Induction of apoptosis by hexaosmium carbonyl clusters

Article abstract of DOI:10.1016/j.jorganchem.2008.07.039

A number of tri- and hexaosmium carbonyl cluster derivatives were screened for cytotoxicity against five cancer cell lines, and the hexaosmium carbonyl clusters Os₆(CO)₁₈ and Os₆(CO)₁₆(NCCH₃)₂

Full text of DOI:10.1016/j.jorganchem.2008.07.039

Journal of Organometallic Chemistry 694 (2009) 834–839
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Induction of apoptosis by hexaosmium carbonyl clusters q
b
Kien Voon Kong a, Weng Kee Leong a, , Lina H.K. Lim
*
a Department of Chemistry, National University of Singapore, Kent Ridge 117543, Singapore
b Department of Physiology, National University of Singapore, Kent Ridge 117597, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 June 2008
Received in revised form 29 July 2008
Accepted 30 July 2008
Available online 6 August 2008
A number of tri- and hexaosmium carbonyl cluster derivatives were screened for cytotoxicity against five
cancer cell lines, and the hexaosmium carbonyl clusters Os6(CO)18 and Os6(CO)16(NCCH3)2 were found to
be active against four of these, viz., ER+ breast carcinoma (MCF-7), ER-breast carcinoma (MDA-MB-231),
metastatic colorectal adenocarcinoma (SW620) and hepatocarcinoma (Hepg2), with IC50 values as low as
6
l
M. Studies on their mode of action with the MDA-MB-231 cell line pointed to the induction of apop-
tosis, as has been found earlier for the trinuclear cluster Os3(CO)10(NCCH3)2.
Ó 2008 Elsevier B.V. All rights reserved.
Keywords:
Osmium
Cluster compounds
Apoptosis
Cancer
1. Introduction
perhaps the molecular requirement for cytotoxicity is the accessi-
bility of two coordination sites on the cluster. To test this hypoth-
esis, we decided to carry out MTS assays on the disubstituted
triosmium cluster derivatives Os3(CO)10(L)2 (1, where L = PPh3
The continuing search for anticancer drugs have led recently to
the development of organometallic compounds as potential candi-
dates [1]. These may comprise organometallic modifications of or-
ganic compounds known to show therapeutic action, such as
ferrocifen [2], and NAMI-A [3], or those which show metal-based
anticancer activity in themselves, such as the titanocenes [4], and
organometallic clusters [5]. In a recent paper, we reported the
screening of six triosmium clusters against five cancer cell lines
and one cluster in particular, Os3(CO)10(NCCH3)2 (1a), was found
to be active against four of these, viz., ER+ breast carcinoma
(MCF-7), ER-breast carcinoma (MDA-MB-231), metastatic colorec-
tal adenocarcinoma (SW620) and hepatocarcinoma (Hepg2) [6].
We have made an initial attempt to study the structure-activity
relationship by screening a larger range of osmium clusters, in
the course of which we discovered two higher nuclearity osmium
clusters which show anticancer activity; the study is reported here.
t
(b), AsPh3 (c), SbPh3 (d), PCy3 (e), P(OMe)3 (f) or BuNC (g)), the
monosubstituted triosmium cluster derivatives Os3(CO)11(L) (2,
where L = PPh3 (b), AsPh3 (c), or SbPh3 (d)), as well as the hexaos-
mium clusters Os6(CO)18 (4) and Os6(CO)16(NCMe)2 (5) (Chart 1).
With the exception of 1a, 4 and 5, all the other clusters had
maximum solubilities (in DMSO) no higher than 10 lM, and their
IC50 values clearly exceeded that (Supporting information Table
S2, Figs. S1 and S2). The computed growth inhibition (IC50) values
for the three active compounds are given in Table 1, and their cyto-
toxicities are also reflected in the morphological changes of the
cells after 24 h incubation with 10 lM solutions (Supporting infor-
mation Figs. S3–7). A comparison of the cytotoxicities was also
made against cisplatin, resveratrol and tamoxifen on both the
MCF-7 and MDA-MB-231 cell lines at both 1 and 10 lM concentra-
tions (Fig. 1). The results show that 1a, 4 and 5 have comparable
efficacies, and they are also comparable in efficacy to cisplatin,
tamoxifen, and resveratrol. In contrast to 1a, however, there is no
significant difference in the cytotoxicity of 4 and 5 towards cancer
vs normal cells. Nevertheless, we sought to confirm that their cyto-
toxicity is due to the induction of apoptosis with several assays
using the MDA-MB-231 cell line.
2. Results and discussion
We have earlier found that besides 1a, the triosmium clusters
Os3(CO)10(l-H)2 and Os3(CO)10(l-OH)(l-H) also exhibited cytotox-
icity. In contrast, the clusters Os3(CO)11(NCMe) (2a), Os3(CO)12 (3),
and Os3(CO)10(cyclooctene)2 exhibited no discernible cytotoxicity
up to their maximum soluble concentrations. This indicated that
MDA-MB-231 cells treated with 10 lM solutions of 1a, 4 and 5
showed significant changes in chromatin distribution (separate
globular structures) compared to the control (rounded, intact nu-
clei) when analysed under fluorescence microscopy with DAPI
staining (Fig. 2). This change is consistent with the induction of
apoptosis. Flow cytometry using FITC-conjugated annexin V and
q
In celebration of Gerard Jaouen’s 65th birthday.
* Corresponding author.
E-mail address: chmlwk@nus.edu.sg (W.K. Leong).
0022-328X/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2008.07.039

K.V. Kong et al. / Journal of Organometallic Chemistry 694 (2009) 834–839
835
The above assays all point to the induction of apoptosis by 4 and
5, just as has been observed earlier for 1a [6]. That 4 showed sim-
ilar activity to 1a, in contrast to the low activity shown by 3, may
be attributed to the difference in their substitutional lability. For
example, the conversion of 4 to 5 via trimethylamine N-oxide acti-
vated substitution can be carried out at low temperatures [9],
while a similar conversion of 3 to 1a required refluxing dichloro-
methane [10].
Os
Os
Os
NCMe
L
Os
Os
Os
Os
Os
Os
L
L
MeCN
1a
1b-g
2b-d
L
L
Os
Os
Os
Os
Os
Os
Os
3. Conclusion
Os
Os
In this study, we set out to test the hypothesis that the molec-
ular requirement for the cytotoxicity exhibited by triosmium clus-
ters such as Os3(CO)10(NCCH3)2 is the accessibility of two
coordination sites on the cluster. The study was limited by the
low solubilities of the compounds but the data does suggest that
labile ligands are required. More significantly, we have uncovered
two hexaosmium clusters Os6(CO)18 and Os6(CO)16(NCCH3)2,
which exhibit good anticancer potential. These clusters show high
cytotoxicity for four of the five types of cancer cells tested, and it
has been demonstrated that the cytotoxicity is the result of induc-
tion of apoptosis. The contrast in cytotoxicity between Os6(CO)18
and Os3(CO)12 suggests that substitutional lability is an important
factor. Interestingly, however, the hexaosmium clusters show
comparable cytotoxicity towards normal cells. The underlying rea-
son for this difference in behaviour from the triosmium cluster is
not understood and awaits studies into their mode of action.
3
4
5
: L = CO
: L = CH3CN
Chart 1.
Table 1
Inhibition of cell growth (IC50
after 24 h incubation, determined by MTS assay
, lM) by 1a, 4 and 5 measured with the six cell lines
Compound MDA-MB-
231
MCF-7
HT29
SW620
Hepg2
MCF-
10A
1a
4
5
4.8 0.7
6.0 0.7
6.8 0.9
7.2 1.2 31.3 1.6 8.7 1.0
6.8 1.5 15.0 1.8 8.2 1.2
8.8 0.8
6.6 1.2 8.6 1.5
10
18
2
9.0 1.0 >40
9.8 0.6 27.7 1.5
2
propidium iodide (PI) staining (24 h incubation with 10 lM solu-
4. Experimental
tions) showed that 1a, 4 and 5 increased the early apoptosis cell
population (FITC stained), reaching ꢀ40–60%, whereas 3 showed
a similar result to control (Fig. 3).
4.1. General procedure
Flow cytometric analysis for DNA fragmentation under the
same conditions also exhibit cell cycle arrest at the G2 phase;
ꢀ35–45% compared to 24.4% and 22.3% for 3 and control, respec-
tively; we did not observe any significant block in other phases
(Supporting information Fig. S8). This is in contrast to resveratrol,
for example, which is known to show S phase cell cycle arrest in
MCF-7 but not in MDA-MB-231 [7]. Finally, proteolytic cleavage
of the 116 kDa PARP to an 89 kDa fragment [8], was determined
by a Western blot analysis on cells treated with varying concentra-
tions of 1a and 5 (Fig. 4). The study showed a concentration-depen-
dent increase in the amount of the 89 kDa fragment.
All manipulations for chemical synthesis were carried out using
standard Schlenk techniques under an argon or nitrogen atmo-
sphere [11]. The osmium carbonyl clusters Os3(CO)10(L)2 (1,
L = NCCH3 (a) [10], PPh3 (b) [12], AsPh3 (c) [13], P(OMe)3 (f) [14],
and tBuNC (g) [15]), Os3(CO)11L (2, L = PPh3 (b) [16], AsPh3 (c)
[17], and SbPh3 (d) [18]) and Os6(CO)16(L)2 (L = CO (4) [19], and
NCCH3 (5) [9]) were prepared according to reported procedures.
1d and 1e were prepared from the reaction of SbPh3 and PCy3,
respectively, with 1a [10]. Os3(CO)12, 3, was purchased from Ox-
kem; all other chemicals were purchased from other commercial
sources and used as supplied.
Fig. 1. Comparison of cell viability of MCF-7 (shaded) and MDA-MB-231 (unshaded) after 24 h incubation with 1 and 10 lM concentrations of 1a, 4, 5, cisplatin (CisPt),
resveratrol (Res) and tamoxifen (Tamox).

836
K.V. Kong et al. / Journal of Organometallic Chemistry 694 (2009) 834–839
Fig. 2. Fluorescence image (100Â) of MDA-MB-231 cells stained with DAPI: (a) control, and after 24 h incubation with 10
lM solution of (b) 1a, (c) 4, and (d) 5. Some cells
showing abnormal chromatin structures are circled in red. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)
4.2. Cell culture and drug treatment
Proliferation Assay (Promega) was added to each well. This was
then left to incubate in a 37 °C incubator with 5% CO2 over 2 h.
The absorbance at 490 nm were then measured and the cell prolif-
eration relative to the control sample was calculated. Each sample
was analyzed in triplicates. The IC50 are absolute values obtained
from plots of percent viability against the logarithm of the dose of
compound added, and fitted with a sigmoidal model.
Experimental cultures of the cell lines MDA-MB-231, MCF-7,
MCF-10A, HT29, SW620 and Hepg2 were obtained from the Amer-
ican Type Culture Collection (ATCC) and cultured in tissue culture
dishes (Nunc Inc., Naperville, IL, USA). The cells were maintained in
Dulbecco’s modified Eagle’s medium (DMEM, Grand Island, NY,
USA) supplemented with 10% fetal bovine serum (FBS), 1% L-gluta-
mate (GIBCO Laboratories), and 1% penicillin–streptomycin (GIBCO
Laboratories) at 37 °C in 5% CO2 atmosphere. Cell cultures were
maintained in an antibiotic-free condition during cell growth and
experiments. Phosphate-buffered saline (PBS) was obtained from
1st BASE.
The osmium carbonyl clusters, cisplatin, tamoxifen and resvera-
trol were dissolved in dimethylsulfoxide (DMSO) with final con-
centration used for treatment being 0.1%. The cells were seeded
in growth medium at the same initial density, and allowed to ad-
here and grow for 24 h before treatment (80% confluence). They
were then washed once with serum-free DMEM and then serum-
starved for 6 h before treatment with the indicated concentrations
of compounds in DMEM. Control cells were treated with vehicle
(0.1% DMSO).
4.4. Morphological analysis
Morphological features of cells were assessed by scoring control
and treated cultures (24 h incubation with 10 lM solutions of
compounds) seeded in 6-well flat bottom plates. Cultures were
fixed in 70% cold ethanol and then stored overnight at À20 °C.
Samples were then examined to identify cells with an inverted Dia-
vert fluorescence microscope (Leica, Wetzlar, Germany).
4.5. Analysis of nuclear staining with DAPI
Nuclear fragmentation of cells was assessed by scoring control
and 24 h incubation with 10 lM of 1a, 4 and 5 treated cultures
on 6-well flat bottom plates. After treatment, the cells were fixed
with 3.7% formalin (RT, 1 h), washed with PBS, and stained with
4,6-diamidino-2-phenylindole (DAPI, 1 mg mLÀ1 in methanol,
5 min at 37 °C), washed, and then examined and photographed
using a Leica SP 5 microscope equipped with a UV light filter.
Apoptotic cells were defined on the basis of chromatin condensa-
tion and nuclear formation (bead-like formation) [20].
4.3. Proliferation assay
The cells were treated with the indicated concentrations of os-
mium carbonyl clusters in serum-free DMEM condition and incu-
bated for 24 h, after which, 20% of Cell Titer 96Ò Aqueous One Cell

K.V. Kong et al. / Journal of Organometallic Chemistry 694 (2009) 834–839
837
Fig. 3. Detection of early and late apoptotic MDA-MB-231 cells after staining with annexin V-FITC and PI. (a) control, and 24 h incubation with 10
(d) 4, and (e) 5.
l
M solution of (b) 3, (c) 1a,
4.6. Annexin V and PI staining for flow cytometry
4.7. Cell cycle by flow cytometry
The percentage of cells actively undergoing apoptosis was deter-
mined using annexin V-PE-based immunofluorescence, as described
previously [21]. Briefly, MDA-MB-231 cells were plated in 6-well flat
bottom plates to yield 80% confluence within 24 h. They were then
MDA-MB-231 cells were plated in 6-well flat bottom plates to
yield 80% confluence after 24 h. They were then incubated with
the compounds (10 lM) for 24 h, trypsinized, washed twice with
cold PBS, resuspended in 70% ethanol, and incubated at RT for
30 min. After this, they were spun down at 2000 rpm (400 g) for
5 min, washed once, and resuspended again with 1 Â PBS contain-
treated with 1a, 4, and 5 (10 lM) and incubated for 24 h, after which
the cells were harvested and then double-labelled with annexin V-
FITC and PI, as described by the manufacturer (BD Biosciences).
The cells were analyzed using a FACScan instrument equipped with
FACStation running Cell Quest software (Becton Dickinson, San Jose,
CA). All experiments were performed in duplicate.
ing 1% FBS (500 lL). This was then incubated (37 °C for 15 min)
after the addition of 1 Â RNase (20
iodide (PI) (50 L) was added, and the samples analyzed by flow
cytometry within 1 h.
lL), after which 1 Â propidium
l

838
K.V. Kong et al. / Journal of Organometallic Chemistry 694 (2009) 834–839
Appendix A. Supplementary material
Compound numbering scheme, table of IC50 values for all
compounds, graphs of cell viability versus concentration for all
compounds on all the cell lines tested, optical images of all the cell
lines after incubation with the compounds, and cellular DNA
content analysis of MDA-MB-231 cells treated with 1a and 3–5.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jorganchem.2008.07.039.
References
[1] (a) B.K. Keppler, Metal Complexes in Cancer Chemotherapy, Wiley-VCH,
Weinheim, 1993;
(b) N. Farrell, Uses of Inorganic Chemistry in Medicine, Royal Society of
Chemistry, Cambridge, UK, 1999;
(c) G. Jaouen, Bioorganometallics: Biomolecules, Labeling, Medicine, Wiley-
VCH, Weinheim, 2006;
(d) C.S. Allardyce, P.J. Dyson, in: G. Simonneaux (Ed.), Bioorganometallic
Chemistry: Topics in Organometallic Chemistry, vol. 17, Springer Verlag,
Berlin-Heidelberg, 2006, p. 177;
Fig. 4. Gel electrophoretogram of MDA-MB-231 cells treated for 24 h with various
concentrations of (a) 1a and (b) 5.
(e) P.J. Dyson, G. Sava, Dalton Trans. (2006) 1929.
[2] (a) E. Hillard, A. Vessieres, L. Thouin, G. Jaouen, C. Amatore, Angew. Chem., Int.
Ed. 45 (2006) 285;
4.8. Cellular protein extraction
(b) A. Vessieres, S. Top, W. Beck, E. Hillard, G. Jaouen, Dalton Trans. (2006) 529;
(c) D. Osella, H. Mahboobi, D. Colangelo, G. Cavigiolio, A. Vessieres, G. Jaouen,
Inorg. Chim. Acta 358 (2005) 1993;
(d) S. Top, A. Vessieres, G. Leclercq, J. Quivy, J. Tang, J. Vaissermann, M. Huche,
G. Jaouen, Chem. Eur. J. 9 (2003) 5223;
(e) K. Kowalski, A. Vessieres, S. Top, G. Jaouen, J. Zakrzewski, Tetrahedron Lett.
44 (2003) 2749;
(f) S. Top, A. Vessieres, C. Cabestaing, I. Laios, G. Leclercq, C. Provot, G. Jaouen, J.
Organomet. Chem. 637–639 (2001) 500.
Harvested cells were washed with cold 1 Â PBS (500
lL) and
immediately placed in a À20 °C bath for 5 min to allow rapid cool-
ing. The cells were then scraped in the presence of 200 lL (per
3 Â 106 cells) of lysis buffer (10 mM HEPES, pH 7.9) (Sigma),
10 mM KCl (Merck), 0.1 mM EDTA, 10% NP40 supplemented with
1 Â protease inhibitor (Pierce Biotechnology), and phosphatase
[3] (a) A. Bergamo, G. Sava, Dalton Trans. (2007) 1267;
(b) C.G. Hartinger, S. Zorbas-Seifried, M.A. Jakupec, B. Kynast, H. Zorbas, B.K.
Keppler, J. Inorg. Biochem. 100 (2006) 891;
inhibitors: 50 lM okadaic acid (Sigma) and 200 mM sodium vana-
date (Sigma). The mixture was vortexed for 1 min every half an
hour over a period of 2 h, and was kept on ice at all times. The ly-
sate was pre-cleared by centrifugation at 13000g at 4 °C, and sub-
sequently stored at À20 °C until evaluation by SDS–PAGE. The
supernatant was collected and protein concentration was deter-
mined by the method of Bradford using a protein assay kit (Bio-
Rad) [22].
(c) G. Zhao, H. Lin, Curr. Med. Chem.: Anti-Cancer Agents 5 (2005) 137;
(d) E. Alessio, G. Mestroni, A. Bergamo, G. Sava, Curr. Top. Med. Chem. 4 (2004)
1525;
(e) E. Alessio, G. Mestroni, A. Bergamo, G. Sava, Met. Ions Biol. Syst. 42 (2004)
323;
(f) M. Galanski, V.B. Arion, M.A. Jakupec, B.K. Keppler, Curr. Pharm. Design 9
(2003) 2078;
(g) L. Messori, A. Casini, D. Vullo, P. Orioli, Rec. Res. Dev. Inorg. Chem. 3 (2003)
19.
[4] (a) P.M. Abeysinghe, M.M. Harding, Dalton Trans. (2007) 3474;
(b) G.I. Pascu, A.C.G. Hotze, C. Sanchez-Cano, B.M. Kariuki, M.J. Hannon,
Angew. Chem., Int. Ed. 46 (2007) 4374;
4.9. SDS gel electrophoresis and Western immunoblot analysis
(c) B. Cebrian-Losantos, A.A. Krokhin, I.N. Stepanenko, R. Eichinger, M.A.
Jakupec, V.B. Arion, B.K. Keppler, Inorg. Chem. 46 (2007) 5023;
(d) A. Dorcier, W.H. Ang, S. Bolanõ, L. Gonsalvi, L. Juillerat-Jeannerat, G.
Laurenczy, M. Peruzzini, A.D. Phillips, F. Zanobini, P.J. Dyson, Organometallics
25 (2006) 4090.
Equal amounts of proteins of each sample were subjected to 8%
and 15% SDS–PAGE electrophoresis and then electrophoretically
transferred to nitrocellulose membranes (Bio-Rad Laboratories)
using a wet transfer apparatus (Bio-Rad Laboratories). Precision
Plus Protein Dual Color Standards (Bio-Rad laboratories) were used
as a molecular mass standard. After transfer, the membranes were
washes in distilled water to remove traces of transfer buffer, and
then air dried for a few hours prior to blotting. Membranes were
blotted with 1 Â Tris-buffered saline with Tween-20 (TBST) con-
taining 5% (w/v) non fat milk (10 mL) overnight at 4 °C. The mem-
branes were then washed with TBST before overnight exposure to
the appropriate primary antibodies at 4 °C. All primary antibodies
(cleaved PARP, b-tubulin and b-actin) were used at a final concen-
[5] (a) B. Therrien, W.H. Ang, F. Chérioux, L. Vieille-Petit, L. Juillerat-Jeanneret, G.
Süss-Fink, P.J. Dyson, J. Clust. Sci. 18 (2007) 741;
(b) D. Colangelo, A.L. Ghiglia, A.R. Ghezzi, M. Ravera, E. Rosenberg, F. Spada, D.
Osella, J. Inorg. Biochem. 99 (2005) 505;
(c) E. Rosenberg, F. Spada, K. Sugden, B. Martin, R. Gobetto, L. Milone, A. Viale,
J. Organomet. Chem. 689 (2004) 4729;
(d) E. Rosenberg, F. Spada, K. Sugden, B. Martin, L. Milone, R. Gobetto, A. Viale,
J. Fiedler, J. Organomet. Chem. 668 (2003) 51;
(e) C.S. Allardyce, P.J. Dyson, J. Clust. Sci. 12 (2001) 563.
[6] K.V. Kong, W.K. Leong, S.P. Ng, T.H. Nguyen, L.H.K. Lim, ChemMedChem,
published online.
[7] (a) A.K. Joe, H. Liu, M. Suzui, M.E. Vural, D. Xiao, I.B. Weinstein, Clin. Cancer
Res. 8 (2003) 893;
(b) E. Pozo-Guisado, A. Alvarez-Barrientos, S. Mulero-Navarro, B. Santiago-
Josefat, P.M. Fernandez-Salguero, Biochem. Pharmacol. 64 (2002) 1375.
[8] (a) S.H. Kaufmann, S. Desnoyers, Y. Ottaviano, N.E. Davidson, G.G. Poirier,
Cancer Res. 53 (1993) 3976;
(b) Y.A. Lazebnik, S.H. Kaufmann, S. Desnoyers, G.G. Poirier, W.C. Earnshaw,
Nature 371 (1994) 346.
[9] B.F.G. Johnson, R.A. Kamarudin, F.J. Lahoz, J. Lewis, P.R. Raithby, J. Chem. Soc.,
Dalton Trans. (1988) 1205.
[10] J.N. Nicholls, M.D. Vargas, Inorg. Synth. 28 (1990) 232.
[11] D.F. Shriver, M.A. Drezdzon, The Manipulation of Air-sensitive Compounds,
second ed., Wiley, New York, 1996.
tration of 1
l
g mLÀ1. All blots were incubated with mouse antibod-
ies b-actin to confirm equal protein loading. Washes with TBST
were carried out again before incubation (RT, 1 h) with the appro-
priate horseradish peroxidase-labelled secondary antibodies. After
secondary body incubation, membranes were washed again with
TBST. The antibody-reactive bands were revealed by chemilumi-
nescence-based detection using West Pico Substrate (Pierce).
Acknowledgements
[12] (a) C.W. Bradford, W. van Bronswijk, R.J.H. Clark, R.S. Nyholm, J. Chem. Soc. A
(1970) 2889;
This work was supported by the University Academic Research
Fund (Grant No. R-143-000-267-112) and a Research Scholarship
to K.V.K. under the MedChem Programme.
(b) M.I. Bruce, M.J. Liddell, C.A. Hughes, B.W. Skelton, A.H. White, J.
Organomet. Chem. 347 (1988) 157;
(c) C.W. Bradford, R.S. Nyholm, J. Chem. Soc., Chem. Commun. (1967) 384.

K.V. Kong et al. / Journal of Organometallic Chemistry 694 (2009) 834–839
[13] C.W. Bradford, R.S. Nyholm, J. Chem. Soc., Dalton Trans. (1973) 529. [18] W.K. Leong, G. Chen, Organometallics 20 (2001) 2280.
839
[14] M.I. Bruce, M.J. Liddell, C.A. Hughes, J.M. Patrick, B.W. Skelton, A.H. White, J.
Organomet. Chem. 347 (1988) 181.
[15] M.J. Mays, P.D. Gavens, J. Chem. Soc., Dalton Trans. (1980) 911.
[16] B.F.G. Johnson, J. Lewis, D.A. Pippard, J. Chem. Soc., Dalton Trans. (1981) 407.
[17] C.T. Tay, W.K. Leong, J. Organomet. Chem. 625 (2001) 231.
[19] C.R. Eady, B.F.G. Johnson, J. Lewis, J. Chem. Soc., Dalton Trans. (1975) 2606.
[20] C. Ferlini, R. Biselli, G. Scambia, A. Fattorossi, Cell Proliferat. 29 (1996) 427.
[21] I. Vermes, C. Haanen, H. Steffens-Nakken, C. Reutellingsperger, Immunol.
Meth. 184 (1995) 39.
[22] M. Bradford, Biochemistry 72 (1976) 248.