Synthesis, characterization, and in vitro antiproliferative activity of [salophene]platinum(II) complexes

Article abstract of DOI:10.1002/cmdc.201402123

A series of methoxy- and fluorine-substituted [salophene]platinum(II) complexes (salophene=N,N′-bis(salicylidene)-1,2-phenylenediamine) were synthesized and characterized by ¹H NMR spectroscopy and mass spectrometry. The structure was confirmed

Full text of DOI:10.1002/cmdc.201402123

CHEMMEDCHEM
FULL PAPERS
DOI: 10.1002/cmdc.201402123
Synthesis, Characterization, and in vitro Antiproliferative
Activity of [Salophene]platinum(II) Complexes
Maria T. Proetto,[a] Wukun Liu,[a] Andrey Molchanov,[a] William S. Sheldrick,[b]
Adelheid Hagenbach,[c] Ulrich Abram,[c] and Ronald Gust*[a, d]
A series of methoxy- and fluorine-substituted [salophene]plati-
as 1-Pt (H), 3-Pt (4-OCH3), and 6-Pt (3-F) exhibited high antipro-
liferative effects (IC50 <2 mm). Drug lipophilicity and cellular ac-
cumulation were analyzed in an attempt to explain the differ-
ences in antitumor potency. To gain insight into their mode of
action, DNA interaction studies were performed, in which com-
pounds such as 1-Pt acted as non-DNA-binding platinum anti-
cancer drugs, as neither intercalation nor DNA covalent bind-
ing were detected.
num(II) complexes (salophene=N,N’-bis(salicylidene)-1,2-phe-
1
nylenediamine) were synthesized and characterized by H NMR
spectroscopy and mass spectrometry. The structure was con-
firmed on the example of [5-OCH3-salophene]platinum(II) (4-Pt)
by crystal structure analysis. The cytotoxicity of all complexes
against MCF-7 cells showed strong dependence on the kind of
substituent and its position on the salicylidene moiety, where-
Introduction
Nowadays cisplatin (cis-diamminedichloroplatinum(II)) plays
a major role in the chemotherapy of a variety of neoplasms.[1]
However, its use is limited by strong side effects and by the
fact that many tumors are either intrinsically resistant or ac-
quire resistance during therapy.[2] Efforts to overcome these
drawbacks induced medicinal chemists to modify the ligands
of cisplatin systematically in order to discover new metal-
based drugs.[3]
leaving group and observed unusual oxidation reactions of the
platinum(II) complexes, resulting in the formation of [o-benzo-
quinonediimine]platinum complexes and hydrogen peroxide.[4]
They proposed that the mutagenic properties of the di-
chloro[1,2-phenylenediamine]platinum(II) complexes[5] were as-
sociated with one or more of their oxidative decomposition
products. To prevent this reaction and to stabilize the com-
plexes, we decided to combine leaving and non-leaving
groups in one molecule. If 1,2-phenylenediamine reacts with
two equivalents of salicylaldehyde, a Schiff base (salophene) is
formed, which has already been used in chelating agents for
several metal ions in various oxidation states.
The toxicity of platinum complexes correlates with their re-
activity, meaning the formation of highly reactive aqua- or dia-
quaplatinum(II) species (the higher the reactivity, the higher
the toxic side effects). Therefore, the chloro ligands in cisplatin
were exchanged by a cyclobutane-1,1-dicarboxylato leaving
group in carboplatin. Another example is oxaliplatin, for which
oxalate was identified as an optimal leaving group.
Outstanding antitumor properties were obtained with
[salophene]iron(III) complexes, which were up to 50-fold more
active than cisplatin against breast cancer cell lines.[6] Investiga-
tions on the mode of action showed that they are able to gen-
erate reactive oxygen species (ROS) and to induce apoptosis,
but they demonstrated only marginal interactions with DNA.[7]
Furthermore, studies on lymphoma and leukemia cells showed
that cell death is induced via the intrinsic mitochondrial path-
way.[8] Further studies on methoxy-substituted [salophene]iron-
(III) complexes gave evidence that the position of methoxy
groups at the salicylidene moiety influences the cytotoxicity.[7]
Promising results were also obtained with a 3-methoxy-substi-
tuted [salophene]nickel(II) complex, which exhibited high cyto-
toxicity against MCF-7 breast cancer cells. In this case, it was
demonstrated that the extrinsic pathway of apoptosis was effi-
ciently triggered.[9]
In an attempt to optimize the pharmacological profile, Kçck-
erbauer and Bednarski used 1,2-phenylenediamine as a non-
[a] Dr. M. T. Proetto, Dr. W. Liu, Dr. A. Molchanov, Prof. Dr. R. Gust
Institute of Pharmacy, Freie Universitꢀt Berlin
Kçnigin-Luise-Straße 2+4, 14195 Berlin (Germany)
[b] Prof. Dr. W. S. Sheldrick
Lehrstuhl fꢁr Analytische Chemie, Ruhr-Universitꢀt Bochum
Universitꢀtsstraße 150, 44780 Bochum (Germany)
[c] Dr. A. Hagenbach, Prof. Dr. U. Abram
Institute of Chemistry and Biochemistry, Freie Universitꢀt Berlin
Fabeckstraße 34-36, 14195 Berlin (Germany)
[d] Prof. Dr. R. Gust
Studies concerning [salophene]platinum(II) complexes were
recently reported by Wu et al.[10] They investigated the inhibi-
tion of the human oncogene c-myc, the overexpression of
which is linked to cellular proliferation in malignant tumors, by
stabilizing the G-quadruplex DNA structure.[11] Thus, a platinu-
m(II) Schiff base complex bearing an amine side chain was sat-
Department of Pharmaceutical Chemistry, Institute of Pharmacy
Center for Molecular Biosciences Innsbruck, University of Innsbruck
CCB-Centrum for Chemistry and Biomedicine
Innrain 80/82, 6020 Innsbruck (Austria)
E-mail: Ronald.Gust@uibk.ac.at
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cmdc.201402123.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1176

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
isfactorily identified after a lead optimization process as
a potent and specific c-myc transcription inhibitor.
It is well known that the phenolic protons in N,N’-bis(salicyli-
dene)-1,2-diaminoethanes form intramolecular hydrogen
bonds to the imine nitrogen (OꢀH···N=C).[15] If substituents in-
fluence the electron density at the OH group, the hydrogen
bonds will have different strength, resulting in a characteristic
In analogy to our recently reported [diarylsalene]platinum(II)
complexes,[12] a series of methoxy- and fluorine-substituted sal-
ophene ligands containing the chelating N2O2 site[6,7,13] were
synthesized and coordinated to platinum(II). The influence of
the substitution pattern on tumor cell growth was evaluated
in cell culture experiments, whereby the relationship between
lipophilicity and cellular uptake with antiproliferative effects
was also investigated. Furthermore, initial insight into their
mode of action was gained, excluding DNA as the main target.
1
shift of the OH signal in the H NMR spectra. Moreover, the res-
onance of the OH group is shifted downfield with respect to
the signal involved in regular hydrogen bonds, characteristic
for resonance-assisted hydrogen bonds (RAHBs).[16] The reso-
nance signals assigned to the phenylene and azomethine pro-
tons (NCH) appeared in similar regions, irrespective of the sub-
stitution pattern at the salicylidene moiety.
The coordination of the salophene ligands to platinum(II),
and the formation of square-planar chelate complexes led to
characteristic changes in the 1H NMR spectra (Figure 1). The
resonance signal of the phenolic OH groups completely disap-
peared, indicating the bonding of the oxygen atom to plati-
num (CꢀOꢀPt). Moreover, H’3, H’6 (Dd~1 ppm) and the azome-
thine protons (Dd~0.5 ppm) were characteristically downfield-
shifted, while the signals of H’4, H’5, and the salicylidene pro-
tons remained nearly unchanged after coordination of the sal-
ophene to platinum(II). The formation of complexes was fur-
ther supported by positive-mode ESI high-resolution mass
spectrometry, which documented base peaks corresponding to
the formulae of the [M+Na]+ or [M+K]+ fragments.
Results and Discussion
Synthesis and characterization
The salophene ligands were synthesized according to a previ-
ously published method by reaction of 1,2-phenylenediamine
with the respective salicylaldehyde.[6] Subsequent treatment
with K2PtCl4 according to Tong et al.[14] yielded the [salophene]-
platinum(II) complexes (Scheme 1).
To gain insight into the coordination chemistry and structur-
al parameters of the [salophene]platinum(II) complexes, 4-Pt
was crystallized by slow evaporation of a concentrated metha-
nol/dichloromethane solution and characterized by X-ray dif-
fraction. Its crystal structure is shown in Figure 2 together with
the numbering scheme for the atoms. A summary of crystal
data, experimental details, and refinement methods is given in
Table 1.
The Schiff base ligand 4 is fourfold coordinated in a square-
planar geometry at the platinum(II) center by the azomethine
nitrogen atoms and the oxygen atoms of the hydroxy groups
with bond angles (N1-Pt1-N2, N1-Pt1-O1, N2-Pt1-O2, and O1-
Pt1-O2) in the range of 82.9–96.18. The respective average Ptꢀ
N and PtꢀO distances of 1.972 and 2.013 ꢁ are similar to those
reported for 1-Pt (1.970 and 1.994 ꢁ)[17] and those of platinum
complexes bearing tetradentate bis(salicylidene)-1,2-diamino-
ethane ligands[18] ([N,N’-bis(salicylidene)-1,2-diaminoethane]pla-
tinum(II): 1.944 and 2.004 ꢁ; [N,N’-bis(salicylidene)-1,1,2,2-tetra-
methyl-1,2-diaminoethane]platinum(II): 1.948 and 1.987 ꢁ). The
spectroscopic data as well as the X-ray structural analysis clear-
ly demonstrate a tetradentate bonding of these types of li-
gands. This finding contradicts that of Gaballa et al., who pre-
viously reported for 1-Pt and similar Schiff base containing
platinum(II) complexes only a coordination of the ligand by
the phenolic oxygen atoms.[19]
Scheme 1. Synthetic pathway to the [salophene]platinum(II) complexes and
their substitution pattern.
The new compounds were fully characterized by 1H NMR
spectroscopy. The resonance signals of the phenolic hydroxy
groups of the ligands were influenced by the position of the
substituents on the salicylidene moiety. The OH signal of the
unsubstituted ligand 1 is located at ~12.9 ppm. F or OCH3 sub-
stituents caused a diamagnetic shift of this signal if they are at
the para position (4 and 8, ~12.3–12.6 ppm). At the meta posi-
tion (3, 5, 7 and 9, ~13.5–13.8 ppm) a downfield shift was ob-
served. The OH signals of the ortho-substituted compounds 2
and 6 resemble that of 1.
The crystal packing of 4-Pt consists of head-to-tail arranged
pairs (Figure 3A), as similarly observed for [salophene]plati-
num(II) complexes reported in the literature.[17] The complexes
are arranged in a zigzag chain formed from intermolecular p–p
stacking interactions, not only between salicylidene, but also
between salicylidene and phenylene moieties (Figure 3B). Fi-
nally, all platinum complexes are sufficiently stable in solid
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1177

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Figure 1. 1H NMR spectra of 3, 3-Pt, 4, and 4-Pt in [D6]DMSO.
Determination of clogP
The [salophene]platinum(II) complexes were studied for lipo-
philicity to establish a possible relationship with biological ef-
fects, e.g., cellular accumulation and cytotoxicity. Lipophilicity
is generally expressed by the octanol/water partition coeffi-
cient (logP), which gives hints on the solubility of compounds
in membranes.[20] In this study, an HPLC method was used
which allows estimation of clogP values and thus a comparison
in this series of complexes. Although this HPLC correlation
method (see Experimental Section for details) is not always reli-
able at estimating logP values, it was demonstrated that the
[salophene]platinum(II) complexes possess high clogP values
in the range from 3.79 (6-Pt) to 6.08 (5-Pt), whereby the me-
thoxy-substituted compounds are more lipophilic than the flu-
orinated analogues (Table 2). The most lipophilic compounds
are those with an R6-substituted position, reaching lipophilicity
values of 6.08 and 5.84 for the methoxy- and fluorine-substi-
tuted compounds, respectively.
Figure 2. X-ray crystallographic structure of 4-Pt. Hydrogen atoms are omit-
ted for clarity. Selected bond lengths [ꢁ] and angles [8]: Pt(1)ꢀN(1) 1.980(8),
Pt(1)ꢀN(2) 1.964(8), Pt(1)ꢀO(1) 2.009(7), Pt(1)ꢀO(2) 2.017(7), N(1)ꢀC(8)
1.301(11), N(1)ꢀC(9) 1.437(12), N(2)ꢀC(14) 1.413(13), N(2)ꢀC(15) 1.313(12),
O(1)ꢀC(1) 1.311(11), O(2)ꢀC(21) 1.303(11); N(1)-Pt(1)-N(2) 82.9(3), N(1)-Pt(1)-
O(1) 96.1(3), N(1)-Pt(1)-O(2) 178.9(3), N(2)-Pt(1)-O(1) 178.4(3), N(2)-Pt(1)-O(2)
96.1(3), O(1)-Pt(1)-O(2) 85.9(2).
state and solution. When the complexes were dissolved in
[D6]DMSO or [D7]DMF, no spectral changes were observed
during storage at room temperature. Furthermore, no degrada-
tion was observed for all complexes in methanol/water (70%)
for 24 h, as determined by HPLC analysis (see Supporting Infor-
mation).
Antiproliferative effects
The antitumor activity of the [salophene]platinum(II) com-
plexes against MCF-7 human breast cancer cells was initially
evaluated in a concentration-dependent manner in a crystal
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1178

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
Table 1. Crystal data and structure refinement for 4-Pt.
Crystal system
Space group
Monoclinic
P21/n
Reflections collected
Independent reflections
16550
4891
Unit cell dimensions
a=13.739(5) ꢁ
b=7.785(5) ꢁ
c=17.102(5) ꢁ
a=908
(Rint=0.0733)
Integration
0.7625/
Absorption correction
Max./min. transmission
0.4817
b=99.448(5)8
g=908
1825.2(14)
4
2.065
7.721
Refinement method
Full-matrix least
squares on F2
4891/0/263
0.921
R1 =0.0584, wR2 =0.1412
R1 =0.0900, wR2 =0.1613
V [ꢁ3]
Data/restrains/parameters
Goodness of fit on F2
Final R indices [I>2s(I)]
R indices (all data)
Largest diff. peak and hole [eꢁꢀ3
Reflections collected
Z
1calcd [gcmꢀ3
]
m(MoKa) [mmꢀ1
F(000)
]
1088
0.16ꢂ0.02ꢂ0.02
]
1.793 and ꢀ3.160
Crystal size [mm3]
16550
20 mm). In the fluorinated series, substituents at R3 increased
the growth inhibitory effects (6-Pt: IC50 =0.7 mm), whereas they
decreased the activity if located at R5 (8-Pt: IC50 =11.2 mm), R4
(7-Pt: IC50 >20 mm) or R6 (9-Pt: IC50 >20 mm).
The most active compounds were further tested in a time-
dependent assay (Figure 4) at optimized concentrations (3-Pt
and 6-Pt: 0.31–5 mm; 1-Pt: 0.63–10 mm; 8-Pt: 1.25–20 mm). All
compounds showed maximum growth inhibition after 48–72 h
of drug incubation. For 1-Pt, a rapid onset of activity was ob-
served, followed by a 50% recovery of the cell population until
the end of the test. Complex 3-Pt showed this effect after 96 h
of drug exposure as well. This rise of the growth curve can be
explained by the damage of only a subpopulation of cancer
cells or by the development of drug resistance.[21]
If the cells were incubated with 6-Pt or 8-Pt, the growth in-
hibition remained nearly constant during the incubation time.
This finding is evidence for the strong influence that substitu-
ents might have on the pharmacological properties of the
compounds.
Accumulation studies
In order to determine whether the differences in growth inhibi-
tion are related to drug accumulation, cell-associated platinum
was measured by graphite furnace atomic absorption spec-
trometry (GF-AAS) after 6 and 24 h of drug incubation. The ac-
cumulation grade of the [salophene]platinum(II) complexes
(obtained by dividing the cellular molar drug concentration by
the concentration used in the extracellular medium)[22] are
listed in Table 2. Notably, this accumulation value represents
the balance between influx and efflux.
Figure 3. Crystal packing diagram of 4-Pt: A) complexes arranged in a head-
to-tail fashion; B) zigzag chain formed from intermolecular interactions.
violet staining assay. The resulting IC50 values determined after
96 h of incubation are listed in Table 2. The unsubstituted com-
plex 1-Pt caused strong growth inhibitory effects (IC50 =
1.1 mm) higher than cisplatin (IC50 ~1.6 mm) tested under our
experimental conditions.[12] Substituents at the salicylidene
moiety changed the activity of the compounds, depending on
their position.
With the exception of the R4-substituted compound 3-Pt
(IC50 =1.8 mm), which was as active as cisplatin, the introduc-
tion of methoxy groups led to inactive complexes (IC50 >
The cell-associated platinum(II) concentration exceeded that
of extracellular platinum(II) by at least sixfold (2-Pt), reaching
a difference of up to 30-fold for the unsubstituted [salophene]-
platinum(II) complex (1-Pt). The accumulation depends on the
nature and position of the substituents, but does not correlate
with the antiproliferative effects. Because reduced antiprolifera-
tive effects can be linked to decreased drug accumulation,
which may be due to increased efflux, diminished uptake, or
even a combination of both processes,[23] the enrichment of
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1179

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
complexes with IC50 values <20 mm was studied in
a time-dependent manner (Figure 5). These data
could be relevant for discussion of the different
growth inhibitory effects and the time–activity curves
depicted in Figure 4.
Table 2. Assay data for the [salophene]platinum(II) complexes: clogP values, antiproli-
ferative activity toward and accumulation grade in MCF-7 cells, and DNA binding effi-
ciency.
Compd
clogP
IC50 [mm][a]
AG[b]
DNA binding efficiency[c]
(pmolmgꢀ1
)
nuc/compd
All complexes were taken up to maximum levels
within 2–6 h. Complex 6-Pt presented an accumula-
tion grade <10 during the whole exposure time,
which suggests the presence of a saturable system of
cellular uptake. It has also been suggested that the
plateau observed in the uptake profiles is due to
damage of the membrane functionality, resulting
from extensive platination.[24] 1-Pt and 8-Pt were simi-
larly accumulated, but showed a slight decrease of
intracellular platinum within 24 h.
[d]
[d]
1-Pt
2-Pt
3-Pt
4-Pt
5-Pt
6-Pt
7-Pt
8-Pt
4.05ꢁ0.01
4.62ꢁ0.11
4.84ꢁ0.01
4.22ꢁ0.01
6.08ꢁ0.01
3.79ꢁ0.01
4.59ꢁ0.01
4.27ꢁ0.01
5.84ꢁ0.01
1.1ꢁ0.2
>20.0
1.8ꢁ0.7
>20.0
>20.0
0.7ꢁ0.1
>20.0
28.9ꢁ1.6
6.7ꢁ0.1
–
–
1.8ꢁ0.7
20.9ꢁ0.3
23.0ꢁ5.8
22.5ꢁ7.5
42.4ꢁ1.1
1935ꢁ757
156ꢁ2
12.4ꢁ0.2
10.0ꢁ0.5
20.3ꢁ2.5
11.0ꢁ3.6
15.5ꢁ2.7
11.7ꢁ4.8
10.6ꢁ5.0
146ꢁ37
153ꢁ51
77ꢁ2
[e]
[e]
–
–
11.2ꢁ1.4
>20.0
50.3ꢁ10.2
236ꢁ113
14.6ꢁ1.8
66ꢁ3
16 ꢁ6
9-Pt
cisplatin
[e]
[e]
–
1.6ꢁ0.1
–
230ꢁ22
[a] Concentration that results in a 50% decrease in cell growth after 96 h incubation;
values are the mean of at least two independent experiments. [b] Accumulation
grade: determined by incubation with test drug at 0.15 mm for 24 h; it is the ratio be-
tween cellular molar drug concentration and the concentration used in the extracellu-
lar medium. [c] Expressed as picomoles of compound bound per microgram of DNA,
also referred to as the amount of nucleoside per bound compound (nuc/compd); re-
sults are the mean of at least two independent experiments, which were performed
with two replicates. [d] Not detected. [e] Not determined.
In contrast, 3-Pt accumulated rapidly and reached
a maximum of 45.5-fold enrichment at 6 h relative to
the cell culture medium. However, toward the end of
the test, the cell-associated platinum was reduced to
nearly half. This could occur through several mecha-
nisms, such as increased drug efflux and decreased
drug transport into the cell. From this point of view,
and for a better understanding of the accumulation
Figure 4. Time-dependent antiproliferative effects of A) 1-Pt, B) 3-Pt, C) 6-Pt, and D) 8-Pt on MCF-7 cells.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1180

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
sive transport through the cell membrane, a correla-
tion between cellular accumulation and lipophilicity
can be observed. Thus, previous investigations with
gold phosphine complexes[26] correlated their antitu-
mor activities with accumulation in tumor cells,
which itself depends on drug lipophilicity. Concern-
ing our [salophene]platinum(II) complexes, no signifi-
cant correlation was observed, neither between lipo-
philicity–accumulation (24 h: r2 =0.27) nor between
accumulation–cytotoxicity (24 h: r2 =0.05). Further-
more, some complexes that were efficiently taken up
by cells (e.g., 7-Pt, which presented a 15.5-fold accu-
mulation after 24 h), exhibited no cytotoxicity on
MCF-7 cells within the concentration range tested. Al-
together, these results suggest that the accumulation
of the complexes is not the main reason for the dif-
ferences observed between cytotoxic potency and in-
dicates the involvement of a specific mode of drug
action. Nevertheless, it is of great interest to study
the qualitative comparison between the time-depen-
dent cellular accumulation and cytotoxicity. In some
cases, as with 1-Pt for example, the accumulation
Figure 5. Time-dependent accumulation of 1-Pt, 3-Pt, 6-Pt, and 8-Pt on MCF-7 cells (each
at 0.3 mm).
pattern of [salophene]platinum(II) complexes, cellular efflux
was investigated. Increased efflux could explain decreased ac-
cumulation during the entire drug exposure and/or the signifi-
cant decrease in intracellular platinum after the maximum has
been reached (3-Pt). The loss of 1-Pt, 3-Pt, 6-Pt, and 8-Pt from
MCF-7 cells was determined by measuring the cellular plati-
num concentration after incubation with the complexes and
subsequent incubation with drug-free medium. The efflux after
an initial exposure to 6-Pt and 3-Pt at 0.3 mm is graphically ex-
pressed in Figure 6A.
profile may explain the observed differences in cytotoxicity as
a function of time. 1-Pt was taken up rapidly and reached its
maximum at 6 h, followed by a slow decrease in cellular con-
centration that continues until the end of the regarded time
(Figure 5). In contrast, the cytotoxicity presented a maximum
at 48 h and over time the effect seems to decrease (Figure 4).
If the intracellular concentration is related with the antitumor
effects, the decrease in the accumulation may explain the de-
crease in cytotoxicity. A similar relation between intracellular
concentration and cytotoxicity can be predicted for 6-Pt, the
accumulation of which reached a plateau after 2 h of incuba-
tion and remained constant. Likewise, the cytotoxicity was ap-
proximately constant during the time-dependent antiprolifera-
tive assay.
Whereas in cells treated with 6-Pt, the platinum concentra-
tion decreased to 70% of the initial value within 120 min, cells
previously incubated with 3-Pt showed very slow efflux. This
may explain the differences between the maximum accumula-
tion levels reached by both complexes: 7- and 45-fold for 6-Pt
and 3-Pt, respectively. Complexes 1-Pt and 8-Pt showed a slow
and similar release of platinum. The cells lost 15–25% of the
accumulated platinum complex after 2 h (Figure 6B). During
the time in which efflux was analyzed (0–2 h), it represented
a considerable influence on accumulation, but can only partly
explain the accumulation profiles. It is important to consider
factors that could decrease the uptake. For example, binding
of the platinum(II) complexes to cellular membrane molecules
could induce changes in the membrane structure and func-
tions, affect its integrity and viability, and as a consequence
modify its permeability.
DNA interaction studies
Circular dichroism (CD) studies
CD is a useful technique for monitoring DNA conformation
changes resulting from changes in environmental conditions,
such as variations in temperature, pH, or as in this case, inter-
actions between the DNA and the platinum complexes. The
UV circular dichroic spectrum of calf thymus DNA (CT-DNA) ex-
hibits a positive band at 270 nm due to base stacking and
a negative band at 239 nm due to the helicity of B-DNA.[27]
These bands are quite sensitive to noncovalent interactions
of metal complexes with B-DNA. Intercalation enhances the in-
tensities of both bands, stabilizing the right-handed B confor-
mation of CT-DNA as observed for the classical intercalator
methylene blue.[28] Therefore, the CD spectra of the interaction
of complexes 1-Pt, 3-Pt, and 5-Pt with CT-DNA were recorded
after 24 h of incubation (Figure 7).
Relationship between lipophilicity, drug accumulation, and
in vitro cytotoxicity
For platinum-based drugs in cancer therapy, DNA platination is
the paramount activity parameter; a relationship between the
overall intracellular platinum uptake and survival of cells ex-
posed to cisplatin and its metabolites has been reported.[25]
Moreover, if the drug enrichment in tumor cells occurs by pas-
Spectral changes of the positive CD signal at 273 nm might
(upward arrow in Figure 7) indicate p–p interactions between
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1181

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
the nucleobases of the DNA and the p system of the metal
complexes. The three tested substances caused an increase in
the intensity of the band in this spectral area. The additional
increase of the band at 246 nm (downward arrow in Figure 7)
in comparison with the band for CT-DNA could reflect an alter-
ation in the original B-helical conformation. The absence of
any induced CD signal caused by the metal complexes, how-
ever, is indicative of only weak or no intercalation of the plati-
num complexes into the DNA double helix.
Thermal denaturation studies
The DNA melting temperature (Tm) is strictly related to the sta-
bility of the double helix, and therefore interactions of com-
pounds with DNA can alter the Tm value. It is known that the
intercalation of small molecules into DNA results in stabiliza-
tion of the double helix due to stabilizing stacking interactions,
increasing the Tm of DNA.[29] The Tm experiment carried out for
CT-DNA in the absence of any added complex revealed a Tm of
67ꢁ18C under our experimental conditions. Addition of com-
plexes 1-Pt, 3-Pt, or 5-Pt to CT-DNA showed none or insignifi-
cant DTm values, the differences of which were within experi-
mental error. As the expected increase in the temperature at
which the helix denatures in the presence of classical intercala-
tors is much greater (8–148C)[30] than the those observed here,
it is possible to consider that the observed changes in Tm are
indicative of a non-intercalative binding mode.
Viscosity measurements
It is known that the viscosity of DNA increases by complete or
partial intercalation of compounds into the DNA base stacking,
whereas electrostatic or covalent binding only lead to mild
modifications.[31] Therefore, viscosity measurements were car-
ried out with 1-Pt and 3-Pt. Figure 8 illustrates the dependence
of the relative specific viscosity (h/h0)1/3, for which h0 and h are
the specific viscosities of CT-DNA in the absence and presence
of the tested compound, respectively, plotted against
the [complex]/[DNA] ratio (r) for complexes 1-Pt and
3-Pt, and ethidium bromide (EtBr). A slope of 1.03
was measured for EtBr, a standard strong intercalator,
in the plot of (h/h0)1/3 versus r using our experimental
setup. This value is in accordance with previous in-
vestigations.[32] Increasing the concentration of 1-Pt
or 3-Pt only led to a small increase (slopes of 0.33
and 0.16, respectively) of the CT-DNA viscosity, when
compared with EtBr. The small increase in DNA
length in the presence of the tested platinum(II)
compounds could possibly be due to covalent or
groove binding, but rules out a significant degree of
intercalation.
Figure 6. Cellular platinum concentration in MCF-7 cells after 2 h incubation
&
&
&
&
with A) 6-Pt ( ) and 3-Pt ( ), or B) 1-Pt ( ) and 8-Pt ( ), and subsequent in-
cubation with drug-free medium. The mean platinum concentration after
2 h incubation was set as 100%.
Atomic absorption studies
The amount of platinum incorporated (by covalent
Figure 7. CD spectra of 1-Pt, 3-Pt, and 5-Pt incubated for 24 h with CT-DNA for 1:10 com-
plex/DNA mixtures [complex=20 mm, DNA concentration (based on nucleotide num-
ber)=200 mm] in 10 mm phosphate buffer, pH 7.2 at 258C.
and intercalative binding) into salmon testes DNA
was determined by using the method described by
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1182

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
antiproliferative effects could not be established, complexes
were taken up with various maximum levels. Furthermore,
they presented distinct time-dependent enrichment profiles,
which can be explained in part by efflux kinetics. Other factors
besides distinct accumulation, such as inactivation of the com-
plexes before reaching the drug target or simple differences in
drug potencies, may have a stronger influence on the cytotox-
icity.
Comparing the obtained cytotoxic effects of methoxy-substi-
tuted salophene ligands coordinated to platinum(II) (2-Pt to 5-
Pt) with those obtained in previous investigations with iron(III)
complexes,[7] the influence of the metal center is quite evident.
While most of the methoxy-substituted platinum(II) complexes
showed no antiproliferative effects within the tested concen-
tration range, iron(III) complexes bearing the same ligands ex-
hibited significant cytotoxic effects at concentrations <5 mm.
Moreover, iron(III) complexes bearing 4- or 6-methoxy-substi-
tuted salophenes demonstrated cytocidal effects even at a con-
centration of 0.5 mm.
It is reasonable to suggest that the metal center is related to
the mechanism of action of the complexes, as was previously
proposed for nickel(II) and iron(III) salophene complexes.[6,9]
Thus, depending on the cellular processing and mechanism of
action, substituents may affect in a different mode the activity
of the compounds.
Figure 8. Dependence of (h/h0)1/3 on r {r=[complex]/[DNA], in which [DNA]
is given in m (base pairs)} for viscosity measurements of CT-DNA with EtBr,
1-Pt, or 3-Pt in 10 mm phosphate buffer (pH 7.2).
Natarajan et al.[33] with some modifications. Salmon testes DNA
was incubated with the platinum complexes, followed by pre-
cipitation and washing of the DNA–PtII adducts with ethanol.
The amount of DNA-bound platinum was measured by GF-
AAS and referenced to the DNA content, which was deter-
mined spectrophotometrically at 260 nm. The DNA binding ef-
ficiencies of the [salophene]platinum(II) complexes are listed in
Table 2.
As shown in Table 2, except for 1-Pt and 2-Pt, all the com-
pounds bound to DNA (>20 pmolmgꢀ1 DNA), in a more effi-
cient manner than cisplatin (~15 pmolmgꢀ1) under the same
experimental conditions. The unsubstituted compound 1-Pt
showed no significant binding affinity (the amount of bound
platinum was below the detection limit of GF-AAS). The me-
thoxy-substituted complexes (with the exception of 2-Pt
Because DNA is a likely target for anticancer platinum-based
coordination compounds, the [salophene]platinum(II) com-
pounds were analyzed for their interactions with isolated DNA.
On the one hand, the platinum ion in the [salophene]plati-
num(II) complexes is part of two planar six-membered chelate
rings, containing aromatic ligands coplanar with the metal co-
ordination sphere. Therefore, an intercalative process for their
mechanism of action was postulated. Unexpectedly, the results
of CD, thermal denaturation, and viscosity studies revealed no
significant degree of intercalation with DNA. On the other
hand, if the platinum complexes underwent a chelate ring-
opening process, they could possibly bind to DNA in a “cispla-
tin-like” mechanism (covalent binding to the DNA). Conse-
quently, atomic absorption studies were selected to quantify
the total amount of DNA-bound platinum, irrespective of the
binding mechanism. Even though some complexes revealed
strong interactions with DNA, it was impossible to find a corre-
lation with cytotoxicity. The fact that some compounds with
significant cellular accumulation and high cytotoxicity showed
no detectable interactions with DNA (1-Pt) suggests that DNA
may not necessarily be the main target of the [salophene]plati-
num(II) complexes. Further investigations related to non-ge-
nomic biomolecules as possible targets, which were already re-
ported for platinum and other metal complexes,[34] are being
carried out to fully understand the mechanism of action of this
promising class of platinum(II) complexes.
(1.8 pmolmgꢀ1 DNA)) showed
a
binding efficiency (~
22 pmolmgꢀ1 DNA) slightly higher than that of cisplatin. The
fluorine-substituted complexes exhibited much higher DNA in-
teractions, with a maximum value of 236 pmolmgꢀ1 DNA deter-
mined for 9-Pt. These data clearly demonstrate a missing corre-
lation between DNA binding efficiency and cytotoxicity. In con-
trast, the compound that exhibits the strongest DNA interac-
tion, 9-Pt, is not cytotoxic in the tested range of concentra-
tions, and 1-Pt, with an IC50 value of 1.1 mm, showed no
significant DNA interaction.
Discussion
The tested [salophene]platinum(II) complexes exhibited cyto-
toxic properties that depend heavily on the type and position
of the substituent at the salicylidene moiety. Some compounds
presented antiproliferative effects similar to and even stronger
than those of cisplatin, whereas others are completely inactive.
Furthermore, the results suggest that fluorine rather than me-
thoxy substituents contribute to the antitumor effects.
To understand the dependence of the cytotoxic profiles on
the substitution patterns, a series of experiments were con-
ducted, including determination of lipophilicity, cellular accu-
mulation and efflux, and DNA interaction studies. Although
a correlation between lipophilicity, cellular accumulation, and
Conclusions
This work introduces novel platinum-based complexes, poten-
tially active as anticancer agents. It is clear that substituents on
the ligand scaffold can modulate the biological properties of
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1183

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
N,N’-Bis(6-fluorosalicylidene)-1,2-phenylenediamine [9]:
9 was
the synthesized platinum(II) complexes. The drug accumulation
is not related to drug lipophilicity, probably due to the pres-
ence of an active cellular uptake mechanism. Furthermore, al-
though a correlation between cytotoxicity (IC50) and intracellu-
lar drug concentration cannot be precisely established, a rela-
tionship in a time-dependent manner is observed for some
compounds. Understanding their antitumor mechanism of
action and target still remains an important challenge in order
to generate functionalized complexes with optimized physico-
chemical and biological properties, and it is an important area
of study for our further investigations.
obtained from 1,2-phenylenediamine (0.31 mmol, 34 mg) and 6-flu-
orosalicylaldehyde (0.63 mmol, 88 mg). Yield: 75 mg (0.21 mmol,
69%) orange powder. 1H NMR ([D6]DMSO): d=6.75–6.86 (m, 4H,
ArH-3, ArH-5), 7.41–7.51 (m, 4H, Ar’H-3, Ar’H-4, Ar’H-5, Ar’H-6),
7.54–7.59 (m, 2H, ArH-4), 9.04 (s, 2H, NCH), 13.62 ppm (s, 2H,
ArOH).
General procedure for the synthesis of the platinum(II) com-
plexes: K2CO3 (2 equiv) was dissolved in water, and the resulting
solution was added dropwise to a solution of the ligand (1 equiv)
in DMSO. Afterward, a solution of potassium tetrachloroplatinate(II)
(1 equiv) in water was added dropwise. The mixture was heated
for 5 h, keeping the temperature of the reaction under 608C. The
mixture was allowed to cool to room temperature, and the solid
matter was filtered off. Finally, the solid was washed with water
and cold methanol and dried under vacuum.
Experimental Section
General
[Pt(L1)] [1-Pt] (H2L1 =N,N’-Bis(salicylidene)-1,2-phenylenedia-
mine): 1-Pt was obtained from N,N’-bis(salicylidene)-1,2-phenylene-
diamine (1) (0.19 mmol, 60 mg) and potassium tetrachloroplatina-
te(II) (0.19 mmol, 79 mg). Yield: 75 mg (0.15 mmol, 78%) red
The following instrumentation was used: 1H NMR spectra: Bruker
ADX 400 spectrometer operated at 400 MHz (internal standard, tet-
ramethylsilane); ESI-TOF spectra: Agilent 6210 ESI-TOF, Agilent
Technologies, Santa Clara, CA, USA; HPLC: 1100 series, Hewlett–
Packard, CA, USA; AAS: Vario 6, Analytik Jena AG, Jena, Germany.
Chemicals were obtained from Sigma–Aldrich (Germany) and used
without further purification. Compounds 1, 2, 3, 4, and 5 were
kindly supplied by Dr. Annegret Hille and have already been re-
ported.[7] Reactions were all monitored by TLC, performed on silica
gel plates 60 F254 (Merck, Darmstadt, Germany). Visualization on
TLC was carried out with UV light. All new complexes were
checked for stability and purity in methanol/water 70%, and purity
>95% was confirmed in each case (see Table S1 in the Supporting
Information).
1
powder. H NMR ([D7]DMF): d=6.77–6.83 (m, 2H, ArH-5), 7.13–7.19
3
3
4
(d, J=8.6 Hz, 2H, ArH-3), 7.45–7.50 (dd, J=6.3 Hz, J=3.2 Hz, 2H,
3
Ar’H-4, Ar’H-5), 7.58–7.64 (m, 2H, ArH-4), 7.91–7.95 (dd, J=8.0 Hz,
4J=1.5 Hz, 2H, ArH-6), 8.50–8.59 (dd, 3J=6.3 Hz, 4J=3.3 Hz, 2H,
Ar’H-3, Ar’H-6), 9.65 ppm (s, 2H, NCH); HRMS (+)-ESI m/z [M+Na]+
calcd for C20H14N2NaO2Pt+: 532.0601, found: 532.0615.
[Pt(L2)] [2-Pt] (H2L2 =(N,N’-Bis(3-methoxysalicylidene)-1,2-phenyl-
enediamine): 2-Pt was obtained from N,N’-bis(3-methoxysalicyli-
dene)-1,2-phenylenediamine (2) (0.10 mmol, 37 mg) and potassium
tetrachloroplatinate(II)
(0.10 mmol,
41 mg).
Yield:
10 mg
(0.02 mmol, 18%) violet powder. 1H NMR ([D6]DMSO): d=3.85 (s,
6H, CH3), 6.68–6.81 (m, 2H, ArH-5), 7.11–7.27 (m, 2H, ArH-4), 7.33–
7.61 (m, 4H, ArH-6, Ar’H-4, Ar’H-5), 8.39–8.63 (m, 2H, Ar’H-3, Ar’H-
6), 9.52 ppm (s, 2H, NCH); HRMS (+)-ESI m/z [M+K]+ calcd for
C22H18KN2O4Pt+: 608.0551, found: 608.0541.
Syntheses
General procedure for the synthesis of N,N’-bis(salicylidene)-1,2-
phenylenediamine derivatives: The respective salicylaldehyde
(2 equiv) in ethanol were added dropwise to a solution of 1,2-phe-
nylenediamine (1 equiv) in ethanol. The mixture was stirred at
reflux for 1–2 h and then allowed to cool to room temperature.
The product was collected, washed with cold ethanol, and dried
(P2O5).
[Pt(L3)] [3-Pt] (H2L3 =(N,N’-Bis(4-methoxysalicylidene)-1,2-phenyl-
enediamine): 3-Pt was obtained from N,N’-bis(4-methoxysalicyli-
dene)-1,2-phenylenediamine (3) (0.10 mmol, 37 mg) and potassium
tetrachloroplatinate(II)
(0.10 mmol,
41 mg).
Yield:
50 mg
1
(0.09 mmol, 88%) orange powder. H NMR ([D6]DMSO): d=3.84 (s,
6H, CH3), 6.45–6.50 (dd, 3J=8.9 Hz, 4J=2.2 Hz, 2H, ArH-5), 6.60–
6.65 (d, 4J=2.0 Hz, 2H, ArH-3), 7.37–7.42 (m, 4H, Ar’H-4, Ar’H-5),
N,N’-Bis(3-fluorosalicylidene)-1,2-phenylenediamine [6]:
6 was
3
3
4
obtained from 1,2-phenylenediamine (0.18 mmol, 20 mg) and 3-flu-
orosalicylaldehyde (0.37 mmol, 52 mg). Yield: 86 mg (0.15 mmol,
82%) orange powder. 1H NMR ([D6]DMSO): d=6.93–7.00 (m, 2H,
ArH-5), 7.37–7.48 (m, 4H, ArH-4, Ar’H-3, Ar’H-6), 7.51–7.56 (m, 4H,
ArH-6, Ar’H-4, Ar’H-5), 9.03 (s, 2H, NCH), 13.30 ppm (s, 2H, ArOH).
7.72–7.77 (d, J=9.0 Hz, 2H, ArH-6), 8.33–8.41 (dd, J=6.2 Hz, J=
3.4 Hz, 2H, Ar’H-3, Ar’H-6), 9.32 ppm (s, 2H, NCH); HRMS (+)-ESI
m/z [M+K]+ calcd for C22H18KN2O4Pt+: 608.0551, found: 608.0539.
[Pt(L4)] [4-Pt] (H2L4 =(N,N’-Bis(5-methoxysalicylidene)-1,2-phenyl-
enediamine): 4-Pt was obtained from N,N’-bis(5-methoxysalicyli-
dene)-1,2-phenylenediamine (4) (0.10 mmol, 37 mg) and potassium
N,N’-Bis(4-fluorosalicylidene)-1,2-phenylenediamine [7]:
7 was
obtained from 1,2-phenylenediamine (0.28 mmol, 31 mg) and 4-flu-
orosalicylaldehyde (0.56 mmol, 79 mg). Yield: 53 mg (0.15 mmol,
54%) yellow powder. 1H NMR ([D6]DMSO): d=6.76–6.93 (m, 4H,
ArH-3, ArH-5), 7.35–7.50 (m, 4H, Ar’H-3, Ar’H-4, Ar’H-5, Ar’H-6),
7.68–7.82 (m, 2H, ArH-6), 8.96 (s, 2H, NCH), 13.56 ppm (s, 2H,
ArOH).
tetrachloroplatinate(II)
(0.03 mmol, 29%) dark violet powder. H NMR ([D6]DMSO): d=3.78
(0.10 mmol,
41 mg).
Yield:
16 mg
1
3
3
(s, 6H, CH3), 7.03–7.11 (d, J=9.3 Hz, 2H, ArH-3), 7.26–7.33 (dd, J=
4
4
9.2 Hz, J=3.2 Hz, 2H, ArH-4), 7.35–7.39 (d, J=3.1 Hz, 2H, ArH-6),
3
4
7.41–7.48 (dd, J=6.3 Hz, J=3.2 Hz, 2H, Ar’H-4, Ar’H-5), 8.37–8.46
(dd, 3J=6.3 Hz, 4J=3.4 Hz, 2H, Ar’H-3, Ar’H-6), 9.51 ppm (s, 2H,
NCH); HRMS (+)-ESI m/z [M+Na]+ calcd for C22H18N2NaO4Pt+:
592.0812, found: 592.0825.
N,N’-Bis(5-fluorosalicylidene)-1,2-phenylenediamine [8]:
8 was
obtained from 1,2-phenylenediamine (0.51 mmol, 55 mg) and 5-flu-
orosalicylaldehyde (1.02 mmol, 143 mg). Yield: 197 mg (0.36 mmol,
71%) orange powder. 1H NMR ([D6]DMSO): d=6.95–7.07 (dd, 3J=
[Pt(L5)] [5-Pt] (H2L5 =(N,N’-Bis(6-methoxysalicylidene)-1,2-phenyl-
enediamine): 5-Pt was obtained from N,N’-bis(6-methoxysalicyli-
dene)-1,2-phenylenediamine (5) (0.10 mmol, 37 mg) and potassium
4
9.1 Hz, J=4.5 Hz, 2H, ArH-3), 7.24–7.31 (m, 2H, ArH-4), 7.41–7.48
(m, 4H, Ar’H-3, Ar’H-4, Ar’H-5, Ar’H-6), 7.50–7.58 (dd, 3J=9.0 Hz,
4J=3.2 Hz, 2H, ArH-6), 8.92 (s, 2H, NCH), 12.58 ppm (s, 2H, ArOH).
tetrachloroplatinate(II)
(0.10 mmol,
41 mg).
Yield:
15 mg
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1184

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
1
(0.03 mmol, 29%) red powder. H NMR ([D6]DMSO): d=3.94 (s, 6H,
ed,[35] were injected onto a reversed-phase HPLC column. The chro-
matographic conditions were: silica-based C18 gel as the stationary
phase; flow rate=0.80 mLminꢀ1; diode array detector; mobile
phase: CH3OH/H2O, 0.1% trifluoroacetic acid (70:30). The logP
values were plotted against the average logk values, and the data
were fitted using a linear regression to create a calibration curve.
Platinum complexes were prepared as 1 mm stock solutions in
DMF. An amount of 30 mL was diluted in 1.470 mL of H2O (twice
distilled), respectively, and chromatographed individually. The
clogP values were obtained by interpolation of the logk values on
the respective calibration regression curve. Results were expressed
as mean values from at least two independent experiments.
CH3), 6.30–6.37 (m, 2H, ArH-5), 6.68–6.74 (m, 2H, ArH-3), 7.36–7.54
3
4
(m, 4H, ArH-4, Ar’H-4, Ar’H-5), 8.21–8.31 (dd, J=6.1 Hz, J=3.1 Hz,
2H, Ar’H-3, Ar’H-6), 9.61 ppm (s, 2H, NCH); HRMS (+)-ESI m/z
[M+K]+ calcd for C22H18KN2O4Pt+: 608.0551, found: 608.0577
[Pt(L6)] [6-Pt] (H2L6 =(N,N’-Bis(3-fluorosalicylidene)-1,2-phenyle-
nediamine): 6-Pt was obtained from N,N’-bis(3-fluorosalicylidene)-
1,2-phenylenediamine (6) (0.10 mmol, 35 mg) and potassium tetra-
chloroplatinate(II) (0.10 mmol, 41 mg). Yield: 25 mg (0.05 mmol,
50%) dark violet powder. 1H NMR ([D6]DMSO): d=6.73–6.83 (m,
2H, ArH-5), 7.48–7.58 (m, 4H, ArH-4, Ar’H-4, Ar’H-5), 7.73–7.81 (d,
3J=8.1 Hz, 2H, ArH-6), 8.47–8.57 (dd, 3J=6.1 Hz, 4J=3.4 Hz, 2H,
Ar’H-3, Ar’H-6), 9.63 ppm (s, 2H, NCH); HRMS (+)-ESI m/z [M+K]+
calcd for C20H12F2KN2O2Pt+: 584.0152, found: 584.0144.
Cytotoxicity
[Pt(L7)] [7-Pt] (H2L7 =(N,N’-Bis(4-fluorosalicylidene)-1,2-phenyle-
nediamine): 7-Pt was obtained from N,N’-bis(4-fluorosalicylidene)-
1,2-phenylenediamine (7) (0.10 mmol, 35 mg) and potassium tetra-
chloroplatinate(II) (0.10 mmol, 41 mg). Yield: 32 mg (0.06 mmol,
56%) dark violet powder. 1H NMR ([D6]DMSO): d=6.69–6.78 (m,
The human MCF-7 breast cancer cell line was obtained from the
American Type Culture Collection. Cells were maintained as a mon-
olayer culture in l-glutamine containing Dulbecco’s modified
Eagle’s medium (DMEM) with 4.5 gLꢀ1 glucose (PAA Laboratories,
Austria), supplemented with 5% fetal bovine serum (FBS; Bio-
chrom, Germany) in a humidified atmosphere (5% CO2) at 378C.
The experiments were performed according to established proce-
dures with some modifications.[36] In 96-well plates, 100 mL of a cell
suspension in culture medium at 7500 cellsmLꢀ1 were plated into
each well and were incubated for three days under culture condi-
tions. After addition of the test compounds at various concentra-
tions in DMF, cells were incubated for the appropriate incubation
times. The medium was then removed, cells were fixed with a glu-
tardialdehyde solution (1%) and stored under phosphate-buffered
saline (PBS) at 48C. Cell biomass was determined by crystal violet
staining, followed by extraction of the bound dye with ethanol
and a photometric measurement at 590 nm. Mean values were cal-
culated, and the effects of the compounds were expressed as per-
cent treated/controlcorr values according to Equation (1):
3
4
2H, ArH-5), 6.83–6.94 (dd, J=12.1 Hz, J=2.2 Hz, 2H, ArH-3), 7.41–
7.50 (dd, 3J=6.2 Hz, 4J=3.1 Hz, 2H, Ar’H-4, Ar’H-5), 7.88–8.00 (m,
3
4
2H, ArH-6), 8.36–8.46 (dd, J=6.1 Hz, J=3.4 Hz, 2H, Ar’H-3, Ar’H-
6), 9.49 ppm (s, 2H, NCH); HRMS (+)-ESI m/z [M+Na]+ calcd for
C20H12F2N2NaO2Pt+: 568.0412, found: 568.0429.
[Pt(L8)] [8-Pt] (H2L8 =(N,N’-Bis(5-fluorosalicylidene)-1,2-phenyle-
nediamine): 8-Pt was obtained from N,N’-bis(5-fluorosalicylidene)-
1,2-phenylenediamine (8) (0.08 mmol, 29 mg) and potassium tetra-
chloroplatinate(II) (0.08 mmol, 34 mg). Yield: 30 mg (0.06 mmol,
69%) dark violet powder. 1H NMR ([D6]DMSO): d=7.08–7.17 (dd,
4
3J=9.4 Hz, J=4.8 Hz, 2H, ArH-3), 7.44–7.56 (m, 4H, ArH-4, Ar’H-4,
3
4
Ar’H-5), 7.61–7.70 (dd, J=9.4 Hz, J=4.8 Hz, 2H, ArH-6), 8.35–8.44
(dd, 3J=6.2 Hz, 4J=3.3 Hz, 2H, Ar’H-3, Ar’H-6), 9.52 ppm (s, 2H,
NCH); HRMS (+)-ESI m/z [M+Na]+ calcd for C20H12F2N2NaO2Pt+:
568.0412, found: 568.0425.
TꢀC0
T
ð1Þ
=
½%ꢂ ¼
ꢃ 100
[Pt(L9)] [9-Pt] (H2L9 =(N,N’-Bis(6-fluorosalicylidene)-1,2-phenyle-
nediamine): 9-Pt was obtained from N,N’-bis(6-fluorosalicylidene)-
1,2-phenylenediamine (9) (0.12 mmol, 42 mg) and potassium tetra-
chloroplatinate(II) (0.12 mmol, 49 mg). Yield: 30 mg (0.06 mmol,
Ccorr
CꢀC0
for which C0 =control cells at the time of compound addition, C=
control cells at the time of test end, and T=probes/samples at the
time of test end. The IC50 value was determined as the concentra-
tion that causes 50% inhibition of cell proliferation and was calcu-
lated as the mean of at least two independent experiments (Ori-
ginPro 8).
1
46%) dark red powder. H NMR ([D6]DMSO): d=6.58–6.65 (m, 2H,
ArH-5), 6.95–6.99 (m, 2H, ArH-3), 7.44–7.48 (m, 2H, Ar’H-4, Ar’H-5),
3
4
7.50–7.57 (m, 2H, ArH-4), 8.40–8.51 (dd, J=6.0 Hz, J=2.9 Hz, 2H,
Ar’H-3, Ar’H-6), 9.47 ppm (s, 2H, NCH); HRMS (+)-ESI m/z [M+Na]+
calcd for C20H12F2N2NaO2Pt+: 568.0412, found: 568.0435.
Cellular accumulation studies
X-ray crystallography
The accumulation studies were performed according to methods
already described by us.[24,37] Briefly, MCF-7 cell cultures were
grown in 75 cm2 flasks at 378C under a 5% CO2 atmosphere until
at least 70% confluency. The medium was removed and replaced
by media containing 0.15 mm of drug. After appropriate incubation
time at 378C under a 5% CO2 atmosphere, the drug-containing
medium was removed, and the cell monolayer was washed with
5 mL of PBS (pH 7.4). The cell pellets were isolated by trypsin treat-
ment and centrifugation (2000 g, 5 min, 48C), resuspended in
150 mL of Triton X-100 solution (1%), lysed by using a sonotrode,
and adequately diluted. An aliquot was removed for the purpose
of protein quantification by the Bradford method. Probes for plati-
num determination by atomic absorption measurement were sta-
bilized by the addition of HCl (18%). The cellular platinum content
was calculated as pmol Pt per mg cellular protein. Furthermore, it
was possible to estimate the cellular platinum concentration using
The intensities for the X-ray determinations were collected on
a STOE IPDS 2T instrument with MoKa radiation (l=0.71073 ꢁ) at
200 K. Standard procedures were applied for data reduction and
absorption correction. Structure solution and refinement were per-
formed with SHELXS97 and SHELXL97 (G. M. Sheldrick, SHELXS-97
and SHELXL-97, programs for the solution and refinement of crys-
tal structures; University of Gçttingen, Germany, 1997). Hydrogen
atom positions were calculated for idealized positions and treated
with the “riding model” option of SHELXL. More details on data
collections and structure calculations are listed in Table 1.
clogP determination
Six chemicals (benzene, naphthalene, biphenyl, anthracene, n-hex-
ylbenzene, and n-octylbenzene) for which logP has been report-
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1185

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
the protein–volume relation (1.0 ng cellular protein in MCF-7 cells
corresponds to a cellular volume of 8.85 pL).[22] By dividing the cel-
lular concentration of the drug by the concentration in the extra-
cellular medium, the accumulation grade was obtained. Results
were expressed as mean values from two independent experi-
ments.
a water bath. Reduced viscosities (h) were calculated by published
methods[42] and plotted as (h/h0)1/3 (h0 =reduced viscosity of the
DNA solution in the absence of complex) against r {r=[complex]/
[DNA], where [DNA] is given in m (base pairs)}.
Platinum binding studies: Precipitation of drug DNA adducts was
performed according to a described method[37] with some modifi-
cations: salmon testes DNA was dissolved in PBS (pH 7.4), and
drugs were added as stock solutions in DMF. The final solutions
contained 40 mm drug and 250 mgmLꢀ1 salmon testes DNA. After
vortexing, the solutions were incubated at 378C in a water bath
for 4 h under continuous and moderate shaking. Aliquots of
200 mL were mixed with 100 mL of 0.9m sodium acetate and three
volumes of ice cold ethanol. Samples were stored at ꢀ208C for
30 min. The pellets were isolated by centrifugation (5000 rpm,
10 min, 48C) and resuspended in 300 mL of 0.3m sodium acetate.
Ice-cold ethanol (900 mL) was added, and the precipitate was col-
lected after centrifugation (5000 rpm, 10 min, 48C). Samples were
washed twice with ice-cold ethanol and were stored at ꢀ208C. The
pellets were dissolved in 500 mL of water (twice distilled), and the
DNA content was determined by absorption reading at 260 nm in
a UV microplate reader. Salmon testes DNA dissolved in water
(twice distilled) was used for calibration purposes. Samples (25 mL)
for GF-AAS platinum determination were stabilized by 25 mL of
Triton X-100 (4%), followed by addition of 25 mL of HCl (18%),
25 mL of DMF, and 50 mL of H2O (twice distilled). The amount of
drug bound to DNA was expressed as pmol drug per mg DNA and
as amount of nucleoside per bound compound (nucleoside/com-
pound). Results were expressed as means from at least two inde-
pendent experiments.
The efflux studies were performed according to a previously de-
scribed procedure.[38] MCF-7 cells were cultivated as described
above. At a confluency of ~70%, the cells were incubated with the
respective platinum complex at 0.3 mm for 2 h. A part of the cells
was collected as described above. The other part of the cells was
washed twice with 5 mL of PBS and incubated for 10, 20, 40, 60,
and 120 min with drug-free medium. Afterward, the cells were col-
lected and analyzed for their protein and platinum concentration
as previously described.
Graphite furnace atomic absorption spectrometry (GF-AAS)
GF-AAS measurements were performed according to a previously
published procedure with some minor modifications.[39] Each com-
pound was measured against its own calibration curve. Therefore,
calibration standards were prepared by making serial dilutions
from analyte stock solutions in DMF. The concentrations chosen for
the calibration curve included 0.000, 10.000, 25.000, 50.000,
100.000, 200.000, 300.000 and 400.000 mgLꢀ1. Working stock solu-
tions were prepared before use with 25 mL of the concentrated
stock solutions, stabilized with 100 mL of Triton X-100 (1%) and
25 mL of HCl (18%). The platinum content of the lysates was as-
sessed by the linear extrapolation method. Platinum was detected
at a wavelength of 265.9 nm with a spectral slit width of 0.2 nm. A
deuterium lamp was used for background correction. Probes were
injected at a volume of 20 mL into regular graphite wall tubes.
Drying, atomization, and tube cleaning steps were performed with
some minor modifications as outlined in greater detail else-
where.[40] The temperature for pyrolysis was set at 16008C. The
mean AUC (area under curve) absorptions of duplicate injections
were used throughout the study.
Acknowledgements
This work was supported by the Deutsche Forschungsgemein-
schaft (DFG) within FOR 630 “Biological function of organometal-
lic compounds”.
Keywords: cytotoxicity
·
DNA
interactions
·
drug
DNA interaction studies
accumulation · platinum complexes · salophene · schiff bases
Circular dichroism spectroscopy: CD spectra were recorded with
a JASCO J-715 instrument in the range of 220–400 nm for 1:10
complex/DNA mixtures [complex=20 mm, DNA concentration in m
(based on nucleotide number)=200 mm] in a 10 mm phosphate
buffer at pH 7.2 with CT-DNA at 258C after an incubation time of
24 h. DMF (~1%) was added to ensure solubility of all substances.
[1] a) N. J. Wheate, S. Walker, G. E. Craig, R. Oun, Dalton Trans. 2010, 39,
8113–8127; b) K. Lloyd, Nat. Rev. Cancer 2007, 7, 573–584.
[2] a) H. S. Zahid, Oncogene 2003, 22, 7265–7279; b) L. P. Martin, T. C. Ham-
ilton, R. J. Schilder, Clin. Cancer Res. 2008, 14, 1291–1295.
[3] a) R. Gust, W. Beck, G. Jaouen, H. Schçnenberger, Coord. Chem. Rev.
2009, 253, 2742–2759; b) R. Gust, W. Beck, G. Jaouen, H. Schçnenberg-
er, Coord. Chem. Rev. 2009, 253, 2760–2779; c) C. Cassino, E. Gabano,
M. Ravera, G. Cravotto, G. Palmisano, A. Vessiꢃres, G. Jaouen, S.
Mundwiler, R. Alberto, D. Osella, Inorg. Chim. Acta 2004, 357, 2157–
2166.
[4] R. Kçckerbauer, P. J. Bednarski, J. Inorg. Biochem. 1996, 62, 281–298.
[5] C. Hansch, B. H. Venger, A. Panthananickal, J. Med. Chem. 1980, 23,
459–461.
[6] A. Hille, I. Ott, A. Kitanovic, I. Kitanovic, H. Alborzinia, E. Lederer, S.
Wçlfl, N. Metzler-Nolte, S. Schꢄfer, W. Sheldrick, C. Bischof, U. Schatz-
schneider, R. Gust, J. Biol. Inorg. Chem. 2009, 14, 711–725.
[7] A. Hille, R. Gust, Eur. J. Med. Chem. 2010, 45, 5486–5492.
[8] S.-Y. Lee, A. Hille, I. Kitanovic, P. Jesse, G. Henze, S. Wçlfl, R. Gust, A.
Prokop, Leuk. Res. 2011, 35, 387–393.
Thermal denaturation studies: The thermal denaturation temper-
atures (Tm) of 1:10 complex/CT-DNA mixtures [complex=20 mm,
CT-DNA concentration in m (based on nucleotide number)=
200 mm] in a 10 mm phosphate buffer at pH 7.2. DMF (1%) was
added to ensure solubility of all substances. Melting curves were
recorded in 28C steps in the range of 50–858C at 260 nm with an
SPECORD 200 spectrometer connected to a Peltier temperature
controller. Tm values were calculated by determining the midpoints
of melting curves from the first-order derivatives. The experimental
DTm values are estimated to be accurate to within ꢁ18C. The con-
centrations of CT-DNA were spectrophotometrically determined
using the molar extinction coefficient e260 =6.600mꢀ1 cmꢀ1 [41]
.
[9] S.-Y. Lee, A. Hille, C. Frias, B. Kater, B. Bonitzki, S. Wçlfl, H. Scheffler, A.
Prokop, R. Gust, J. Med. Chem. 2010, 53, 6064–6070.
[10] P. Wu, D.-L. Ma, C.-H. Leung, S.-C. Yan, N. Zhu, R. Abagyan, C.-M. Che,
Chem. Eur. J. 2009, 15, 13008–13021.
Viscosity measurements: Viscosities for complex/sonicated DNA
mixtures were determined using a Cannon-Ubbelhode semi-micro-
dilution viscometer held at a constant temperature of 258C in
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1186

CHEMMEDCHEM
FULL PAPERS
www.chemmedchem.org
[11] D. J. Patel, A. T. Phan, V. Kuryavyi, Nucleic Acids Res. 2007, 35, 7429–
7455.
[27] R. H. Garrett, C. M. Grisham, Biochemistry, Brooks/Cole, Cengage Learn-
ing, Boston, 2010.
[12] M. Proetto, W. Liu, A. Hagenbach, U. Abram, R. Gust, Eur. J. Med. Chem.
2012, 53, 168–175.
[28] B. Nordꢅn, F. Tjerneld, Biopolymers 1982, 21, 1713–1734.
[29] a) J. B. Lepecq, C. Paoletti, J. Mol. Biol. 1967, 27, 87–106; b) J. M. Kelly,
A. B. Tossi, D. J. McConnell, C. OhUigin, Nucleic Acids Res. 1985, 13,
6017–6034.
[30] G. A. Neyhart, N. Grover, S. R. Smith, W. A. Kalsbeck, T. A. Fairley, M. Cory,
H. H. Thorp, J. Am. Chem. Soc. 1993, 115, 4423–4428.
[31] a) L. S. Lerman, J. Mol. Biol. 1961, 3, 18–30; b) G. Yang, W. J. Zhong, L.
Wang, J. L. Nian, X. Tian, J. Inorg. Biochem. 1997, 66, 141–144.
[32] R. Bieda, I. Ott, R. Gust, W. S. Sheldrick, Eur. J. Inorg. Chem. 2009, 3821–
3831.
[13] a) A. Ray, S. Banerjee, G. Rosair, V. Gramlich, S. Mitra, Struct. Chem. 2008,
19, 459–465; b) A. Hille, R. Gust, Arch. Pharm. 2009, 342, 625–631.
[14] W.-L. Tong, L.-M. Lai, M. C. W. Chan, Dalton Trans. 2008, 1412–1414.
[15] a) P. Gilli, V. Bertolasi, V. Ferretti, G. Gilli, J. Am. Chem. Soc. 2000, 122,
10405–10417; b) A. Szady-Chelmieniecka, E. Grech, Z. Rozwadowski, T.
Dziembowska, W. Schilf, B. Kamienski, J. Mol. Struct. 2001, 565–566,
125–128; c) A. Makal, W. Schilf, B. Kamienski, A. Szady-Chelmieniecka, E.
Grech, K. Wozniak, Dalton Trans. 2011, 40, 421–430.
[16] a) J. Chin, F. Mancin, N. Thavarajah, D. Lee, A. Lough, D. S. Chung, J. Am.
Chem. Soc. 2003, 125, 15276–15277; b) H. Kim, Y. Nguyen, C. P.-H. Yen,
L. Chagal, A. J. Lough, B. M. Kim, J. Chin, J. Am. Chem. Soc. 2008, 130,
12184–12191.
[17] C.-M. Che, C.-C. Kwok, S.-W. Lai, A. F. Rausch, W. J. Finkenzeller, N. Zhu,
H. Yersin, Chem. Eur. J. 2010, 16, 233–247.
[18] a) W. Sawodny, U. Thewalt, E. Potthoff, R. Ohl, Acta Crystallogr. Sect. C
1999, 55, 2060–2061; b) C.-M. Che, S.-C. Chan, H.-F. Xiang, M. C. W.
Chan, Y. Liu, Y. Wang, Chem. Commun. 2004, 1484–1485.
[19] A. S. Gaballa, M. S. Asker, A. S. Barakat, S. M. Teleb, Spectrochim. Acta
Part A 2007, 67, 114–121.
[20] E. H. Kerns, L. Di, Drug Discovery Today Technol. 2004, 1, 343–348.
[21] G. Bernhardt, H. Reile, H. Birnbçck, T. Spruß, H. Schçnenberger, J. Cancer
Res. Clin. Oncol. 1992, 118, 35–43.
[33] G. Natarajan, R. Malathi, E. Holler, Biochem. Pharmacol. 1999, 58, 1625–
1629.
[34] a) P. J. Dyson, G. Sava, Dalton Trans. 2006, 1929–1933; b) L. Cubo, M.
Groessl, P. J. Dyson, A. G. Quiroga, C. Navarro-Ranninger, A. Casini,
ChemMedChem 2010, 5, 1335–1343; c) S. A. De Pascali, F. Lugoli, A. De
Donno, F. P. Fanizzi, Met. Based Drugs 2011, 763436; d) R. Sasanelli, A.
Boccarelli, D. Giordano, M. Laforgia, F. Arnesano, G. Natile, C. Cardellic-
chio, M. A. M. Capozzi, M. Coluccia, J. Med. Chem. 2007, 50, 3434–3441.
[35] J. Sangster, J. Phys. Chem. Ref. Data 1989, 18, 1111–1229.
[36] a) H. Reile, H. Birnbçck, G. Bernhardt, T. Spruß, H. Schçnenberger, Anal.
Biochem. 1990, 187, 262–267; b) W. Liu, K. Bensdorf, M. Proetto, U.
Abram, A. Hagenbach, R. Gust, J. Med. Chem. 2011, 54, 8605–8615.
[37] I. Ott, K. Schmidt, B. Kircher, P. Schumacher, T. Wiglenda, R. Gust, J. Med.
Chem. 2005, 48, 622–629.
[22] R. Gust, B. Schnurr, R. Krauser, G. Bernhardt, M. Koch, B. Schmid, E.
Hummel, H. Schçnenberger, J. Cancer Res. Clin. Oncol. 1998, 124, 585–
597.
[23] D. P. Gately, S. B. Howell, Br. J. Cancer 1993, 67, 1171–1176.
[24] G. Bernhard, H. Reile, M. Koch, H. Schçnenberger, M. Hollstein, F. Lux,
Cancer Chemother. Pharmacol. 1992, 30, 113–122.
[38] I. Buß, D. Garmann, M. Galanski, G. Weber, G. V. Kalayda, B. K. Keppler,
U. Jaehde, J. Inorg. Biochem. 2011, 105, 709–717.
[39] T. Kapp, A. Dullin, R. Gust, J. Med. Chem. 2006, 49, 1182–1190.
[40] T. Kapp, A. Dullin, R. Gust, Bioconjugate Chem. 2010, 21, 328–337.
[41] H.-Q. Liu, S.-M. Peng, C. M. Che, J. Chem. Soc. Chem. Commun. 1995,
509–510.
[25] F. H. Shirazi, J. M. Molepo, D. J. Stewart, C. E. Ng, G. P. Raaphorst, R.
Goel, Toxicol. Appl. Pharmacol. 1996, 140, 211–218.
[42] G. Cohen, H. Eisenberg, Biopolymers 1969, 8, 45–55.
[26] M. J. McKeage, S. J. Berners-Price, P. Galettis, R. J. Bowen, W. Brouwer, L.
Ding, L. Zhuang, B. C. Baguley, Cancer Chemother. Pharmacol. 2000, 46,
343–350.
Received: April 11, 2014
Published online on May 20, 2014
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemMedChem 2014, 9, 1176 – 1187 1187