Synthesis, characterization, structure, molecular modeling studies and biological activity of sterically crowded Pt(II) complexes containing bis(imidazole) ligands

Article abstract of DOI:10.1016/j.jinorgbio.2010.12.002

The syntheses, structures and biological evaluation of a series of cisplatin-like complexes containing bis(imidazole) derivatives - the so-called Joseph ligands - are described. Their cytotoxicity is discussed in terms of their polar surface area, rate of aquation, and lipophilicity. The X-ray crystal structure of the platinum diiodido derivative of dimethyl 2-(di(1H-imidazol-2- yl)methyl)malonate) is reported and compared to those of related systems. Molecular modeling studies are focused on the hydrogen bonding properties of such systems, and their relevance to antitumor activity.

Full text of DOI:10.1016/j.jinorgbio.2010.12.002

Journal of Inorganic Biochemistry 105 (2011) 400–409
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Synthesis, characterization, structure, molecular modeling studies and biological
activity of sterically crowded Pt(II) complexes containing bis(imidazole) ligands
a
a
a
b
b
Mauro Ravera , Elisabetta Gabano , Manuele Sardi , Giuseppe Ermondi , Giulia Caron ,
c
c
d
d
a,
Michael J. McGlinchey , Helge Müller-Bunz , Elena Monti , Marzia B. Gariboldi , Domenico Osella ⁎
Dipartimento di Scienze dell'Ambiente e della Vita, Università del Piemonte Orientale “Amedeo Avogadro”, Viale Michel 11, 15100 Alessandria, Italy
CASSMedChem, Dipartimento di Scienza e Tecnologia del Farmaco, c/o Centro dell'Innovazione, Università di Torino, Via Quarello 11, 10135 Torino, Italy
School of Chemistry and Chemical Biology, University College Dublin, Belfield, Dublin 4, Ireland
a
b
c
d
Dipartimento di Biologia Strutturale e Funzionale, Università dell'Insubria, Via Alberto da Giussano 10, 21052 Busto Arsizio (VA), Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 July 2010
Received in revised form 1 December 2010
Accepted 3 December 2010
Available online 14 December 2010
The syntheses, structures and biological evaluation of a series of cisplatin-like complexes containing bis
(imidazole) derivatives – the so-called Joseph ligands – are described. Their cytotoxicity is discussed in terms
of their polar surface area, rate of aquation, and lipophilicity. The X-ray crystal structure of the platinum
diiodido derivative of dimethyl 2-(di(1H-imidazol-2-yl)methyl)malonate) is reported and compared to those
of related systems. Molecular modeling studies are focused on the hydrogen bonding properties of such
systems, and their relevance to antitumor activity.
Keywords:
©
2010 Elsevier Inc. All rights reserved.
Pt(II) derivatives
X-ray diffraction
Cytotoxicity
Lipophilicity
Molecular interaction fields
1
. Introduction
protein binding of cisplatin is massive in vivo (N90%), in particular
with Cys, Met and His residues [5]. It has been estimated that only
about 1–5% of the cisplatin administered manages to reach the DNA
[6]. Moreover, cisplatin is inactivated in cells by reduced glutathione
(GSH, present in millimolar concentrations in cells) and by metal-
lothioneins (MTs, small proteins having 20 Cys groups per molecule),
the main cellular detoxification agents [7–9]. A bulky carrier ligand
coordinated to a platinum core can reduce the level of undesired
substitution reaction in the square-planar Pt(II) complexes [10]. This
rationale prompted the design and the clinical tests of cis-amminedi-
chlorido(2-methylpyridine)platinum (II) (picoplatin), an active Pt(II)
antitumor drug [11] able to circumvent cellular resistance mechanism
by virtue of the presence of the sterically demanding 2-methylpyr-
idine (2-picoline). This ligand hinders the axial approach to the
platinum centre, resulting in less exhaustive side-reactions than
cisplatin [12–14]; nevertheless, picoplatin showed the same level of
DNA platination as cisplatin, albeit with a slower rate of adduct
formation. The slower rate of cross-link formation may have positive
implications in terms of DNA repair (a mechanism adverse to the
cytotoxic activity of alkylating drugs).
2 3 2
Cis-diamminedichloridoplatinum(II) (cis-[PtCl (NH ) ], cisplatin)
is a highly successful platinum-based antitumor agent that has been
used clinically for over 30 years [1]. Nevertheless, its spectrum of
activity is narrow, and its clinical use is limited by severe systemic
toxicity. For these reasons, the quest for new cisplatin analogues, by
modifying the ligands around the Pt centre, has been actively pursued
with the aim of discovering new antitumor drugs possessing three
fundamental advantages relative to cisplatin: lower toxicity, better
spectrum of activity, and higher water solubility [2,3].
The mechanism of action of cisplatin is reasonably well known. In
an oversimplified view, the high chloride concentration (≥100 mM)
in plasma prevents its hydrolysis, but as it enters the cell the lower
chloride concentration (ca. 5 mM) in cytosol prompts its aquation
reaction giving positively charged aqua species. These active com-
plexes are attracted to the negatively charged DNA backbone, then
complexation to N-7 atoms of two vicinal guanines follows leading to
intrastrand cross-linking, which deforms the DNA structure, prevent-
ing DNA replication and transcription.
However, platinum(II) species react indiscriminately with many
other nucleophiles, especially S-donor biomolecules [4]. Thus, the
Another example of sterically crowded Pt centres has been
reported by Reedijk et al. [15]. In particular, two complexes con-
taining a bis(imidazole) ligand showed interesting features (Fig. 1):
cis-[Pt(bmic)Cl
nol) was reported to have significant cytotoxicity in L1210 leukemia-
bearing mice, while the less sterically hindered cis-[Pt(bmi)Cl ] (2,
2
] (1, bmic=bis(1-methyl-1H-imidazol-2-yl)metha-
⁎
Corresponding author. Tel.: +39 0 131 360266; fax: +39 0 131 360250.
E-mail address: domenico.osella@mfn.unipmn.it (D. Osella).
2
0162-0134/$ – see front matter © 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2010.12.002

M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
401
molecular ion peaks [M+H]+ or sodiated [M+Na]+ peaks were
assigned on the basis of the m/z values and, in the case of Pt
complexes, of the simulated isotope distribution patterns. The
solution behavior was studied on a HPLC Waters 2695 separations
module with 2487 dual λ absorbance detector, using the silica-based
C18 Phenomenex column Gemini 250×3.00 mm 5 μm as the
−
1
stationary phase, flow 0.5 mL min , isocratic elution, UV–visible
Waters 2487 dual λ absorbance detector set at 210 and 230 nm, and
3
3
0:70 15 mM HCOOH/CH OH mixture as eluant. The same method has
2
Fig. 1. Sketch of picoplatin and cis-[Pt(L)Cl ] (1, L=bmic, bis(1-methyl-1H-imidazol-2-
yl)methanol; 2, bmi=1,1′-dimethyl-1H,1′H-2,2′-biimidazole) complexes cited in the
text.
been used to assess the purity of the compounds. On the basis of this
criteria, all compounds possess purity at N98%.
The Pt content in saturated water solutions of the 8, 10–15
complexes was assessed by means of a Spectro Genesis ICP-OES
spectrometer (Spectro Analytical Instruments, Kleve, Germany)
equipped with a crossflow nebulizer. In order to quantify the platinum
concentration the Pt 299.797 line was selected. A platinum standard
bmi=1,1′-dimethyl-1H,1′H-2,2′-biimidazole), was found to be inac-
tive (the dihedral angle between the planes of the two imidazole rings
was 30.6° and 3.1°, for 1 and 2, respectively).
These structural differences resulted in different reactivity: 1 is
less susceptible than 2 to deactivation by cellular thiols. The observed
reactivity of 1 may have application in cisplatin-resistant tumors, but,
in spite of the interesting results obtained, no further development of
this kind of complex has been reported to date.
−
1
stock solution of 1000 mg L
prepare calibration standards.
was diluted in 1.0% v/v nitric acid to
2.2. Synthesis and characterization of the complexes
2.2.1. Synthesis of cis-[Pt(Im) I ], 7 (Im=imidazole)
2 2
Starting from Reedijk's original class of complexes, herein we
describe the synthesis, characterization, structure, molecular model-
ing studies and biological evaluation of a series of cisplatin-like
complexes containing bis (imidazole) derivatives (the so-called
Joseph ligands) [16]. These ligands have been extensively used as
chelators for metal-ions, to study the binding sites and catalytic
activities of metalloenzymes (in particular copper and zinc) [17].
More recently, Joseph ligands have been employed as versatile linkers
to connect metal complexes to biomolecules [18]. In fact, the
methylene bridge between the two imidazoles is easily modifiable
and allows the incorporation of a range of different functionalities.
2 4 2
K [PtCl ] (0.500 g, 1.20 mmol) was dissolved in 7 mL of H O, in the
dark at room temperature. KI (1.190 g, 7.17 mmol) was then added
and the mixture was allowed to react for 30 min. After filtration
imidazole (0.270 g, 3.96 mmol) was added to the filtrate and the
resulting mixture reacted 24 h in the dark at room temperature to
form a yellow–brown precipitate that was isolated by centrifugation.
It was then washed with cold water, ethanol and diethyl ether and
dried under vacuum. Yield: 0.639 g, 91%. 1H NMR (DMF-d ): 13.11
7
(2H, m, H ), 8.28 (2H, s, H ), 7.35 (2H, s, H or H ), 7.10 (2H, s, H or
3
2
4
5
4
13
H ) ppm; C NMR (DMF-d ): 138.14 (C ), 128.43 (C or C ), 117.53
(C or C ) ppm;
with 1% DMF) m/z (%) [M+Na] calcd for C H I N PtNa : 606.84
6 8 2 4
5
7
2
4
5
195
4
5
Pt NMR (DMF-d ): −3109 ppm. ESI MS (CH OH
7
3
+ +
Several substitutions of the CH
2
moiety have been considered in order
to modulate the steric hindrance, water solubility, and to prepare
functionalized ligands for drug targeting and delivery [19].
(90.07), 607.84 (100.00), 608.84 (77.20), 609.84 (6.00), 610.84
(19.93), found: 606.23 (89.95), 607.25 (100.00), 608.32 (77.05),
6
09.50 (5.99), 610.33 (19.56).
2
. Experimental section
2
2.2.2. Synthesis of cis-[Pt(Im) (CBDC)], 8
2
.1. Chemistry
Ag SO (0.166 g, 0.535 mmol) was dissolved in 20 mL of water in the
2
4
dark at 50 °C with stirring. After 30 min, 7 (0.319 g, 0.545 mmol) was
added and the mixture allowed to react for 24 h. The mixture was then
filtered and barium 1,1-cyclobutanedicarboxylate (prepared from Ba
K
2
[PtCl
4
] (Johnson Matthey and Co.) and all other chemicals
(
Aldrich) were analytical grade and used without further purification.
Milli-Q grade water (18 MΩ cm) was used for the preparation of
aqueous solutions. The syntheses of ligands 3–6 were carried out as
described by Joseph et al. [16]. Ligand names were generated with
ChemDraw software, version 11. The NMR spectra were measured on
2 2
(OH) •8 H O, 0.168 g, 0.535 mmol, and 1,1-cyclobutanedicarboxylic
acid, 0.081 g, 0.562 mmol) was added to the filtrate and the mixture
allowed to react for 24 h in the dark at 50 °C. The mixture was filtered to
remove BaSO
and washed with CH
(s, 2H, H ), 7.90 (d, 2H, H
2.96 (t, 4H, H12), 1.93 (q, 2H, H13) ppm; C NMR (D
(C10), 137.38 (C ), 127.77 (C or C ), 118.07 (C or C ), 56.46 (C11), 31.24
(C12), 15.52 (C13) ppm; 1 Pt NMR (D
O): δ=−1653 ppm. ESI MS
4
, precipitated and the filtrate was evaporated to dryness
1
1
a JEOL Eclipse Plus spectrometer operating at 400 MHz ( H),
3
OH. Yield: 0.100 g, 39%. H NMR (D
), 7.22 (d, 2H, H or H ), 6.98 (d, 2H, H
O): δ=181.85
2
O): δ=13.64
1
00.5 MHz ( C) and 85.9 MHz ( Pt), respectively. 1H NMR and
1
3
195
3
2
4
5
4
or H ),
5
1
3
13
C NMR chemical shifts were reported in parts per million referenced
2
to residual solvent proton resonances. 1 Pt NMR spectra were
95
2
4
5
4
5
95
recorded using a solution of K
reference. The shift for K [PtCl
Na [PtCl ] (δ=0 ppm). For H and C NMR the resonances of the
2 4
[PtCl ] in aqueous KCl as the external
2
+
+
2
4
] was adjusted to −1628 ppm from
3 2 14 4 4
(CH OH with 1% H O) m/z (%) [M+Na] calcd for C12H N O PtNa :
1
13
2
6
495.05 (84.86), 496.05 (100.00), 497.06 (80.02), 498.06 (11.67), 499.06
(19.94), found: 495.15 (84.66), 496.16 (100.00), 497.35 (80.22), 498.36
(11.79), 499.89 (20.01).
imidazole atoms were named according to the standard organic
nomenclature (atom 1 is the coordinated nitrogen, then the atoms are
numbered toward the second nitrogen). The splitting of proton
1
resonances in the reported H NMR spectra is defined as s=singlet,
2.2.3. Synthesis of cis-[Pt(4)I
2
], 9 (4=di(1H-imidazol-2-yl)methane)
O, in the
d=doublet, t=triplet, q=quartet, and quint=quintet.
K
2
[PtCl ] (0.676 g, 1.63 mmol) was dissolved in 8 mL of H
4
2
Electrospray mass spectra (ESI MS) were obtained using a
Micromass ZMD mass spectrometer. Typically, a dilute solution of
dark at room temperature. KI (1.623 g, 9.78 mol) was then added and
the mixture was allowed to react for 30 min. After filtration 4 (0.338 g,
2.28 mol) was added to the filtrate and the resulting mixture allowed
to react 24 h in the dark at room temperature to form a yellow–brown
precipitate that was isolated by centrifugation. It was then washed
2 3
compound in H O/CH OH 20:80 was delivered directly to the
−
1
spectrometer source at 0.01 mL min , using a Hamilton microsyr-
inge controlled by a single-syringe infusion pump. The nebulizer tip
operated at 3000–3500 V and 150 °C, with nitrogen used both as a
drying and a nebulizing gas. The cone voltage was usually 30 V. Quasi-
with cold water, ethanol and diethyl ether and dried under vacuum.
1
Yield: 0.913 g, 85%. H NMR (acetone-d
6
): δ=12.04 (2H, s, H
3
), 7.83

402
M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
(
(
C
−
C
6
4H, t, J=1.83 Hz, H
4
or H
5
), 7.21 (4H, t, J=1.83 Hz, H
): δ=139.25 (C ), 131.25 (C
Pt NMR (acetone-d ): δ=
4
or H
5
), 4.42
(100.00), 567.33 (56.56), 568.99 (59.87), 569.32 (15.41), 570.52
(15.00).
1
3
2H, s, H
), 116.16 (C
3297 ppm. ESI MS (CH
6
) ppm; C NMR (acetone-d
or C ), 25.92 (C ) ppm;
OH with 1% acetone) m/z (%) [M+K] calcd
PtK : 634.81 (89.34), 635.81 (100.00), 636.81 (83.98),
37.81 (14.02), 638.81 (25.39), found: 634.56 (89.60), 635.75
6
2
5
or
195
4
4
5
6
6
+
3
2.2.7. Synthesis of cis-[Pt(5)(oxalato)], 13
+
11
H
I
20 2
N
4
3
AgNO (0.187 g, 1.10 mmol) was dissolved in 20 mL of water in the
dark, with stirring at room temperature After 30 min, 11 (0.400 g,
0.550 mmol) was added and the resulting mixture was allowed to
react for 24 h at 50 °C. After filtration, sodium oxalate (0.083 g,
(
100.00), 636.96 (84.05), 637.23 (14.25), 638.66 (25.30).
0
.62 mmol) was added to the filtrate and the mixture was left for 24 h
in the dark at 50 °C. The mixture was centrifuged to isolate the product
and washed with hot water and CH OH to remove impurities. Yield:
): δ=13.14 (2H, m, H ), 7.51 (2H, d,
), 7.20 (2H, d, J=1.46 Hz, H or H ), 5.62 (1H, d,
), 4.58 (1H, d, J=8.06 Hz, H ), 3.73 (6H, s, H
) ppm; 13
): δ=166.66 (C ), 140.15 (C ), 127.22 (C or C ), 119.09
), 56.15 (C ), 53.06 (C ), 34.92 (C
) ppm; 1 Pt NMR (DMF-
): δ=−1948 ppm. ESI MS (CH OH with 0.1% DMF) m/z (%) [M+
2
.2.4. Synthesis of cis-[Pt(4)(CBDC)], 10
Ag SO (0.153 g, 0.490 mmol) was dissolved in 20 mL of water in the
dark, with stirring at room temperature. After 30 min, 9 (0.298 g,
.453 mmol) was added and the resulting mixture allowed to react for
4 h at 50 °C. After filtration barium 1,1-cyclobutanedicarboxylate
2
4
3
1
0
.112 g, 36%. H NMR (DMF-d
J=1.46 Hz, H or H
J=8.06 Hz, H
NMR (DMF-d
or C
7
3
0
2
4
5
4
5
6
7
9
C
(
2 2
prepared from Ba(OH) •8 H O, 0.168 g, 0.535 mmol, and 1,1-cyclobu-
7
8
2
4
5
tanedicarboxylic acid, 0.081 g, 0.562 mmol) was added to the filtrate
and the mixture reacted for 24 h in the dark at 50 °C. The mixture was
95
(
C
4
5
7
9
6
d
7
3
filtered to remove BaSO
dryness and washed with CH
δ=13.64 (2H, s, H ), 7.21 (2H, d, J=1.83 Hz, H
J=1.83 Hz, H or H ), 4.55 (2H, s, H ), 2.90 (4H, t, J=7.87 Hz, H12), 1.92
2H, q, J=7.87 Hz, H13) ppm; 13C NMR (D
O): δ=180.35 (C10), 139.60
), 56.06 (C11), 36.40 (C ), 28.62
Pt NMR (D O): δ=−1901 ppm. ESI MS
4
, precipitated and the filtrate was evaporated to
+
+
1
Na] calcd for C14
5
found: 583.13 (83.45), 584.60 (100.00), 585.63 (81.58), 586.99
(14.37), 587.40 (20.91), 588.09 (3.49).
14 4 8
H N O PtNa : 583.03 (83.18), 584.036 (100.00),
3
OH. Yield: 0.128 g, 58%. H NMR (D
2
O):
85.036 (81.44), 586.04 (14.13), 587.04 (20.48), 588.042 (3.45),
3
4
or H ), 7.13 (2H, d,
5
4
5
6
(
(
(
(
2
C
C
2
), 126.42 (C
12), 15.16 (C13) ppm;
CH OH with 1% H O) m/z (%) [M+H] calcd for C13
85.07 (84.05), 486.07 (100.00), 487.07 (80.35), 488.08 (12.44), 489.07
4
or C
5
), 117.39 (C
4
or C
5
6
1
95
2.2.8. Synthesis of cis-[Pt(5)(CBDC)], 14
2
+
+
Ag
dark, with stirring at room temperature. After 30 min, 11 (0.360 g,
.495 mmol) was added and the resulting mixture was allowed to
react for 24 h at 45 °C. After filtration, barium 1,1-cyclobutanedicar-
boxylate (prepared from Ba(OH) •8 H O, 0.151 g, 0.480 mmol, and
,1-cyclobutanedicarboxylic acid, 0.079 g, 0.55 mmol) was added to
the filtrate and the mixture left for 24 h in the dark at 50 °C. The
mixture was filtered to remove BaSO , precipitated, and the filtrate
was evaporated to dryness and washed with CH OH. Yield: 0.125 g,
2 4
SO (0.150 g, 0.48 mmol) was dissolved in 20 mL of water in the
3
2
15 4 4
H N O Pt :
4
0
(
(
19.88), 490.08 (3.13), found: 485.40 (84.15), 486.32 (100.00), 487.56
80.35), 488.89 (12.36), 489.78 (19.80), 490.80 (3.30).
2
2
1
2
.2.5. Synthesis of cis-[Pt(5)I
yl)methyl)malonate)
2
[PtCl ] (0.500 g, 1.20 mmol) was dissolved in 7 mL of H O in the
2
], 11 (5=dimethyl 2-(di(1H-imidazol-2-
4
K
2
4
3
dark at room temperature with stirring. KI (1.19 g, 7.17 mmol) was
then added and the mixture was allowed to react for 30 min. After
filtration 5 (0.400 g, 1.44 mmol) was added to the filtrate and the
mixture was left for 24 h in the dark at room temperature to form a
yellow–brown precipitate. It was isolated by centrifugation and
1%. 1H NMR (CD
OD): δ=7.22 (4H, m, H and H ), 5.27 (1H, d,
J=2.93 Hz, H ), 4.66 (1H, d, J=2.93 Hz, H ), 3.67 (6H, s, H ), 2.93 (2H,
t, J=7.60 Hz, H12 or H12′), 2.83 (2H, t, J=7.60 Hz, H12 or H12′) 1.89 (2H,
4
3
4
5
6
7
9
13
quint, J=7.60 Hz, H13) ppm; C NMR (CD
66.43 (C ), 139.80 (C ), 126.82 (C or C ), 117.63 (C
), 55.85 (C11), 52.60 (C ), 36.45 (C ), 31.66 (C12 or C12′), 29.88 (C12
Pt NMR (CD OD): δ=−1886 ppm. ESI
3
OD): δ=180.51 (C10),
1
8
2
4
5
4
or C ), 56.14
5
washed with cold water, ethanol and diethyl ether and dried under
vacuum. Yield: 0.610 g, 70%. H NMR (acetone-d
(
C
7
9
6
1
6
): δ=12.23 (2H, m,
or H ), 5.53 (1H, d,
), 3.66 (6H, s, H ) ppm;
), 139.27 (C ), 131.53 (C or C ),
), 52.86 (C ), 37.01 (C
) ppm; 1 Pt NMR
): δ=−3313 ppm. ESI MS (CH OH with 1% acetone) m/z
195
or C12′), 15.15 (C13) ppm;
MS (CH OH) m/z (%) [M+Na] calcd for C18
80.18), 638.08 (100.00), 639.084 (82.95), 640.08 (17.95), 641.09
20.42), 642.089 (4.24), found: 637.28 (80.96), 638.53 (100.00),
3
H
3
), 7.79 (2H, s, H
J=10.38 Hz, H ), 5.36 (1H, d, J=10.38 Hz, H
C NMR (acetone-d ): δ=166.72 (C
17.02 (C or C ), 54.52 (C
acetone-d
4
or H
5
), 7.27 (2H, s, H
4
5
+
+
3
20 4 8
H N O PtNa : 637.08
6
7
9
(
(
6
13
6
8
2
4
95
5
1
4
5
7
9
6
39.05 (82.75), 640.40 (17.85), 641.63 (20.10), 642.52 (4.57).
(
(
(
(
(
6
3
+
+
14 2 4 4
%) [M+K] calcd for C12H I N O PtK : 764.84 (84.97), 765.84
2.2.9. Synthesis of cis-[Pt(5)(OH
2 4
)](SO ), 15
100.00), 766.84 (86.13), 767.84 (18.90), 768.84 (25.71), 769.84
3.78), found: 764.64 (84.15), 765.77 (100.00), 766.33 (86.19), 767.80
18.62), 768.44 (25.05), 769.33 (3.33).
Ag SO (0.126 g, 0.404 mmol) was dissolved in 15 mL of water in the
2
4
dark at 50 °C with stirring. After 30 min, 11 (0.300 g, 0.410 mmol) was
added and the mixture was allowed to react for 24 h. The mixture was
then filtered and the filtrate was evaporated to dryness, washed with
1
2
.2.6. Synthesis of cis-[Pt(5)Cl
2
], 12
cold water and precipitated from acetone. Yield: 0.080 g, 32%. H NMR
AgNO (0.166 g, 0.977 mmol) was dissolved in 20 mL of water in
3
(D
or H
(6H, s, H
or C ), 117.59 (C
NMR (D
2
O): δ=7.27 (2H, d, J=1.83 Hz, H
), 5.48 (1H, d, J=7.32 Hz, H ), 4.76 (1H, d, J=7.32 Hz, H
) ppm; 13C NMR (D
O): δ=166.40 (C ) 139.75 (C ), 126.78 (C
or C ), 56.12 (C ), 52.55 (C ), 36.40 (C
O/CH OH 1:1) m/z (%) [M−
Pt : 489.06 (84.83), 490.06
4
or H
5
), 6.90 (2H, d, J=1.83 Hz, H
4
the dark at 50 °C with stirring. After 30 min, 11 (0.360 g, 0.495 mmol)
was added and the mixture was left to react for 24 h. The mixture was
then filtered, and KCl (1.86 g, 25.0 mmol) was added to the filtrate
and the mixture left for 24 h in the dark at room temperature. The
precipitated product was isolated by centrifugation, dissolved in DMF,
5
6
7
), 3.94
9
2
8
2
4
), ppm; 1 Pt
95
5
4
5
7
9
6
2
O): δ=−1396 ppm. ESI MS (H
2
3
+
2− +
+
H
3
O –SO
4
]
15 4 5
calcd for C12H N O
precipitated with acetone, and dried under vacuum. Yield: 0.120 g,
(100.00), 491.07 (80.22), 492.07 (11.91), 493.07 (20.10), found:
489.60 (84.56), 490.45 (100.00), 491.89 (80.33), 492.23 (11.51),
493.44 (20.49).
1
4
5%. H NMR (DMF-d
7
): δ=13.10 (2H, m, H
), 7.40 (2H, d, J=1.46 Hz, H
), 5.07 (1H, d, J=10.10 Hz, H
ppm; 13C NMR (DMF-d
): δ=166.56 (C ), 139.62 (C
), 117.60 (C or C ), 55.46 (C ), 53.08 (C
NMR (DMF-d ): δ=−2216 ppm. ESI MS (CH
3
), 7.50 (2H, d,
or H ), 5.42 (1H,
), 3.71 (6H, s, H
J=1.46 Hz, H
d, J=10.10 Hz, H
4
or H
5
4
5
6
7
9
)
7
8
2
), 128.28 (C
4
or
2.3. Determination of log k′
0
95
C
5
4
5
7
9
), 36.55 (C
6
) ppm; 1 Pt
7
+
3
OH with 0.1% DMF) m/z
PtNa : 564.99 (64.07), 565.99
The log k′ values of the complexes were determined by means of
0
+
(
(
(
%) [M+Na] calcd for C12
H
14Cl
2
N
4
O
4
RP-HPLC [20] using a silica-based C18 gel Merck Lichrospher 100 RP-18
74.52), 566.99 (100.00), 567.99 (56.24), 568.99 (59.33), 569.99
15.34), 570.99 (15.83), found 564.79 (64.65), 565.75 (74.25), 566.45
250×4 mm (5 μm) column. The chromatographic conditions were [21]:
silica-based C18 gel as the stationary phase; flow rate=0.75 mL min−
1
;

M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
403
isocratic elution, UV–visible detector set at 210 nm; KI was the internal
reference to determine the column dead-time (t ); a mobile phase
containing different percentage of 15 mM HCOOH and CH OH (a
chromatogram for each complex with each different eluant composi-
tion, the CH OH fraction ranging from 20% to 50% has been performed).
Starting from these results, the extrapolation of the capacity factors to
glutathione levels [22]. The three cell lines were maintained in RPMI-
1640 (Sigma, Italy) supplemented with 10% fetal calf serum (Celbio,
Italy), 2 mM L-glutamine and 1% antibiotic mixture (streptomycin/
0
3
penicillin; Sigma, Italy) under standard culture conditions (95% O
and 5% CO at 37 °C in a humidified atmosphere). Cell survival
2
3
2
following exposure to platinum complexes was evaluated using the
MTT ([3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-
mide]) assay, based on the reduction of MTT by living cells [23].
0
3 0
% CH OH (log k′ ) for each compound has been calculated.
3
Briefly, 1×10 cells per well were plated onto 96-well sterile plates
2
.4. Single-crystal X-ray structure determination of 11
and allowed to attach and grow for 24 h. All Pt-complexes but 15 were
dissolved in DMSO to obtain 10 mM stock solutions that were diluted
with complete medium for cell treatment (15 has been dissolved
directly in the complete medium). The final range of Pt concentrations
was 10–500 μM; the DMSO concentration never exceeded 0.1%. Three
days later, MTT was added to each well (final concentration
0.4 mg mL ) and plates were incubated for 3 h at 37 °C. Cell viability
was determined by measuring the absorbance (λ=570 nm) in
individual wells, using the Universal Microplate Reader (Bio-Tek
Instruments, Winooski, VT, USA). The cytotoxic effects of Pt
complexes were quantitated by calculating the drug concentration
inhibiting tumor cell growth by 50% (IC50), based on non-linear
regression analysis of dose–response data, performed using the
Calcusyn software (Biosoft, Cambridge, UK).
Yellow–orange crystals of 11 were obtained from slow diffusion of
ether into an acetone solution at room temperature for about 2 days.
Suitably sized crystals were selected for X-ray single-crystal diffrac-
tion measurements. Crystallographic data were collected using a
Bruker SMART APEX CCD area detector diffractometer. A full sphere of
the reciprocal space was scanned by phi-omega scans. A semi-
empirical absorption correction was performed based on redundant
reflections. Table 1 contains a summary of crystal data, data collection
parameters and structural analysis. Crystallographic data for the
structural analysis have been deposited with the Cambridge Crystal-
lographic Data Centre, CCDC No. 781162. Copies of this information
may be obtained free of charge on application to CCDC, 12 Union Road,
Cambridge CB2 1EZ, UK (fax: 44 1223 336 033; e-mail: deposit@ccdc.
cam.ac.uk or www: http://www.ccdc.cam.ac.uk).
−
1
2.6. Computational methods
2
.5. Cytotoxicity tests
The 3D structures of investigated Pt complexes were obtained by
Three human tumor cell lines were used for the present study:
modification of the X-ray structure of 11 and imported into the
Spartan '08 molecular modeling software (Spartan '08; Wavefunction,
Inc.: Irvine, CA, 2009). Following the literature procedure [24], the
geometries of the complexes were fully optimized without symmetry
constraints using the semiempirical quantum-mechanical PM3
Hamiltonian as implemented in Spartan '08 and retained for surface
properties calculation (the total, CPK, and the polar, PSA, surface
area). The GRID/BIOCUBE4mf procedure [25] was used for the MIF
calculations: the program BIOCUBE4mf (v. 2.0 [26]) filters MIF points
A2780 (ovarian carcinoma) and HCT116 (colon adenocarcinoma),
were obtained from ATCC (American Type Culture Collection,
Manassas, VA, USA); A2780Cp8 cells (so-called because of their
ability to grow in medium containing 8 μM cisplatin), were developed
by chronic exposure of the parent cisplatin-sensitive line to increasing
concentrations of cisplatin, and were originally obtained from Dr. R.
Ozols (Fox Chase Cancer Center, Philadelphia, PA, USA). A2780Cp8
cells differ from the parental cell line in a number of features that
contribute to the resistant phenotype, including alterations in
mismatch and nucleotide excision DNA repair systems and increased
calculated with GRID (GRID v. 22b [27]) according to user-specified
−1
threshold values (here −3.0 kcal mol
both for N1 and O).
Table 1
Crystal data and structure refinement for complex 11.
3
. Results and discussion
Empirical formula
12 14 4 4 2
C H N O I Pt
727.16
100(2)
M (g mol−
1
)
3.1. Preliminary design of the complexes
Temperature (K)
Wavelength (Å)
Crystal system
Space group
a (Å)
0.71073
Monoclinic
P2 /c (#14)
In a previous report, the role of polar surface area (PSA, defined as
the surface sum over all polar atoms, including attached hydrogens) in
governing passive cellular uptake in tumor cell lines was suggested
[28]. Therefore, PSA for the whole series of complexes investigated
has been calculated (Table 2), as a simple, preliminary pharmacoki-
netic filtering procedure.
1
10.1571(7)
14.9467(11)
12.9129(9)
112.940(1)
1805.3(2)
4
b (Å)
c (Å)
β (°)
Volume (Å )
3
Z
3
Density (calculated, Mg/m )
2.675
11.215
Absorption coefficient (mm−1)
Table 2
Polar surface area (PSA) values of the complexes
investigated.
F(000)
Crystal size (mm)
1320
0.26×0.20×0.13
θ range for data collection (°)
Index ranges
2.18 to 29.91°.
−14≤h≤13, −20≤k≤20, −17≤l≤17
18,603
4846 [R(int)=0.0340]
99.9%
Compound
PSA (Ų)
Reflections collected
Independent reflections
Completeness to θ=28.00°
Absorption correction
Max. and min. transmission
Refinement method
Data / restraints / parameters
Cisplatin
Oxaliplatin
Carboplatin
1
69.99
98.50
123.20
30.52
82.77
58.93
67.89
69.47
120.45
116.71
132.89
Semi-empirical from equivalents
0.3234 and 0.1676
8
Full-matrix least-squares on F2
4846 / 0 / 210
10
11
12
13
14
15
2
Goodness-of-fit on F
1.045
Final R indices [IN2σ(I)]
R indices (all data)
Largest diff. peak and hole
R1=0.0310, wR2=0.0704
R1=0.0363, wR2=0.0724
1.979 and −0.913 e Å−
3

404
M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
Table 2 shows that the PSA values of the complexes investigated
range from 83 to 133 Å , and suggest that compounds 8–15 can enter
the diiodido synthons 7, 9, and 11. The diiodido complex was then
transformed into the corresponding cisplatin-like dichlorido deriva-
2
the cancer cells by passive mechanisms since poor cell penetration is
tives (by reaction with AgNO
like oxalato derivatives (by reaction with AgNO
oxalate), and carboplatin-like 1,1-cyclobutanedicarboxylato (CBDC)
derivatives (by reaction with Ag SO and then the barium salt of 1,1-
cyclobutanedicarboxylic acid) (Fig. 3). However, the dichlorido and
oxalato complexes obtained from 7 and 9 exhibited poor solubility in
both water (Table 4) and DMSO, the solvents used for biological tests.
Moreover, the complexes obtained after several attempts gave poor
elemental analysis and so only the CBDC derivatives 8 and 10 were
selected for further study.
3
and then excess chloride), oxaliplatin-
2
associated with PSA≥140 Å . As a comparison, the PSA was also
3
and then sodium
2
calculated for 1 (PSA=30.52 Å ). Moreover, according to a previous
report [29], active transport via an organic cationic transporter can be
operative in the uptake of the aquated, cationic species 15 [30].
In the already cited report [28] a further relationship between PSA
and IC50 was suggested (i.e., complexes having low PSA exhibited
higher cytotoxicity) for a homogeneous series of compounds (i.e.,
compounds having similar dicarboxylato leaving groups). However, in
the actual case, such a relationship may not be applied because of the
differences both in leaving and carrier groups, that is reflected in
different pharmacokinetics and pharmacodynamics.
2
4
The coordination of ligands 3 and 6 was attempted both by Dhara's
2
−
method and by direct reaction with [PtCl
4
]
[32,33]. However,
neither synthesis afforded acceptable yields; moreover, they formed
several by-products, and so study of these particular Joseph
derivatives was discontinued.
3
.2. Synthesis of ligands and complexes
Ligand 5 turned out to be the most promising in the series since its
diiodido synthon 11 could be obtained in good yield and high purity,
and the corresponding dichlorido, 12, oxalato, 13, and CBDC, 14,
complexes were soluble enough to allow their complete character-
ization and biological testing.
The synthesis of the Joseph ligands starts from the reactive
intermediate 2,2′-(nitromethylene)bis(1H-imidazole), 3 (Fig. 2),
where the –NO group is readily substituted by suitable nucleophiles,
2
or removed after a redox reaction.
Ligand 4 represents the simplest ligand (without substitution on
the methylene bridge), while 6 (bearing a carboxylic acid group
obtained by hydrolysis of the dimethyl malonate ester intermediate
To complete the series, the [Pt(Im) (CBDC)] (Im=imidazole)
2
derivative 8, and the diaqua derivative 15, (being the common
hydrolysis product of the 11–14 series), were also synthesized.
5
) was chosen because it may increase the water solubility of the final
Pt(II) complex. The syntheses of ligands 3–6 were carried out as
described by Joseph et al. [16].
3.3. X-ray structure of 11
The coordination of Pt(II) to two independent imidazole moieties
and the Joseph ligands 4 and 5 following Dhara's method [31] afforded
A number of the Joseph-type complexes have been characterized
by X-ray crystallography, and these studies have revealed several
interesting features. In dichloridobis(N-methylimidazolyl)platinum
(
II) the orientation of the heterocycles is not restricted and, in the
solid state, the N–Pt–N angle is 91.5°, the N•••N distance is 2.889 Å and
the dihedral angle between the imidazole ligands is 62.4° [34]. In
contrast, in 2, and in its monomethyl analogue 2b, the imidazoles
deviate from coplanarity only by 3°, and in 2b the N•••N distance is
markedly reduced to 2.573 Å. This parallels the behavior of platinum
complexes of 4,5-diazafluorenone in which the Pt–N distances are
also rather short and the N–Pt–N angle is reduced [35]. Moreover,
molecule 2b cocrystallized with tetraethylammonium chloride and
the NH moiety of the unsubstituted imidazolyl ligand interacts with
the neighbouring chloride [36]. When the chelating fragment is
enlarged to a six-membered ring, the molecule adopts a less strained,
shallow boat-like geometry such that the platinum and the hydroxyl-
bearing carbon are folded away from the central C N plane by 21°
2 2
and 26°, respectively. The dihedral angle between the imidazole
ligands is now 149.3°, the N•••N distance is 2.842 Å, and the N–Pt–N
angles are once again close to 90° [15]. In cis-[MCl (big)] (big=bis(1-
2
methylimidazol-2-yl)glyoxal), which possesses a seven-membered
chelate ring, the boat-like character is much more pronounced such
that the platinum and the diketone moiety are bent out of the central
C N plane by 43° and 36°, respectively. The dihedral angle between
2 2
the imidazole ligands is now 113.6°, the N•••N distance is 2.819 Å, and
the N–Pt–N angles are once again close to 90° [37].
The structure of 11 appears as Fig. 4 and reveals a gentle boat-like
geometry whereby the platinum and the malonate-bearing carbon are
bent out of the central C N plane by 33° and 36°, respectively. The
2 2
dihedral angle between the imidazole ligands is now 131°, the N•••N
distance is 2.772 Å, and the N–Pt–N and Cl–Pt–Cl angles are now 86.3°
and 91.8°, respectively. In the solid state, 11 exists as centrosymmetric
hydrogen bonded dimers mediated by interactions between an
imidazolyl N–H unit in one molecule and the double-bonded oxygen
of an ester in the nearest neighbour (Fig. 5). Further details of the
crystal structure investigation may be obtained from the Cambridge
Crystallographic Data Centre (see Experimental section).
Fig. 2. Sketch of the Joseph ligands 3–6 used in this work. a) Na
for 45 min., CO ; b) NaOMe, diethylmalonate, CH OH, refluxed for 2 h, CO
c) NaOH (80 °C for 5 min.), 37% HCl (80 °C for 15 min.), NaHCO
2
S
2
O
4, NaOH, steam bath
2
3
2
; and
3
.

M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
405
Fig. 3. Sketch of the complexes under study. The numbering scheme, used to assign the NMR signals, is reported (see Experimental section).
3
.4. Solution behavior
the Joseph ligand 5. RP-HPLC was used to monitor the solution behavior
of the complexes in a cell-free system under several different
experimental conditions. The final solution products were identified
by ESI-MS.
The behavior in solution was studied for 12, 13, and 14, i.e. the
cisplatin-, oxaliplatin-, and carboplatin-like complexes, respectively, of
In order to better simulate the conditions used in the cytotoxicity
tests, the solutions of complexes 12–14, freshly prepared in DMSO,
were diluted to a final 1% (12 and 14) or 2% (13) of organic co-solvent,
and maintained at 37 °C for three days. The used DMSO percentage is
Fig. 5. In the solid state, complex 11 exists as a dimer linked by a hydrogen bond
between an imidazolyl N–H unit in one molecule and the double-bonded oxygen of an
ester in the other.
Fig. 4. X-ray crystal structure of complex 11. Thermal ellipsoids are drawn at the 50%
probability level.

406
M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
slightly higher than in the cytotoxicity tests (0.1%) in order to reach
mM concentration for all the complexes and to have good signal-to-
noise ratio in the chromatograms. At regular time intervals, the
solutions were injected into the HPLC system, and the area of the
chromatographic peak was used to calculate the pseudo-first-order
reported by Dabrowiak et al. on carboplatin (t
½
in water=378 h; in
−
1
23.8 mM carbonate = 94.4 h; in 140 mM Cl = 250 h; in
100 mM PB = 448 h, temperature = 37 °C, pH= 7–7.5 [41]), 14
shows a lower half-life, with a further, marked decrease caused by
the presence of salts. It is noteworthy to observe in particular the
strong effect caused by the presence of carbonate on the stability of
14.
half-life, t
½
.
Table 3 summarizes the t
½
values for the reaction of the Pt
complexes in water and in the two most important buffers maintain-
ing physiologic pH values in blood, i.e. phosphate (PB, 1 mM, pH=7.4)
and carbonate buffers (23.8 mM, the total concentration of all
carbonate forms in equilibrium in blood [38], pH=7.4), at two
chloride concentrations, i.e. 5 and 100 mM NaCl, to simulate the intra-
and extracellular conditions, respectively.
The rate of aquation of Pt-drugs is the key step in their activation,
but the timescale is extremely wide, ranging from hours for cisplatin
to days for carboplatin.
As already reported, the formation of active Pt-species, by loss of
chloride groups in the case of cisplatin, or by ring opening in the case
of the dicarboxylato species, is required before the Pt centre can
interact with DNA. It has been suggested that nucleophiles, such as
endogenous bicarbonate and phosphate, displace the carboxylato
groups in the Pt complexes forming unstable products that are more
rapidly converted into aqua complexes [39,40]. Dabrowiak et al.
showed that carbonate anions at the concentration typical of culture
media displace one oxygen donor atom of the 1,1-cyclobutanedicar-
boxylate ligand in carboplatin to form a ring-opened product having a
monodentate linked carbonate [41,42]. Also, in the case of cisplatin, a
carbonate-containing intermediate is produced in culture media and
taken up by cells [43–45]. The use of an electrochemical biosensor
confirmed that carbonate buffer significantly enhances binding of
carboplatin and oxaliplatin to DNA [46].
Stock solutions in DMSO of 8, 10, 13 and 14, i.e. the dicarboxylato
complexes, are stable over 24 h at room temperature (evaluated by
means of multinuclear NMR, ESI-MS and conductivity measure-
ments), whereas the stock solutions of 11 (iodido complex) and 12
(chlorido complex) undergo ligand substitution. A 2 h-old solution of
12 contains about 30% of the solvated-species and that of 11 contains
about 60% of the solvated-species. Notably, a 2 h-old DMSO solution of
the prototypal drug cisplatin contains about 50% of DMSO mono-
solvated species. This corroborates the initial rationale that complexes
containing Joseph ligands are less prone to undergo substitution
reactions, where DMSO represents a model for S-donor biomolecules.
It is important to recall that the stock solutions are used as soon as
prepared; therefore, the percentage of solvated-species is lower. In
5 min (time to dissolve the complex and to dilute the solution) the
solvolysis of 12 is only 5% and that of 11 is 30%. When the DMSO
mother solutions are diluted with complete medium, the formation of
the aqua species immediately starts.
ESI-MS analyses after 3 days in water containing 1–2% DMSO
indicated that the three complexes underwent hydrolysis/solvolysis
to a different extent according to the corresponding kinetics. Complex
12 was totally converted into its mono-DMSO-derivative (one DMSO
molecule has replaced a chloride as a ligand), and about half of the
complexes 13 and 14 were transformed into both aquated and DMSO-
solvated species in which the dicarboxylato ligand (i.e. oxalato or
CBDC, respectively), has been completely lost. It is important to recall
that ESI-MS highlights the presence of the most readily ionizable
species, not necessarily the most abundant in solution.
The presence of chloride ions also increases the aquation rate of Pt-
dicarboxylato complexes by virtue of an exchange between the
dicarboxylate and chlorides, thus forming cisplatin that in turn
undergoes faster activation by aquation [21].
These considerations may apply also in the case of complexes 12–
3.5. Lipophilicity determination
1
4. The salts present in solution have a limited effect on the half-life of
the chlorido complex 12, because it is already rather short. The PB and
the carbonates decrease slightly the stability of the complex, while the
presence of chloride inhibits the aquation reaction (higher inhibition
The log Po/w (i.e. the partition coefficient or the ratio between the
concentrations of a compound dissolved in two immiscible solvents,
normally n-octanol and water) is widely used as an index of molecular
lipophilicity. In particular, the lipophilicity of a drug is related to its
ability to cross cell membranes by means of passive diffusion since n-
octanol is a rough model of the lipid bilayer of a cell membrane and
water represents the fluid in and out of cells.
Considering that RP-HPLC retention is due to partitioning between
the (polar) mobile and (non-polar) stationary phases, there is
a correlation between partition coefficient P and HPLC capacity factor
−
with higher [Cl ]), as expected. The aquation half-life of 12 is slightly
lower than that observed for cisplatin (t
½
=2.4 h in 0.1 M PB pH=7
[
47], but several kinetic studies under different experimental
conditions reported t
½
values ranging from 1.7 to 9.6 h [48]).
−
The presence of PB and Cl in solutions of 13 increases the rate of
ring opening. The carbonate has a stronger activation effect, even in
the presence of chlorides. The aquation reaction of 13 is considerably
more rapid than that observed for the analogous oxaliplatin complex
(k′=(t , where t
R
−t
0
)/t
0
0
is the retention time for an unretained
(
t
½
=160 h, at pH 7.4 and 37 °C) [49].
compound and t is the retention time of the analyte). Therefore, we
R
A similar trend is observed for 14, but in this case the effect of salts
is even more remarkable. In fact, when compared with the data
have performed a set of measurements of capacity factors for the Pt
(II) complexes under study and for cisplatin, carboplatin and
oxaliplatin. Pt(II) complexes are generally hydrophilic, as is usual
for coordination complexes containing polar coordination bonds.
According to the values reported in Table 3, the Joseph ligands impart
Table 3
a certain lipophilicity to the Pt(II) complexes, resulting in log k′
0
values higher (less negative, or even positive) than that of reference
compounds cisplatin, carboplatin and oxaliplatin. Since there is an
Pseudo-first-order half-life time (in h) for the reaction of Pt-complexes (1 mM) in
various media at 37 °C. The solutions contain 1% (12 and 14) or 2% (13) of DMSO.
medium
t
½
(h)
0
exponential relationship between log Po/w or log k′ and cellular
uptake [50,51], the experimental data support the initial hypothesis
based on the use of PSA, and so the Joseph Pt-complexes should have a
good cellular uptake.
1
2
13
14
H
1
1
1
2
O
1.26
1.00
1.65
3.20
0.83
1.33
1.62
12.4
11.2
10.3
191.5
12.7
37.5
26.6
mM phosphate buffer (pH=7.4, PB)
mM PB+5 mM NaCl
mM PB+100 mM NaCl
9.77
3.6. Cytotoxicity assay
23.8 mM carbonate (pH=7.4))
23.8 mM carbonate+5 mM NaCl
23.8 mM carbonate+100 mM NaCl
3.57
4.29
4.76
5.15
5.42
5.50
The cytotoxic properties of compounds 8 and 10–15 were
evaluated together with those of cisplatin, oxaliplatin, and

M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
407
Table 4
Water solubility, cytotoxicity and lipophilicity of the complexes 8 and 10–15.
Compound
Water solubility (mM)
IC50 (μM) A2780
IC50 (μM) A2780Cp8
Resistance factor RFa
IC50 (μM) HCT116
log k′
0
Cisplatin
Oxaliplatin
Carboplatin
8
2.64± 0.61
0.63± 0.11
19.95± 2.03
38.64± 5.33
85.56± 0.9
181.03± 47.6
81.94± 1.29
54.33± 4.8
72.45± 5.8
18.7± 4.22
14.99± 0.44
0.55± 0.11
78.32± 10.8
41.32± 6.8
143.94± 5.32
94.24± 6.9
72.09± 6.35
56.42± 3.37
80.19± 13.94
20.86± 2.49
5.68
0.87
3.93
1.07
1.68
0.54
0.88
1.04
1.11
1.11
10.98± 1.9
0.661± 0.035
50.9± 7.15
65.58± 10.14
N500
154.46± 1.21
136.48± 20.35
126.46± 6.51
123.32± 17.29
93.59± 3.1
−0.935
−0.120
−0.343
1.637
1.970
2.676
1.166
1.516
2.377
0.035
0.451± 0.010
0.361± 0.018
0.004± 0.001
0.159± 0.012
0.032± 0.008
0.325± 0.021
12.06± 0.11
10
11
12
13
14
15
a
RF=IC50 (A2780Cp8)/IC50 (A2780).
carboplatin. The cytotoxicity of all of the complexes under investiga-
tion was evaluated on the cisplatin-sensitive A2780 human ovarian
adenocarcinoma cell line, on its cisplatin-resistant subline A2780Cp8,
and on the cisplatin-insensitive HCT116 colon carcinoma cell line.
Cytotoxicity tests were performed using the MTT assay, following
continuous exposure of the cells to Pt derivatives for 3 days.
For complexes 8 and 10–14 DMSO was used for the preparation of
the mother solutions, which were diluted with complete medium as
soon as prepared so that the co-solvent (DMSO) percentage never
exceeded 0.1%. Only the water-soluble compound 15 was dissolved
directly in the complete medium.
is quite active (IC50 =19.95 μM) on the A2780 cell line, whereas
8 (IC50 =38.64 μM) is about one half as active, and 10 and 14 show
even lower cytotoxicity. In other words, the identity of the carrier
group is expected to be the main factor responsible for the different
IC50 values registered for carboplatin-like compounds, and this
difference can be traced back to the different stabilities and structures
2+
of the adducts formed between the [Pt(NH
)
3 2
]
fragment and DNA.
The X-ray structure of a double-stranded DNA decamer containing
2+
the [Pt(NH
3
)
2
]
fragment generated by cisplatin (PDB code: 1A2E
[52]) shows the presence of a hydrogen bond (2.9 Å) between the
ammine group N10 and the O2 of T7; in contrast, this type of
interaction does not exist for the N20 ammine group which is linked
to water molecules (Fig. 6).
This observation suggests that not only the different size of the
carrier group, but also the different hydrogen bonding pattern of the
nitrogen atoms, could explain the loss in activity registered for 8, 10
and 14 compared to carboplatin. Molecular Interaction Fields (MIFs)
were thus used to verify this hypothesis by a methodology already
used in previous studies [25].
Briefly, the interaction energy between a small molecule or
particle (called the probe) and a compound (here a Pt complex) is
calculated by positioning the probe in a 3D regular lattice of points
surrounding the Pt complex. The set of interaction energy values
generates a MIF typical of the compound and the probe, and thus
different MIFs can be obtained to explore different properties of the Pt
complex by using different probes. Two probes were used in this
study, N1 and O to describe the hydrogen bonding acceptor (HBA) and
donor (HBD) interaction profile, respectively, of the complexes.
Results are shown in Fig. 7A–D where a) only carrier groups are
considered (the leaving groups were not taken into account, e.g.
The results of the cytotoxicity tests are reported in Table 4.
All complexes show IC50 values higher than those of the reference
compounds cisplatin, oxaliplatin and carboplatin; in the case of
HCT116 the biological activity is even less satisfactory. However, they
show a better (lower) resistance factor RF (RF=IC50 (A2780Cp8) /
IC50 (A2780)): i.e. the IC50 values on cisplatin-resistant A2780Cp8 are
similar (in some cases better) than those on A2780 cell line.
Surprisingly, the highly water-soluble cation 15 shows the best
performance within the series. This behavior is particularly remarkable
because preformed cationic Pt(II) complexes are not commonly consid-
ered active species, owing to their inability to diffuse through the
(hydrophobic) cell membrane. However, this complex may be regarded
as the final, reactive product of hydrolysis of all the 12–14 complexes, able
to interact directly with DNA without further activation steps.
3
.7. Molecular modeling studies
Complexes 8, 10 and 14 share the same leaving group as
carboplatin (i.e. CBDC) but differ in their carrier group. Carboplatin
Fig. 6. The X-ray structure of a double-stranded DNA decamer containing the [Pt(NH
3
)
2
]2+ core (PDB code: 1A2E): the HB patterns of Pt–nitrogen atoms (N10 and N20) are discussed
in the text.

408
M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
Fig. 7. In red (grey) the HBD, in blue (black) the HBA properties extracted from MIFs (see text for details) for Pt(II) and the carrier group of: A: cisplatin; B: 8; C 10; D: 14.
Fig. 7A shows an analogue of cisplatin) and b) the grey point and
cubes are extracted from the MIF calculated using the O probe (and
thus show the HBD properties of the complex), whereas the black
ones are those from the MIF due to the N1 probe (and thus show HBA
properties of the complex). MIFs suggest that a) the directionality of
the HBD properties of nitrogen atoms should be taken into
consideration during the design of new Pt complexes and b) the
introduction of HBA properties in the carrier portion of the complex
seems to be detrimental for antitumoral activity. The importance of
hydrogen bonding for complexes 1 and 2 was highlighted by
lipophilicity of the overall complex, thereby facilitating cell penetra-
tion by passive diffusion through the cell membrane, and protects the
Pt centre from deactivation by non-nucleoside nucleophiles (as
indicated by the lower resistance factor), the overall biological
performance of such derivatives was somewhat lower than expected,
probably because of an unfavorable interaction with DNA, as
suggested by molecular modeling studies.
Within the series, cation 15 shows the highest potency on A2780
and its cisplatin-resistant analogue A2780Cp8 cell lines. This behavior
is repeated also on the HCT116 cell line, at least considering only the
series of complexes having the same carrier group (i.e., 11–15).
This activity may be related to a combination of different factors: i)
the presence of the Joseph ligand may balance the positive charge of
2
Bloemink et al. [15]. Molecular modeling of [Pt(bmic)(GMP-N7) ]
revealed a number of hydrogen bonds between the OH group of the
complex and the nucleobase resulting in a stabilization of the adduct.
This type of interaction, obviously absent in 2, induces significantly
different distortions to the DNA structure [53].
0
15 resulting in a final log k′ higher than those of the reference
compounds cisplatin, carboplatin and oxaliplatin; ii) Pt cationic
species may act as substrates for the organic cationic transporters
increasing their cellular accumulation bypassing the passive diffusion
through the (hydrophobic) cell membrane [30,54]; and iii) inside the
cell, the aquated cationic 15 is already in its activated form and may
readily react with its final target, the polyanionic DNA [55].
4
. Conclusion
In conclusion, although the presence of the bulky Joseph ligands
modulates the lability of the Pt–X (leaving) bonds, increases the

M. Ravera et al. / Journal of Inorganic Biochemistry 105 (2011) 400–409
409
[
[
[
[
13] J.F. Holford, F.I. Raynaud, B.A. Murrer, K. Grimaldi, J.A. Hartley, M.J. Abrams, L.R.
Abbreviations
Kelland, Anti-Cancer Drug Des. 13 (1998) 1–18.
14] F.I. Raynaud, F.E. Boxall, P.M. Goddard, M. Valenti, M. Jones, B.A. Murrer, M.J.
Abrams, L.R. Kelland, Clin. Cancer Res. 3 (1997) 2063–2074.
big
bis(1-methylimidazol-2-yl)glyoxal
bmi
1,1′-dimethyl-1H,1′H-2,2′-biimidazole
bis(1-methyl-1H-imidazol-2-yl)methanol)
1,1-cyclobutanedicarboxylato
total surface area
15] M.J. Bloemink, H. Engelking, S. Karentzopoulos, B. Krebs, J. Reedijk, Inorg. Chem.
bmic
CBDC
CPK
DMF
35 (1996) 619–627.
16] M. Joseph, T. Leigh, M.L. Swain, Synthesis (1977) 459–461.
[17] K. Várnagy, T. Csorba, D. Kiss, E. Garribba, G. Micera, D. Sanna, Eur. J. Inorg. Chem.
dimethylformammide
(2007) 4884–4896.
[
18] R. Hamzavi, T. Happ, K. Weitershaus, N. Metzler-Nolte, J. Organomet. Chem. 689
2004) 4745–4750.
19] E. Gabano, M. Ravera, D. Osella, Curr. Med. Chem. 16 (2009) 4544–4580.
ESI-MS electrospray ionization mass spectrometry
(
HBA
HBD
hydrogen bonding acceptor
hydrogen bonding donor
[
[20] W.J. Lambert, J. Chromatogr. A 656 (1993) 469–484.
[
[
[
[
21] O. Heudi, S. Mercier-Jobard, A. Cailleux, P. Allain, Biopharm. Drug Dispos. 20
1999) 107–116.
ICP-OES inductively coupled plasma–optical emission spectroscopy
(
Im
imidazole
22] B.C. Behrens, T.C. Hamilton, H. Masuda, K.R. Grotzinger, J. Whang-Peng, K.G. Louie,
T. Knutsen, W.M. McKoy, R.C. Young, R.F. Ozols, Cancer Res. 47 (1987) 414–418.
23] M.C. Alley, D.A. Scudiero, A. Monks, M.L. Hursey, M.J. Czerwinski, D.L. Fine, B.J.
Abbott, J.G. Mayo, R.H. Shoemaker, M.R. Boyd, Cancer Res. 48 (1988) 589–601.
24] E. Monti, M. Gariboldi, A. Maiocchi, E. Marengo, C. Cassino, E. Gabano, D. Osella,
J. Med. Chem. 48 (2005) 857–866.
k′
HPLC capacity factor
MIF
MTT
molecular interaction field
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium
bromide
[
[
25] G. Caron, A. Nurisso, G. Ermondi, Chem. Med. Chem. 4 (2009) 29–36.
26] http://www.casmedchem.unito.it/theory.html8 (last accessed November 21,
PB
phosphate buffer
P
PSA
RF
o/w
n-octanol / water partition coefficient
polar surface area
resistance factor (RF=IC50(A2780Cp8) / IC50(A2780))
2010).
[27] http://www.moldiscovery.com8 (last accessed November 21, 2010).
[
[
28] G. Caron, G. Ermondi, M.B. Gariboldi, E. Monti, E. Gabano, M. Ravera, D. Osella,
Chem. Med. Chem. 4 (2009) 1677–1685.
29] M.D. Hall, M. Okabe, D.-W. Shen, X.-J. Liang, M.M. Gottesman, Annu. Rev.
Pharmacol. Toxicol. 48 (2008) 495–535.
RP-HPLC reverse-phase high-performance liquid chromatography
t
t
0
HPLC retention time for an unretained compound
HPLC retention time of the analyte
[30] K.S. Lovejoy, R.C. Todd, S. Zhang, M.S. McCormick, J.A. D'Aquino, J.T. Reardon, A.
Sancar, K.M. Giacomini, S.J. Lippard, Proc. Natl. Acad. Sci. U. S. A. 105 (2008)
R
8902–8907.
[
[
[
[
31] S.C. Dhara, Indian J. Chem. 8 (1970) 193–194.
32] B. Miller, S. Wild, H. Zorbas, W. Beck, Inorg. Chim. Acta 290 (1999) 237–246.
33] Y. Kageyama, Y. Yamazaki, H. Okuno, J. Inorg. Biochem. 70 (1998) 25–32.
34] B.J. Graves, D.J. Hodgson, C.G. van Kralingen, J. Reedijk, Inorg. Chem. 17 (1978)
3007–3011.
Acknowledgments
Financial support for this work was from the Regione Piemonte
[
35] L. Maguire, C.M. Seward, S. Baljak, T. Reumann, Y. Ortin, E.V. Banide, K. Nikitin, H.
Müller–Bunz, M.J. McGlinchey, Eur. J. Inorg. Chem. (2009) 3250–3258.
36] S. Karentzopoulos, H. Engelking, B. Bremer, N. Paschke, B. Krebs, Acta Crystallogr.,
Sect. C: Cryst. Struct. Commun. 53 (1997) 172–174.
(
CIPE 2006 project-code A 370 and Ricerca Sanitaria Finalizzata 2009)
and the ATF Association (Alessandria, Italy). The research was carried
out within the framework of the European Cooperation COST D39
[
(
Metallo-Drug Design and Action) and Consorzio Interuniversitario di
[37] E. Bulak, M. Leboschka, B. Schwerderski, O. Sarper, T. Varnali, J. Fiedler, F. Lissner,
T. Schleid, W. Kaim, Inorg. Chem. 46 (2007) 5562–5566.
Ricerca in Chimica dei Metalli nei Sistemi Biologici (CIRCMSB, Bari,
Italy).
[
[
38] E. Gross, I. Kurtz, Am. J. Physiol. Renal Physiol. 283 (2002) F876–F887.
39] F. Luo, T.-Y. Yen, S.D. Wyrick, S.G. Chaney, J. Chromatogr., B: Anal. Technol.
Biomed. Life Sci. 724 (1999) 345–356.
[
[
[
[
[
40] S.K. Mauldin, M. Plescia, F.A. Richard, S.D. Wyrick, R.D. Voyksner, S.G. Chaney,
Biochem. Pharmacol. 37 (1988) 3321–3333.
41] A.J. Di Pasqua, J. Goodisman, D.J. Kerwood, B.B. Toms, R.L. Dubowy, J.C. Dabrowiak,
Chem. Res. Toxicol. 19 (2006) 139–149.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
doi:10.1016/j.jinorgbio.2010.12.002.
42] A.J. Di Pasqua, C. Centerwall, D.J. Kerwood, J.C. Dabrowiak, Inorg. Chem. 8 (2009)
1192–1197.
43] C.R. Centerwall, J. Goodisman, D.J. Kerwood, J.C. Dabrowiak, J. Am. Chem. Soc. 127
(2005) 12768–12769.
44] C.R. Centerwall, K.A. Tacka, D.J. Kerwood, J. Goodisman, B.B. Toms, R.L. Dubowy, J.C.
Dabrowiak, Mol. Pharmacol. 70 (2006) 348–355.
References
[
1] A. Bonetti, R. Leone, F.M. Muggia, S.B. Howell (Eds.), Platinum and Other Heavy
Metal Compounds in Cancer Chemotherapy (Molecular Mechanisms and Clinical
Applications), Humana Press / Springer Science, New York, 2009.
[45] A. Binter, J. Goodisman, J.C. Dabrowiak, J. Inorg. Biochem. 100 (2006) 1219–1224.
[46] M. Ravera, G. Bagni, M. Mascini, J.C. Dabrowiak, D. Osella, J. Inorg. Biochem. 101
(2007) 1023–1027.
[47] R.J. Knox, F. Friedlos, D.A. Lydall, J.J. Roberts, Cancer Res. 46 (1986) 1972–1979.
[48] J.-L. Jestin, B. Lambert, J.-C. Chottard, J. Biol. Inorg. Chem. 3 (1998) 515–519.
[49] E. Jerremalm, P. Videhult, G. Alvelius, W.J. Griffiths, T. Bergman, S. Eksborg, H.
Ehrsson, J. Pharm. Sci. 91 (2002) 2116–2121.
[50] A.R. Ghezzi, M. Aceto, C. Cassino, E. Gabano, D. Osella, J. Inorg. Biochem. 98 (2004)
73–78.
[51] J.A. Platts, D.E. Hibbs, T.W. Hambley, M.D. Hall, J. Med. Chem. 44 (2001) 472–474.
[52] F. Coste, J.M. Malinge, L. Serre, W. Shepard, M. Roth, M. Leng, C. Zelwer, Nucl. Acids
Res. 27 (1999) 1837–1846.
[
[
[
[
[
2] M. Galanski, Recent Pat. Anti-Cancer Drug Discovery 1 (2006) 285–295.
3] M. Galanski, M.A. Jakupec, B.K. Keppler, Curr. Med. Chem. 12 (2005) 2075–2094.
4] J. Reedijk, Chem. Rev. 99 (1999) 2499–2510.
5] E. Gabano, M. Ravera, D. Colangelo, D. Osella, Curr. Chem. Biol. 1 (2007) 278–289.
6] E. Gabano, C. Cassino, S. Bonetti, C. Prandi, D. Colangelo, A. Ghiglia, D. Osella, Org.
Biomol. Chem. 3 (2005) 3531–3539.
[
[
7] M.A. Fuertes, C. Alonso, J.M. Pèrez, Chem. Rev. 103 (2003) 645–662.
8] J. Reedijk, J.M. Teuben, in: B. Lippert (Ed.), Cisplatin. Chemistry and Biochemistry
of a Leading Anticancer Drug, Wiley-VCH, Weinheim, 1999, pp. 339–362.
9] M.G. Cherian, S.B. Howell, N. Imura, C.D. Klaassen, J. Koropatnick, J.S. Lazo, M.P.
Waalkes, Toxicol. Appl. Pharmacol. 126 (1994) 1–5.
[
[53] T.W. Hambley, A.R. Jones, Coord. Chem. Rev. 212 (2001) 35–59.
[54] K.S. Lovejoy, S.J. Lippard, Dalton Trans. (2009) 10651–10659.
[55] R. Gust, W. Beck, G. Jaouen, H. Schönenberger, Coord. Chem. Rev. 253 (2009)
2760–2779.
[10] F. Basolo, F.J. Chatt, H.B. Gray, R.G. Pearson, B.L.J. Shaw, J. Chem. Soc. (1961)
2207–2215.
[11] L. Kelland, Nat. Rev. Cancer 7 (2007) 573–584.
[12] Y. Chen, Z. Guo, S. Parsons, P.J. Sadler, Chem. Eur. J. 4 (1998) 672–676.