Combining aspects of the platinum anticancer drugs picoplatin and BBR3464 to synthesize a new family of sterically hindered dinuclear complexes; Their synthesis, binding kinetics and cytotoxicity

Article abstract of DOI:10.1039/c2dt31313h

Picoplatin is a sterically hindered mononuclear platinum drug undergoing clinical trials. The 2-methylpyridine ring provides steric hindrance to the drug, preventing attack from biological nucleophiles. BBR3464 is a trinuclear platinum drug which was recently in Phase II clinical trials, and is highly cytotoxic both in vitro and in vivo; it derives this activity through the flexible adducts it forms with DNA. In this work we sought to combine the properties of both drugs to synthesise a family of sterically hindered, dinuclear platinum complexes as potential anticancer agents. The bis-pyridyl-based ligands were synthesised through a peptide coupling reaction using diaminoalkanes of differing lengths (n = 2, 4 or 8) and 4-carboxypyridine or 2-methyl-4-carboxypyridine. The resultant dinuclear platinum complexes were synthesised by reacting two equivalents of transplatin or mono-aquated transplatin to each ligand, followed by purification by precipitation with acetone. The unprotected complexes react faster with 5′-guanosine monophosphate (drug to nucleotide ratio 1:2; t_1/2 = 2 h), glutathione (1:10, t_1/2 = 55 min) and human serum albumin (HSA) (1:1, t _1/2 = 24 h) compared to their hindered, protected equivalents (5′-guanosine monophosphate, t_1/2 = 3.5 h; glutathione = 1.7 h; HSA, t_1/2 = 110 h). The complexes were tested for in vitro cytotoxicity in the A2780 and A2780/cp70 ovarian cancer cell line. The unprotected platinum complexes were more cytotoxic than their protected derivatives, but none of the complexes were able to overcome resistance. The results provide important proof-of-concept for the development of a larger family of sterically hindered multinuclear-based platinum complexes.

Full text of DOI:10.1039/c2dt31313h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 11330
www.rsc.org/dalton
PAPER
Combining aspects of the platinum anticancer drugs picoplatin and BBR3464
to synthesize a new family of sterically hindered dinuclear complexes; their
synthesis, binding kinetics and cytotoxicity†
Sarah D. Brown,a Katherine D. Trotter,a Oliver B. Sutcliffe,b Jane A. Plumb,c Bruce Waddell,a
Naomi E. B. Briggsa and Nial J. Wheate*d
Received 12th June 2012, Accepted 12th July 2012
DOI: 10.1039/c2dt31313h
Picoplatin is a sterically hindered mononuclear platinum drug undergoing clinical trials. The
2-methylpyridine ring provides steric hindrance to the drug, preventing attack from biological
nucleophiles. BBR3464 is a trinuclear platinum drug which was recently in Phase II clinical trials, and is
highly cytotoxic both in vitro and in vivo; it derives this activity through the flexible adducts it forms with
DNA. In this work we sought to combine the properties of both drugs to synthesise a family of sterically
hindered, dinuclear platinum complexes as potential anticancer agents. The bis-pyridyl-based ligands
were synthesised through a peptide coupling reaction using diaminoalkanes of differing lengths (n = 2, 4
or 8) and 4-carboxypyridine or 2-methyl-4-carboxypyridine. The resultant dinuclear platinum complexes
were synthesised by reacting two equivalents of transplatin or mono-aquated transplatin to each ligand,
followed by purification by precipitation with acetone. The unprotected complexes react faster with
5′-guanosine monophosphate (drug to nucleotide ratio 1 : 2; t1/2 = 2 h), glutathione (1 : 10, t1/2 = 55 min)
and human serum albumin (HSA) (1 : 1, t1/2 = 24 h) compared to their hindered, protected equivalents
(5′-guanosine monophosphate, t1/2 = 3.5 h; glutathione = 1.7 h; HSA, t1/2 = 110 h). The complexes were
tested for in vitro cytotoxicity in the A2780 and A2780/cp70 ovarian cancer cell line. The unprotected
platinum complexes were more cytotoxic than their protected derivatives, but none of the complexes were
able to overcome resistance. The results provide important proof-of-concept for the development of a
larger family of sterically hindered multinuclear-based platinum complexes.
causes severe toxic side effects7 and the ability of some tumours
Introduction
to develop resistance to drug treatment.1,8,9
The use of small molecule drugs for the treatment of human and
animal cancers remains an important component of treating the
disease and is no better demonstrated than by the drug cisplatin
(Fig. 1), the parent compound of the platinum-based family of
anticancer drugs.1,2 The metal complex binds to its target, DNA,
via coordination bonds and distorts the normal three-dimensional
structure of the double helix.3,4 This subsequently induces cell
death through apoptosis.5,6 The success of cisplatin is limited,
however, by its poor selectivity for cancerous cells, which
Since the approval of cisplatin in 1979, thousands of ana-
logues have been examined in an effort to produce less toxic
derivatives, drugs that can be administered orally, or drugs that
are able to circumvent tumour resistance.2,10,11 Whilst 23 plati-
num-based drugs have undergone clinical trials, only two
aStrathclyde Institute of Pharmacy and Biomedical Sciences, University
of Strathclyde, 161 Cathedral Street, Glasgow, G4 0RE, UK
bDivision of Chemistry and Environmental Science, School of Science
and the Environment, Manchester Metropolitan University, Chester
Street, Manchester, M1 5GD, UK
cInstitute of Cancer Sciences, College of Medical, Veterinary and Life
Sciences, University of Glasgow, Cancer Research UK Beatson
Laboratories, Garscube Estate, Glasgow, G61 1BD, UK
dFaculty of Pharmacy, The University of Sydney, NSW 2006, Australia.
E-mail: nial.wheate@sydney.edu.au
Fig. 1 The chemical structures of (a) cisplatin, (b) picoplatin and (c)
BBR3464. Counter ions for BBR3464 are usually either 4(Cl−) or
4(NO3−).
†CCDC 870360. For crystallographic data in CIF or other electronic
format see DOI: 10.1039/c2dt31313h
11330 | Dalton Trans., 2012, 41, 11330–11339
This journal is © The Royal Society of Chemistry 2012

View Article Online
(carboplatin and oxaliplatin) have gained worldwide marketing
approval.1,2 One drug currently in Phase II and III clinical trials
is cis-amminedichlorido(2-methylpyridine)platinum(II), also known
as ZD0473 or picoplatin (see Fig. 1).1,12,13 This drug was
designed specifically to overcome glutathione-mediated drug
resistance. Picoplatin’s reactivity towards soft nucleophiles is
attributed to its stereochemistry;12 the 2-methylpyridine (pico-
line) ring sits almost perpendicular to the plane of the platinum
ligands, which places the methyl group above the platinum
atom.14 The location of the methyl in this position provides
direct steric protection of the platinum atom, effectively slowing
both drug aquation and binding by cysteine/methionine residues
in peptides and proteins.14–16
Another drug of particular interest is BBR3464, a trinuclear
complex (see Fig. 1), which displayed excellent activity in cis-
platin sensitive and resistant cancers in both in vitro and in vivo
models.10,17,18 Whilst cisplatin forms short range, rigid, intra-
strand cross-links, BBR3464 forms flexible, long and short
range, inter- and intrastrand cross-links upon binding,19 as well
as inducing B- to Z- and B- to A-type DNA conformational
changes.18,20 Preclinical trials revealed cytotoxicity at concen-
trations one thousand times lower than the concentration required
by cisplatin;18 however, in human Phase I trials, systemic toxi-
city was found to be so severe that the maximum tolerated dose
was just 0.9–1.1 mg m−2, compared to cisplatin with an adminis-
trable dose of 20–120 mg m−2.1 In Phase II trials, BBR3464 was
also found to be ineffective in all tumours examined.18,21,22
Model studies suggest that this might be due to a combination of
the low doses that can be administered and the high reactivity
and rapid degradation of the drug in vivo.23,24
We have therefore hypothesised that a new family of plati-
num-based drugs could be developed that incorporate the steric
protection of picoplatin with the flexible long range DNA adduct
formation of BBR3464 (Scheme 1). Such drugs may be highly
cytotoxic but, because they are less susceptible to degradation
and protein binding, may also have less severe systemic side-
effects with more drug reaching tumours intact. In this paper, we
describe the synthesis of our first proof-of-concept compounds
starting with a family of dinuclear platinum complexes with pyri-
dine- and 2-methylpyridine-based bridging linkers of different
lengths. These complexes were characterised by 1H NMR,
195Pt NMR, electrospray ionisation mass spectrometry (ESI-MS)
and elemental analysis. Their reaction kinetics using 5′-guano-
sine monophosphate, glutathione and human serum albumin
were examined to see the effectiveness of the bridging ligand in
providing steric protection to the complexes. Finally, the di-
nuclear complexes were tested for in vitro cytotoxicity via
growth inhibition assays using the human ovarian cancer cell
line A2780 and its cisplatin resistant sub-line A2780/cp70.
Scheme 1 The synthetic method of the bispyridyl ligands and their
subsequent dinuclear platinum complexes, showing the hydrogen label-
ling scheme used for the aromatic protons 1, n = 2, R = H; 2, n = 4, R =
H; 3, n = 8, R = H; 4, n = 2, R = CH3; 5, n = 4, R = CH3; 6, n = 8, R =
CH3; 7, n = 2, R = H; 8, n = 4, R = H; 9, n = 8, R = H; 10, n = 2, R =
CH3; 11, n = 4, R = CH3; 12, n = 8, R = CH3.
evaluated. Flexibility is also important so that the pyridyl groups
have free rotation thus allowing them to position themselves
with the 2-methyl groups over the platinum atom. In addition,
it was preferable to have two versions of the ligand; one with a
2-methyl group, and one without, so that the amount of steric
protection could be fully evaluated. We therefore chose ligands
1–6 and synthesised them from the peptide coupling of 4-car-
boxypyridine or 2-methyl-4-carboxypyridine to diaminoalkanes
of variable length (n = 2, 4 or 8; Scheme 1). The synthesis of the
unmethylated ligands has previously been reported,25 but their
2-methyl derivatives have not. The benefit of these ligands is
their ease of synthesis, as peptide coupling reactions are routine
to conduct and have high yields. More importantly, synthesis via
peptide coupling produces ligands with two amide groups. For
multinuclear platinum complexes it has been shown that hydro-
gen bonding and electrostatic binding between the bridging
ligand and the DNA helix is important in stabilising and
directing the type of DNA adducts formed.19,26–28 The ligands
synthesised in this work are therefore capable of acting as both
hydrogen bond acceptors/donors and when pre-associated in the
Results and discussion
Ligand synthesis
No suitable ligand was commercially available for use as the
bridging ligand of the sterically hindered platinum complexes.
Such a ligand requires two pyridyl groups linked by a flexible
chain. It was preferable that the chain of the ligand come in
various lengths so that its effect on cytotoxicity could be
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 11330–11339 | 11331

View Article Online
DNA minor groove, can potentially form hydrogen bonds to the
bases as well as ion–dipole interactions with the anionic DNA
backbone.
by powder X-ray diffraction and the XRPD pattern used to deter-
mine if there were differences to the structure determined by
single crystal X-ray diffraction. Pawley-type fit to the data (data
not shown), in which background, zero-point, peak shape and
unit cell parameters describing the profile were refined, yielded a
unit cell of dimensions (a, b, c = 26.7143, 5.6299, 10.0695 Å;
α, β, γ (°) = 90.0, 101.496, 90.0). Comparison with the single
crystal data25 (a, b, c = 26.519, 5.637, 10.084 Å; α, β, γ (°) = 90,
101.36, 90) confirms that the polycrystalline sample is consistent
with the single crystal X-ray structure. Background subtracted
XRPD data returns a χ2 of 1.207, and a flat difference plot indi-
cates that the sample was pure phase.
For one ligand, 5, single crystals suitable for X-ray analysis
were grown allowing its solid-state structure to be accurately
determined. The asymmetric unit of the sterically hindered linker
contains two crystallographically independent molecules and
two waters of hydration. One linker molecule is well ordered
(Fig. 4 and Table 1) whereas the butyldicarboxamide linker in
the second is disordered over two locations. A feature of the
solid-state packing is the 180° rotation about the carbonyl–
pyridyl sigma bond in the disordered moiety, causing the methyl
groups to orientate in the opposite direction. In comparison to
the unmethylated linker, which features a doubly penetrated
network, the methylated linker behaves differently due to the
loss of inherent symmetry whereby it forms discrete alternating
layers linked by a hydrogen bonding network via solvated water
molecules in the channels. This structural change arises due to
the pyridyl proton involved in the hydrogen bonding network
being replaced by a methyl group and is therefore no longer able
to participate in the same fashion.
All four ligands were characterised by 1H and 13C NMR,
ESI-MS and their purity (>99%) confirmed by elemental analy-
sis. The free unmethylated bispyridyl ligands (1–3) display two
1
doublets in the H NMR aromatic region and either two or four
resonances in the aliphatic region depending on the diamino-
alkane used. For the methylated bispyridyl ligands (4–6) the loss
of symmetry due to the 2-methyl group gives rise to two doublet
resonances (Scheme 1; Ha and Hb) and one singlet resonance
(Hc), as well as the expected aliphatic resonances, including a
large resonance for the methyl group at 1.29 or 1.56 ppm. For
both the unmethylated and methylated free ligands, formation of
the peptide bond is observed by the large shift of the amine reso-
nance from a broad peak around 3–4 ppm for the free diamino-
alkane to a broad triplet upon coupling between 8.65 to
8.80 ppm for the amide proton.
Ligand pseudo-polymorphism
The commercial development of any drug is not made solely on
the basis of its activity in vitro and in vivo, but will also include
elements of: the potential market and revenue, its ease/cost of
synthesis, its stability during manufacture and storage, its solu-
bility, drug polymorphism or pseudo-polymorphism, and the
drug’s ability to be made into appropriate dosage forms. Whilst
making the ligands used for the synthesis of the platinum com-
plexes it was found that all could be recrytallised from hot water
and formed either plate-like sheets or rectangular crystals and the
type of crystalline ligands formed was dependent on the rate of
crystallisation. Elemental analysis of the different crystalline
materials found that, where there was a difference, this was due
to the formation of hydrates. Powder X-ray diffraction was used
to show that these hydrates are pseudo-polymorphs which can
convert to the anhydrous form upon heating or standing for short
periods of time (for example see Fig. 2). As this could affect
their production during manufacture, we sought to examine this
in more detail.
Ligand 3 was analysed by differential scanning calorimetry
(DSC) and thermogravimetric analysis (Fig. 3). As can be seen,
the sample shows water loss in the first heat cycle, with onset
mass loss (TGA curve) occurring at the same temperature
(92 °C) as the first endothermic transition in the DSC curve. The
DSC curve stabilises before the next endothermic transition
which corresponds to a melting temperature of 131 °C. On the
following heat cycle an exothermic transition begins at 44 °C
which corresponds to the crystallisation temperature. In this
cycle, the TGA curve remains stable and no mass loss occurs,
with the only endothermic transition this time observed for the
sample melting, again at 131 °C. This result therefore shows that
the initial formation of hydrates is unlikely to be a problem
during manufacture if batches are properly dried before use in
the synthesis of the platinum complexes, as the ligands them-
selves are not hygroscopic once dried.
Metal complex synthesis
Formation of the dinuclear platinum complexes 7–12 was
achieved through the reaction of either transplatin, or mono-
aquated transplatin, with the ligands 1–6 in hot water for 6–12 h.
The coordination of the platinum to the methylated ligands is
considerably more difficult, presumably due to the steric protec-
tion of the pyridyl-nitrogen atom by the methyl group. As such,
the platination reactions were usually conducted for longer time
periods with mono-aquated transplatin and even then resulted in
lower yields compared to the unprotected complexes. The plati-
num complexes were purified by dissolving them in a minimum
amount of hot water, cooling and then fractionally precipitating
with acetone. Acetone is added slowly until the solution just
turns opaque; this first precipitate is the impurity. Addition of
more acetone (10–20 mL) makes the solution very cloudy and
filtration through a nylon filter (0.2 or 0.45 μm) removes what
appears to be impurity leaving the platinum complexes in solu-
tion; which are then recovered by freeze drying or rotary evapor-
ation. Sometimes the acetone precipitation purification step
needs to be repeated 2–3 times, especially for the protected com-
plexes, to get the metal complexes to better than 99% purity by
elemental analysis and 1H NMR.
It is important to note that both the purification of the free
ligands and the dinuclear platinum complexes is based on their
precipitation from the solvent (in most cases water). The yield
and the ease by which this is done is highly dependent on both
the concentration of the product and the quantity being
In some cases different crystal types were formed by a few
ligands that did not correspond to hydrates. To examine if this is
due to polymorphism, the powder form of ligand 2 was analysed
11332 | Dalton Trans., 2012, 41, 11330–11339
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 2 The X-ray powder diffraction pattern of 2 showing (bottom) the hydrated form of the ligand produced from its rapid crystallisation from
water, (middle) the anhydrous crystalline form of the ligand produced from its slow crystallisation from water and (top) the same sample from the
bottom spectrum upon standing for 2 days showing its conversion to the anhydrous state.
Fig. 3 Simultaneous and thermogravimetric analysis (green) and differential scanning calorimetry (red line) showing change in input energy (milli-
watts per milligram) as a function of temperature, giving (1) the start of the DSC trace, (2) water loss from the crystal, (3) ligand melting and (4)
sample recrystallisation, for ligand 3.
1
synthesised/purified. The larger the quantity/higher the concen-
tration the easier and more successful the synthesis becomes. At
low concentrations/quantities (e.g. <50 mg) both the free ligands
and the platinum complexes can be difficult to precipitate and
obtain as pure compounds.
The dinuclear platinum complexes were characterised by H,
195Pt NMR, ESI-MS and their purity (>98%) confirmed by
elemental analysis. Coordination of platinum to all four ligands
is observed by the downfield shifts of the aromatic resonances of
the bispyridyl ligands in the 1H NMR spectra. For the
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 11330–11339 | 11333

View Article Online
Fig. 4 The single crystal X-ray structure of the methylated bis-pyridyl
ligand 5, showing the cis- and trans-orientations of the carbonyl groups
to the 2-methyl groups in the two independent molecules of the unit
cell, with the slight disorder of two of the methylene carbon atoms
demonstrated on the bottom structure.
Fig. 5 A plot of the percentage free of 5′-guanosine monophosphate as
a function of time during its reaction with the unprotected dinuclear
platinum complex 8, solid line, and the sterically protected derivative 11,
dashed line.
Table 1 Single crystal X-ray diffraction data for the methylated ligand
5
aromatic amine trans to a chlorido ligand.29,30 No resonances in
the region of −2100 or −1800 ppm are observed, which indi-
cates there is no free transplatin, or free aquated-transplatin
remaining.30
Crystal data
Values
Empirical formula
Formula weight
Temperature
Wavelength
Crystal system, space group Triclinic, P1
C36H52N8O8
725.00
123(2) K
0.71073 Å
Platinum complex binding kinetics
Unit cell dimensions
a = 7.3486(3) Å, α = 66.270(5)°
b = 12.1524(7) Å, β = 77.808(4)°
c = 12.3530(6) Å, γ = 72.732(4)°
959.05(8) Å3
In designing our sterically hindered multinuclear platinum com-
plexes we sought to combine the steric protection of the
2-methyl pyridine ligand of picoplatin with the flexible DNA
adduct formation of multinuclear platinum drugs. Whilst the
incorporation of steric bulk is mainly to prevent degradation by
thiol peptides, i.e. glutathione,31 our sterically hindered com-
plexes should display slower reaction rates with all biological
nucleophiles, including DNA.15,16 As such, we conducted three
different kinetic experiments to examine the effect of the
2-methyl group on the reaction rates of the platinum complexes.
Volume
Z, Density (calculated)
Absorption coefficient
F (000)
Crystal size
Theta range for data
collection
Limiting indices
Reflections collected/unique 12 777/4185 [R(int) = 0.0160]
Completeness to theta =
27.00
1, 1.255 Mg m−3
0.090 mm−1
388
0.2 × 0.2 × 0.1 mm
2.92° to 27.00°
−9 ≤ h ≤ 9, −15 ≤ k ≤ 15, −15 ≤ l ≤ 15
99.9%
Absorption correction
Max and min transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2σ(I)]
R indices (all data)
Semi-empirical from equivalents
1.00000 and 0.81603
Full-matrix least-squares on F2
4185/79/273
5′-Guanosine monophosphate. As a first experiment, and for
simplicity, we chose 5′-guanosine monophosphate (5′-GMP) as a
model for DNA binding using H NMR. Upon coordination of
1
1.029
platinum at the N7 of guanosine, the nucleotide’s H8 resonance
shifts from ∼8.1 to ∼8.7 ppm. Both resonances are clearly resol-
vable in 1H NMR spectra and so the amount of each species can
be determined by simply integrating the relative size of each res-
onance. The unprotected dinuclear platinum complex 8 which
contains the butane-based linker (n = 4) and its protected deriva-
tive 11 were incubated with two equivalents of 5′-guanosine
monophosphate at 37 °C for time periods up to 7 h. The 1H
NMR spectra, which were then used to generate a plot of the
percent of free guanosine as a function of time (Fig. 5), demon-
strate that the protected platinum complex 11 reacts nearly two-
fold slower compared with the unprotected complex 8. Whilst
the sterically protected complex 11 has a reaction half-life of
∼3.5 h, the unprotected complex 8 has a half-life of just ∼2 h
and has fully reacted with the guanosine by 7 h.
R1 = 0.0450, wR2 = 0.1090
R1 = 0.0521, wR2 = 0.1153
0.301 and −0.443 e Å3
Largest diff. peak and hole
unprotected complexes the Ha resonance (see Scheme 1 for
proton labelling details) moves downfield by 0.25 ppm and the
Hb by 0.03 ppm. For the protected complexes, the Ha, Hb and Hc
resonances shift 0.5, 0.04 and 0.2 ppm, respectively. Broad plati-
num coupling is also seen at the base of the Ha resonance for all
six dinuclear platinum complexes and a broad peak representing
the –NH3 groups is sometimes observed around 4 ppm in freshly
prepared samples, before it disappears from solvent exchange.
The 195Pt NMR spectra also clearly indicate coordination of
the platinum to all six ligands. Both the unprotected and pro-
tected platinum complexes show a single resonance in the region
of ∼−2300 ppm which is in the region observed for similar plati-
num complexes that contain two trans-ammine ligands and an
Glutathione. Next, the rate of reaction of 8 and 11 with gluta-
1
thione was analysed by H NMR spectroscopy. Glutathione was
incubated with the metal complexes in a 10 : 1 ratio to mimic the
11334 | Dalton Trans., 2012, 41, 11330–11339
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 6 A plot of the percentage amount of unbound platinum complex
as a function of time with glutathione at 37 °C, showing the faster rate
of reaction of the unprotected dinuclear platinum complex 8, solid line,
compared with its equivalent sterically hindered, protected complex 11,
dashed line.
Fig. 7 A plot of the percentage amount of unbound platinum complex
as a function of time with human serum albumin at 37 °C, showing the
faster rate of reaction of the unprotected dinuclear platinum complex 8,
solid line, compared with its equivalent sterically hindered, protected
complex 11, dashed line.
high concentrations of the peptide inside the cell. As glutathione
has no aromatic resonances, the degradation of the metal com-
plexes can easily be followed through the appearance of
additional resonances in the aromatic 1H NMR region. The
metal complexes have either two or three resonances for the free
drug, but as the complexes react with glutathione additional res-
onances are observed between 8 and 9 ppm. As the reaction pro-
ceeds these peaks increase, then decrease, in size until only
resonances representing uncoordinated bridging ligands are
observed. Relative integration of the resonances allowed us to
plot the percentage of free metal complex as a function of time
and determine the reaction half-lives of complexes 8 and 11
(Fig. 6). As expected, the unprotected complex 8 reacted quick-
est with a half-life of 55 min. The sterically protected complex
11 in contrast reacts almost two-fold slower with a half-life of
1.7 h.
Table 2 The in vitro cytotoxicity of the protected and unprotected
dinuclear platinum complexes in the human ovarian carcinoma cell line
A2780 and its cisplatin-resistant derivative A2780/cp70, given as the
IC50; the concentration of metal complex required to inhibit cell growth
by 50%
IC50 (μM)
Methyl-
protection
Linker
length
Complex
A2780
A2780/cp70
Cisplatin
N/A
N/A
N/A
No
Yes
No
Yes
No
Yes
N/A
N/A
N/A
2
2
4
4
8
8
0.24
2.55
BBR346447
0.01 0.007
1.15 0.4
0.90 0.15
1.75 0.11
0.028 0.002
0.14 0.01
0.054 0.008
1.54 0.2
n.d.a
Picoplatin47
7
10
8
11
9
n.d.a
12.5 1.8
31.3 4.5
2.19 0.3
14.3 0.06
1.38 0.2
9.85 0.5
12
Human serum albumin. As well as binding to intracellular
nucleophiles, a large amount of administered platinum drug is
bound by proteins in blood serum. Human serum albumin
(HSA) is the most abundant protein in human blood serum and
has been shown to readily bind to platinum drugs.32–36 It was
therefore of interest to examine the reaction kinetics of the pro-
tected and unprotected dinuclear platinum drugs with HSA and
see if any protection against HSA binding was afforded to the
protected complex.
The unprotected dinuclear platinum complex 8 and its pro-
tected derivative 11 were incubated with a stoichiometric equi-
valent of HSA (mole of metal complex to mole of protein) in
phosphate-buffered saline (pH 7.4) at 37 °C for time periods up
to 168 h. At intervals, aliquots were taken from the reaction
mixture and immediately centrifuged on Viva-spin columns. All
HSA and HSA-bound platinum remained in the top of the
column, with all unbound platinum complex washed to the
bottom. The quantity of unbound platinum was then determined
using UV-visible spectrophotometry (248/249 nm) which then
allowed the percentage free platinum in solution to be deter-
mined at each time point. A plot of free complex as a function of
time demonstrates that the unprotected and protected complexes
display similar kinetics to those seen in the 5′-GMP and
a n.d. = not determined.
glutathione experiments as the sterically hindered complex 11
reacted significantly more slowly with HSA than the unprotected
complex 8. The unprotected platinum complex 8 has a HSA
reaction half-life of 24 h, whereas the protected complex 11
reacts more than 4.5-fold slower, with a half-life of approxi-
mately 110 h (Fig. 7).
In vitro cytotoxicity
The cytotoxicity of the six dinuclear platinum complexes was
determined using in vitro growth inhibition assays with the
human ovarian carcinoma cell line A2780 and its cisplatin resis-
tant sub-line A2780/cp70 (Table 2). The results suggest two
clear trends: first that there is an ideal linker length. The most
active dinuclear complexes (8 and 11) contain the n = 4 linker
followed by the complexes made by the n = 8 linker (9 and 12).
The complexes made using the shortest linker (7 and 10; n = 2)
were least active. This is consistent with structure–activity
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 11330–11339 | 11335

View Article Online
studies of other multinuclear platinum complexes from the
BBR3464 family which has found that the ideal methylene
chain length is six carbon units.18 Linkers shorter or longer than
this length were found to produce metal complexes that were
less cytotoxic. Both our n = 4 and n = 8 ligands are longer than
this ideal length, although the shorter length of the n = 4 ligand
means it is closer to the preferred end-to-end distance, and
hence, results in more cytotoxic metal complexes.
data for the development of drugs that are able to overcome drug
resistance, but are also resistant to deactivation and degradation.
In particular, we hypothesise that with suitable modification of
the bis-pyridine linkers, such that only one side of the ligand is
protected, we can synthesize the trinuclear platinum-based
complexes. Such trinuclear complexes would better mimic the
structure of BBR3464 while still having steric protection to the
terminal platinum groups. We hope to make a family of these tri-
nuclear complexes in the future.
Secondly, the protected complexes are less active than the
unprotected derivatives. This result is not unexpected given the
short incubation periods used in in vitro assays and the reduced
rate of reaction that the protected complexes will have in com-
parison to the unprotected derivatives; their slower reaction rates
mean the onset of apoptosis is later and hence the cells reproduce
more before the effects of the complexes are felt. This reduced
cytotoxicity is also observed for picoplatin when compared with
cisplatin, and has been demonstrated in a panel of 11 different
cell lines, where picoplatin displayed IC50 values that were
3.1-fold higher than those for cisplatin.12 This is due to picopla-
tin’s slower aquation and rate of reaction; about 2-fold slower
than cisplatin and similar to the slower rate of reaction observed
for the protected complexes in this work. This reduced cyto-
toxicity of picoplatin was not a concern as the reduced reactivity
also led to much reduced toxicity.37 This gives picoplatin a
higher therapeutic index compared with cisplatin; an effect we
are trying to mimic with these sterically hindered dinuclear com-
plexes. Unfortunately, none of the six platinum complexes are
able to overcome resistance, although this result is not necessa-
rily surprising. The dinuclear derivative of BBR3464, called
BBR3005, is also not able to overcome resistance in many
different matched cell lines including A2780 pairs.38–40
Experimental
Materials
All reagents were purchased from Sigma Aldrich with the excep-
tion of 1-propylphosphonic acid cyclic anhydride (T3P) and
anhydrous dimethylformamide which were purchased from Alfa
Aesar. Viva-spin columns with a molecular mass cut off of 5000
daltons were purchased from VWR.
Instrument methods
Elemental analyses were performed on a Perkin Elmer 240
Elemental Analyser. 1H and 13C NMR were recorded on a JEOL
400 MHz spectrometer, whereas 195Pt and kinetics studies were
recorded on a Bruker DPX 400 MHz spectrometer. 195Pt NMR
resonances were referenced externally to a sample of K2PtCl4 in
D2O at −1631 ppm.41 Positive ion ESI-MS were recorded on a
Finnigan LTQ Orbitrap. Samples were dissolved in H2O to a
concentration of 100 μM and injected into the instrument in 90%
CH3CN/10% H2O at a flow rate of 400 μL min−1. The capillary
temperature and voltage were 200 °C and 40 V, respectively,
with a source voltage of 4000 V. UV-vis experiments were con-
ducted on a Varian Cary 50 Bio spectrophotometer at a wave-
length of 249 nm for (8; ε = 13 665 M−1 cm−1) and 248 nm for
(11; ε = 16 182 M−1 cm−1).
Conclusions
In this work we have prepared a family of six dinuclear platinum
anticancer complexes based on the chemical structures of the
sterically hindered platinum drug picoplatin and the confor-
mationally flexible multinuclear platinum drugs, of which
BBR3464 is the lead agent. Two versions of each ligand were
made; with and without methylation at the 2-position of the pyri-
dine ring to examine the effect of steric hindrance on the sub-
sequent platinum complexes. The protected dinuclear platinum
complex 11, reacts significantly more slowly with 5′-guanosine
monophosphate, glutathione and human serum albumin com-
pared with the equivalent unprotected dinuclear platinum
complex 8. The reduced rate of reaction with HSA by the pro-
tected platinum complex may be both a function of slower reac-
tion kinetics, and also, different pre-association effects with the
protein.
The reduced rates of reaction of the protected complexes are
similar in magnitude to that observed for picoplatin compared
with cisplatin. In vitro cytotoxicity assays demonstrate that there
is an ideal linker length (around four methylene groups in the
bridging linker), and the protected complexes are less active than
the unprotected complexes. The latter result is consistent with
their slow reaction with nuclear DNA and is similar to the
reduced in vitro cytotoxicity of picoplatin compared with cis-
platin. Overall these results provide important proof-of-concept
Synthesis of pyridyl-based linkers
Diaminoalkane (1.13 mmol) and 4-carboxypyridine (4.19 mmol,
3 eq.) were dissolved in anhydrous dimethylformamide (2 mL).
To this, triethylamine (945 μL, 6.8 eq.) was added and the
reaction stirred at room temperature, for 3 h, under a nitrogen
atmosphere. The flask was then cooled to 0 °C and 1-propylphos-
phonic acid cyclic anhydride (1.5 mL, 2.35 eq.) was added
slowly with stirring. The flask was allowed to warm to room
temperature and stirred for 2 days. The reaction mixture was
diluted with water and extracted against diethyl ether (3 ×
100 mL) and the aqueous layer was separated (at this point the
linker made using 1,4-diaminobutane precipitates from solution).
Sodium hydrogen carbonate solution (1% w/v, 100 mL) was
added to the aqueous layer and stirred overnight at room temp-
erature to yield a fine, yellow precipitate which was collected by
filtration, washed with diethyl ether and air dried.
N,N′-(Ethane-1,2-diyl)diisonicotinamide (1). Yield: 87%.
Elemental analysis: C14H14N4O2 requires: C, 62.21; H, 5.22; N,
20.73%. Found: C, 61.88; H, 5.41; N, 20.74%. 1H NMR
(400 MHz, d6-DMSO, ppm) 8.91 (2H, t), 8.72 (4H, d), 7.73
11336 | Dalton Trans., 2012, 41, 11330–11339
This journal is © The Royal Society of Chemistry 2012

View Article Online
(4H, d), 3.47 (4H, m). 13C NMR (100 MHz, d6-DMSO, ppm)
165.5, 150.8, 142.0, 121.8, 39.9. ESI-MS [M − H]+ found:
271.1 m/z, requires 271.1.
118.9, 27.0, 24.6 ppm. ESI-MS [M + H]+ requires: 327.40 m/z,
found: 327.18.
N,N′-(Octane-1,8-diyl)bis(2-methylisonicotinamide) (6). On
addition of water (100 mL), the product precipitated out of solu-
tion. This was then collected by filtration, washed with diethyl
ether and allowed to air dry to produce a fine, silky powder.
Yield: 91%. Elemental Analysis: C22H30N4O2 requires: C,
69.08; H, 7.91; N, 14.65%. Found: C, 69.00; H, 7.88; N,
N,N′-(Butane-1,4-diyl)diisonicotinamide (2). Yield: 79%.
Elemental analysis: C16H18N4O2 requires: C, 64.41; H, 6.08;
N, 18.78%. Found: C, 64.27; H, 6.02; N, 18.65%. 1H NMR
(400 MHz, d6-DMSO, ppm): 8.77 (2H, t), 8.71 (4H, d), 7.73
(4H, d), 1.58 (4H, m). 13C NMR (100 MHz, d6-DMSO, ppm)
165.1, 150.8, 142.1, 121.8, 26.6 (2C). ESI-MS [M + H]+ found:
299.13 m/z, requires: 299.15; [M + Na]+ found: 321.07 m/z,
requires: 321.13.
1
14.31%. H NMR (400 MHz, d6-DMSO, ppm): 8.67 (2H, t),
8.55 (2H, d), 7.60 (2H, s), 7.52 (2H, d), 3.24 (4H, q), 1.51 (4H,
t), 1.29 (6H, s). 13C NMR (100 MHz, d6-DMSO, ppm) 165.3,
159.2, 150.0, 142.5, 121.1, 118.9, 29.4, 29.2, 27.0, 24.6.
ESI-MS [M + H]+ requires: 383.47 m/z, found: 383.25.
N,N′-(Octane-1,8-diyl)diisonicotinamide (3). Yield: 91%.
Elemental analysis: C20H26N4O2 requires: C, 67.77; H, 7.39; N,
15.81%; Found: C, 67.29; H, 7.42; N, 15.56%. 1H NMR
(400 MHz, d6-DMSO, ppm) 8.73 (2H, t), 8.70 (4H, d), 7.74
(4H, d), 3.25 (4H, q), 1.52 (4H, t), 1.30 (8H, s). 13C NMR
(100 MHz, d6-DMSO, ppm) 165.1, 150.7, 142.1, 121.7, 29.4,
29.3, 26.9 (2C). ESI-MS: [M + H]+ requires: 355.21 m/z, found
355.24.
Synthesis of unprotected platinum complexes
Ligand 1–3 (100 mg) and transplatin (2.1 eq.) were dissolved in
water and heated to 55–60 °C with stirring for 5–6 h during
which the solution turned colourless. Next, the solution was
stirred at room temperature for 1–2 days in the dark. Water was
then removed in vacuo. The remaining solid was dissolved in
the minimum amount of hot water and acetone was added until
the solution became cloudy. The precipitate was removed by
filtration, the filtrate collected and the solvent removed by rotary
evaporation. The process was repeated until the product, a pale
yellow powder, was pure by 1H NMR.
Synthesis of 2-methylpyridyl-based linkers
Diaminoalkane (1 mmol) and 2-methyl-4-carboxypyridine
(411 mg, 3 eq.) were dissolved in anhydrous dimethylformamide
(4 mL). To this, triethylamine (833 μL, 6 eq.) was added and the
reaction stirred at room temperature, for 3 h, under a positive
pressure of nitrogen. The flask was then cooled to 0 °C and
1-propylphosphonic acid cyclic anhydride added (1.4 mL, 2.2 eq.)
The flask was allowed to warm to room temperature and stirred
for 2 days.
[Bis-{trans-diammineplatinum(II)}{μ-N,N′-(ethane-1,2-diyl)bis-
(isonicotinamide)}] chloride hydrate, complex (7). Yield: 59%.
Elemental analysis: C14H26Cl4N8O2Pt2·H2O requires: C, 18.93;
H, 3.18; N, 12.61%. Found: C, 19.18; H, 3.28; N, 12.36%.
1H NMR (400 MHz, D2O, ppm) 8.97 (4H, d), 7.76 (4H, d),
3.65 (4H, s). 195Pt NMR (85 MHz, D2O) −2299 ppm. ESI-MS:
[M + H]+ requires: 800.49 m/z, found 801.13.
N,N′-(Ethane-1,2-diyl)bis(2-methylisonicotinamide) (4). The
reaction mixture was diluted with water and NaOH added. Upon
cooling a white precipitate formed, which was collected by
vacuum filtration, washed with water and air dried. Slow evapo-
ration of a saturated aqueous solution yielded crystalline
material. Yield: 96%. Elemental analysis: C16H18N4O2·3H2O
requires: C, 54.53; H, 6.86; N, 15.90%. Found: C, 54.33; H, 6.66;
[Bis-{trans-diammineplatinum(II)}{μ-N,N′-(butane-1,4-diyl)bis-
(isonicotinamide)}] chloride, complex (8). Yield: 65%. Elemental
analysis: C16H30Cl4N8O2Pt2 requires: C, 21.39; H, 3.37; N,
12.47%. Found C, 21.11; H, 3.21; N, 12.72%. 1H NMR
(400 MHz, D2O, ppm) 8.95 (4H, d), 7.76 (4H, d), 3.41 (4H, s),
1.66 (4H, s). 195Pt NMR (85 MHz, D2O) −2299 ppm. ESI-MS:
[M]2+ requires: 413.07 m/z, found: 413.53; [M]+ requires:
826.12 m/z, found: 826.00.
1
N, 15.76%. H NMR (400 MHz, d6-DMSO, ppm) 8.85 (2H, t),
8.56 (2H, d), 7.61 (1H, s), 7.53 (2H, d), 3.45 (4H, m) 2.52 (6H,
s). 13C NMR (100 MHz, d6-DMSO, ppm) 165.8, 159.2, 150.0,
142.4, 121.2, 118.9, 39.6, 24.63. ESI-MS [M + H]+ requires:
299.35 m/z, found: 299.1.
[Bis-{trans-diammineplatinum(II)}{μ-N,N′-(octane-1,8-diyl)bis-
(isonicotinamide)}] chloride trihydrate, complex (9). Yield:
63%. Elemental analysis: C20H38Cl4N8O2Pt2·3H2O requires: C,
23.82; H, 4.40; N, 11.11%. Found C, 23.95, H, 4.14, N, 11.20%.
1H NMR (400 MHz, D2O, ppm) 8.95 (4H, d), 7.76 (4H, d), 3.35
(4H, t), 1.56 (4H, t), 1.30 (8H, s). 195Pt NMR (85 MHz, D2O)
−2299 ppm. ESI-MS m/z [M]+ requires: 882.18 m/z, found:
882.00; [M]2+ requires: 441.09 m/z, found: 441.80.
N,N′-(Butane-1,4-diyl)bis(2-methylisonicotinamide) (5). The
reaction mixture was diluted with water and extracted against
diethyl ether (3 × 150 mL). The aqueous layer was separated and
20% NaOH solution (w/v, 50 mL) was added. The solution was
allowed to sit at room temperature for 3 days until a precipitate
appeared. This was then filtered and dried to yield a fine, yellow
powder. Slow evaporation of a saturated aqueous solution
yielded single crystals suitable for X-ray diffraction. Yield: 89%.
Elemental analysis: C18H22N4O2·2.5H2O requires: C, 58.21, H,
7.33 N, 15.08%. Found: C, 58.17; H, 7.26; N, 14.99%. 1H NMR
(400 MHz, d6-DMSO, ppm) 8.73 (2H, t), 8.55 (2H, d), 7.61
(2H, s), 7.53 (2H, d), 3.29 (4H, d), 1.56 (6H, s). 13C NMR
(100 MHz, d6-DMSO, ppm) 165.3, 159.2, 150.0, 142.5, 121.1,
Synthesis of protected platinum complexes
Ligands 4–6 (50 mg) and mono-aquated transplatin (2.1 eq.)
were dissolved in hot water (100 mL) and stirred in darkened
conditions at 55–60 °C for 3 h. To the reaction mixture, THF
was added (30–50 mL) and was allowed to stir with heating for
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 11330–11339 | 11337

View Article Online
a further 3 h, then at room temperature for a further 2 days.
Solvent was removed in vacuo. The remaining solid was dis-
solved in the minimum amount of hot water and acetone was
added until the solution became cloudy. The solid was removed
by filtration, the filtrate collected and the solvent removed by
rotary evaporation. The process was repeated until the product
was pure, giving a pale fluffy powder or glass-like solid.
calculated positions riding on the parent atoms and were refined
isotropically. The structure solution and refinement used the pro-
grams SHELX-86,42 SHELX-9742 and the graphical interface
WinGX.43 Absorption corrections were made with SADABS44
and figures generated using Mercury 2.4. Selected crystallo-
graphic parameters are in Table 1 and full data and esd’s are pro-
vided in the ESI.†
[Bis-{trans-diammineplatinum(II)}{μ-N,N′-(ethane-1,2-diyl)bis-
(2-methylisonicotinamide)}] chloride dihydrate, complex
(10). Yield: 30%. Elemental analysis: C16H30Cl4N8O2Pt2·2H2O
requires: C, 18.53; H, 3.44; N, 12.35%. Found: C, 18.01; H,
3.29; N, 12.46%. 1H NMR (400 MHz, D2O, ppm): 8.94 (2H, d),
7.69 (2H, s) 7.51 (2H, d), 3.63 (4H, s), 3.12 (6H, s). 195Pt NMR
(85 MHz, D2O) −2300 ppm. ESI-MS [M]2+ requires: 826.12
m/z, found: 826.05.
Powder X-ray diffraction
A sample was lightly ground in an agate mortar and pestle and
filled into a 0.7 mm borosilicate glass capillary. The sample was
mounted and aligned on a Bruker-AXS D8 Advance diffracto-
meter and data were collected at room temperature in the range
4–35° 2θ (2 kW; Cu Kα1, λ = 1.54056 Å; step size 0.015° 2θ
step time = 1 s). The data from the polycrystalline sample were
compared to the known single-crystal structure25 using the soft-
ware Dash 3.2 via a Pawley refinement45 as implemented in
TOPAS Academic v4.1.
[Bis-{trans-diammineplatinum(II)}{μ-N,N′-(butane-1,4-diyl)bis-
(2-methylisonicotinamide)}]
chloride
dihydrate,
complex
(11). Yield: 49%. Elemental analysis: C18H34Cl4N8O2Pt2·2H2O
requires: C, 22.46, H, 3.98, N, 11.64%. Found: C, 22.17, H,
1
3.95, N, 11.83%. H NMR (400 MHz, D2O, ppm) 9.07 (4H, d),
7.80 (2H, s, CH), 7.57 (4H, d), 3.42 (2H, t), 3.13 (6H, s), 1.68
(4H, d). 195Pt NMR (85 MHz, D2O) −2316 ppm. ESI-MS
[M + Na]+ requires: 877.14 m/z, found: 878.00.
Guanosine binding kinetics
Stock solutions of complexes 8 and 11 were prepared in D2O, as
was a solution of 5′-guanosine monophosphate. Complex 8 and
11 (1 mM) was added to an NMR tube followed by guanosine
monophosphate (2 eq.) giving a total volume of 600 μL. Upon
addition of guanosine monophosphate solution, a 1D water sup-
pression proton scan was run at 37 °C and repeated every 30 min
for 7.5 h.
[Bis-{trans-diammineplatinum(II)}{μ-N,N′-(octane-1,8-diyl)bis-
(2-methylisonicotinamide)}] chloride tetrahydrate, complex
(12). Yield: 55%. Elemental analysis: C22H42Cl4N8O2Pt2·4H2O
requires: C, 25.05; H, 4.78; N, 10.62%. Found C, 25.18; H,
1
4.71; N, 10.43%. H NMR (400 MHz, D2O, ppm) 9.07 (4H, d),
7.57 (4H, d), 3.36 (4H, t), 3.13 (6H, s), 1.59 (4H, t), 1.34 (8H,
s). 195Pt NMR (85 MHz, D2O) −2320 ppm. ESI-MS [M + H]+
requires 911.00 m/z, found 912.00.
Human serum albumin binding kinetics
Solutions containing stoichiometric equivalents of platinum
complexes (8 and 11; 2 μmol) and HSA (2 μmol) were stirred
and incubated at 37 °C in a phosphate buffer solution (pH 7.4)
containing 100 mM NaCl for a maximum of 7 days (total
sample volume 4.166 mL). Samples were taken periodically
(250–350 μL; 0, 1, 2, 4, 8, 24, 48, 72, 168 h) and transferred to
2 mL Viva-spin columns with a 5000 dalton molecular mass cut-
off filter. These were centrifuged at 8000 rpm for 25 min before
the supernatant fluid containing the unbound platinum complex
(200–250 μL) was collected and diluted to 750 μL with phos-
phate-buffered saline for analysis.
Differential scanning calorimetry/thermogravimetric analysis
Thermal behaviour, mass loss and melting point were monitored
using simultaneous thermal analysis (STA), comprising DSC
(differential scanning calorimetry) and TGA (thermal gravi-
metric analysis) analyses. This was carried out on a Netzsch
STA 449 C thermocouple, equipped with a Netzsch CC 200
liquid nitrogen supply system and a Netzsch CC 200 C control
unit. Analysis was performed using the Netzsch Proteus data
analysis package. About 3 mg of sample was placed in a sealed
crucible. This was then heated from 20–145 °C at 10 °C min−1
cooled to 20 °C at 20 °C min−1 and again heated to 145 °C at
10 °C min−1
,
.
Glutathione binding kinetics
Solutions containing platinum complex (8 and 11) in D2O were
added to a solution of glutathione in D2O to make a total volume
of 600 μL with a platinum complex concentration of 2 mM and
a glutathione concentration of 20 mM. The NMR tube was
placed in the NMR spectrometer at 37 °C and kinetics runs were
completed using 32 scans and a delay between each scan set of
30 min. The total amount of free and bound metal complex
was determined by relative integration of the free platinum
complex peaks compared to the new peaks which arise between
8 and 9 ppm.
Single crystal X-ray diffraction
Crystals were coated in mineral oil and mounted on glass fibres.
Data were collected at 123 K on a Bruker Nonius Apex II CCD
diffractometer using monochromated Mo Kα radiation. The
heavy atom positions were determined by Patterson methods and
the remaining atoms located in difference electron density maps.
Full-matrix least-squares refinement was based on F2, with all
non-hydrogen atoms anisotropic. While hydrogen atoms were
mostly observed in the difference maps, they were placed in
11338 | Dalton Trans., 2012, 41, 11330–11339
This journal is © The Royal Society of Chemistry 2012

View Article Online
10 J. B. Mangrum and N. P. Farrell, Chem. Commun., 2010, 46,
6640–6650.
In vitro growth assays
11 F. Arnesano and G. Natile, Coord. Chem. Rev., 2009, 253, 2070–2081.
12 M. Hay, Curr. Opin. Oncol., Endocr. Metab. Invest. Drugs, 1999, 1,
443–447.
13 J. Treat, J. Schiller, E. Quoix, A. Mauer, M. Edelman, M. Modiano,
P. Bonomi, R. Ramlau and E. Lemarie, Eur. J. Cancer, 2002, 38,
S13–S18.
14 Y. Chen, Z. Guo, S. Parsons and P. J. Sadler, Chem.–Eur. J., 1998, 4,
672–676.
15 J. Holford, F. Raynaud, B. A. Murrer, K. Grimaldi, J. A. Hartley,
M. Abrams and L. R. Kelland, Anti-Cancer Drug Des., 1998, 13, 1–18.
16 V. P. Munk, C. I. Diakos, L. T. Ellis, R. R. Fenton, B. A. Messerle and
T. W. Hambley, Inorg. Chem., 2003, 42, 3582–3590.
17 N. J. Wheate and J. G. Collins, Coord. Chem. Rev., 2003, 241, 133–145.
18 N. J. Wheate and J. G. Collins, Curr. Med. Chem.: Anti-Cancer Agents,
2005, 5, 267–279.
19 A. Hegmans, S. J. Berners-Price, M. S. Davies, D. S. Thomas,
A. S. Humphreys and N. Farrell, J. Am. Chem. Soc., 2004, 126,
2166–2180.
In vitro growth assays using the human ovarian cancer cell line
A2780 were conducted using published methods.46 Cells were
grown in RPMI media containing 10% foetal calf serum, penicil-
lin, streptomycin and L-glutamate in a 5% CO2 atmosphere. The
cells were trypsinised, counted and adjusted to 500–1000 cells
per well. Cisplatin or metal complex stock solutions were diluted
with RPMI to prepare a dilution series based on the total plati-
num concentration (0.1–100 μM). 10 μL aliquots were taken
from these solutions and added in triplicate to each well along
with RPMI only. The plates were further incubated for 24 h
before the medium and drug was removed and replaced by fresh
medium and incubated for a further 48 h. Plates were then fed
with fresh medium and MTT solution (50 μL), wrapped in tinfoil
and incubated for 4 h. The medium and MTT were removed
from the wells leaving the purple MTT-formazan crystals. These
were dissolved by addition of DMSO (200 μL) and glycine
buffer (25 μL) to each well. Plates were then read by their absor-
bance at 570 nm with the resulting dose–response curve display-
ing absorbance (y-axis) with respect to drug concentration in μM
(x-axis).
20 M. B. G. Kloster, J. C. Hannis, D. C. Muddiman and N. Farrell, Biochem-
istry, 1999, 38, 14731–14737.
21 D. I. Jodrell, T. R. J. Evans, W. Steward, D. Cameron, J. Prendiville,
C. Aschele, C. Noberasco, M. Lind, J. Carmichael, N. Dobbs, G. Camboni,
B. Gatti and F. De Braud, Eur. J. Cancer, 2004, 40, 1872–1877.
22 T. A. Hensing, N. H. Hanna, H. H. Gillenwater, M. G. Camboni,
C. Allievi and M. A. Socinski, Anti-Cancer Drugs, 2006, 17, 697–704.
23 M. E. Oehlsen, Y. Qu and N. Farrell, Inorg. Chem., 2003, 42,
5498–5506.
24 N. Summa, J. Maigut, R. Puchta and R. van Eldik, Inorg. Chem., 2007,
46, 2094–2104.
Abbreviations
25 M. Sarkar and K. Biradha, Cryst. Growth Des., 2006, 6, 202–208.
26 R. A. Ruhayel, J. J. Moniodis, X. Yang, J. Kasparkova, V. Brabec,
S. J. Berners-Price and N. P. Farrell, Chem.–Eur. J., 2009, 15, 9365–9374.
27 Y. Qu, M.-C. Tran and N. P. Farrell, J. Biol. Inorg. Chem., 2009, 14,
969–977.
28 S. Komeda, T. Moulaei, K. K. Woods, M. Chikuma, N. P. Farrell and
L. D. Williams, J. Am. Chem. Soc., 2006, 128, 16092–16103.
29 N. J. Wheate, C. Cullinane, L. K. Webster and J. G. Collins, Anti-Cancer
Drug Des., 2001, 16, 91–98.
5′-GMP 5′-guanosine monophosphate
DSC
differential scanning calorimetry
human serum albumin
thermogravimetric analysis
X-ray powder diffraction
HSA
TGA
XRPD
30 B. M. Still, P. G. Anil Kumar, J. R. Aldrich-Wright and W. S. Price,
Chem. Soc. Rev., 2007, 36, 665–686.
Acknowledgements
31 J. Reedijk, Chem. Rev., 1999, 99, 2499–2510.
32 E. I. Montero, B. T. Benedetti, J. B. Mangrum, M. J. Oehlsen, Y. Qu and
N. P. Farrell, Dalton Trans., 2007, 4938–4942.
33 B. P. Esposito and R. Najjar, Coord. Chem. Rev., 2002, 232, 137–149.
34 C. Møller, H. S. Tastesen, B. Gammelgaard, I. H. Lambert and S. Stürup,
Metallomics, 2010, 2, 811–818.
35 L. Trynda-Lemiesz, H. Kozlowski and B. K. Keppler, J. Inorg. Biochem.,
1999, 77, 141–146.
This worked was funded in part by a University of Strathclyde,
Research Development Fund Grant, No. RDF1539. We thank
Dr Alan Kennedy, Department of Pure and Applied Chemistry,
for his assistance in modelling the disorder in the single crystal
X-ray diffraction and Prof Alastair Florence for access to the
Glasgow Physical Organic Chemistry centre facilities.
36 A. R. Timerbaev, C. G. Hartinger, S. S. Aleksenko and B. K. Keppler,
Chem. Rev., 2006, 106, 2224–2248.
37 F. Raynaud, F. Boxall, P. Goddard, M. Valenti, M. Jones, B. Murrer,
M. Abrams and L. Kelland, Clin. Cancer Res., 1997, 3, 2063–2074.
38 Y. Qu, H. Rauter, A. P. S. Fontes, R. Bandarage, L. R. Kelland and
N. Farrell, J. Med. Chem., 2000, 43, 3189–3192.
39 J. D. Roberts, G. Beggiolin, C. Manzotti, L. Piazzoni and N. Farrell,
J. Inorg. Biochem., 1999, 77, 47–50.
References
1 N. J. Wheate, S. Walker, G. E. Craig and R. Oun, Dalton Trans., 2010,
39, 8113–8127.
2 L. Kelland, Nat. Rev. Cancer, 2007, 7, 573–584.
40 J. D. Roberts, J. Peroutka and N. Farrell, J. Inorg. Biochem., 1999, 77, 51–57.
41 O. McNoleg, Comput. Geosci., 1996, 22, 585–588.
42 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 2008,
A64, 112–122.
43 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837–838.
44 G. M. Sheldrick, SADABS (Version 2.03), 2002.
45 W. I. F. David, K. Shankland, J. van de Streek, E. Pidcock,
W. D. S. Motherwell and J. C. Cole, J. Appl. Crystallogr., 2006, 39, 910–915.
46 G. J. Kirkpatrick, J. A. Plumb, O. B. Sutcliffe, D. J. Flint and
N. J. Wheate, J. Inorg. Biochem., 2011, 105, 1115–1122.
47 S. Sharp, C. O’Neill, P. Rogers, F. Boxall and L. Kelland, Eur. J. Cancer,
2002, 38, 2309–2315.
3 D. R. Boer, A. Canals and M. Coll, Dalton Trans., 2009, 399–414.
4 J. Reedijk, Proc. Natl. Acad. Sci. U. S. A., 2003, 100, 3611–3616.
5 V. Cepeda, M. A. Fuertes, J. Castilla, A. Carlos, C. Quevedo and
J. M. Perez, Anti-Cancer Agents Med. Chem., 2007, 7, 3–18.
6 R. C. Todd and S. J. Lippard, Metallomics, 2009, 1, 280–291.
7 K. Barabas, R. Milner, D. Lurie and C. Adin, Vet. Comp. Oncol., 2008, 6,
1–18.
8 C. R. Centerwall, D. J. Kerwood, J. Goodisman, B. B. Toms and
J. C. Dabrowiak, J. Inorg. Biochem., 2008, 102, 1044–1049.
9 G. L. Beretta, V. Benedetti, G. Cossa, Y. G. A. Assaraf, E. Bram, L. Gatti,
E. Corna, N. Carenini, D. Colangelo, S. B. Howell, F. Zunino and
P. Perego, Biochem. Pharmacol., 2010, 79, 1108–1117.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 11330–11339 | 11339