Synthesis and characterization of new transition metal complexes containing DNA intercalators of the acridine family

Article abstract of DOI:10.1039/b304288j

The coordinating ability of two DNA intercalates of the acridine family, 4-hydroxyacridine (4-OH-Acrid) and acridine orange (AO), with the transition metal ions Ni(II), Pd(II) and Pt(II), are investigated. Octahedral complexes, [Ni(4-O-Acrid)₂(en)] (1), and [Ni(4-O-Acrid)₂(H ₂O)₂] (2), are obtained when the 4-OH-Acrid ligand reacts with [Ni(Cl)₂(en)₂] and NiCl₂·6H ₂O, respectively. Single crystal X-ray analysis of complex 1 has shown that two acridine molecules are bonded to nickel through their nitrogen and oxygen atoms in a chelate fashion. Moreover, when AO reacts with different Pd(II) and Pt(II) precursors, new triamine complexes of the general formula cis-[M(A)_n(Cl)(AO)]⁺X^- (M = Pd(II), Pt(II); A = en, NH₃; n = 1, 2; X^- = BF₄^-, NO₃^-, PF₆^-) (3-6) are synthesized. Full characterization by IR, ¹H NMR and electrospray ionization mass spectrometry in solution has demonstrated the coordination of the AO ligand to the metal ion through the endocyclic nitrogen atom. The biological activity of these new acridine orange derivatives is conducted on complex cis-[Pt(NH ₃)₂Cl(AO)][NO₃] (5). The cytotoxic activity of complex 5 is compared with that of cisplatin (cis-DDP) in DU145, A2780 and A2780-cp8 cancer cell lines and the results show that its activity is twice as effective as that of cisplatin in the two cis-DDP resistant cell lines DU145 and A2780-cp8, respectively.

Full text of DOI:10.1039/b304288j

View Article Online / Journal Homepage / Table of Contents for this issue
Synthesis and characterization of new transition metal complexes
containing DNA intercalators of the acridine family
Alessandra Crispini,*a Daniela Pucci,a Stefania Sessa,a Antonella Cataldi,a Anna Napoli,a
Alessandra Valentinib and Mauro Ghedinia
a
`
`
LASCAMM, Unita INSTM della Calabria, Dipartimento di Chimica, Universita della
Calabria, 87030 Arcavacata di Rende (CS), Italy. E-mail: a.crispini@unical.it;
Fax: (+39)0984492044
b
Laboratory of Experimental Oncology, Oncology Institute of Southern Switzerland,
C/O IRB, Via Vela 6, CH-6500 Bellinzona, Switzerland
Received (in London, UK) 17th April 2003, Accepted 1st August 2003
First published as an Advance Article on the web 28th August 2003
The coordinating ability of two DNA intercalates of the acridine family, 4-hydroxyacridine (4-OH-Acrid)
and acridine orange (AO), with the transition metal ions Ni(II), Pd(II) and Pt(II), are investigated. Octahedral
complexes, [Ni(4-O-Acrid)2(en)] (1), and [Ni(4-O-Acrid)2(H2O)2] (2), are obtained when the 4-OH-Acrid
ligand reacts with [Ni(Cl)2(en)2] and NiCl2ꢀ6H2O, respectively. Single crystal X-ray analysis of complex 1
has shown that two acridine molecules are bonded to nickel through their nitrogen and oxygen atoms in a
chelate fashion. Moreover, when AO reacts with different Pd(II) and Pt(II) precursors, new triamine complexes
of the general formula cis-[M(A)n(Cl)(AO)]+Xꢁ (M ¼ Pd(II), Pt(II); A ¼ en, NH3 ; n ¼ 1, 2; Xꢁ ¼ BF4
,
ꢁ
NO3ꢁ, PF6ꢁ) (3–6) are synthesized. Full characterization by IR, H NMR and electrospray ionization mass
spectrometry in solution has demonstrated the coordination of the AO ligand to the metal ion through the
endocyclic nitrogen atom. The biological activity of these new acridine orange derivatives is conducted on
complex cis-[Pt(NH3)2Cl(AO)][NO3] (5). The cytotoxic activity of complex 5 is compared with that of cisplatin
(cis-DDP) in DU145, A2780 and A2780-cp8 cancer cell lines and the results show that its activity is twice as
effective as that of cisplatin in the two cis-DDP resistant cell lines DU145 and A2780-cp8, respectively.
1
In addition, the positive charge of the Pt(II) complexes may
Introduction
contribute to their binding with the negatively charged phos-
phate oxygen atoms of DNA through electrostatic interac-
tions.17 Acridine is an important aza-aromatic compound
and its derivatives have been widely used in pharmacology.18
Their polyaromatic nature facilitates insertion or intercalation
between base pairs of double stranded DNA.19 The driving
force of intercalation is based on intermolecular interaction
between the plane of base pairs of double stranded DNA
and the aromatic plane of the polyaromatic intercalator.
On this basis Lippard and co-workers have synthesized and
analyzed new Pt(II) adducts containing, as the Am ligand,
intercalators of the acridine family. The intercalators used in
their studies were 9-aminoacridine and 3,6-bis-(dimethylami-
no)acridine (acridine orange) quaternized at the endocyclic
nitrogen atom.20,21 None of these ligands coordinated the
metal ion through the endocyclic nitrogen atom despite its
greater basicity compared with the exocyclic one. In the case
of 9-aminoacridines intercalation in DNA was found with
the endocyclic nitrogen atom placed in the major groove and
the exocyclic one placed in the minor groove.22 Therefore
coordination between the platinum and the endocyclic nitro-
gen atom could bring the metal in the major groove close to
the nitrogen atom of an adjacent purine residue, allowing them
to interact. Recently the endocyclic coordination to platinum
in nitro-9-aminoacridines was shown to depend on both the
position of the nitro group with respect to the endocyclic nitro-
gen atom and the charge of the metal complexes used as
precursors.23
Cisplatin (cis-diaminedichloroplatinum(II), cis-DDP) is a well-
established antitumor agent effective especially against ovar-
ian, testicular and head cancers.1,2 Cis-DDP is believed to
exert its cytotoxicity by coordinating bifunctionally to DNA
through the N(7) atoms of two adjacent guanines on the same
strand, arresting DNA replication.3 Despite its clinical success,
cis-DDP exhibits only a narrow spectrum of activity and has
been associated with severe side effects and acquired resis-
tance.4 Cisplatin is often administered in combination with
other intercalative drugs such as adriamycin and actinomy-
cin.5,6 The general idea behind modifying the cisplatin struc-
ture to derive platinum-DNA adducts of high cytotoxicity is
to utilise a suitable carrier ligand to better ‘‘target’’ the plati-
num moiety to DNA.7–9 Examples of DNA-targeted platinum
drugs showing improved selectivity with respect to cisplatin-
resistant cell lines include doxorubicin,10,11 9-anilinoacridine12
and the more soluble acridinecarboxamide derivatives.13
Moreover, a study of the antitumoral activity of platinum
complexes tethered to variously coordinated anthraquinones
has given promising results.14
Anticancer activity has also been observed in complexes of
the general formula cis-[Pt(A)n(Cl)(Am)]+, where A is either
NH3 (n ¼ 2) or en (n ¼ 1) and Am is a heterocyclic amine
ligand based on pyridine, pyrimidine, purina and piperidine.15
Despite the lack of structural features characteristic of stan-
dard cisplatin analogues, the unanticipated activity of these
complexes can be ascribed to the possibility that the aromatic
ligand might intercalate between base pairs adjacent to the site
of platinum coordination, thereby forming pseudo-bifunc-
tional adducts which are capable of inhibiting replication.16
Metallointercalation has also been extended to three dimen-
sions using octahedral complexes of Ru(II), Co(III), Os(II),
Rh(III) and Ni(II), which, through their photophysical,
DOI: 10.1039/b304288j
New J. Chem., 2003, 27, 1497–1503
1497
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2003

View Article Online
(corresponding to [Ni(4-O-Acrid)2(H2O)H]+); m/z 447.08,
[C26H17N2O2Ni]+ (corresponding to [Ni(4-O-Acrid)2H]+).
Found: C, 65.73; H, 4.15; N, 5.79%. Required for
C26H20N2NiO4 (483.16): C, 65.63; H, 4.12; N, 5.80%.
[Pd(en)Cl(AO)][BF4] (3). [Pd(en)Cl2] (30 mg, 0.13 mmol) and
Ag(BF4) (24.52 mg, 0.13 mmol) were stirred in 10 mL of water,
at room temperature, for 12 h. The resulting AgCl precipitate
was removed by filtration and a solution of acridine orange
(33.39 mg, 0.13 mmol) in water (3 mL) was added to the fil-
trate. After the mixture was stirred for 24 h, the resulting solid
was filtered, recrystallized from ethanol and dried in vacuo
(31.1 mg, 43.2%).
Chart 1
photochemical and redox properties, offer the possibility of
becoming sensitive spectroscopic and reactive DNA probes.24
Octahedral complexes capable of being metallointercalators
contain an extended aromatic heterocyclic ligand which can
insert and stack between base pairs.
The aim of the present study is to give few more insight
about the coordination between transition metal ions and 4-
hydroxy-acridine (4-OH-Acrid) and acridine orange (AO),25
two intercalators of the acridine family whose coordinating
ability has been less explored.
While in both cases an endocyclic nitrogen atom is present,
offering a potential strong binding site for metal coordination,
4-OH-Acrid has the potential of working both as a monoden-
tate and bidentate (chelating) ligand. The coordinating ability
of these ligands with the transition metal ions Ni(II), Pd(II) and
Pt(II), starting with different metal precursor complexes, was
tested with the generation of both octahedral and square
planar metallointercalators.
IR (KBr): n 1074 cmꢁ1 (BF4). ESI (+) MS: m/z 484.0994,
[C19H29N5OClPd]+ (calculated mass 484.10, ppm error
ꢁ20.94); CID spectrum of [M]+ (m/z 484.09): m/z 466.2,
[M–H2O]+; m/z 441.87, [M–C2H4N]+; m/z 423.88 [M–
C2H4N–H2O]+; m/z 401.04, [M–C4H8N2]+; m/z 381.90 [M–
C4H8N2–H2O]+; m/z 363.01 [M–C4H8N2–H2O–Cl]+; m/z
223.20 [(AO–C2H4N)H]+.
Found: C, 40.93; H, 4.83; N, 11.52%. Required for
C19H27BClF4N5Pd (554.12): C, 41.18; H, 4.91; N, 11.64%.
[Pt(en)Cl(AO)][BF4] (4). [Pt(en)Cl2] (30 mg, 0.09 mmol) and
Ag(BF4) (17.47 mg, 0.09 mmol) was stirred in 3 mL of water,
at room temperature, for 12 h. The resulting AgCl precipitate
was removed by filtration and a solution of acridine orange
(23.87 mg, 0.09 mmol) in water (3 mL) was added to the fil-
trate. After the mixture was stirred for 24 h, the solid formed
was filtered, recrystallized from ethanol and dried in vacuum
(29.8 mg, 52%).
IR (KBr): n 1074 cmꢁ1 (BF4). ESI (+) MS: m/z 573.2088,
[C19H29N5OClPt]+ (calculated mass 573.17, ppm error
67.16), m/z 520.84, [C19H28N5Pt]+ (corresponding to
[Pt(AO)(en)H]+); m/z 478.83, [C17H22N3OPt]+ (corresponding
to [Pt(AO)(H2O)H]+); m/z 460.78 [C17H20N3Pt]+ (corre-
sponding to [Pt(AO)H]+); m/z 266.19 [C17H20N3]+ (corre-
sponding to [AOH]+). Found: C, 36.06; H, 4.00; N, 10.06%.
Required for C19H27BClF4N5Pt (642.81): C, 35.50; H, 4.23;
N, 10.89%.
Experimental
General details
4-Hydroxyacridine, acridine orange, NiCl2ꢀ(H2O)6 and
[Pd(en)Cl2] were used without further purification. [Ni(en)2-
Cl2],26 [Pt(en)Cl2],27 and cis-[Pt(NH3)2Cl2]28 were synthetized
as reported in the literature. cis-DDP was dissolved in a
0.9% (w/v) NaCl solution, prepared fresh prior to use. The
infrared spectra were recorded on a Spectrum One FT-IR
1
PerkinElmer spectrometer in diffuse reflectance mode. The H
Cis-[Pt(NH3)2Cl(AO)][NO3] (5). Cis-DDP (70 mg, 0.23
mmol) and AgNO3 (39.60 mg, 0.23 mmol) was stirred in 10
mL of water, at room temperature, for 16 h. After the AgCl
was filtered, a solution of acridine orange (61.74 mg, 0.23
mmol) in water (3 mL) was added to the filtrate and the mix-
ture was stirred for 24 h. The resulting orange solid was fil-
tered, recrystallized from ethyl acetate and dried in vacuo
(92.2 mg, 68%).
NMR spectra were recorded on a Bruker WH-300 spectro-
meter in CDCl3 solution, with TMS as internal standard.
Elemental analyses were performed with a PerkinElmer 2400
analyzer.
Preparations
[Ni(4-O-Acrid)2(en)] (1). [Ni(en)2Cl2] (30 mg, 0.12 mmol)
was dissolved in water (3 mL) and a solution of 4-hydroxyacri-
dine (46.86 mg, 0.24 mmol) and KOH (6.73 mg, 0.12 mmol) in
a mixture water–ethanol (2 mL + 1 mL) was added. The mix-
ture was stirred at room temperature for 3 h. The red solid was
filtered, recrystallized from a mixture of chloroform–hexane
and dried in vacuo (43.0 mg, 71%). Found: C, 66.38; H, 5.00;
N, 10.85%. Required for C28H24N4NiO2 (507.23): C, 66.30;
H, 4.77; N, 11.05%.y
1H NMR (300 MHz, CDCl3 , 25 ꢂC, TMS): d 8.38 (s, 1H,
H9), 7.68 (d, 2H, H1,8), 7.23 (d, 2 H, H4,5), 7.03 (dd, 2H,
H
2,7), 3.24 (s, 12H, CH3). IR (KBr): n 1384 cmꢁ1 (NO3). ESI
(+) MS: m/z 565.17 [C17H29N5O2ClPt]+ (calculated mass
565.16, ppm error 17.70)]; CID spectrum of [M]+ (m/z
565.17): m/z 547.98, [M–NH3]+; m/z 530.94, [M–2NH3]+;
m/z 511.99, [M–2NH3–H2O]+; m/z 266.2 [AOH]+. Found:
C, 30.15; H, 3.89; N, 14.89%. Required for C17H25ClN6O3Pt
2H2O (628.0): C, 32.51; H, 4.65; N, 13.38%.
[Ni(4-O-Acrid)2(H2O)2] (2). NiCl2ꢀ(H2O)6 (30 mg, 0.13
mmol) was added to a solution of 4-hydroxyacridine (49.19
mg, 0.25 mmol) and piperidine (0.024 mL, 0.25 mmol) in water
(25 mL). The resulting suspension was stirred at room tem-
perature for 2 h. The orange solid was filtered, recrystallized
from ethanol and dried in vacuum (31.5 mg, 50%).
Cis-[Pt(NH3)2Cl(AO)][PF6] (6). cis-DDP (0.05 g, 0.17 mmol)
and AgPF6 (41.96 mg, 0.17 mmol) was stirred in 3 mL of water
for 12 h. The resulting AgCl was filtered off and a solution of
acridine orange (43.99 mg, 0.17 mmol) in water (3 mL) was
added to the filtrate. After the mixture was stirred for 72 h,
the resulting orange solid was filtered, recrystallized from ethyl
acetate and dried in vacuum (63.4 mg, 55%).
ESI (+) MS: m/z 483.0939, [C26H21N2O4Ni]+ (calculated
mass 483.08, ppm error 18.56); m/z 465.09 [C26H19N2O3Ni]+
IR (KBr): n 851 cmꢁ1 (PF6); ESI (+) MS: m/z 565.17
[C17H29N5O2ClPt]+ (calculated mass 565.16, ppm error
17.70). CID spectrum of [M]+ (m/z 565.17): m/z 547.98,
[M–NH3]+; m/z 530.93, [M–2NH3]+; m/z 511.99, [M–2NH3–
y CCDC reference number 207089. See http://www.rsc.org/supp-
data/nj/b3/b304288j for crystallographic data in CIF or other electro-
nic format.
1498
New J. Chem., 2003, 27, 1497–1503

View Article Online
H2O]+; m/z 266.20 [AOH]+. Found: C, 31.03; H, 4.24; N,
10.14%. Required for C17H25ClF6N5PPt (674.93): C, 30.25;
H, 3.73; N, 10.38%.
ium for an additional 48 and 72 h at 37 ꢂC in 5% CO2 humidi-
fied atmosphere. The cells were then trypsinized and counted
in a particle counter (Coulter Counter, Coulter Electronics,
Luton, Bedfordshire, UK). The percentage of adherent viable
cells was determined through trypan blue dye exclusion. Each
experimental sample was run in triplicate. The results were
expressed as the total number of adherent cells in treated sam-
ples compared with untreated (control) samples. In vitro drug
activity was expressed in terms of the concentration which
inhibited cell proliferation by 50% (IC50).
Molecular mechanics
Universal force field (UFF)29 was employed as implemented
in the Cerius modelling package.30 All calculations were per-
formed on a Silicon Graphics RS 4000 Indigo workstation
and carried out with atomic charges. The nickel, palladium
and platinum centres were taken as force-field atom type
M4+2. Initial trials on model systems were found to give
reasonable agreement with the CSD-derived results noted
below without main changes in the force field used for the
calculations.
Cell cycle and apoptosis analysis
Cells treated as described above, were trypsinized, harvested,
and washed once in PBS and then fixed in 80% ethanol. After
1 h of incubation at 4 ꢂC cells were stained with propidium
iodide (PI, 50 mg mLꢁ1, Sigma) in PBS containing Rnase-A
(75 kU mLꢁ1, Sigma) and analyzed for DNA content. The
analysis of cell cycle and apoptosis was performed, using a
FACScan flow cytometer (Becton Dickinson, USA) and the
Cell Quest software package (Becton Dickinson).
X-ray crystallography
Crystal data for 1: C28H24N4O2Ni, M ¼ 507.2, monoclinic,
ꢂ
˚
a ¼ 18.180(2), b ¼ 12.346(3), c ¼ 10.680(2) A, b ¼ 108.95(2) ,
3
˚
U ¼ 2267.3(8) A , T ¼ 298 K, space group C2/c, Z ¼ 4,
m(Mo Ka) ¼ 8.91 cmꢁ1, 2133 reflections measured, 2013
reflections unique (R(int) ¼ 0.032), 156 variable parameters,
R1 ¼ 0.0386 (I > 2s(I)), wR2 ¼ 0.0977, GOF(F2) ¼ 0.945.
Dark red crystals of 1 were crystallized at room temperature
from a chloroform/hexane mixture.
Results and discussion
Chemical studies of octahedral complexes
In order to obtain octahedral metallointercalators Ni(II) hexa-
coordinated precursors dichloro-bis-ethylenediamine nichel(II),
[Ni(Cl)2(en)2], and NiCl2ꢀ6H2O, were used in reaction with two
equivalents of 4-hydroxyacridine ligand (Scheme 1).
Data collection was carried out at room temperature on a
Siemens R3m/v diffractometer using graphite-monochromated
Mo Ka (l ¼ 0.71073 A) radiation. The data were corrected for
˚
Lorentz and polarization effects. Absorption correction was
applied (XABS2).31 Since it was not possible to distinguish
between the Cc and C2/c space groups from the analysis of
the systematic absences, attempts to solve the structure in both
acentric and centric systems were made. The best solution was
found in the C2/c space group. The structure solution and full-
matrix least-squares refinements based on F2 were performed
with XS and XL routines in the SHELXTL-NT program pack-
age. All non-hydrogen atoms were refined anisotropically, with
nickel atom lies on a mirror plane. Hydrogen atoms were
included as idealized atoms riding on the respective nitrogen
and carbon atoms, with N–H and C–H bond lengths appropri-
ate to the atom hybridization.
The reaction between [Ni(Cl)2(en)2] and two equivalents of
4-hydroxyacridine was conducted in a basic aqueous solution
to deprotonate the acridine ligand, which then acts as a chelat-
ing ligand, coordinating with both the nitrogen and the oxygen
atoms. A new Ni(II) complex of the general formula [Ni(4-O-
acrid)2(en)], 1, was synthesized, in which two chlorine and
one ethylenediamine ligands were substituted by two acridine
ligands. Single crystal X-ray analysis confirmed the stoichio-
metry initially postulated from the analytical data. A view of
the structure of complex 1 is shown in Fig. 1. Selected bond
distances and angles are reported in Table 1.
The coordination about Ni, which is situated on the C2 axis,
is distorted octahedron . Two acridine molecules are bonded to
nickel through their nitrogen and oxygen atoms in a chelate
Electrospray ionization mass spectrometry
˚
fashion. The Ni–N(1) bond distance is found to be 0.098 A
longer than Ni–O(1) and both distance values as well as the
bite angle of 79.31(8)ꢂ are in agreement with those observed
in 8-quinolinate Ni(II) octahedral complexes searching
ESI experiments were carried out on a hybrid Q-Star Pulsar-i
(PE-SCIEX) mass spectrometer equipped with an ionspray
ionization source. Complexes 2–6 were dissolved in a solution
of water–methanol (98 + 2) and analyzed by direct infusion (5
mL minꢁ1) at the optimum ion spray voltage of 4800 V. Com-
plex 5 was also been analyzed by direct infusion of diluting
mother solution (2 mL to 1 mL of water).
MS–MS experiments were performed in the collision cell q
on the isotopically pure peak of the selected precursor ions
by keeping the CE at 20 eV and unit resolution, and scanning
the TOF analyser. All acquisitions were averaged over 60 scans
at 8000 TOF resolution.
NiNOCC five membered chelated rings in the CSD (Cam-
bridge Structural Database, version 5.23, April 2002). The
two chelated acridine molecules are not coplanar, as demon-
strated by the dihedral angle of 92.2ꢂ between the two five-
membered rings. The coordination sphere is completed by
one ethylenediamine molecule.
As shown in Fig. 2, the supramolecular structure of complex
1 can be described in terms of both intermolecular hydrogen
bonding and p-p interactions.
The hydrogen bonding interactions involve the oxygen
atoms of the acridine ligands and the NH2 fragment of
the ethylenediamine molecules [O(1)ꢀ ꢀ ꢀH(2a)* ¼ 2.27 A,
Cytotoxicity
˚
The human prostatic carcinoma cell line, DU145, and the
human ovarian carcinoma cell line A2780 and its cisplatin-
resistant subline A2780-cp8,32 were cultured in RPMI-1640
medium supplemented with fetal calf serum (10%), gentamycin
(0.1%) and L-glutamine (1%) at 37 ꢂC in a 5% CO2–95% air
atmosphere in humidified incubator. After harvesting in the
logarithmic growth phase, cells were seeded in six-well plates
and treated with varying doses of complex 5 and cis-DDP
for 1 h. The medium was then removed and cells were rinsed
in phosphate-buffered saline (PBS) and incubated in fresh med-
Scheme 1 Synthetic route to the octahedral complexes 1 and 2.
New J. Chem., 2003, 27, 1497–1503
1499

View Article Online
Fig. 2 Crystal packing of [Ni(4-O-Acrid)2(en)] (1), viewed down the
a axis.
Fig. 1 Ortep drawing of the [Ni(4-O-Acrid)2(en)], (1), complex
molecule with the labelling scheme. The ellipsoids enclose 50% of
probability.
NO3ꢁ, PF6 at 1074, 1384 and 851 cmꢁ1, respectively, in the
ꢁ
IR spectra. High resolution mass spectrometry confirmed
the expected coordination sphere around the metal through
the formation of mono-charged molecular ions, organic ligands
and some ionic fragments, resulting from the direct and/or
consecutive elimination of neutral ligands under ESI-MS
condition. Moreover, the collision induced dissociation (CID)
behaviour of these complexes characterized by a sequential
loss of ligands has clarified the structure of the cations, even
in solution: one acridine orange, one chloride and one diamine
or two ammonia ligands per metal ion. It was only possible to
carry out 1H NMR analysis on complex 5 which was the most
soluble of complexes 3–6.
O(1)ꢀ ꢀ ꢀN(1)* ¼ 3.168(3) A, N(1)*–H(2a)*ꢀ ꢀ ꢀO(1) ¼ 156ꢂ,
˚
* ¼ ꢁx, ꢁy, ꢁz]. The second structural motif that charac-
terises the crystal packing is the p-p interaction between aro-
matic rings which are not involved in the coordination of the
acridine ligands. The shortest interplanar distance of 3.38 A
is found between the C(5)/C(10) phenyl ring and its centro-
symmetrical one.
An analogous Ni(II) complex, [Ni(4-O-Acrid)2(H2O)2] (2),
was obtained by reacting NiCl2ꢀ6H2O with 4-hydroxyacridine
in a basic aqueous solution using a weak aprotic base such as
piperidine and working at pH 8.04 to avoid precipitation of
Ni(OH)2 (Scheme 1).
Complex 2 was characterized in an aqueous solution using
high-resolution mass spectrometry. The results (see Experi-
mental section) confirmed the chelation of two acridine mole-
cules per metal ion, with the octahedral coordination sphere
completed by two water molecules. The minimum energy
structure, obtained performing MM calculations using UFF,
confirm the coplanarity of the two chelated acridine ligands,
with the two water molecules occupying the axial positions
of the octahedral sphere (Fig. 3).
˚
In the 1H NMR spectrum of 5 the resonance attributable to
the H4 and H5 protons is low-field shifted by about 0.5 ppm
and appears as a doublet with respect to the singlet of the
uncomplexed acridine orange. This observed shift can be
attributed to the close interaction between the platinum centre
and protons above and below the coordination plane,23 con-
firming the coordination of AO to the platinum through the
N atom. A slight downfield shift and a splitting into a doublet
of doublets also occurs for the H1 and H8 protons as observed
in the resonance of the corresponding protons in the ligand.
˚
The Ni–OH2 bond distances of 2.06 A and the H2O–Ni–
OH2 bond angle of 179ꢂ are in good agreement with values
found in other octahedral Ni(II) aqua complexes which contain
two coplanar N,O-chelated ligands (CSD).
Chemical studies of square-planar complexes
The coordinating capability of acridine orange was tested
using Pd(II) and Pt(II) complexes. Cationic triamine complexes
3–6 of the general formula cis-[M(A)n(Cl)(AO)]+ (A ¼ en,
NH3 ; n ¼ 1, 2) were synthesized using dichloro-ethylenedi-
amine Pd(II) and Pt(II) as well as cis-DPP as precursors.
The reactions, conducted in water solution, were performed
in two steps: (i) synthesis of the intermediate reactive water
derivatives; (ii) reaction in water solution with one equivalent
of acridine orange (Scheme 2).
Fig. 3 Minimized structure of the [Ni(4-O-Acrid)2(H2O)2] (2)
complex.
All the products isolated were ionic complexes as confirmed
ꢁ
by the presence of the typical bands of the counter ions BF4
,
ꢂ
˚
Table 1 Selected bond distances (A) and angles ( ) for complex 1
Ni–O(1)
Ni–N(2)
2.0567(18)
2.112(2)
Ni–N(1)
O(1)–C(14)
2.153(2)
1.302(3)
79.29(8)
O(1)–Ni–O(1a)
O(1)–Ni–N(2)
O(1)–Ni–N(2a)
N(1)–Ni–N(2)
N(2)–Ni–N(2a)
172.02(11)
89.32(8)
84.71(8)
O(1)–Ni–N(1)
O(1)–Ni–N(1a)
N(1)–Ni–N(1a)
N(1)–Ni–N(2a)
Ni–O(1)–C(14)
106.41(8)
91.93(12)
93.47(9)
168.38(8)
83.14(13)
114.45(17)
Scheme 2 Synthetic route to the square-planar complexes 3–6.
1500
New J. Chem., 2003, 27, 1497–1503

View Article Online
Moreover, the products 5 and 6, which were obtained start-
ing with cisplatin as precursor, are unique isomers, which most
probably consist, in both cases, of the cis isomer. Figs. 4 (a)
and 4 (b) show the minimum energy structures obtained per-
forming MM calculations of the two cations [Pt(en)(AO)Cl]+
and [Pt(NH3)2(AO)Cl]+. The acridine orange ligand is coordi-
nated to the metal centre via the nitrogen atom (bond distance
solution of 5 (Fig. 5b) and the initial addition of the first
and the second water molecule with the rapid disappearance
of the parent ion complex was observed. Positive product
ion spectrum (CID) of [C17H29N5O2ClPt]+ of 5 showed the
typical rearrangement reaction,37,38 which involves the loss
of the small neutral ligands and the formation of the ionized
AO (see Experimental section), providing further details about
the coordination sphere.
˚
Pt–Nacrid ¼ 2.06 A in both cases), with the mean plane nearly
orthogonal to the mean coordination plane (mean torsion
angle Cl–Pt–Nacrid–C of 88ꢂ and 78ꢂ, respectively). This geo-
metrical feature, as well as the bond distances and angles, is
found to be in agreement with those of other ionic complexes
of the type [M(A)n(Cl)(Am)]+ (A ¼ en, NH3 ; n ¼ 1, 2) where
Am is a DNA base N-bounded.15,31,32
Biological studies of cis-[Pt(NH3)2Cl(AO)][NO3] complex (5)
The biological activity of these new acridine orange derivatives
was conducted with complex 5 which was the most water solu-
ble of the 3–6 complexes.
Both complexes 5 and 6 are water soluble (5 more so than 6)
and stability in water was assessed using ESI-MS (electrospray
ionization mass spectrometry) experiments in positive ioniza-
tion mode.35 A typical mass spectrum obtained by direct infu-
sion of a water solution containing complex 5 is shown in Fig.
5a. Clearly, the expected parent ion at 528 m/z is not present;
in fact, the formation of [C17H29N5O2ClPt]+ at 565.17 m/z
indicates that the complex adds two molecules of water during
the ionization processes.36 Therefore, some ESI (+) experi-
ments were performed by direct infusion of a diluted mother
Cytotoxicity
The cytotoxic activity of complex 5 was compared with that of
cis-DDP, in three human cancer cell lines including the DU145
human prostatic cancer cell line, which is inherently resistant
to cis-DDP,39 the A2780 cis-DDP-sensitive ovarian cancer cell
line and A2780-cp8 sub-line, which manifests experimentally
induced-resistance to cis-DDP.40
As shown in Table 2, A2780 cells were equally sensitive to
both Pt-based drugs. On the other hand, the activity of
Fig. 4 Minimized structures of the (a) [Pt(en)(AO)Cl]+ and (b) [Pt(NH3)2(AO)Cl]+ cations of complexes 4 and 5–6.
Fig. 5 ESI(+) spectrum of complex 5; expanded region of molecular ion obtained by direct infusion (a) of a water solution containing 5, (b) of
diluted mother solution of 5.
New J. Chem., 2003, 27, 1497–1503
1501

View Article Online
Table 2 Mean IC50 values obtained for cis-DDP and complex 5
against the three tumors cell lines
IC50/mM ꢃ SD, cell lines
Complex
DU145
A2780
A2780-cp8
5
16 ꢃ 0.4
33 ꢃ 1.0
3.4 ꢃ 0.3
4.0 ꢃ 0.5
Values in parentheses are resistance factors, IC50 resistant/IC50 sensitive
31 + 0.5 (9.1)a
Cis-DDP
60 ꢃ 1.0 (15)
a
.
complex 5 was found to be twice as effective as that of cis-DDP
in the two cis-DDP resistant cell lines.
Specifically, in the A2780-cp8 cells the resistance factor was
lowered from 15 to 9.1 when going from cisplatin to complex
5. Overall, these results indicate that complex 5 is able to par-
tially lower cis-DDP resistance. This enhanced cytotoxic activ-
ity has been observed in cell lines characterized by different
mechanisms of cis-DDP resistance including decreased uptake,
increased GSH (glutathione) level and DNA repair (as in the
case of A2780-cp8)41 and alteration in the apoptotic pathways,
which compromise the ability to undergo cis-DDP induced
apoptosis (as in the case of DU145).42
Fig. 6 Effect of complex 5 and cis-DDP on A2780-cp8 and DU145
cell cycle. Representative fluorescence-activated cell sorter analysis of
DNA content (propridium iodide [PI] fluorescence).
phase fraction (43%) was observed as well as a persistent accu-
mulation of cells in the G2/M compartment. In addition, at 72
h recovery, about 25% of apoptotic cells were estimated. Cis-
DDP treatment failed to perturb the cell cycle as evaluated
at both 48 h and 72 h recovery time.
Cell cycle perturbation and apoptosis induction
DNA flow cytometric analysis was used to determine whether
complex 5 enhanced cis-DDP cytotoxicity through perturba-
tion of the cell cycle in these cis-DDP resistant cell lines.
Cell cycle analysis was performed on DU145 and A2780-cp8
cells 48 and 72 h after a 1 h exposure to equitoxic concentra-
tions (IC50) of complex 5 and cis-DDP. The kinetics of cell
cycle distribution in untreated and drug-treated cells are shown
in Table 3.
These results show that in A2780-cp8 and DU145 cells,
compared with cis-DDP, the higher cytotoxic activity observed
for complex 5 could be attributed to the specific induction of
cell cycle impairment and apoptosis. Consequently, complex 5
appears to activate pathways of cellular response for DNA
damage different from those caused by cis-DDP in the same
tumor models.
In A2780-cp8 cells, an S-phase cell accumulation was
observed after 48 h treatment with complex 5 (32.2% versus
24% untreated cells and 28% cis-DDP treated cells). In addi-
tion apoptosis, evaluated as the sub-G1 peak (arrow in
Fig. 6), was found in the 20% of A2780-cp8 cells treated with
complex 5.
After 72 hours of treatment, the S-phase accumulation of
A2780-cp8 cells treated with complex 5 was constant, and a
complete depletion of cells percentage in G2/M phase was
recorded. Moreover, at this time point the presence of 33%
apoptotic cells was observed, which indicates that the apopto-
sis induced by complex 5 in A2780cp8 is a consequence of
S-phase cell accumulation. Cis-DDP treatment did not perturb
the cell cycle at any time of recovery. In DU145 cells, 48 h
incubation with complex 5 caused 54% of the cells to accumu-
late in the G2/M phase and 15% of the cells recovered were
apoptotic. After 72 h of treatment a substantial increase in S
Conclusions
In conclusion, the present work has demonstrated the coordi-
nating ability of 4-OH-Acrid and AO ligands with the tran-
sition metal ions Ni(II), Pd(II) and Pt(II), leading to the
synthesis of new octahedral [(Ni(II)] and square-planar [Pd(II)
and Pt(II)] complexes, potentially definable as metallointercala-
tors. Between the two octahedral Ni(II) derivatives, complex 1,
[Ni(4-O-Acrid)2(en)], is probably the best candidate to work as
a metallointercalator. The presence of one molecule of ethyl-
enediamine, together with two extended aromatic heterocyclic
ligands, make this complex similar to other octahedral metal
amine complexes whose capabilities of interaction with the
DNA base pairs through both direct hydrogen bonding and/
or specific van der Waals contacts has been already tested.24
Therefore, deeper studies about the mode of interaction of
these types of complex with DNA are under investigation. In
the case of acridine orange ligand, the new triamine complexes
3–6 of the general formula cis-[M(A)n(Cl)(AO)]+Xꢁ (M ꢁ¼
Table 3 Cell cycle perturbation induced by complex 5 and cis-DDP
Pd(II), Pt(II); A ¼ en, NH3 ; n ¼ 1, 2; Xꢁ ¼ BF4ꢁ, NO3
,
48 h
72 h
PF6ꢁ) synthesized, fall in the general class of platinum triamine
cations with potential antitumor activity.15,16 The biological
studies conducted on complex 5 have demonstrated the poten-
tial clinical utility of this type of derivative opening the way for
further studies of DNA-binding able to clarify the structure–
activity relationships of this class of acridine complexes.
G1
(%)
S
(%)
G2/M
(%)
G1
(%)
S
(%)
G2/M
(%)
A2780-cp8
untreated
5a
55
24
18
65
21
14
46
52
54
32.2
28
21.8
19.5
20
70
60
66
27.2
22
2.4
21.7
15
Cis-DDPa
DU145
untreated
5a
25
19
25
50.6
21
29.4
54
20
3
67
43.3
20
54
13
Acknowledgements
Cis-DDPa
This work was partially supported by CIPE grants (Cluster 14
and 26) from MIUR.
a
Specific IC50
.
1502
New J. Chem., 2003, 27, 1497–1503

View Article Online
375; (b) F. Charmantray, A. Duflos, J. Lhome and M.
Demeunynck, J. Chem. Soc., Perkin Trans. 1, 2001, 2962.
20 W. I. Sundquist, D. P. Bancroft and S. J. Lippard, J. Am. Chem.
Soc., 1990, 112, 1590.
21 B. E. Bowler, K. J. Ahmed, W. I. Sundquist, L. S. Hollis, E. E.
Whang and S. J. Lippard, J. Am. Chem. Soc., 1989, 111, 1299.
22 S. A. Woodson and D. M. Crothers, Biochemistry, 1988, 27, 8904.
23 L. Maresca, C. Pacifico, M. C. Pappadopoli and G. Natile, Inorg.
Chim. Acta, 2000, 304, 274.
References
1
2
3
E. R. Jamieson and S. Lippard, Chem. Rev., 1999, 99, 2467.
P. J. Loehrer and L. H. Einhorn, An. Inten. Med., 1984, 100, 704.
(a) S. E. Sherman and S. J. Lippard, J. Chem. Rev., 1987, 87, 1153;
(b) J. Reedijk, Pure Appl. Chem., 1987, 59, 181.
M. A. Fuertes, C. Alonso and G. M. Perez, Chem. Rev., 2003,
103, 645.
4
5
6
G. Pizzocaro, R. Salvioni, M. Pazi, F. Zanoni, A. Milani, S.
Pilotti and S. Monfardini, Cancer, 1985, 56, 249.
D. Vugrin, W. F. Whitmore Jr. and R. B. Golbey, Cancer, 1983,
1, 211.
24 (a) K. E. Erkkila, D. T. Odom and J. K. Barton, Chem. Rev.,
1999, 99, 2777; (b) M. E. Nun˜ez and J. K. Barton, Curr. Opinion
Chem. Biol., 2000, 4, 199.
25 A. H. J. Wang, G. J. Quigley and A. Rich, Nucleic Acids Res.,
1979, 6, 3879.
7
8
A. W. A. Denny, Anti-Cancer Drug Des., 1989, 4, 241.
T. A. Gourdie, K. K. Valu, G. L. Gravatt, T. J. Boritzki, B. C.
Baguley, L. P. G. Wakelin, W. R. Wilson, P. D. Woodgate and
W. A. Denny, J. Med. Chem., 1990, 33, 1177.
K. W. Kohn, J. A. Hartley and W. B. Mattes, Nucleic Acids Res.,
1987, 15, 10 531.
26 H. M. State, in Inorganic Synthesis, ed. E. G. Rochow, McGraw-
Hill Book Co., New York, 1960, vol. VI, pp. 198–199.
27 B. J. McCormick and E. N. Jaynes, in Inorganic Synthesis,
ed. H. F. Holtzclaw, McGraw-Hill Book Co., New York, 1960,
vol. VIII, pp. 242–244.
9
10 F. Zunino, G. Savi and A. Pasini, Cancer Chemother. Pharmacol.,
1986, 18, 180.
11 A. Pasini, Gazz. Chim. Ital., 1987, 117, 763.
28 B. Kauffman and D. O. Cowan, in Inorganic Synthesis,
ed. J. Kleinerg, 1960, vol. VII, pp. 239–245.
`
29 A. K. Rappe, C. J. Casewit, K. S. Colwell and W. A. Goddard III,
J. Am. Chem. Soc., 1992, 114, 10 024.
12 B. D. Palmer, H. H. Lee, P. Johnson, B. C. Baguley, G. Wickham,
L. P. G. Wakelin, W. D. McFadyen and W. A. Denny, J. Med.
Chem., 1990, 33, 3008.
13 (a) H. H. Lee, B. D. Palmer, B. C. Baguley, M. Chin, W. D.
McFadyen, G. Wickham, D. Thorsbourne-Palmer, L. P. G.
Wakelin and W. A. Denny, J. Med. Chem., 1992, 35, 2983; (b)
M. D. Temple, W. D. McFadyen, R. J. Holmes, W. A. Denny
and V. Murray, Biochemistry, 2000, 39, 5593.
14 (a) D. Gibson, K. F. Gean, R. Ben-Shoshan, A. Ramu, I. Ringel
and J. Katzhendler, J. Med. Chem., 1991, 34, 414; (b) K. F. Gean,
R. Ben-Shoshan, I. Ramu, I. Ringel and J. Katzhendler, Eur.
J. Med. Chem., 1991, 26, 5593; (c) D. Gibson, N. Mansur and
K. F. J. Gean, Inorg. Biochem., 1995, 58, 79.
30 Molecular Simulation Inc., CERIUS, Version 3.4. Molecular
Simulation Inc., The Quorum, Barnwell Road, Cambridge,
England, 1994.
31 S. Parkin, B. Moezzi and H. J. Hope, J. Appl. Crystallogr., 1995,
28, 53.
32 L. Orlandi, G. Colella, A. Bearzatto, G. Abolafio, C. Manzotti,
M. G. Dandone and N. Zaffaroni, J. Europ. Canc., 2001, 37, 649.
33 (a) S. Neidle, G. L. Taylor and A. B. Robins, Acta Crystallogr.,
1978, B34, 1838; (b) R. Faggiani, B. Lippert and C. J. L. Lock,
Inorg. Chem., 1980, 19, 295.
34 J. D. Orbell, C. Solorzano, L. G. Marzilli and T. J. Kistenmacher,
Inorg. Chem., 1982, 21, 3806.
35 R. W. Vachet, J. A. Hartman, J. W. Gertner and J. H. Callahan,
Int. J. Mass Spectrom., 2001, 204, 101.
36 S. Blair, B. Goolsby and J. Brodbelt, Int. J. Mass Spectrom., 1999,
185, 49.
15 L. S. Hollis, A. R. Amundsen and E. W. Stern, J. Med. Chem.,
1989, 32, 128.
16 L. S. Hollis, W. I. Sundquist, J. N. Burstyn, W. J. Heiger-Bernays,
S. F. Bellon, K. J. Ahmed, A. R. Amundsen, E. W. Stern and S. J.
Lippard, Cancer Res., 1991, 51, 1866.
37 M. Kohler and J. A. Leary, Int. J. Mass Spectrom. Ion Processes,
1997, 162, 17.
38 M. Kohler and J. A. Leary, Am. Soc. Mass Spectrom., 1997,
8, 1124.
39 R. Uslu, U. A. Sanli, C. Sezgin, B. Karabulut, E. Tezioglu, S. B.
Omay and E. Goker, Clin. Cancer Res., 2000, 6, 4957.
40 B. A. J. Jansen, J. Brouwer and J. Reedijk, J. Inorg. Biochem.,
2002, 89, 197.
41 W. Wong and C. M. Giandomenico, Chem. Rev., 1999, 99,
2451.
42 M. Kartalou and J. M. Essigmann, Mutat. Res., 2001, 478, 23.
17 (a) A. Odani, R. Shimata, H. Masuda and O. Yamauchi, Inorg.
Chem., 1991, 30, 2133; (b) A. Odani, H. Masuda, O. Yamauchi
and S. Ishiguro, Inorg. Chem., 1991, 30, 4484.
18 (a) W. A. Denny, B. F. Cain, G. J. Atwell, C. Hansch, A.
Panthananickal and A. Leo, J.Med. Chem., 1982, 25, 276; (b)
S. A. Gamage, N. Tepsiri, P. Wilairat, S. J. Wojcik, D. P. Figgitt,
R. K. Ralph and W. A. Denny, J. Med. Chem., 1994, 37, 1486; (c)
T.-W. Hahn, T. P. O’Brien, W.-J. Sah and J.-H. Kim, Jpn J.
Ophthalmol., 1998, 42, 108.
19 (a) M. Takagi, H. Yokoyama, S. Takenaka, M. Yokoyama and
H. Kondo, J. Incl. Phenom. Mol. Recognit. Chem., 1998, 32,
New J. Chem., 2003, 27, 1497–1503
1503