
21 Nanyang Link, Singapore 637371 (Singapore)
E-mail: chmlwk@ntu.edu.sg
[b] Dr. S. Top, Dr. A. Vessiꢀres
UMR CNRS 7223, Laboratoire Charles Friedel
Ecole Nationale Supꢁrieure de Chimie de Paris
11 rue Pierre et Marie Curie, 75231 Paris Cedex 05 (France)
E-mail: siden.top@chimie-paristech.fr
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201300394.
Figure 1. Molecular structures of osmium clusters 1–5.
ChemMedChem 2014, 9, 1453 – 1457 1453
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
This route, however, was inadequate for the preparation of an-
alogues in which the nitrile was a solid or a high boiling point
liquid. In principle, 1 or 3a could serve as precursors, and an
excess of the desired nitrile could be used to displace the ace-
tonitrile ligands; however, distinguishing the product and pre-
cursor spectroscopically was difficult as they showed almost
identical patterns of carbonyl stretches in the IR spectrum. A
synthetic route based on the cluster Os3(CO)11(COE) (2) as the
intermediate was thus used for the syntheses of
Os3(CO)11(NCCH2CH2OH) (3b) and Os3(CO)11(NCC6H4OH-p) (3c)
(Scheme 1). Cluster 2 had a distinct IR pattern that was differ-
these compounds was therefore estimated using their solubili-
ty in DMSO instead (Table S1).
The cationic cluster 4a, which has better solubility in DMSO,
shows enhanced activity over its neutral analogue 3a (saturat-
ed concentrations of 19.9 and 3.7 mm, respectively). This also
Table 1. Inhibition of cell growth by triosmium carbonyl clusters after
24 h, as determined by MTS assay.
Compd
IC50 [mm][a]
MCF-7
MDA-MB-231
MCF-10A
1
30Æ9
16Æ2
28Æ6
10Æ3
19Æ2
14Æ6[b]
8Æ1
18Æ3
>30
19Æ2
22Æ1
23Æ10
3b
3c
4a
5
13Æ3[b]
7Æ1
3Æ1
Scheme 1. Synthetic route to osmium clusters with nitrile ligands. Reagents
and conditions: a) Me3NO, CH2Cl2, 2 h, 408C, 98% yield; b) NCR, CH2Cl2, 1 h,
room temperature, 87–93% yield; COE=cis-cyclooctene.
[a] Data represent the mean ÆSD of n=9 independent experiments per-
formed in triplicate. [b] Calculated from the results of a separate assay.
ent from the nitrile-substituted osmium clusters, and the cyclo-
octene (COE) ligand was more weakly bound than the acetoni-
trile ligand, making it a better leaving group for substitution
reactions. The identity of 2 was established by its IR spectrum,
which showed a similar pattern for the CO stretches as that of
the ethene analogue Os3(CO)11(C2H4), and by its reaction with
PPh3 to form the known derivative Os3(CO)11(PPh3).[8] Os3(CO)11-
appears to be a factor for the cytotoxicity of 1 (saturated con-
centration of 15.0 mm). On the other hand, the cationic cluster
4b, which contains a non-labile phosphine ligand, is inactive.
The importance of this is further supported by the observation
that 3b and 3c, which contain labile nitrile ligands with hy-
droxy groups that improve their solubility (saturated concen-
trations of 14.0 mm and 9.5 mm, respectively), also show good
antiproliferative activity. Good solubility (cell permeability)
alone, however, is insufficient; a labile ligand is also required.
Furthermore, neither a positive charge on the cluster nor on
the BF4À anion contributes to the activity. The activity of cluster
5, in which the weak bonding interaction between the carbon-
yl group of the pyrone and the cluster core can lead to the for-
mation of a vacant coordination site, also demonstrates that
activity is not limited to compounds with nitrile ligands; the
availability of a vacant coordination site at the metal center is
the only requirement.
À
(NCCH3)(m-H)+BF4 (4a) was prepared by protonating 3a with
HBF4 acid,[9] while Os3(CO)9(m-H)(m-g-C6H5O3) (5) was synthe-
sized by the reaction of 1 and 2-methyl-3-hydroxy-g-pyrone.[10]
Biochemical studies
Cell viability was assessed using the MTS assay in a MDA-MB-
231 cell line at a concentration of 20 mm for triosmium carbon-
yl clusters 3–5, with 1 as the positive control (Figure 2). A
factor that is expected to influence the cytotoxicity of a com-
pound is its ability to permeate the cell membrane. This is usu-
ally measured as the water/octanol partition coefficient via re-
versed-phase HPLC. However, the instability of some of the
compounds precluded this method. The cell permeability of
The IC50 values of 1, 3b, 3c, 4a, and 5 were determined and
showed significant antiproliferative activity at 20 mm in three
cell lines, namely, MCF-7 (ER+ breast cancer), MDA-MB-231
(ERÀ breast cancer) and MCF-10A (normal breast epithelial) cell
lines (Table 1). These results showed that clusters 3b, 3c, 4a,
and 5 have either similar or higher cytotoxicity (lower IC50
values) than 1. In particular, cluster 5, the most active com-
pound, has an IC50 value of 3 mm in a short incubation time of
24 h. One clear structural difference between cluster 5 and the
other clusters is that it contains a maltolato instead of a nitrile
ligand. Maltolato–metal complexes, especially those that con-
tain iron, have been found to generate reactive oxygen species
(ROS) in cells. In many cancer models, it has been found that
this causes DNA strand breakage, leading to antineoplastic ac-
tivity.[11] It is therefore possible that, in addition to Os-induced
activity, the activity of 5 may be enhanced by release of the
maltolato ligand upon reaching the target. This may then bind
to any metal ions, such as iron, that are present in the cell to
form ROS species. This hypothesis requires further investiga-
tion.
Figure 2. MTS assay of osmium clusters 1 and 3–5 on MDA-MB-231 breast
cancer cells; 20 mm, 24 h. DMSO concentration: 0.2% (v/v).
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1453 – 1457 1454

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Figure 3. MDA-MB-231 cells stained with Annexin V–FITC and PI after 15 h
incubation with a) vehicle only (control), b) 10 mm 1 (positive control), and
c) 10 mm 4a. FL1-H and FL2-H refer to Annexin V–FITC and PI, respectively.
Q2: necrotic cells (stained with both Annexin V–FITC and PI); Q3: apoptotic
cells (stained with Annexin V–FITC only); Q4: live cells (no stain).
Interestingly, all of the compounds were also found to be
more active against MDA-MB-231 than against MCF-7 cells.
This is in contrast to tamoxifen, the drug currently used for the
treatment of breast cancer, which is active only against MCF-7
cells. It would therefore appear that these compounds do not
act through estrogen receptors but rather act on a different
biological target that is specific to MDA-MB-231 cells.
The cytotoxicity exhibited by cationic cluster 4a was particu-
larly interesting as, unlike its unprotonated analogue 3a, it did
not lose its acetonitrile ligand easily. For example, 4a failed to
react upon stirring with triphenylphosphine, even up to 808C.
Charged species are known to be able to induce necrosis
through rupture of the cell membrane. This was unlikely in the
case of 4a, however, as 4b, which is also charged, was not cy-
totoxic. Nevertheless, assays to show that 4a induced apopto-
sis were carried out. For instance, flow cytometric analysis,
using FITC-conjugated annexin V and propidium iodide (PI)
staining of MDA-MB-231 cells that were incubated for 15 h
with 10 mm of 4a showed that 50% of the cells were dead, of
which 80% were in early apoptosis (Figure 3). Induction of
apoptosis was further supported by significant changes in
chromatin distribution of MDA-MB-231 cells treated with
a 5 mm solution of 4a relative to the control (Figure 4; more
images in the Supporting Information), as well as observation
of the characteristic sub-G1 peak in the DNA content frequency
histogram for cells treated with a 10 mm solution of 4a for
15 h (14.8%, relative to 2.11% with control).[12]
A possible biological target for 4a is the mitochondria, remi-
niscent of other lipophilic cationic compounds such as the al-
kyltriphenylphosphine (TPP) cations, which have been em-
ployed to target the mitochondria.[13] TPP cations are able to
pass directly through the phospholipid bilayer of the cell mem-
brane because of their large hydrophobic surface area. They
also tend to accumulate in the mitochondria because of the
large mitochondria membrane potential (higher negative
charge inside the mitochondria);[14] selective targeting of
cancer cells rests on the fact that these cells exhibit higher mi-
tochondrial transmembrane potential than normal cells. Mode
of action studies to test this hypothesis for 4a are currently
underway.
Conclusions
In this study, we have shown that the cytotoxicity of clusters
of the structural type Os3(CO)12Àn(L)n require good solubility
and at least one labile ligand. We also found some clusters
with different structural types that were cytotoxic. In particular,
they were more cytotoxic against MDA-MB-231 ERÀ breast
cancer cells than against MCF-7 ER+ breast cancer cells or
MCF-10A normal breast epithelial cells. This indicated that they
did not act against the hormone receptor but via a different
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1453 – 1457 1455

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
with a magnetic stir bar. The reaction mixture was stirred for 2 h at
408C under argon, and formation of the product was monitored
by IR spectroscopy. The mixture was then filtered through silica gel
to remove any unreacted Me3NO, and the solvent was removed in
vacuo to give Os3(CO)11(COE) (2) as a yellow oil (21.3 mg, 98%):
1H NMR: d=3.66–3.69 (m, 2H, vinylic protons), 2.49–2.54 (m, 2H,
allylic protons), 1.88–1.97 (m, 2H, allylic protons), 1.68–1.82 (m, 4H,
4H of C8H14), 1.43–1.53 ppm (m, 4H, 4H of C8H14); IR (cyclohexane):
n˜CO =2114 (w), 2061 (s), 2041 (s), 2025 (vs), 2006 (w), 2000 (m),
1989 (w), 1979 (w), 1974 (w), 1962 cmÀ1 (w).
Synthesis of Os3(CO)11(PPh3): Cluster 2 (21 mg, 21 mmol) and PPh3
(5.6 mg, 21 mmol) were dissolved in CH2Cl2 (5 mL). The mixture was
then stirred for 1 h at room temperature. Formation of the product
was monitored by IR spectroscopy. The CH2Cl2 was removed in
vacuo, and the crude product was purified by TLC (hexane/CH2Cl2,
1.75:1, v/v) to give Os3(CO)11(PPh3) as a yellow solid (15.8 mg,
66%): IR: n˜CO =2107 (m), 2055 (s), 2034 (ms), 2018 (s), 1988 (m),
1978 cmÀ1 (m); Lit. values (cyclohexane):[8] n˜CO =2108 (m), 2055 (s),
2035 (s), 2019 (s), 2000 (ms), 1989 (m), 1978 cmÀ1 (m).
Synthesis
of
Os3(CO)11(NCCH2CH2OH)
(3b)
and
Os3(CO)11(NCC6H4OH-p) (3c): An identical procedure to that used
for the synthesis of Os3(CO)11(PPh3) was followed, with the excep-
tion that no further purification was carried out. Removal of the
solvent afforded a yellow solid.
3b (18.5 mg, 93%): mp: 1218C; 1H NMR: d=3.92 (s, broad, 2H,
CH2OH), 3.22–3.25 ppm (t, 2H, J=6.4 Hz, NCCH2); IR: n˜CO =2106
(w), 2067 (m), 2053 (vs), 2040 (vs), 2017 (s, sh), 2008 (vs), 1980 cmÀ1
(m); ESI: 951.52 [M+H]+.
1
3c (18.2 mg, 87%): mp: 1438C; H NMR: d=7.53–7.57 (m, 2H, 2H
of C6H4), 6.90–6.98 (m, 2H, 2H of C6H4), 5.61 ppm (OH); IR: n˜CO
=
2106 (w), 2068 (m), 2053 (vs), 2040 (vs), 2017 (s, sh), 2007 (vs), 1982
(m) cmÀ1; ESI: 998.55 [M+H]+.
Figure 4. Fluorescence image (60ꢂ) of MDA-MB-231 cells stained with DAPI:
a) control and b) after incubation with 5 mm 4a for 3 h.
Biological methods
General: Experimental cultures of MCF-7, MDA-MB-231, and MCF-
10A cell lines were obtained from American Type Culture Collec-
tion (ATCC) and cultured in tissue culture dishes (Nunc Inc., Naper-
ville, IL, USA). MCF-7 and MDA-MB-231 cells were maintained in
Dulbecco’s modified Eagle’s medium (DMEM; Grand Island, NY,
USA), supplemented with 10% fetal bovine serum (FBS), 1% l-glu-
tamine (Gibco Laboratories), and 1% penicillin (Gibco Laboratories)
at 378C in 5% CO2 atmosphere. MCF-10A cells were maintained in
DMEM:F12 supplemented with 7.5% FBS, 1% l-glutamine, 0.4%
gentamicin (Gibco Laboratories), and growth factors EGF (Invitro-
gen), insulin (Invitrogen), hydrocortisone (Sigma–Aldrich) and chol-
era toxin (Sigma–Aldrich) at 378C in 5% CO2 atmosphere. Phos-
phate-buffered saline (PBS) was obtained from Gibco. For all apop-
tosis assays, the control refers to cells incubated with DMSO (vehi-
cle), and positive control refers to cells incubated with 1.
target specific to MDA-MB-231 cells. Although apoptotic assays
confirmed that the cationic cluster 4a induced apoptosis, its
biological target has not been established, with further re-
search on this aspect currently underway.
Experimental Section
Synthesis
General: Synthetic steps were carried out under an argon atmos-
1
phere using standard Schlenk techniques. H NMR spectra were re-
corded on a Jeol 400 NMR spectrometer as CDCl3 solutions unless
otherwise stated; chemical shifts reported were referenced against
the residual proton signals of the solvent. IR spectra were recorded
on a Bruker Alpha FTIR spectrometer as CH2Cl2 solutions unless
otherwise stated. The starting material, Os3(CO)12, was obtained
from Oxkem Ltd. (Reading, UK); all other chemicals were purchased
from other commercial sources and used as supplied. The clusters
1,[7] 3a,[7] 4a,[9] 4b,[15] and 5[10] were synthesized according to pub-
lished methods.
Proliferation assay: The osmium carbonyl clusters were first dis-
solved in sterile DMSO to make 10 mm stock solutions. They were
then diluted with serum-free medium, with final concentrations in
a range of 0.1–200 mm used for treatment. The cells were then
treated with the indicated concentrations of osmium carbonyl clus-
ters and incubated for 24 h at 378C in 5% CO2, after which, Cell
Titer 96_Aqueous One Cell Proliferation Assay (20 mL, Promega)
was added to each well. This was then left to incubate at 378C in
5% CO2 for 2 h. Absorbance at 490 nm was then measured, and
cell proliferation relative to the control sample was calculated.
Synthesis of Os3(CO)11(COE) (2): A sample of Os3(CO)12 (20 mg,
22 mmol) and Me3NO (2 mg, 26 mmol) were added to CH2Cl2
(10 mL) and cis-cyclooctene (1 mL) in a reaction vessel equipped
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1453 – 1457 1456

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Chem. 1997, 40, 3715; d) F. Dubar, J. Khalife, J. Brocard, D. Dive, C. Biot,
Molecules 2008, 13, 2900.
Curves were then generated using a sigmoidal dose–response (var-
iable slope) equation and GraphPad Prism 5 software.
[2] a) E. A. Hillard, A. Vessiꢀres, G. Jaouen in Medicinal Organometallic
Chemistry, Vol. 32 (Eds.: G. Jaouen, N. Metzler-Nolte), Springer, Heidel-
berg, 2010, pp. 81–117; b) A. Nguyen, A. Vessiꢀres, E. A. Hillard, S. Top,
P. Pigeon, G. Jaouen, Chimia 2007, 61, 716; c) A. Vessiꢀres, J. Organomet.
Chem. 2013, 734, 3; d) S. Top, A. Vessiꢀres, G. Leclercq, J. Quivy, J. Tang,
J. Vaissermann, M. Huchꢃ, G. Jaouen, Chem. Eur. J. 2003, 9, 5223; e) S.
Top, A. Vessiꢀres, C. Cabestaing, I. Laios, G. Leclercq, C. Provot, G.
Jaouen, J. Organomet. Chem. 2001, 637, 500.
[3] a) P. J. Dyson, G. Sava, Dalton Trans. 2006, 1929; b) E. Alessio, G. Mestro-
ni, A. Bergamo, G. Sava, Curr. Top. Med. Chem. 2004, 4, 1525; c) C. G.
Hartinger, S. Zorbas-Seifried, M. A. Jakupec, B. Kynast, H. Zorbas, B. K.
Keppler, J. Inorg. Biochem. 2006, 100, 891; d) F. Lentz, A. Drescher, A. Lin-
dauer, M. Henke, R. A. Hilger, C. G. Hartinger, M. E. Scheulen, C. Dittrich,
B. K. Keppler, U. Jaehde, Anti-Cancer Drugs 2009, 20, 97.
Annexin V–FITC and PI staining for flow cytometry: MDA-MB-231
cells were plated in six-well plates at the same initial density and
were allowed to adhere and grow for 24 h at 378C in 5% CO2.
They were then serum-starved for 24 h before treatment with
10 mm of 1 or 4a. The cells were harvested after the incubation
period (15 h) and washed in cold PBS (5 mL). The washed cells
were centrifuged to obtain a pellet, the supernatant was discarded,
and the pelleted cells were washed with PBS (2ꢂ5 mL) before re-
suspension in 1ꢂ annexin-binding buffer (Invitrogen). Annexin V–
FITC (5 mL) and PI (1 mL) stains were then added, and the stained
cells were analyzed using a BD FACSCalibur flow cytometer.
Nuclear staining with 4,6-diamidino-2-phenylindole (DAPI): MDA-MB-
231 cells were plated in single-well plates (Ibidi) at the same initial
density and were allowed to adhere and grow for 24 h at 378C in
5% CO2. They were then serum-starved for 24 h before treatment
with 5 mm of 4a for 3 h at 378C. After treatment, the cells were
fixed with 3.7% formalin (room temperature, 1 h), washed with
PBS (2 mL), stained with 4,6-diamidino-2-phenylindole (DAPI,
1 mgmLÀ1 in PBS, 5 min at 378C), washed, and then photographed
using a confocal microscope equipped with a UV light filter
(Nikon D Eclipse C1).
[4] a) P. Kçpf-Maier, Anticancer Res. 1999, 19, 493; b) M. Singh, Curr. Pharm.
Des. 1999, 5, 443; c) K. Strohfeldt, M. Tacke, Chem. Soc. Rev. 2008, 37,
1174; d) M. A. Jakupec, M. Galanski, B. K. Keppler, Rev. Physiol. Biochem.
Pharmacol. 2003, 146, 1.
[5] a) E. Rosenberg, F. Spada, K. Sugden, B. Martin, R. Gobetto, L. Milone, A.
Viale, J. Organomet. Chem. 2004, 689, 4729; b) D. Colangelo, A. Ghiglia,
A. Ghezzi, M. Ravera, E. Rosenberg, J. Inorg. Biochem. 2005, 99, 505; c) B.
Therrien, W. H. Ang, F. Chꢃrioux, L. V. P. Lucienne, J. Jeanneret, G. S. Fink,
P. J. Dyson, J. Cluster Sci. 2007, 18, 741; d) V. D. Reddy, D. Dayal, D. J.
Szalda, S. C. Cosenza, M. V. R. Reddy, J. Organomet. Chem. 2012, 700,
180; e) A. Maillet, S. Yadav, Y. L. Loo, K. Sachaphibulkij, S. Pervaiz, Cell
Death Dis. 2013, 4, e653.
DNA fragmentation using PI: MDA-MB-231 cells were plated in T25
flasks at the same initial density and were allowed to adhere and
grow for 24 h at 378C in 5% CO2. They were then serum-starved
for 24 h before treatment with 10 mm of 4a for 15 h. After treat-
ment, the cells were fixed with 70% EtOH (0.7 mL, À208C, over-
night). They were then centrifuged (200 g, 5 min), suspended in
PBS (0.8 mL), and centrifuged again (300 g, 5 min). The cell pellet
was then suspended in PBS (0.05 mL), DNA extraction buffer
(0.04 mL) was added, and the mixture was incubated for 5 min at
room temperature to facilitate the extraction of low molecular
weight DNA. The cells were then centrifuged, PI (0.15 mL) was
added, and they were incubated at room temperature in the dark
for 30 min. The stained cells were then analyzed using a BD FACS-
Calibur flow cytometer.
[6] a) K. V. Kong, W. K. Leong, S. P. Ng, T. H. Nguyen, L. H. K. Lim, ChemMed-
Chem 2008, 3, 1269; b) K. V. Kong, W. K. Leong, L. H. K. Lim, J. Organo-
met. Chem. 2009, 694, 834; c) K. V. Kong, W. K. Leong, L. H. K. Lim, Chem.
Res. Toxicol. 2009, 22, 1116.
[7] J. Y. Jung, D. K. Kempe, L. J. Loh, S. E. Shoultz, G. L. Powell, J. Organomet.
Chem. 2012, 700, 219.
[8] a) J. Evans, G. S. McNulty, J. Chem. Soc. Dalton Trans. 1984, 79; b) B. F. G.
Johnson, J. Lewis, D. A. Pippard, J. Chem. Soc. Dalton Trans. 1981, 407.
[9] C. E. Anson, E. J. Ditzel, M. Fajardo, H. D. Holden, B. F. G. Johnson, J.
Lewis, J. Puga, P. R. Raithby, J. Chem. Soc. Dalton Trans. 1984, 2723.
[10] Q. Lin, W. K. Leong, Organometallics 2003, 22, 3639.
[11] a) M. Hironishi, R. Kordek, R. Yanagihara, RM. Garuto, Neurodegeneration
1996, 5, 325; b) K. Murakami, K. Ishida, K. Watakabe, R. Tsubouchi, M.
Haneda, M. Yoshino, BioMetals 2006, 19, 253; c) K. Murakami, K. Ishida,
K. Watakabe, R. Tsubouchi, M. Naruse, M. Yoshino, Toxicol. Lett. 2006,
161, 102; d) S. Amatori, G. Ambrois, M. Fanelli, M. Formica, V. Fusi, L.
Giorgi, E. Macedi, M. Micheloni, P. Paoli, R. Pontellini, P. Rossi, J. Org.
Chem. 2012, 77, 2207; e) S. Amatori, I. Bagaloni, E. Macedi, M. Formica,
L. Giorgi, V. Fusi, M. Fanelli, Br. J. Cancer 2010, 103, 239.
Acknowledgements
[12] Flow cytometry: Z. Darzynkiewicz, Methods Cell Biol. 1994, 41, 27.
[13] a) G. Cheng, J. Zielonka, B. P. Dranka, D. McAllister, A. G. Mackinnon, Jr.,
J. Joseph, B. Kalyanaraman, Cancer Res. 2012, 72, 2634; b) D. Dindo, F.
Dahm, Z. Szulc, A. Bielawska, L. M. Obeid, Y. A. Hamun, R. Graf, P. A.
Clavien, Mol. Cancer Ther. 2006, 5, 1520; c) J. S. Modica-Napolitano, J. R.
Aprille, Adv. Drug Delivery Rev. 2001, 49, 63.
[14] a) M. P. Murphy, Biochim. Biophys. Acta Bioenerg. 2008, 1777, 1028;
b) M. P. Murphy, Trends Biotechnol. 1997, 15, 326; c) M. P. Murphy, R. A. J.
Smith, Annu. Rev. Pharmacol. Toxicol. 2007, 47, 629; d) C. M. Porteous, A.
Logan, C. Evans, E. C. Ledgerwood, D. K. Menon, F. Aigbirhio, R. A. J.
Smith, M. P. Murphy, Biochim. Biophys. Acta Gen. Subj. 2010, 1800, 1009.
[15] A. J. Deeming, S. D. Mtunzi, S. E. Kabir, M. B. Hursthouse, K. M. A. Malik,
N. P. C. Walker, J. Chem. Soc. Dalton Trans. 1987, 1869.
Financial support from the Nanyang Technological University
(NTU), Singapore through a start-up grant (no. M4080504) to
W.K.L. and a Research Scholarship under the NTU-ParisTech (Paris
Institute of Technology, France) joint PhD Program to H.Z.S.L. are
gratefully acknowledged.
Keywords: apoptosis · carbonyl clusters · estrogen receptors ·
osmium · structure–activity relationships
[1] a) M. A. L. Blackie, P. Beagley, S. L. Croft, H. Kendrick, J. R. Moss, K. Chi-
bale, Bioorg. Med. Chem. 2007, 15, 6510; b) F. Dubar, T. J. Egan, B. Pra-
dines, D. Kuter, K. K. Ncokazi, D. Forge, J. F. Paul, C. Pierrot, H. Kalamou,
J. Khalife, E. Buisine, C. Rogier, H. Vezin, I. Forfar, C. Slomianny, X. Trivelli,
S. Kapishnikov, L. Leiserowitz, D. Dive, C. Biot, ACS Chem. Biol. 2011, 6,
275; c) C. Biot, G. Glorian, L. Maciejewski, J. Brocard, O. Domarle, G.
Blampain, P. Millet, A. J. Georges, H. Abessolo, D. Dive, J. Lebibi, J. Med.
Received: October 8, 2013
Revised: December 5, 2013
Published online on January 20, 2014
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1453 – 1457 1457