Cytotoxic hydrophilic iminophosphorane coordination compounds of d 8 metals. Studies of their interactions with DNA and HSA

Article abstract of DOI:10.1016/j.jinorgbio.2012.06.017

The synthesis and characterization of a new water-soluble N,N-chelating iminophosphorane ligand TPAN - C(O) - 2-NC₅H₄ (N,N-IM) (1) and its d⁸ (Au^III, Pd^II and Pt^II) coordination complexes are reported. The structures of cationic [AuCl ₂(N,N-IM)]ClO₄ (2) and neutral [MCl₂(N,N-IM)] M=Pd (3), Pt(4) complexes were determined by X-ray diffraction studies or by means of density-functional calculations. While the Pd and Pt compounds are stable in mixtures of DMSO/H₂O over 4 days, the gold derivative (2) decomposes quickly to TPAO and previously reported neutral gold(III) compound [AuCl₂(N,N-H)] 5 (containing the chelating N,N-fragment HN - C(O) - 2-NC₅H₄). The cytotoxicities of complexes 2-5 were evaluated in vitro against human Jurkat-T acute lymphoblastic leukemia cells and DU-145 human prostate cancer cells. Pt (4) and Au compounds (2 and 5) are more cytotoxic than cisplatin to these cell lines and to cisplatin-resistant Jurkat sh-Bak cell lines and their cell death mechanism is different from that of cisplatin. All the compounds show higher toxicity against leukemia cells when compared to normal human T-lymphocytes (PBMC). The interaction of the Pd and Pt compounds with calf thymus and plasmid (pBR322) DNA is different from that of cisplatin. All compounds bind to human serum albumin (HSA) faster than cisplatin (measured by fluorescence spectroscopy). Weak and stronger binding interactions were found for the Pd (3) and Pt (4) derivatives by isothermal titration calorimetry. Importantly, for the Pt (4) compounds the binding to HSA was reversed by addition of a chelating agent (citric acid) and by a decrease in pH.

Full text of DOI:10.1016/j.jinorgbio.2012.06.017

Journal of Inorganic Biochemistry 116 (2012) 204–214
Contents lists available at SciVerse ScienceDirect
Journal of Inorganic Biochemistry
journal homepage: www.elsevier.com/locate/jinorgbio
Cytotoxic hydrophilic iminophosphorane coordination compounds of d8 metals.
Studies of their interactions with DNA and HSA
Monica Carreira a, Rubén Calvo-Sanjuán b, Mercedes Sanaú c, Xiangbo Zhao a, Richard S. Magliozzo a,
Isabel Marzo b, María Contel a,
⁎
a
Department of Chemistry, Brooklyn College and The Graduate Center, The City University of New York, Brooklyn, NY 11210, USA
Department of Biochemistry and Molecular and Cellular Biology, University of Zaragoza, 50009, Spain
Departamento de Química Inorgánica, Universidad de Valencia, Burjassot, Valencia, 46100, Spain
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis and characterization of a new water-soluble N,N-chelating iminophosphorane ligand
TPA_N―C(O)―2-NC5H4 (N,N-IM) (1) and its d8 (AuIII, PdII and PtII) coordination complexes are reported.
The structures of cationic [AuCl2(N,N-IM)]ClO4 (2) and neutral [MCl2(N,N-IM)] M=Pd (3), Pt(4) complexes
were determined by X-ray diffraction studies or by means of density-functional calculations. While the Pd and
Pt compounds are stable in mixtures of DMSO/H2O over 4 days, the gold derivative (2) decomposes quickly to
TPA_O and previously reported neutral gold(III) compound [AuCl2(N,N-H)] 5 (containing the chelating N,
N-fragment HN―C(O)―2-NC5H4). The cytotoxicities of complexes 2–5 were evaluated in vitro against human
Jurkat-T acute lymphoblastic leukemia cells and DU-145 human prostate cancer cells. Pt (4) and Au compounds
(2 and 5) are more cytotoxic than cisplatin to these cell lines and to cisplatin-resistant Jurkat sh-Bak cell lines and
their cell death mechanism is different from that of cisplatin. All the compounds show higher toxicity against leu-
kemia cells when compared to normal human T-lymphocytes (PBMC). The interaction of the Pd and Pt com-
pounds with calf thymus and plasmid (pBR322) DNA is different from that of cisplatin. All compounds bind to
human serum albumin (HSA) faster than cisplatin (measured by fluorescence spectroscopy). Weak and stronger
binding interactions were found for the Pd (3) and Pt (4) derivatives by isothermal titration calorimetry. Impor-
tantly, for the Pt (4) compounds the binding to HSA was reversed by addition of a chelating agent (citric acid) and
by a decrease in pH.
Received 18 April 2012
Received in revised form 29 June 2012
Accepted 30 June 2012
Available online 6 July 2012
Keywords:
Cytotoxic
Leukemia
Iminophosphoranes
Water-soluble
Platinum
Gold
© 2012 Elsevier Inc. All rights reserved.
1. Introduction
that stabilizes the resulting square-planar d8 transition metal complex.
An extra advantage is that the P atom in the PR3 fragment can be used
Cisplatin combination chemotherapy is the cornerstone of treatment
for many cancers. Cisplatin and the follow-on drugs carboplatin and
oxaliplatin are used to treat 40–80% of cancer patients [1]. The high effec-
tiveness of cisplatin in the treatment of several types of tumors is
severely hindered by some clinical problems such as normal tissue toxic-
ity and the frequent occurrence of initial and acquired resistance to
the drug [1–5]. A growing number of so-called cisplatin rule-breaker
compounds have been reported [3–6] along with some Pt(IV) pro-
drugs that can be photoactivated by visible light [7–8]. These findings
have paved the way for studying other metal-based chemotherapeutic
compounds [3] including relevant examples of gold [9], palladium [10],
titanium [11], ruthenium [12], osmium [13] and iron complexes [14].
We and others have reported on the biological [15–18] and catalytic
[19–21] applications of iminophosphorane or iminophosphine deriva-
tives of gold(III) (selected examples in Fig. 1) [15,16]. The main advantage
of iminophosphorane ligands is that they provide a C,N- or N,N-backbone
as a “spectroscopic marker” to study the in vitro stability (and oxidation
state) by 31P{1H} NMR [16]. Gold(III) organometallic complexes (with a
pincer C,N-iminophosphorane ligand) displayed a high cytotoxicity in
vitro against human ovarian cancer and leukemia cell lines while
being less toxic to normal T-lymphocytes by a mode of action different
from that of cisplatin [15,16]. The compounds did not interact with DNA
[15]. ROS (reactive oxygen species) production at the mitochondrial
level was a critical step in the cytotoxic effect of these compounds
[15]. We wanted to increase the hydrophilicity of this type of IM
(iminophosphorane) ligands and also to explore the biological activity
of other d8 metal (Pd(II) and Pt(II)) IM complexes. We envisioned
an iminophosphorane N,N- chelating ligand as an ideal platform to con-
fer stability to d8 transition metal centers while simplifying the synthet-
ic steps required to prepare organometallic compounds (avoiding
transmetallation for metals not prone to easy C―H bond activation).
We had used such a strategy to prepare gold(III) coordination catalysts
for C―C and C―heteroatom bond formation [20]. The challenge here
was the incorporation of a water-soluble phosphine in the skeleton of
the ligand since it is well known that IMs in contact with water or
⁎
Corresponding author. Fax: +1 7189514607.
E-mail address: mariacontel@brooklyn.cuny.edu (M. Contel).
0162-0134/$ – see front matter © 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2012.06.017

M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
205
3
2
polar solvents break down to the corresponding amine and phosphine
oxide [22].
8 Hz, C(O)CCHCH), 7.37 (1H, t, JHH 7 Hz, NCHCH), 4.35 (6H, d, JPH
8 Hz, PCH2N), 4.42 (AB system, 6H, NCH2N). 13C{1H} NMR (CDCl3): δ
177.7 (s, C O), 152.0 (s, C(O)C(N)(CH)), 149.1 (s, NCH), 136.7 (s, C(O)
CCHCH), 125.4 (s, NCHCH), 123.5 (s, C(O)CCH), 72.8 (d, JPC 9 Hz,
NCH2N), 52.7 (d, JPC 46 Hz, PCH2N). IR (cm−1): ν 1582 (C O), 1360 (P N).
[AuCl2(N,N-IM)]ClO4 (2) to K[AuCl4] (0.057 g, 0.15 mmol) in dry
MeCN (10 mL), AgClO4 (0.068 g, 0.33 mmol) in dry MeCN (2 mL)
was added and the reaction mixture was stirred in the dark for
30 min, after which it was filtered through Celite (to remove AgCl).
1 (0.042 g, 0.15 mmol) in CH2Cl2 (1 mL) was added and the yellow
solution became instantly brown. After 50 min stirring, the reaction
mixture was again filtered through celite (to remove KClO4) and
then the solvent was reduced in vacuo to a minimum. Upon addition
of Et2O (10 mL), a red-brown solid was obtained, which was washed
with MeCN (b1 mL at a time) and Et2O and dried in vacuo. Yield:
By the use of a picolinamide and water-soluble TPA (1,3,5-
triaza-7-phosphaadamantane) phosphine through the Pomerantz
method [23], we prepared a stabilized water-soluble N,N-chelating
iminophosphorane ligand TPA_N―C(O)―2-NC5H4 1 (Scheme 1).
To the best of our knowledge, this is the first water-soluble
iminophosphorane reported. We describe here the preparation
(Scheme 1), characterization, stability in solution and study of the
cytotoxic properties in vitro of its coordination complexes containing
Au(III), Pd(II) and Pt(II) centers. The study of the interactions of
these complexes with different biomolecules (DNA and human
serum albumin HSA) is also reported and compared to that of
cisplatin.
2. Experimental section
0.050 g (52%). MS (HR-ESI+) [m/z]: 544.0126 [M]+ 31P{1 H} NMR:
.
δ 6.5 (s, DMSO-d6), 5.7 (s, CD3CN). 1H NMR (CD3CN): δ 9.39 (1H, d,
3
2.1. Materials and methods
3JHH 6 Hz, NCH), 8.60 (1H, t, JHH 8 Hz, C(O)CCHCH), 8.19 (2H, m,
2
C(O)CCH+NCHCH), 5.08 (6H, JPH 10 Hz, PCH2N), 4.52 (AB system,
13
All manipulations involving air-free syntheses were performed using
standard Schlenk-line techniques under an argon atmosphere or in a
glove-box MBraun MOD System. Solvents were purified by use of
a PureSolv purification unit from Innovative Technology, Inc. The phos-
phine substrate TPA was purchased from Sigma-Aldrich, K[AuCl4], PtCl2
and AgClO4 were purchased from Strem chemicals and used without fur-
ther purification. [PdCl2(COD)] [24] and [AuCl2(HN―C(O)―2-NC5H4)] 5
[25] were prepared as previously reported. NMR spectra were recorded
in a Bruker AV400 (1H NMR at 400 MHz, 13C NMR at 100.6 MHz, 31P
NMR at 161.9 MHz). Chemical shifts (δ) are given in ppm using CDCl3,
d6-DMSO or D2O as solvent, unless otherwise stated. 1H and 13C chemi-
cal shifts were measured relative to solvent peaks considering TMS
(tetramethylsilane)=0 ppm; 31P{1H} was externally referenced to
H3PO4 (85%). Infrared spectra (4000–250 cm−1) were recorded on a Ni-
colet 6700 FT-IR spectrophotometer on KBr pellets. Elemental analyses
were performed on a Perkin Elmer 2400 CHNS/O Analyzer, Series II.
Mass spectra HR-ESI (high-resolution electrospray ionization) or
MALDI (matrix-assisted laser desorption/ionization) were performed
on an Agilent Analyzer or a Bruker Analyzer. Conductivity was measured
in an OAKTON pH/conductivity meter in CH3CN solution (10−3 M). Cir-
cular Dichroism spectra were recorded using a Chirascan CD Spectrome-
ter equipped with a thermostated cuvette holder. Electrophoresis
experiments were carried out in a Bio-Rad Mini sub-cell GT horizontal
electrophoresis system connected to a Bio-Rad Power Pac 300 power
supply. Photographs of the gels were taken with an Alpha Innotech
FluorChem 8900 camera. Fluorescence intensity measurements were
carried out on a PTI QM-4/206 SE Spectrofluorometer (PTI, Birmingham,
NJ) with right angle detection of fluorescence using a 1 cm path length
quartz cuvette. ITC measurements were carried out at 25.0 ( 0.2) °C
on a MicroCal VP-ITC calorimeter.
6H, NCH2N).
C{1H} NMR (CD3CN): δ 146.1 (s, C(O)CCHCH), 132.2
(s, NCHCH), 130.4 (s, C(O)CCH), 71.2 (d, JPC 9 Hz, NCH2N), 54.5 (d,
JPC 37 Hz, PCH2N) (unable to identify remaining carbons due to poor
signal to noise ratio because of the low solubility and stability of com-
pound). IR (cm−1): ν 1654 (C O), 1292 (P N), 1089 (ClO4), 254 and
302 (M-Cl). Conductivity: Λ=97.8 μS/cm.
[PdCl2(N,N-IM)] (3) [PdCl2(COD)] (0.041 g, 0.15 mmol) and 1
(0.042 g, 0.15 mmol) were dissolved in dry and degassed CH2Cl2
(4 mL) and left to react for 3 h at room temperature after which an or-
ange precipitate formed. The solvent was then reduced to a minimum
in vacuo and upon addition of Et2O a pale orange solid was obtained
which was washed twice with CHCl3 and dried in vacuo. Yield: 0.035 g
(52%). Anal. Calcd for C12H16Cl2N5OPPd (454.59): C, 31.71; H, 3.55; N,
15.41. Found: C, 31.63; H, 3.23; N, 15.41. MS (MALDI+) [m/z]: 420.0
[M-Cl]. 31P{1H} NMR(DMSO-d6): δ −6.4 (s). 1H NMR (DMSO-d6): δ
3
3
9.03 (1H, d, JHH 5 Hz, NCH), 8.20 (1H, t, JHH 7 Hz, C(O)CCHCH), 7.85
3
3
(1H, d, JHH 8 Hz, C(O)CCH), 7.78 (1H, t, JHH 7 Hz, NCHCH), 4.39 (AB
system, 6H, NCH2N), 4.42 (6H, JPH 8 Hz, PCH2N). 13C{1 H} NMR
2
(DMSO-d6): δ 173.8 (s, C O), 150.5 (s, C(O)C(N)(CH)), 149.1 (s, NCH),
141.4 (s, C(O)CCHCH), 129.6 (s, NCHCH), 127.1 (s, C(O)CCH), 71.4 (d,
JPC 9 Hz, NCH2N), 53.7 (d, JPC 37 Hz, PCH2N). IR (cm−1): ν 1659 (C O),
1290 (P N), 290 (M-Cl).
[PtCl2(N,N-IM)] (4) PtCl2 (0.080 g, 0.30 mmol) and 1 (0.083 g,
0.30 mmol) were refluxed for 2 days in a mixture of CH2Cl2 (10 mL)
and acetone (2 mL). The reaction mixture was filtered though Celite
to remove any unreacted PtCl2, giving a yellow solution. The solvent
was reduced to a minimum in vacuo and upon addition of Et2O a yel-
low solid was obtained, which was washed twice with CHCl3 and
dried in vacuo. Yield: 0.025 g (15%). Anal. Calcd for C12H16Cl2N5OPPt
(542.01): C, 26.53; H, 2.97; N, 12.89. Found: C, 25.90; H, 2.48; N,
13.16. MS (MALDI+) [m/z]: 502.1 [M-Cl]. 31P{1H} NMR(DMSO-d6):
3
2.2. Synthesis
δ −4.4 (s). 1H NMR (DMSO-d6): δ 9.35 (1H, d, JHH 5 Hz, NCH), 8.33
3
3
(1H, t, JHH 8 Hz, C(O)CCHCH), 7.83 (2H, d, JHH 8 Hz, C(O)
2
TPA_N-C(O)-2-NC5H4 (N,N-IM) (1) TPA (0.157 g, 1.0 mmol) and
picolinamide (0.122 g, 1.0 mmol) were placed in a Schlenk flask under
nitrogen. Dry, degassed THF tetrahydrofuran (15 mL) was added and to
this solution, tBuDAD N,N-bis(tert-butyl)1,4-diazabutadiene (0.230 g,
1.0 mmol) in dry and degassed THF (4 mL) was added dropwise at
0 °C. The reaction was left stirring at RT for 15 h. After this period, the
yellow reaction mixture was colourless with some white precipitate.
The solvent was reduced in vacuo to a minimum (b2 mL) and Et2O
(10 mL) was added, giving a white solid that was filtered and dried in
vacuo. Yield: 0.254 g (93%). Anal. Calcd for C12H16N5OP (277.11): C,
51.98; H, 5.82; N, 25.26. Found: C, 51.76; H, 5.79; N, 25.24. MS (ESI+)
CCH+NCHCH), 4.95 (6H, d, JPH 8 Hz, PCH2N), 4.42 (6H, s, NCH2N).
13C{1H} NMR (DMSO-d6): δ 173.8 (s, C O), 151.1 (s, C(O)C(N)(CH)),
150.0 (s, NCH), 141.9 (s, C(O)CCHCH), 130.6 (s, NCHCH), 128.0 (s,
C(O)CCH), 71.3 (d, JPC 7 Hz, NCH2N), 55.1 (d, JPC 50 Hz, PCH2N). IR
(cm−1): ν 1667 (C O), 1282 (P N), 290 (M-Cl).
2.3. X-ray crystallography
Single crystals of 3 and 4 (see details below) were mounted on a
glass fiber in a random orientation. Data collection was performed
at room temperature on a Kappa CCD diffractometer using graphite
monochromated Mo-Kα radiation (λ=0.71073 Å). Space group assign-
ments were based on systematic absences, E statistics and successful
refinement of the structures. The structures were solved by direct
[m/z]: 278.12 [M]+
.
31P{1H} NMR: δ −29.4 (s, CDCl3), −31.4 (s,
3
DMSO-d6), −24.8 (s, D2O). 1H NMR (CDCl3): δ 8.68 (1H, d, JHH
3
3
4 Hz, NCH), 8.10 (1H, d, JHH 8 Hz, C(O)CCH), 7.75 (1H, t, JHH

206
M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
methods with the aid of successive difference Fourier maps and were
refined using the SHELXTL 6.1 software package. All non-hydrogen
atoms were refined anisotropically. Hydrogen atoms were assigned to
ideal positions and refined using a riding model. Details of the crystallo-
graphic data are given in Table S1 (Supplementary Information (SI)).
These data can be obtained free of charge from The Cambridge Crystal-
lographic Data Center via www.ccdc.cam.ac.uk/data_request/cif. (CCDC
858692 for compound 3, and 858693 for compound 4).
3: Crystals of 3 (orange prisms with approximate dimensions
0.28×0.24×0.24 mm) were obtained from a solution of 3 in DMSO
by slow diffusion of Et2O at RT. 4: Crystals of 4 (yellow prisms with
approximate dimensions 0.26×0.24×0.20 mm) were obtained from
a solution of 4 in DMSO by slow diffusion of Et2O at RT.
of the MTT-reduction method [28]. Total cell number and cell viabil-
ity were determined by the Trypan-blue exclusion test. Apoptosis/
necrosis hallmarks were analyzed by simultaneously measuring the
mitochondrial membrane potential, exposure of phosphatydilserine
and 7-AAD uptake. In brief, 2.5×105 cells in 200 μL were incubated in
ABB (140 mM NaCl, 2.5 mM CaCl2, 10 mM Hepes/NaOH, pH 7.4), with
either 5 nM DiOC6(3) or 60 nM TMRE (both from Molecular Probes) at
37 °C for 10 min. AnnexinV-PE or AnnexinV-FITC (Invitrogen) at a con-
centration of 0.5 μg/mL or 7-AAD (50 ng/μL) was added to samples and
incubated at 37 °C for an additional 15 min. In all cases, cells from each
well were diluted to 1 mL with ABB to be counted by flow cytometry
(FACScan, BD Bioscience, Spain).
2.6. Interaction of metal complexes with CT DNA by circular dichroism
spectroscopy
2.4. Computational methods
All calculations reported here were carried out using the Gaussian 09
program package [26]. All gas phase structures were optimized at the
B3LYP/6-31G* level of density functional theory (DFT) in combination
with relativistic effective-core potential cc-pVDZ-PP for heavy metals
(platinum and gold) where appropriate. Aqueous solution structures
are found with PCM (polarizable continuum model) full geometry
optimization at the B3LYP/6-31G* level of DFT in combination
with relativistic effective-core potential cc-pVTZ-PP for heavy metals
(platinum and gold) where appropriate. Computed frequencies were
positive, in all cases, indicating that the optimized structures represent
the minima of ground-state potential-energy surfaces. In addition,
a simple visual comparison of computed IR spectra with experimen-
tal spectra was performed to further validate reliability of computed
structures.
Stock solutions (5 mM) of each complex were freshly prepared in
DMSO-H2O (1:99) prior to use. Aliquots were added to a solution of
CT (calf thymus) DNA (195 μM) also freshly prepared in buffer
(5 mM Tris/HCl, 50 mM NaClO4, pH=7.39) to achieve molar ratios
of 0.1, 0.25, 0.5, 1.0 drug/DNA while keeping a total volume of 3 mL.
The samples were incubated at 37 °C for 20 h in the dark. All CD spec-
tra of DNA and of the DNA–drug adducts were recorded at 25 °C over
a range 220–320 nm and finally corrected with a blank and noise
reduction. Noise reduction was performed using the Savitzky–Golay
smoothing filter incorporated in the Chirascan software. The spectra
are given in molar ellipticity (millidegrees).
2.7. Interaction of metal complexes with plasmid (pBR322) DNA by elec-
trophoresis (mobility shift assay)
2.5. Determination of compound cytotoxicity
10 μL aliquots of pBR322 plasmid DNA (20 μg/mL) in buffer
(5 mM Tris/HCl, 50 mM NaClO4, pH=7.39) were incubated with
molar ratios between 0.25 and 4.0 of the compounds at 37 °C for
20 h in the dark. Samples of free DNA and cisplatin-DNA adduct
were prepared as controls. After the incubation period, 2 μL of loading
dye were added to the samples and of these mixtures, 7 μL were final-
ly loaded onto the 1% agarose gel. The samples were separated by
electrophoresis for 1.5 h at 80 V in Tris-acetate/EDTA buffer (TAE).
Afterwards, the gel was stained for 30 min with a solution of GelRed
Nucleic Acid stain.
The human T-cell leukemia Jurkat (clone E6.1) from the ATCC
collection and the prostate carcinoma DU-145 were routinely cul-
tured in RPMI 1640 medium supplemented with 5% FCS (fetal calf
serum), L-glutamine and penicillin/streptomycin (hereafter, com-
plete medium). Blood samples from healthy donors were obtained
from the Banco de Sangre y Tejidos de Aragón. All subjects gave writ-
ten informed consent and the Ethical Committee of Aragón approved
the study. Normal T-lymphocytes (peripheral blood mononuclear
cells: PBMC) were freshly isolated by Ficoll-Paque density centrifu-
gation. After isolation, normal PBMC were kept in RPMI 1640 medium
supplemented with 10% decomplemented FCS, L-glutamine and penicil-
lin/streptomycin. Jurkat cells lacking Bak were generated using RNA
interference techniques, as described previously [27]. For toxicity
assays, Jurkat cells (5×105 cells/mL), DU145 cells (1×105 cells/
mL) or normal T-lymphocytes PBMC (3×106 cells/mL) were seeded
in flat-bottom 96-well plates (100 μL/well) in complete medium.
DU145 cells were allowed to attach prior to addition of cisplatin or
tested compounds. Compounds were added at different concentra-
tions in triplicate. Cells were incubated with cisplatin or compounds
for 24 h and then cell proliferation was determined by a modification
2.8. Interaction of metal complexes with HSA by fluorescence
spectroscopy
The excitation wavelength was set to 295 nm, and the emission
spectra were recorded at room temperature in the range of 300 to
450 nm. The fluorescence intensities of the metal compounds, the
buffer and the DMSO are negligible under these conditions, and so
is the effect of additions of pure DMSO on the fluorescence of HSA.
An 8 mM solution of each compound in DMSO was prepared and
ten aliquots of 2.5 μL were added successively to a solution of HSA
(10 μM) in phosphate buffer (pH=7.39), the fluorescence being
Fig. 1. Examples of selected iminophosphorane organogold(III) complexes with cytotoxic properties prepared by our group [13]. ROS production at the mitochondrial level was a
critical step in the cytotoxic effect of these compounds [13a].

M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
207
Scheme 1. Synthesis of d8 metal coordination complexes containing new chelating ligand TPA_N―C(O)―2-NC5H4 (N,N―IM) 1.
measured 240 s after each addition. The data were analyzed using the
classical Stern–Volmer equation F0/F=1+KSV[Q].
binding reversibility studies, to a 2 mL sample of saturated HSA-3 or
HSA-4 incubated previously for 20 h at 37 °C, was added either
20 μL of concentrated HCl to reach pH 4.00 or the appropriate amount
of a 60 mM solution of citric acid in the same buffer to reach a 1:3
complex:chelator ratio. The CD spectra were recorded in the visible
region after 1 and 20 h.
2.9. Interaction of metal complexes 3 and 4 with HSA by isothermal titra-
tion calorimetry
The HSA and titrant solutions were prepared in phosphate buffer
(pH=7.39) and degassed prior to titration. In a standard experiment,
the protein (5 μM) was titrated with the metal complex solution
(1.5–3 mM) by a series of 34 successive injections of 7 μL each, with
an interval of 240 s between injections. A background titration of
titrant solution into buffer was subtracted from each experimental
titration to account for heat of dilution. The data were analyzed
with a two-site binding model using Origin 50 software supplied
with the microcalorimeter. Equilibrium was reached after each addi-
tion of ligand to protein during the ITC titrations as evidenced by the
return to baseline of the trace of microcalories heat released per unit
time in the raw data collected for the binding experiments. Concen-
trations are not directly measured but calculated based upon the
starting total protein concentration and the actual titration curve gen-
erated from the integrated heat released for each injection of ligand
solution of known concentration into the known volume of protein
solution in the ITC cell.
3. Results and discussion
3.1. Synthesis, characterization and stability of the new compounds
The ligand 1 and metallic compounds (2–3) can be obtained in mod-
erate to high yields following the procedure depicted in Scheme 1. All
the new compounds (1–4) were fully characterized (see Experimental
section) and are air-stable red (2), orange (3) and yellow (4) solids.
The structure of cationic Au compound 2 was confirmed by IR (bands
that can be assigned to anionic ClO4−), conductivity measurements
(1:1 electrolyte) and high resolution mass spectrometry. The stabilities
of the ligand and the complexes were evaluated by 31P{1H} and 1H NMR
spectroscopy in deuterated solvents and mixtures of deuterated sol-
vents. Ligand 1 is soluble in H2O, and solvents such as DMSO and
CHCl3. It is stable in DMSO solution at room temperature for 14 days
whereas in H2O it is stable for 3 h.
The cationic Au derivative 2 is soluble in CH3CN and DMSO and insol-
uble in chlorinated solvents, Et2O, most non-polar organic solvents and
H2O. At room temperature it is stable in CH3CN solution for 24 h while
it decomposes in DMSO solution after 4 h. In a mixture 50:50 DMSO:
H2O compound 2 decomposes after 30 min to phosphine oxide
TPA_O and to compound 5 (Eq. (1)). Compound 5 (containing
deprotonated uninegative bidentate ligand picolinamide) had been pre-
viously reported as a cytotoxic agent against MOLT-4 (human leukemia)
and C2C12 (mouse myoblast cancer) cell lines [25]. Neutral complexes 3
and 4 are soluble in DMSO in which they are stable for over 7 days and
they are also stable in mixtures 50:50 DMSO:H2O for over 4 days.
2.10. Reversibility of binding to HSA by circular dichroism
The interaction of complexes 3 and 4 with HSA was studied by CD
spectroscopy in the visible region (300–550 nm) in phosphate buffer
(pH=7.39). 2 mL samples of 1.12×10−4 M HSA were incubated at
37 °C for 20 h with increasing amounts of compound 3–4 (1–7 eq);
the interaction was monitored by the appearance of a band in the vis-
ible region at 360 nm for 3 and 395 nm for 4. All spectra were
corrected by subtraction of a blank consisting of a solution of drug
in buffer at the appropriate concentration for each sample. For
Equation 1. Decomposition products of compound 2 in DMSO solution or a 50:50 mixture of DMSO:H2O at RT after 4 h or 30 min respectively.

208
M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
Fig. 2. Molecular structure of the compound [PdCl2(N,N-IM) ] 3 with the atomic numbering scheme. The structure for compound [PtCl2(N,N-IM) ] 4 is analogous and selected bond
and angles for 3 and 4 are collected in Table 1.
Compounds 2, 3 and 4 are soluble at micromolar concentration in mix-
tures 1:99 DMSO:H2O or DMSO:buffer.
(N(2)―Au(1)―N(1) angle of 80.62°). The distances calculated for
Au―N(1) (Au―N(iminic)) and Au―N(2) (Au-N(pyridine) and
Au―Cl are in the range of those found for related complexes like
[Au(Ph2PyP_NC(O)Ph)Cl2]ClO4 [18] and those obtained for previously
described compound 5 [AuCl2(HN―C(O)―2-NC5H4)] [25]. The longer
distances P―N(1) of 1.700 Å and N(1)―C(1) of 1.392 Å in 2 (compared
to those in 3 and 4) are indicative of a greater delocalized charge density
[30] than in neutral Pd and Pt compounds (3 and 4). This may be the
cause for the lower stability of compound 2 in water or in polar solvents
compared to complexes 3 and 4.
Frequencies of selected normal vibrational modes were also deter-
mined by DFT methods (Table S2) for the optimized structures of
compounds 2–4. The calculated and the experimental frequencies of
M―Cl and P N and C O stretching modes are in reasonable agreement
[32].
The crystal structures of 3 and 4 (Structure of 3 in Fig. 2) were
compared to the molecular structures obtained by means of density
functional calculations (see Experimental section and Supplementary
Material). Selected bond lengths and angles are collected in Table 1
and in Table S2. The geometry about the Pd(II) and Pt(II) centers is
pseudo-square planar with the N(2)–M(1)–N(1) angle of 80.86(8)°
(Pd 3) and 80.4(3)° (Pt 4) suggesting a rigid ‘bite’ angle of the chelat-
ing ligand. The Pd and Pt centers are on an almost ideal plane with
negligible deviations from the least-squares plane. The coordination
plane for Pd and Pt and the metallocyclic plane are only slightly twist-
ed to give an angle of 2.6° (3) and 2.4° (4) between them. The main
distances M–N(1) or M–N(iminic) and M–N(2) found for 3 and 4
are very similar to those found in related iminophosphorane com-
plexes like [Pd {k2-C,N-C6H4(PPh2_NC6H4Me-4′)-2}{μ-OAc)}]2,
[Pd{k2―C,N―C6H4(PPh2 NC6H4Me-4′)-2}(tmeda)]ClO4 [29], [Pd(C6H4
CH2NMe2)(Ph3PNC(O)―2-NC5H4)]ClO4 [30] and [PtCl(PMe2Ph)
{(N(p-Tol)_PPh2)2CHCH3}]Cl [31]. We were unable to get crystals
from the Au(III) compound 2 useful for a single crystal X-ray dif-
fraction study but we got good estimates for its structural parame-
ters from density functional calculations (see Table 1 and Table S2).
The molecular structure of the cation in the Au(III) compound
2 is depicted in Fig. 3. Like in the Pd(II) and Pt(II) complexes 3
and 4, the metal center is in a pseudo-square planar arrangement
3.2. Cytotoxicity studies
Cytotoxicity results for ligand 1 and the new metallic compounds
(2–4) are collected in Table 2. Values for the decomposition products
of ligand 1 and compound 2 (the phosphine oxide TPA_O and gold com-
pound 5, Eq. (1)) are also collected for comparison purposes. The
cytotoxicity (by a modification of the MTT (3-(4,5-dimethulthiazol-2-yl)-
2,5-diphenyltetrazolium bromide)-reduction method, see Experimental
section) was evaluated against two selected cell lines: human Jurkat-T
acute lymphoblastic leukemia cells and DU-145 human prostate cancer
cells. Cells were incubated in the presence of the compounds for 24 h.
The sensitivity of T-cell leukemia Jurkat cells to compounds cisplatin, 1–
5, and TPA_O was compared to that of normal T-lymphocytes (PBMC).
Table 1
Selected structural Parameters of complexes 2–4 obtained from DFT calculations and
the comparison with experimental values.a Bond lengths in Å and angles in °. (Full
Table S2 in supplementary material).
2
3
4
Calcd
Exp
Calcd
Exp
Calcd
M-N(1)
2.097
2.073b
2.086
2.064b
2.304
2.321b
2.301
2.309b
1.700
1.694b
1.392
1.390b
80.62
80.84b
88.88
88.30b
2.052(2)
2.112
2.083b
2.069
2.049b
2.311
2.349b
2.329
2.342b
1.663
1.678b
1.377
1.378b
79.54
80.59b
90.52
88.86b
2.058(6)
2.098
2.076b
2.040
2.031b
2.326
2.358b
2.339
2.354b
1.667
1.681b
1.379
1.379b
79.64
80.39b
89.46
87.92b
M-N(2)
2.024(2)
2.2943(8)
2.2922(8)
1.658(2)
1.373(3)
80.86(8)
88.74(3)
2.004(7)
2.300(2)
2.299(2)
1.661(7)
1.366(11)
80.4(3)
M-Cl(1)
M-Cl(2)
P-N(1)
N(1)-C(1)
N(1)-M-N(2)
Cl(1)-M-Cl(2)
88.17(9)
a
Gaussian 09 program package (see Experimental section), calculated in gas phase.
Calculated in PCM aqueous solution (see Experimental section).
Fig. 3. Optimized structure of the cation in [AuCl2(N,N-IM)]ClO4 (2) obtained from DFT
calculations.
b

M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
209
Table 2
the cells is very similar to that of the controls (Fig. 4, selected exam-
ples Au (5) and Pt (4) compounds), with neither signs of nuclear
condensation nor chromatin fragmentation. Also, DU-145 showed
not phosphatydilserine exposure or membrane permeabilization
(Fig. 5), confirming that compounds 4 and 5 do not induce cell
death in DU-145 at 24 h. In contrast, analysis of nuclear morphology
revealed that compounds 4 and 5 induced cell death in Jurkat cells.
Both apoptotic, condensed and fragmented nuclei, and necrotic, swollen
nuclei were observed (Fig. 4). Flow cytometry after AnnexinV-APC/
7-AAD staining (Fig. 5) confirmed that a significant percentage of cells
were necrotic (7-AAD+ cells, upper right quadrant). Thus the mode of
action of the gold compound [AuCl2(HN-C(O)-2-NC5H4)] 5 and new
hydrophilic iminophosphorane platinum complex 4 in Jurkat cells
seems different from that of cisplatin (which is via apoptosis) [34]. It
is estimated that about 1% of intracellular cisplatin binds to DNA pri-
marily by forming intrastrand cross-links between adjacent purines
[3,35] and triggers the DNA-damage response that activates canonical
apoptotic pathways [36].
We found that iminophosphorane gold(III) complexes a and b
(Fig. 1) induced both necrosis and apoptosis and that the necrotic cell
death induced by a and b was Bax/Bak- and caspase-independent
[15]. We demonstrated that compounds a and b act mainly through
ROS generation that leads to necrosis [15]. Similarly, compounds 4
and 5 seem to induce both necrotic and apoptotic cell death in Jurkat
leukemia cells. Necrotic cell death that we observed in a fraction of
cells could correspond to necroptosis, a “programmed” necrosis, since
it has been demonstrated by others that ROS accumulation can be
implicated in this type of cell death [37].
IC50 (μM) of cisplatin, ligand 1, phosphine oxide TPA_O and metal complexes 2–5
complexes in human cell lines. All compounds were dissolved in 1% of DMSO and dilut-
ed with water before addition to cell culture medium for a 24 h incubation period. Cis-
platin was dissolved in water.
Jurkat
PBMC
DU-145
Cisplatin
TPA_O
7.4 2.1
>500
>500
0.66 0.07
19.6 2.6
3.5 0.7
0.97 0.4
500 18
>500
>500
7.9 2.9
213 22
23.4 0.9
7.3 3.3
79.2 15.8
>500
>500
17.4 5.1
301 27
65.3 7.0
19.9 1.7
1
2
3
4
5
Compound 3 (Pd) was less cytotoxic than cisplatin for Jurkat cells but Pt
(4) and Au compounds 2 and 5 were two (4), seven (5) or even eleven
(2) times more cytotoxic than cisplatin (Table 2). The IC50 values
displayed by the gold compounds are similar to that reported for an
organogold(III) iminophosphorane complex described by us (b in Fig. 1
[16]) to Jurkat leukemia cells (0.2 μM) and CLL (B-cell chronic lymphocyt-
ic leukemia cells) derived from patients; 60 nM to 100 nM) [15]. Com-
pound 5 was reported to have an IC50 value of 3.1 μM for MOLT-4
human leukemia cell line [25]. All compounds were less toxic to PMBC
than to T-Jurkat cell lines. The compounds were twelve (2), ten (3) or
seven times (4, 5) less toxic to the normal lymphocytes than to the leuke-
mia cell line. However we have found here that cisplatin is 70 times less
toxic to PMBC than to Jurkat cell lines.
The cytotoxicity to the DU-145 human prostate cancer cell line
was evaluated and while Pd (3) was less cytotoxic than cisplatin for
this cell line, the Pt (4) compound (65.3 μM) had an IC50 value similar
to that of cisplatin (79.2 μM). Au compounds 2 and 5 were more
cytotoxic (17.4 μM (2) and 19.9 μM (5), 4.5 and 4 times more cyto-
toxic than cisplatin). The value obtained for compounds 2 and 5 is
in line with values obtained with heterometallic Ti-Au2 complexes
recently described by our lab (14–27 μM) with a live imaging method
after 24 h [33].
The activity of compounds 3–5 and cisplatin against apoptosis-
resistant cells was also tested (Fig. S1). Multidomain proapoptotic
proteins Bax and Bak are essential for the onset of mitochondrial perme-
abilization and apoptosis through the intrinsic pathway. Interestingly,
we observed that compounds 4 and 5 exhibited some toxicity to Jurkat
sh-Bak cells (Bax/Bak-deficient Jurkat cells) similar to previous results
with iminophosphorane–organogold(III) complexes (compounds
a
and b in Fig. 1) [15]. This could be an advantage over cisplatin, whose ac-
In addition, the cytotoxicity of ligand 1 and its decomposition
product phosphine oxide TPA_O is notably low in all cell lines stud-
ied having IC50 values greater than 500 μM. This indicates that the cy-
totoxic effect is due to the metal complex and that the cytotoxicity on
the cancer cell lines follows the order Au>Pt>Pd for the metal cen-
ter. Compound 2 decomposes quickly in solution to compound 5
and therefore the IC50 values obtained for both derivatives are very
similar.
tivity is totally dependent on the presence of functional Bax and Bak.
3.3. Interaction with DNA
Since DNA replication is a key event for cell division, it is among
critically important targets in cancer chemotherapy. Most cytotoxic
platinum drugs form strong covalent bonds with the DNA bases
[38,39]. However, a variety of platinum compounds act as DNA
intercalators upon coordination to the appropriate ancillary ligands
[40]. There are also reports on palladium derivatives interacting
with DNA in covalent [41,42] and noncovalent ways [43–45]. While
Preliminary results on the cell death pathway caused by com-
plexes 2–5 showed an inhibition of cell growth in the period studied
(24 h) in DU-145 human prostate cancer cells. The morphology of
Fig. 4. Compounds 3–5 induce apoptosis and necroptosis. Jurkat and DU-145 cells were cultured for 24 h in the presence of compounds 3–5 or left untreated. Nuclei were stained
with Hoechst 333248 (10 μg/mL) and cells were photographed under UV light. The figure shows the images of controls and selected compounds 4 and 5.

210
M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
Fig. 5. Jurkat cells (5×105 cells/mL) or DU145 cells (1×105 cells/mL) were treated for 24 h with compounds 4 or 5 (concentrations for IC50 values). Apoptosis/necrosis hallmarks
were analyzed by simultaneously measuring the exposure of phosphatidylserine and 7-Amino-Actinomycin D (7-AAD) uptake as described in Experimental section. The percentage
of cells in each region is indicated.
most gold(III) and gold(I) compounds display reduced affinity for
DNA [9,15,16], gold(I)-phosphine derivatives with weakly bound li-
gands (such as halides) bind in a nondenaturing fashion to DNA
[35,46]. DNA conformational alterations can be detected by means
of circular dichroism spectroscopy [37]. When Calf Thymus DNA is in-
cubated with increasing amounts of Au derivative (5), a slight decrease
(Fig. 6) of the intensities of the negative and especially the positive
bands are observed, indicating a decrease in helicity (destabilizing effect)
[47]. Those changes in the stacking and the helicity of CT DNA at low
drug:nucleotide ratios (r=0.1/0.5) are probably related to electrostatic
effects or to the ability of the compounds to intercalate DNA. Similar
CD spectra were observed with other cationic square-planar gold(III)
compounds with N-containing ligands such as [Au(phen)Cl2]Cl or
[AuCl(terpy)]Cl2 [48]. Au(III) complexes structurally related to
compound 5 derived from aminoquinoline such as [Au(Quinpy)Cl]
Cl, [Au(Quingly)Cl]Cl and [Au(Quinala)Cl]Cl were proposed to inter-
act by intercalation, although no CD spectra were provided [49].
In contrast, when CT DNA is incubated for 20 h with increasing
amounts of Pd (3), and particularly Pt (4) there are not noticeable
changes at low DNA–drug ratios (up to r=0.5 or even 1 for 4) and
only at ratios of 1 is there a small increase of the positive band indicating
a small stabilizing effect most plausibly by electrostatic effects (Fig. S2).
We also performed agarose gel electrophoresis studies on the
effects of compounds 3–5 on plasmid (pBR322) DNA (Fig. 7). This
plasmid has two main forms: OC (open circular or relaxed form,
Form II) and CCC (covalently closed or supercoiled form, Form I).
Changes in electrophoretic mobility of both forms are usually taken
as evidence of metal-DNA binding. Generally, the larger the retardation
of supercoiled DNA (CCC, Form I), the greater the DNA unwinding
produced by the drug [50]. Binding of cisplatin to plasmid DNA, for
instance results in a decrease in mobility of the CCC form and an
increase in mobility of the OC form (see lanes a, b and c for cisplatin
in Fig. 7) [51]. Treatment with increasing amounts of ligand 1 does
not cause any shift for either form, consistent with no unwinding or
other change in topology under the chosen conditions. Treatment with
increasing amounts of compounds 3–5 retards the mobility of the
faster-running supercoiled form (Form I) especially at high molar ratios.
Similar patterns of mobility have been described for some Pt(IV) com-
plexes in the presence of glutathione as reducing agent [51]. In these
cases, although the resulting Pt(II) products bind to closed circular DNA,
their effect in the mobility of Form I (or CCC) DNA is different from that
produced by cisplatin [51]. These results are in agreement with the
small stabilizing effects observed in the interaction of the Pd (3) and Pt
(4) compounds with calf thymus DNA by CD spectroscopy. These effects
are different from those observed in the CD spectra of cisplatin and other
Pd and Pt derivatives which bind in covalent cis-bidentate fashion to
DNA [33].
In conclusion, the experiments probing DNA–drug interactions
showed that new Pd (3) and Pt (4) IM complexes do not interact
strongly with Calf Thymus DNA at physiological pH in vitro and that
Fig. 6. CD spectra of CT incubated for 24 h at 37 °C with Au derivative (5) at 0, 0.1, 0.25,
0.5 and 1 ratios. The ordinate indicates CD intensity (ellipticity) in millidegrees.

M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
211
Fig. 7. Electrophoresis mobility shift assays for cisplatin and compounds 1, 3–5 (see Experimental section for details). DNA refers to untreated plasmid pBR322. a, b, c and d corre-
spond to metal/DNA ratios of 0.25, 0.5, 1.0 and 2.0 respectively.
gold compound (5) interacts more strongly with DNA most likely by
intercalation. All the complexes interact with plasmid (pBR322)
DNA, most plausibly in a way different from that of cisplatin.
being the major contributor to the intrinsic fluorescence of HSA. This
Trp fluorescence is sensitive to the environment and binding of substrates
or changes in conformation that can result in quenching. Quenching can
occur by two different mechanisms, a dynamic mechanism caused by dif-
fusional collisions between the protein and the quencher, and a static
mechanism resulting from the formation of a non‐fluorescent complex.
The fluorescence spectra of HSA in the presence of increasing amounts
of 3–5 and cisplatin were recorded in the range of 300–450 nm upon
excitation of the tryptophan residue at 295 nm. The compounds caused
a concentration dependent quenching of fluorescence without changing
the emission maximum or shape of the peaks, as seen in Fig. 8 for com-
pound 3. The fluorescence data were analyzed by the Stern–Volmer equa-
tion. While a linear Stern–Volmer plot is indicative of a single quenching
mechanism, either dynamic or static, the positive deviation observed in
the plots of F0/F versus [Q] of our compounds (Fig. 8) is indicative of the
presence of different binding sites in the protein. These binding sites
may have naturally different affinities for the quencher, or the effects
may be due to changes in the conformation of the protein after initial
interactions that expose new binding sites previously inaccessible to the
quencher [53]. In this graph higher quenching by the iminophosphorane
3.4. Interaction with HSA
Human serum albumin is the most abundant carrier protein in plas-
ma and is able to bind a variety of substrates including metal cations,
hormones and most therapeutic drugs. It has been demonstrated that
the distribution, the free concentration and the metabolism of various
drugs can be significantly altered as a result of their binding to the pro-
tein [52]. HSA possesses three fluorophores, these being tryptophan
(Trp), tyrosine (Tyr) and phenylalanine (Phe) residues, with Trp214
Fig. 8. (A) Fluorescence titration curve of HSA with compound 3. Arrow indicates the
increase of quencher concentration. (B) Stern–Volmer plot for quenching with com-
pounds 3–5 and cisplatin.
Fig. 9. ITC titration of HSA with complex 3. Data were corrected for heat of dilution (see
Experimental section).

212
M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
Table 3
studied by a variety of analytical techniques [52] and despite varying
the techniques and conditions (such as incubation time, initial concentra-
tion of protein, drug:protein ratio and the nature of incubation medium)
there seems to be a common strong and irreversible binding of cisplatin
to HSA most plausibly to methionine and histidine residues [54].
In an attempt to gather more information about the nature of
the interaction between the iminophosphorane complexes and
HSA, isothermal titration calorimetry (ITC) experiments were car-
ried out. This sensitive tool allows for a complete characterization
of the thermodynamics of drug binding. ITC has been used for the
study of the interaction of Ni2+ [55], Co2+ [56], Er3+ [57] and
Ru2+-chloroquine [58] compounds with HSA but to our knowledge,
the binding of palladium or platinum compounds has not been assessed
with this technique. The data obtained from the titrations (illustrated in
Thermodynamic parameters obtained from ITC experiments for compounds 3 and 4.
3
4
K1 (M−1
)
(5.29 0.87)×104
(−1.06 0.16)×104
−14.00
(1.43 0.13)×104
(−4.69 0.41)×104
−138.2
371.6 20.05
(−1.89 0.10)×106
−6321
ΔH1 (Kcal/mol)
ΔS1
K2 (M−1
)
(1.41 0.04)×103
(−3.96 0.08)×105
−1313
ΔH2 (Kcal/mol)
ΔS2
complexes and 5 was observed compared to that of cisplatin under the
chosen conditions. This is probably an indication of the faster hydrolysis
in aqueous solution of our compounds compared to that of cisplatin.
The interaction of cisplatin and related compounds with HSA have been
Fig. 10. A) CD spectra (visible region) of HSA with 4. B) Effect of the addition of 15 equivalents of citric acid on the spectrum of the HSA-4 complex over time. c) Effect of low pH on
the spectrum of the HSA-4 complex over time. Y axis indicates ellipticity in millidegrees. X axis indicates wavelength in nm.

M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
213
Fig. 9 for complex 3) were best fitted to a model of two sequential bind-
4. Conclusion
ing sites with different affinities for the drug. A sequential binding
model was required to fit the data based upon the very large errors in
fits using a single site, and the excellent fit to data using 2 sites. The er-
rors are obtained from the iterated fitting routines provided in the soft-
ware for the ITC apparatus such that the quality of fits can be evaluated
both visually and from parameters like errors in KD and n, and
X-squared values (see Experimental section). The thermodynamic pa-
rameters for the interaction of compounds 3 and 4 with the protein
are collected in Table 3 with K1 of (5.29 0.87)×104 (3) and (1.43
0.13)×104 (4). The first binding site has a binding affinity approximate-
ly 40 times larger than the second one for both Pd(II) and Pt(II) com-
We have prepared new hydrophilic iminophosphorane complexes of
d8 metals. While we did not get stable gold(III) complexes containing the
water-soluble ligand reported here, Pd(II) and Pt(II) complexes were sta-
ble in the used conditions for biological experiments. The cytotoxicity,
cell death pathway and reactions with DNA and HSA of the gold(III)
compound were due to the generation of a cytotoxic decomposition
product [AuCl2(HN―C(O)―2-NC5H4)] 5, previously reported. The Pd
compound (3) was found to be less cytotoxic than cisplatin and to have
a related mode of action. We found that the Pt(II)–iminophosphorane co-
ordination derivative (4) was more cytotoxic than cisplatin to leukemia
cell lines (both T-Jurkat and cisplatin-resistant Jurkat sh-Bak). 4 showed
higher toxicity against leukemia cells when compared to normal human
T-lymphocytes or PBMC and its cell death pathway seems to be different
from that of cisplatin. Interestingly, this compound did not interact
strongly with CT DNA. It was shown that its interaction with plasmid
(pBR322) DNA (binding to CCC or Form I) is different from that of cisplat-
in. It was demonstrated that it interacted with HSA in a moderate and re-
versible way which may be an advantage to the antitumor activity of this
complex in vivo. These and previous results obtained in our laborato-
ries warrant further studies on iminophosphorane complexes of
gold(III) and platinum(II) as anticancer agents.
plexes ([1.41 0.04]×103
3 and [371.6 20.05] 4), hence the
sequential interaction. Binding constant values for other metal ions
(Ni2+ [55], Co2+ [56], Er3+ [57]) or metal complexes (Ru2+ [58]) bind-
ing to HSA are also in the mM to microM range.
The ITC results may explain the lack of linearity observed in the fluo-
rescence quenching studies, as the Stern–Volmer method assumes all
binding sites to be equivalent. Unfortunately, ITC experiments with the
gold(III) compound 5 could not be performed due to poor solubility at
the required concentrations. Similarly, experiments with cisplatin did
not yield useful data. The slow hydrolysis of the chloride ligands requires
incubation of the drug with the protein for several hours or days [52],
which is incompatible with the ITC technique.
Attempts to correlate the Ka of a drug with the percentage plasma
protein binding have been carried out [59,60]. The first binding con-
stant for the Pt compound (4) (1.43 0.13)×104 is correlated with
a roughly 90% plasma protein binding which is considered the
upper limit for effective chemotherapeutics (values higher than 90%
are considered to be disadvantageous). More important however, is
to asses if this binding can be reversed. If the interaction to HSA is
too strong, the drug might not be able to reach the cellular target.
Reversibility can be achieved by contact with a low pH environment,
as in tumor tissues, or by chelators present in the cytosol that may
displace the amino acids in typical HSA–metallodrug complexes. We
studied the reversibility of the binding of HSA to compounds 3 and
4 by CD spectroscopy. The protein was first titrated with compounds
3 and 4, achieving saturation with 4 and 6 equivalents respectively as
determined by the appearance and increase in intensity of a visible
band at 360 nm for 3 (negative) and 395 nm for 4 (positive) in
their CD spectra (Fig. 10A and S3a).
Abbreviations
ABB
annexin binding buffer
7-AAD
CD
7-Amino-actinomycin
circular dichroism
COD
cyclooctadiene
CT
calf thymus
DFT
density functional theory
FCS
fetal calf serum
Hepes
HQuinala
HQuinpy
HSA
4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
N-(8-quinolyl)alanine-carboxamide
N-(8-quinolyl)pyridine-2-carboxamide
human serum albumin
ITC
mTOR
MTT
isothermal titration calorimetry
mammalian target of rapamycin
3-(4,5-dimethulthiazol-2-yl)-2,5-diphenyltetrazolium
By using a metal ion chelator like citric acid, a decrease of the in-
tensity of the visible bands was observed, especially in the case of
platinum compound 4, with a 50% reduction after only 1 h of incuba-
tion with a 3-fold excess of the chelating agent (Fig. 10b). The effect of
citric acid on the HSA-3 adduct was less significant with only a 12% de-
crease after the same period of time (Fig. S3b). Decreases of 25–30% in
the same conditions were achieved with arene-Ru(II)-chloroquine anti-
malarial and antitumor agents whose binding to HSA was considered
reversible [58]. In a different experiment with our new complexes, the
pH of the buffered solutions was decreased from 7.40 to 4.00 by
addition of HCl. An effect on the spectrum of the HSA-4 adduct was ob-
served again, by the decrease of the visible band intensity of 52% after
1 h (Fig. 11c), while the CD spectrum of the HSA-3 adduct showed
large shifts and shape modifications of the visible band but no genuine
indication of the Pd center being released from the protein (Fig. S3c). A
decrease of 18%–40% after 96 h incubation was achieved with the previ-
ously described arene-Ru(II)-chloroquine derivatives whose binding to
HSA was reversed by a decrease of pH [58]. Thus, we propose that the
reversibility of the binding 4 to HSA by chelators present in the cytosol
and/or contact with a low pH environment is possible. However, for the
Pd (3) complex, some permanent bands were seen after long incubation
in the presence of citrate (Fig. S3b) or at low pH (Fig. S3c) that could re-
flect modification of HSA by this metal. Therefore, no evidence for re-
lease (reversibility) was seen using CD spectroscopy for the Pd (3)
derivative.
bromide
phen
ROS
TAE
1,10-phenantroline; PS: phosphatidylserine
reactive oxygen species
tris-acetate/EDTA buffer
tert-BuDAD N,N′-bis(2,6-diisopropylphenyl)2,3-butanediimine
terpy
THF
tmeda
TMRE
TPA
2,2′,6′,2′′-terpyridine
tetrahydrofuran
N,N,N′,N′′-tetramethylethylenendiamine
tetramethylrhodamine ethyl ester
1,3,5-triaza-7-phosphaadamantane
Acknowledgements
We thank the financial support of a grant from the NIH-National
Institute of General Medical Sciences (NIGMS), SC2GM082307 (M.C.)
and a grant from the Spanish Ministry of Science and Innovation
SAF2010-14920 (I.M.). We thank Prof. Andrzej Jarzecki (Brooklyn Col-
lege) for his help with the DFT calculations of compounds 2–4 and stu-
dents Farrah Benoit and Malgorzata Frik for their assistance with some
fluorescence measurements. We also thank Dr. Esteban P. Urriolabeitia
for helpful discussions on the synthesis of ligand 1.

214
M. Carreira et al. / Journal of Inorganic Biochemistry 116 (2012) 204–214
F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R.
Appendix A. Supplementary Material
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi,
M. Cossi, N. Rega, J.N. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli,
J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador,
J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö Farkas, J.B. Foresman, J.V. Ortiz, J.
Cioslowski, D.J. Fox. Gaussian, Inc., Wallingford CT, 2009.
Supplementary data to this article: crystallographic data and refine-
ment for 3 and 4; table with selected structural parameters and
frequencies of selected normal modes (IR Spectroscopy) for 2–4; data
on the implication of Bax/Bak in the toxicity of 3–5; graphs on the int-
eractions of 3 and 4 with Calf Thymus DNA by circular dichroism
spectroscopy; graphs on the studies on the reversibility of the binding
of 3 to HSA by circular dichroism spectroscopy. See http://dx.doi.org/
10.1016/j.jinorgbio.2012.06.017.
[27] N. Lopez-Royuela, P. Perez-Galan, P. Galan-Malo, V.J. Yuste, A. Anel, S.A. Susin, J.
Naval, I. Marzo, Biochem. Pharmacol. 79 (2010) 1746–1758.
[28] 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.
[29] J. Vicente, J.A. Abad, R. Clemente, J. Lopez-Serrano, M.C. Ramirez de Arellano, P.G.
Jones, D. Bautista, Organometallics 22 (2003) 4248–4259.
[30] L.R. Falvello, M.M. Gracia, I. Lazaro, R. Navarro, E.P. Urriolabeitia, New J. Chem.
(1999) 227–235.
[31] M.W. Avis, C.J. Elsevier, N. Veldman, H. Kooijman, A.L. Spek, Inorg. Chem. 35 (1996)
1518–1528.
[32] J. Baker, A.A. Jarzecki, P. Pulay, J. Phys. Chem. A (1998) 1412–1424.
[33] J.F. González-Pantoja, M. Stern, A.A. Jarzecki, E. Royo, E. Robles-Escajeda, A.
Varela-Ramírez, R.J. Aguilera, M. Contel, Inorg. Chem. 50 (2011) 11099–11110
(and refs. therein).
[34] B.T. Elie, C. Levine, Iban Ubarretxena-Belandia, A. Varela, R. Aguilera, R. Ovalle, M.
Contel, Eur. J. Inorg. Chem. (2009) 3421–3430 (and refs. therein).
[35] G. Natile, F. Arnesano, Coord. Chem. Rev. 253 (2009) 2070–2081 (and refs. therein).
[36] Z.H. Siddik, Oncogene 22 (2003) 7265–7279.
[37] R.J. Thapa, S.H. Basagoudanavar, S. Nogusa, K. Irrinki, K. Mallilankaraman, M.J. Slifker,
A.A. Beg, M. Madesh, S. Balachandran, Mol. Cell. Biol. 14 (2011) 2934–2946.
[38] J.P. Caradonna, S.J. Lippard, M.J. Gait, M. Singh, J. Am. Chem. Soc. 104 (1982)
5793–5795.
[39] J.C. Dabrowiak, Metals in medicine, John Wiley and Sons, 2009.
[40] H.-K. Liu, P.J. Sadler, Acc. Chem. Res. 44 (2011) 349–359.
[41] J. Ruíz, N. Cutillas, C. Vicente, M.D. Villa, G. López, J. Lorenzo, F.X. Avilés, V.
Moreno, D. Bautista, Inorg. Chem. 44 (2005) 7365–7376.
[42] E. Gao, M. Zhu, L. Liu, Y. Huang, L. Wang, C. Shi, Z. Chuyue, W. Sun, Inorg. Chem. 49
(2010) 3261–3270.
[43] A.G. Quiroga, J.M. Pérez, I. López-Solera, J.R. Masaguer, A. Luque, P. Roman, A.
Edwards, C. Alonso, C. Navarro-Ranninger, J. Med. Chem. 41 (1998) 1399–1408.
[44] H. Mansouri-Torshizi, M. I-Moghaddam, A. Divsalar, A.-A. Saboury, Bioorg. Med.
Chem. 16 (2008) 9616–9625.
[45] A.K. Paul, H. Mansouri-Torshizi, T.S. Srivastava, S.J. Chavan, M.P. Chitnis, J. Inorg.
Biochem. 50 (1993) 9–20.
[46] C.E. Blank, J.C. Dabrowaik, J. Inorg. Biochem. 21 (1984) 21–29.
[47] S.M. Cutts, A. Masta, C. Panousis, P.G. Parsons, R.A. Sturm, D.R. Phillips, Drug-DNA
Interact, Protocols 90 (1997) 95–106.
[48] L. Messori, P. Orioli, C. Tempi, G. Marcon, Biochem. Biophys. Res. Commun. 281
(2001) 352–360.
[49] T. Yang, C. Tu, J. Zhang, L. Lin, X. Zhang, Q. Liu, J. Ding, Q. Xu, Z. Guo, Dalton Trans.
(2003) 3419–3424.
[50] S.E. Sherman, S.J. Lippard, Chem. Rev. (1987) 1153–1181.
[51] Y. Shi, S.-A. Liu, D.J. Kerwood, J. Goodisman, J.C. Dabrowiak, J. Inorg. Biochem. 107
(2012) 6–14.
[52] A.R. Timerbaev, C.G. Hartinger, S.S. Aleksenko, B.K. Keppler, Chem. Rev. 106
(2006) 2224–2248.
[53] J.R. Lacowicz, Principles of Fluorescence Spectroscopy, In: Springer, 2006,
pp. 238–266, Ch 8.
References
[1] A.M. Thayer, Platinum drugs take their toll, Chem. Eng. News 88 (26) (2010) 24–28.
[2] L. Kelland, Nat. Rev. Cancer 7 (2007) 573–584.
[3] G. Sava, A. Bergamo, P.J. Dyson, Dalton Trans. 40 (2011) 9069–9075 (and refs.
therein).
[4] S. Dhar, S.J. Lippard, Current status and mechanism of action of platinum-based
drugs, In: in: E. Alessio (Ed.), Bioinorganic Medicinal Chemistry, Wiley-VCH, 2011,
pp. 79–96, Ch 3.
[5] X. Wang, X.Z. Guo, New trends and future developments of platinum-based antitumor
drugs, In: in: E. Alessio (Ed.), Bioinorganic Medicinal Chemistry, Wiley-VCH, 2011,
pp. 97–149, Ch 4.
[6] J. Aleman, V. del Solar, A. Alvarez-Valdes, C. Rios-Luci, J.M. Padron, C.
Navarro-Ranninguer, MedChemComm. 2 (2011) 789–793 (and refs. therein).
[7] S.J. Berners-Price, Angew. Chem. Int. Ed. 50 (2011) 804–805.
[8] N.J. Farrer, J.A. Woods, L. Salassa, Y. Zhao, K.S. Robinson, G. Clarkson, F.S. Mackay,
P.J. Sadler, Angew. Chem. Int. Ed. 49 (2010) 8905–8908.
[9] S.J. Berners-Price, Gold-based therapeutic agents. A new perspective, In: in: E.
Alessio (Ed.), Bioinorganic Medicinal Chemistry, Wiley-VCH, 2011, pp. 197–222,
Ch 7.
[10] F.A. Serrano, A.L. Matsuo, P.T. Monteforte, A. Bechara, S.S. Smaili, D.P. Santana, T.
Rodrigues, F.V. Pereira, L.S. Silva, J. Machado Jr., E.L. Santos, J.B. Pesquero, R.M.
Martins, L.R. Travassos, A.C.F. Caires, E.G. Rodrigues, BMC Cancer 11 (2011)
296–311 (and refs. therein).
[11] U. Olzewski, J. Clafeey, M. Hogan, M. Tacke, R. Zeillinger, P. Bednarski, G.
Hamilton, Invest. New Drugs 29 (2011) 607–614 (and refs. therein).
[12] D. de Oliveira Silva, Anticancer Agents Med. Chem. 10 (2010) 312–323 and refs.
therein.
[13] S.D. Shyder, Y. Fu, A. Habtemariam, S.H. van Rijt, P.A. Cooper, P.M. Loadman, P.J.
Sadler, Med. Chem. Commun. 2 (2011) 666–668 (and refs. therein).
[14] G. Gasser, I. Ott, N. Metzler-Nolte, J. Med. Chem. 54 (2011) 3–25 (and refs. therein).
[15] L. Vela, M. Contel, L. Palomera, G. Azaceta, I. Marzo, J. Inorg. Biochem. 105 (2011)
1306–1313.
[16] N. Shaik, A. Martinez, I. Augustin, H. Giovinazzo, A. Varela, R. Aguilera, M. Sanaú,
M. Contel, Inorg. Chem. 48 (2009) 1577–1587.
[17] K.J. Kilpin, W. Henderson, B.K. Nicholson, Inorg. Chim. Acta 362 (2009)
3669–3676.
[18] S.D.J. Brown, W. Henderson, K.J. Kilpin, B.K. Nicholson, Inorg. Chim. Acta 360 (2007)
1310–1315.
[19] D. Aguilar, M. Contel, E.P. Urriolabeitia, Chem. A Eur. J. 16 (2010) 9287–9296.
[20] D. Aguilar, M. Contel, R. Navarro, T. Soler, E.P. Urriolabeitia, J. Organomet. Chem.
694 (2009) 486–493.
[21] A. Aguilar, M. Contel, R. Navarro, E.P. Urriolabeitia, Organometallics 26 (2007)
4604–4611.
[22] S.S. van Berkel, M.B. van Eldijk, J.C.M. van Hest, Angew. Chem. Int. Ed. 50 (2011)
8806–8827.
[54] A.I. Ivanov, J. Christodoulou, J.A. Parkinson, K.J. Barnham, A. Tucker, J. Woodrow,
P.J. Sadler, J. Biol. Chem. 273 (1998) 14721–14730.
[55] A.A. Saboury, J. Chem. Thermodyn. (2003) 1975–1981.
[56] M. Sokolowska, M. Wszelaka-Rylik, J. Poznański, W. Bal, J. Inorg. Biochem. (2009)
1005–1013.
[57] G. Rezaei Behbehani, A. Divsalar, A.A. Saboury, F. Faridbod, M.R. Ganjali, J. Therm.
Anal. Calorim. 96 (2009) 663–668.
[58] A. Martínez, J. Suárez, T. Shand, R.S. Magliozzo, R.A.J. Sánchez-Delgado, J. Inorg.
Biochem. (2011) 39–45.
[23] S. Bittner, Y. Assaf, P. Krief, M. Pomerantz, B.T. Ziemnicka, C.G. Smith, J. Org. Chem.
50 (1985) 1712–1718.
[24] D. Drew, J.R. Doyle, A.G. Shaver, Inorg. Synth. 13 (2007) 47–55.
[25] D. Fang, C.-T. Yang, J.D. Randford, J.J. Vittal, Dalton Trans. (2003) 4749–4753.
[26] Gaussian 09, Revision A.1, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H.
Nakatsuji, M. Caricato, X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L.
Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, R. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery, J.E. Peralta Jr.,
[59] N.A. Kratochwil, W. Huber, F. Muller, M. Kansy, P.R. Gerber, Biochem. Pharmacol.
64 (2002) 1355–1374.
[60] L. Yan, X. Wang, Y. Wang, Y. Zhang, Y. Li, Z. Guo, J. Inorg. Biochem. (2012) 46–51.