Chemico-Biological Interactions 173 (2008) 84–96

Contents lists available at ScienceDirect

Chemico-Biological Interactions
journal homepage: www.elsevier.com/locate/chembioint

Copper–adenine complex, a compound, with multi-biochemical

targets and potential anti-cancer effect
Hassan H. Hammud a , Georges Nemer b , Walid Sawma b , Jhonny Touma b , Pascale Barnabe b ,
Yolla Bou-Mouglabey b , Amer Ghannoum a , Jida El-Hajjar b , Julnar Usta b,∗
a Beirut Arab University, Department of Chemistry, Beirut, Lebanon
b American University of Beirut, Department of Biochemistry, School of Medicine, Abdul-Aziz Street, Hamra, Beirut, Lebanon

a r t i c l e i n f o a b s t r a c t

Article history: A series of adenine–copper complexes (1–6) with various ligands (Cl− , SCN− , BF4 − and
Received 14 September 2007 acac [acetylacetonate ion]) have been synthesized and characterized by elemental analysis,
Received in revised form 20 February 2008
infrared spectroscopy and thermal analysis. Among the six complexes only complex (1),
Accepted 12 March 2008
Cu2 (adenine)4 Cl4 ·2EtOH (abbreviated as Cu–Ad), demonstrated some toxic effect on differ-
Available online 21 March 2008
ent cell lines. In vitro investigations of the biological effect of Cu–Ad complex have shown
that it: (1) binds genomic DNA; (2) decreases significantly, the viability of cells in culture in a
Keywords:
Copper–adenine concentration (15–125 ␮M)-dependant manner; an estimated IC50 of: 45 ␮M with HepG2;
HepG2 73 ␮M with C2C12; 103 ␮M with NIH3T3; and 108 ␮M with MCF7. Cu–Ad had no effect on
PCR A549 cells; (3) inhibits Taq polymerase-catalyzed reaction; (4) inhibits the binding of the
GATA-5 transcription factor GATA-5 to labeled DNA probes; (5) inhibits mitochondrial NADH-UQ-
TbX20 reductase with an estimated IC50 of 2.8 nmol, but had no effect on succinate dehydrogenase
WND activity; (6) increases reactive oxygen species (60%) at 45 ␮M Cu–Ad; and (7) decreases ATP
(80%) at 50 ␮M Cu–Ad. The new compound Cu2 (adenine)4 Cl4 ·2EtOH (Cu–Ad), belongs to a
class of copper–adenylate complexes that target many biochemical sites and with potential
anti-cancer activity.
© 2008 Elsevier Ireland Ltd. All rights reserved.

1. Introduction dination of the metal to residues at/or outside the active

Metal coordination complexes have been extensively
used in clinical applications as enzyme inhibitors [1],
anti-bacterial [2,3], antiviral [4–6] and as anti-cancerous
[7–9]. Different kinds of metals have been employed in
these complexes including platinum, gold, vanadium, iron,
molybdenum, cobalt, tin, gallium, copper and many others
[8].
Metals may have structural role in many enzymes. This
makes them natural targets in drug therapy and develop-
ment of enzyme inhibitors. The structural integrity and
hence function of the enzyme is influenced by the coor-
∗ Corresponding author.
E-mail address: justa@aub.edu.lb (J. Usta).
site of the enzyme, which may block the substrate binding
or interfere in the catalytic cycle. Alternatively metal com-
plexes targeting enzymes in the life cycle of the virus have
been used as antiviral [5,6]. The human immunodeficiency
virus for example, encodes three critical enzymes: a reverse
transcriptase, an integrase and a protease. The metallo-
thiosemicarbazones were recently characterized and found
to inhibit the reverse transcriptase; thus interfering with
virus life cycle.
Organo-arsenicals were used as anti-bacterial in the
treatment of syphilis and as anti-parasitic as feed additives
in livestock [2,10,11].
In addition metal coordination complexes have been
employed in the treatment of cancer. The hunt for new
drugs designed against specific proteins that are differ-
entially expressed in cancer cells remains an active field

0009-2797/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.cbi.2008.03.005
 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96
of research. Cisplatin discovered in 1965, has been use-
ful in the treatment of testicular carcinomas and ovarian,
head and neck, bladder and lung cancers [8]. Gallium
complexes inhibited specifically ornithine decarboxylase
activity inducible by phorbol esters [1,11]. Seleno com-
pounds have been found to exhibit an inhibitory effect on
protein kinase-C (PKC) one of the widely tackled targets of
anti-tumor agents [12–14]. Copper(II)–anthracyclin com-
plex exerted various effects on DNA topoisomerase-II [1,15].
An interesting group of chemotherapeutic agents used
in cancer therapy comprises molecules that interact with
DNA. The covalent or non-covalent association of agents
with DNA may inhibit mitotic progression and thus would
be used as antineoplastic drug [15,16]. Alternatively, the
cytotoxic effect exhibited by some cancer therapeutics may
be attributed to damage at nuclear DNA level, thereby inter-
fering with the replication and transcription machinery.
In addition, the anti-tumor effect may involve forma-
tion of DNA-adduct, site specific cleavage of DNA-strand,
alkylation of DNA [17,19] and shape distortion of the
genetic material which would interfere with RNA syn-
thesis [20]. Among the most common metal binding
sites to DNA, are the hetero-atoms of the nucleoside
bases that form strong complexes with transition metal
ions [8].
Copper is an essential trace element required for
enzymes and animal tissues in biological systems in both
the +1/+2 valence states; hence involved in redox met-
alloenzymes and in oxygenation, and oxygen-carrying
proteins.
The binding of copper ions to specific sites can modify
the conformational structures of proteins, polynucleotides,
DNA and bio-membranes [18,19]. The binding of Cu to
DNA occurs with high affinity. The crystal structure of the
complex formed between CuCl2 and DNA forms a pseudo-
octahedral geometry and involves copper binding to N7 of
guanine residue. No interaction with ribose hydroxyls is
observed. It has been found that in aqueous solution coor-
dination of Cu to the N7 and N1 sites of purine rings is
pH dependant that diminishes as the pH of the solution
increases [21].
In this study we describe the synthesis of six
copper–adenine complexes (1–6). Their ligands and struc-
tural components were identified by elemental analysis,
infrared spectroscopy and thermal analysis. The cytotoxic
effect of these compounds was assessed initially on dif-
ferent cell lines using trypan blue exclusion dye test.
Only compound (1), Cu2 (adenine)4 Cl4 ·2EtOH, exhibited a
cytotoxic effect and resulted in significant cell death in
HepG2 cells, verified as well by 3-(4,5-dimethyl thiazol-
2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. In
addition the compound was found to bind genomic DNA
and to exhibit significant inhibitory effect on different
biochemical targets including: Taq polymerase-catalyzed
reaction, binding of DNA to the transcription factor GATA-
5 and complex I of the mitochondrial electron transport
chain. Cu–Ad is a compound with potential anti-cancer
effects; further design and synthesis of structural analogues
is needed to establish proper structure function relation
that would increase the specificity towards biochemical
targets.
85
2. Methods
2.1. Chemical preparation of the copper–adenine
complexes (1–6)
Scheme 1 summarizes the starting metal salt, the lig-
and(s) and the solvent used in the synthesis of compounds
(1–6). The experimental details are described below. The
formula of compounds (1–6) were identified and ver-
ified by: (a) C, H, N elemental analysis performed by
Microanalytical Service, Department of Chemistry, Uni-
versity of Surrey, Guilford, Surrey, GU2 5XH and the
American University of Beirut, Environmental CORE Lab-
oratory, Beirut, Lebanon, (b) IR spectra were recorded
with Shimadzu 8300 FTIR spectrophotometer using the
KBr pellet method, (c) thermal analysis was performed on
Setaram-Labsys instrument using thermogravimetric and
differential scan calorimetric curves (TG–DSC). The heating
rate was 3 ◦ C/min. Dry nitrogen was passed over 15 mg of
the sample placed in a platinum crucible, and (d) UV–vis
spectra was obtained on Shimadzu UV-2101 PC UV–vis
scanning spectrophotometer.
2.1.1. Preparation of Cu2 (adenine)4 Cl4 ·2EtOH:
complex (1) (Cu–Ad)
Adenine (0.20 g, 1.48 mmol) was dissolved in boil-
ing ethanol (30 ml). To this solution was added 0.04 g
(0.74 mmol) of NH4 Cl solution in 10 ml ethanol and 0.39 g
(1.49 mmol) of copper acetylacetone solution, {Cu(acac)2 },
in 20 ml chloroform. The turbid mixture was refluxed for 2 h
and the greenish-blue precipitate was filtered, washed with
ethanol and dried in an oven under vacuum (38% yield). Ele-
mental analysis for Cu2 C24 H32 N20 O2 Cl4 , % Calcd: C, 31.66;
H, 3.55; N, 31.08; and % Found: C, 30.30; H, 3.36; N, 30.35
(MW: 901.1 g/mol). Complex (1) has low solubility in both
DMF and dimethyl sulfoxide (DMSO). Throughout the study
complex (1) was referred to as copper–adenine complex or
Cu–Ad.
2.1.2. Preparation of Cu(acac)2 (adenine)·EtOH:
complex (2)
The complex is prepared by refluxing adenine and cop-
per acetylacetone (1 mmol) in a mixture of chloroform
(50 ml) and ethanol (25 ml) for 2 h. The blue solution was
evaporated for few weeks at room temperature to yield blue
crystals (25% yield). Elemental analysis for CuC17 H25 N5 O5 ,
% Calcd: C, 46.10; H, 5.69; N, 15.81; and % Found: C, 49.96; H,
5.77; N, 16.05 (MW: 442.96 g/mol). Complex (2) is soluble
in both DMF and DMSO.
2.1.3. Preparation of Cu(acac)(adenine)·H2 O: complex (3)
To adenine (0.20 g, 1.48 mmol) in 40 ml boiling water,
a solution of copper acetylacetone (0.39 g, 1.49 mmol) in
20 ml chloroform was added drop wise. An immediate
violet precipitate was formed; the mixture was refluxed
for 1 h following which the precipitate was filtered and
dried under vacuum (17% yield). Elemental analysis for
CuC10 H14 N5 O3 , % Calcd: C, 38.03; H, 4.44; N, 22.19; and %
Found: C, 38.45; N, 24.1 (MW: 315.5 g/mol). Complex (3)
was neither soluble in DMF nor in DMSO.
86 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96
Scheme 1. Synthesis of copper–adenine complexes.
2.1.4. Preparation of Cu(adenine)2 Cl2 ·2H2 O: complex (4)
To adenine (1.35 g, 1 mmol) solution in 30 ml of boiling
ethanol, a solution of copper chloride dihydrate (0.086 g,
0.50 mmol) was added in 5 ml ethanol. The mixture was
refluxed for 1 h and the blue precipitate was filtered and
dried under vacuum (32% yield). Elemental analysis for
CuC10 H14 N10 Cl2 O2 , % Calcd: C, 27.24; H, 3.1; N, 31.79; and %
Found: C, 25.34; H, 3.04; N, 29.81 (MW: 440.45 g/mol). The
complex is soluble in both DMF and DMSO.
2.1.5. Preparation of Cu(adenine)2 (BF4 )2 ·2EtOH:
complex (5)
To adenine (1.35 g, 1 mmol) solution in boiling ethanol
(30 ml), tetrabutyl ammonium tetrafluoroborate (0.68 g,
3 mmol) and copper chloride (0.17 g, 1 mmol) in 30 ml
ethanol were added. The resulting bright yellow solution
was refluxed for half hour, and the precipitated green com-
plex was filtered and dried (28% yield). Elemental analysis
for CuC14 H22 N10 O2 B2 F8 , % Calcd: C, 28.04; H, 3.67; N, 23.37;
and % Found: C, 28.13; H, 3.6; N, 23.51 (MW: 599.17 g/mol).
The complex is soluble in both DMF and DMSO.
2.1.6. Preparation of
Cu(acetate)(adenine)(SCN)·0.5CH3 OH: complex (6)
To copper acetate dihydrate (0.60 g, 3.00 mmol) in 50 ml
methanol, adenine (0.44 g, 3.26 mmol) in 25 ml methanol
and ammonium thiocyanate (1.00 g, 13 mmol) in 25 ml
methanol were added. The resulting mixture was refluxed
for 1 h and the green precipitate was filtered and dried. Ele-
mental analysis for CuC8.5 H9 N6 O2.5 S, % Calcd: C, 30.86; H,
2.72; N, 25.42; and % Found: C, 29.1; H, 2.8; N, 26.49 (MW:
315.55 g/mol). The complex is sparingly soluble in DMF or
DMSO.
2.2. In vitro assays
2.2.1. Trypan blue exclusion assay
The cytotoxicity of the various complexes on different
cell lines was initially assessed qualitatively using trypan
blue exclusion test. In 12-well plates, six cell lines (C2C12,
HepG2, A549, NIH3T3, MCF7 and MCF7-BcL2) were cul-
tured in appropriate media (1 ml) under standard incubator
conditions of humidified atmosphere, 95% air, 5% CO2 and
temperature 37 ◦ C. Cu–Ad complex at varying concentra-
tion (15–125 ␮M) was added to cells (5 × 104 ). Cell toxicity,
expressed as %cell death was estimated after 24 h, by deter-
mining the ratio of dead cells (acquiring a blue color) to the
total number of cells and compared to a control treated with
the vehicle DMSO/ethanol not exceeding 2.5%.
2.2.2. MTT viability assay
Cell viability was assessed using MTT, as described by
the Kit manual (Roche). HepG2 cells (1 × 104 ) cultured in
96-well plates, were treated for 24 h with varying con-
centrations of: Cu–Ad (15–125 ␮M), adenine (15, 50 and
125 ␮M) and copper chloride (15, 50 and 125 ␮M), fol-
lowing which the formation of purple formazan by the
metabolically active cells was quantified at 595 nm using
 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96
an ELISA reader, Thermo Electron Corporation Multiskan
Ex multi-well spectrophotometer.
2.2.3. Reactive oxygen species (ROS) measurement
The intracellular ROS generation was measured as
described [22] using the nitroblue tetrazolium (NBT) reduc-
tion to formazan assay. NBT, at a final concentration of
1 mg/ml, was added to the control and Cu–Ad-treated
(24 h) HepG2 cells (1 × 104 ) in culture. Cells were lysed and
formazan was dissolved with 2 M KOH and 1.4 V DMSO.
The absorbance was monitored spectrophotometrically at
630 nm.
2.2.4. Intracellular ATP level
The intracellular ATP level was determined in control
and Cu–Ad-treated cells following the instruction manual
of the ATP Bioluminescence Assay Kit (Roche). In 12-well
plates, seeded HepG2 cells (5 × 104 per well) were treated
for 24 h with different concentrations of Cu–Ad. The cells
were then collected, lysed using the kit lysis reagent, then
50 ␮l were transferred to white MTP plates followed by
the addition of the luciferase reagent (50 ␮l). ATP biolu-
minescence was measured, compared to a control, as per
kit instruction manual using the Luminometer; Ascent FL,
Fluroskan-Thermo Lab Systems.
2.3. Binding of Cu–Ad to DNA
The binding of Cu–Ad to genomic DNA was assessed by
monitoring the change in absorption of Cu–Ad complex at
260 nm in the presence or absence of genomic DNA.
Genomic DNA was prepared following standard pro-
cedures. Blood sample (5 ml) was collected, added to
red blood cell lysis buffer (6 ml RCLB) and incubated at
37 ◦ C for 10–12 min. The mixture was then centrifuged
for 2 min at 4000 rpm and the supernatant was decanted.
To the pellet, RCLB (2 ml) was added, incubated again
for 10–12 min at 37 ◦ C and re-centrifuged for 2 min at
4000 rpm, the supernatant was discarded and the white
cells pellet was re-suspended in white cell lysis buffer
(1.8 ml) containing sodium dodecyl sulfate (24 ␮l SDS 10%)
and proteinase K (18 ␮l, 20 mg/ml). The suspension was
incubated 2 h at 55 ◦ C (or overnight at 37 ◦ C) following
which sodium chloride (3.3 mM) was added, vortexed and
centrifuged for 15 min at maximal speed. The supernatant
was transferred into a tube to which 3 ml of absolute
ethanol were added. The DNA flocculent was transferred
into 1.5 ml eppendorf tube containing 200 ␮l of ethanol
(70%), centrifuged for 5 min at maximal speed and the
supernatant was decanted. DNA was left to dry at room
temperature, then re-suspended in TE (1×) buffer (200 ␮l
of 1 mM Tris:0.1 mM EDTA). DNA concentration was
determined using UV–vis spectrophotometer SHIMADZU
UV-1201 at wavelength of 260 nm.
To determine if Cu–Ad binds genomic DNA, the UV–vis
spectra for each of Cu–Ad complex (10 nmol); genomic DNA
(800 ng); and copper–adenine:DNA (10 nmol:800 ng) mix-
ture was monitored in Tris–HCl (1 M) pH 7.0 using quartz
micro-cuvette (1 ml). All samples were premixed, incu-
bated for 15 min at 37 ◦ C prior to spectral scanning. No
peaks were observed in the visible region between 400 and
87
800 nm with 10 or 50 nmol of Cu–Ad; however, because of
possible low absorptivity at higher wavelength and lim-
ited solubility, Cu–Ad spectrum at higher concentration
(2.2 ␮mol) was performed by adsorbing it on a filter paper
saturated with 1 M Tris buffer pH 7.0, in the absence and
presence of DNA (1600 ng).
2.3.1. Gel shift assay
The effect of Cu–Ad on the binding of DNA probes to
nuclear factors, hence on transcription, was determined
using electrophoretic mobility shift assay (EMSA). In this
experiment, we investigated the effect of Cu–Ad on the
binding of GATA-5 to 32 P-labeled double-stranded oligonu-
cleotides harboring the GATA-5 consensus binding motifs.
2.3.2. Preparation of DNA probe
The double-stranded primer of GATA-5 was phosphory-
lated using T4 kinase that adds 32 P on the 5 -end. Briefly,
the phosphorylation reaction contained in a final volume
of 5 ␮l: 2 ␮l of the annealed GATA-5 primers {5 pmol, of
each of the forward: 5 -AGCCAGAGATAAGACCCG-3 ; and
reverse: 5 -CGGGTCTTATCTCTGGCT-3 primers}, 1 ␮l of T4
kinase, 1 ␮l kinase buffer (10×), 20 ␮l of 32 P-␥ATP and 1 ␮l
of water; all were mixed and incubated for 1 h at 37 ◦ C.
The probe was then loaded on 12% bis-acrylamide gel in
TBE 1× buffer (0.1 M Tris:0.1 M borate:2 mM EDTA pH 8.00)
at 110 V for 30 min. The pure double-stranded probe was
located after exposure to an XOMAT film and cut. The acry-
lamide gel containing the probe was incubated overnight
in 400 ␮l TES buffer (Tris:EDTA:NaCl) at 37 ◦ C with agita-
tion. The content of the tube was then added to a quick
spin column (Costar) and centrifuged for 10 s at maximum
speed. The probe was precipitated in 2.5× volume of probe
in 1 ml of ethanol, i.e. about 1 ml ethanol (100%) and cen-
trifuged at 4 ◦ C at 14,000 rpm for 15 min. The supernatant
was discarded and the pellet was washed with 700 ␮l of
70% ethanol and centrifuged at 14,000 rpm for 10 min. The
supernatant was discarded and ethanol was allowed to air
dry, then finally suspended in 50 ␮l TE (1×). The radioactiv-
ity of the probe was measured by adding 2 ␮l of the probe
to 4 ml scintillation liquid and counted using a ␤ counter.
2.3.3. Assembly for gel casting: sample preparation
The EMSA unit was assembled using 6% polyacrylamide
gel. A total volume of 100 ml of the gel was prepared
by adding 19 ml of acrylamide:bis-acrylamide (29:1) to
2.5 ml of 10× TBE (1 M Tris base:1 M borate:20 mM EDTA,
pH 8.00), 4 ml APS (1.6%), 74.5 ml water and finally 50 ␮l
TEMED. After polymerization, the gel was pre-run for
45 min at 205 V (or current intensity 15 mA) in a 0.25×
TBE buffer. Sample was prepared in a final volume of 20 ␮l
containing 4 ␮l of the binding buffer (4 mM Tris:24 mM
KCl:0.4 mM EDTA:0.4 mM DTT:5 mM MgCl2 :glycerol 5%),
1 ␮l poly-dI/dC (Amersham 1 ␮g/␮l), 3 ␮l of the inhibitor
of non-specific binding and the labeled probe (50,000 cpm)
then finally 3 ␮l of the nuclear extract and water. The
contents were incubated for 15 min at room temperature
following which samples (20 ␮l) were loaded on the gel.
Separation was carried by running the current (2–3 h) at
a constant voltage (205 V). The gel was then dried using
a Biorad drying system for 90 min at 80 ◦ C under vacuum
88 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96

and finally it was exposed and developed using X-ray film
(XOMAT). The effect of Cu–Ad complex at 10 and 25 nmol
was compared to controls of adenine, copper chloride and
DMSO.
2.3.4. Effect of Cu–Ad on Taq polymerase
The effect of Cu–Ad on Taq polymerase-catalyzed
gene amplification reaction was investigated. Two exons
namely: exon-11 of the ATP7B gene (Wilson gene) and
exon-1 of the transcription factor TbX20 gene were ampli-
fied in the presence and absence of the Cu–Ad complex
and compared to controls of DMSO, copper chloride and
adenine. Briefly, the amplification reaction, of final vol-
ume of 50 ␮l, contained 25 ␮l of IQ-Supermix (containing
100 mM KCl, 40 mM Tris–HCl, pH 8.4, 1.6 mM dNTPs, iTaq
DNA polymerase (Biorad) 50 units/ml, 6 mM MgCl2 ), 22 ␮l
of autoclaved H2 O, 1 ␮l of the forward and reverse primers
(250 ng/␮l, listed below) for exon-11 of ATP7B and exon-
1 of TbX20 genes and 1 ␮l of genomic DNA (250 ng/␮l).
The amplification program was as follows: Taq activation
(2 min at 94 ◦ C) and 38 cycles of denaturation (30 s at 94 ◦ C),
annealing (30 s at 60 ◦ C) and extension (30 s at 72 ◦ C) and a
final extension (7 min at 72 ◦ C). The following primers were
used in the amplification of E-11: ATP7B: forward: 5 -GCT
GTC AGG TCA CAT GAG TGC-3 ; reverse: 5 -CTG ATT TCC
CAG AAC TCT TCA CAT-3 ; E-1: TbX20: forward: 5 -TGT TTC
GGG TCT TTG TCT CC-3 ; reverse: 5 -TGC ACA TTC ACA GCA
TTC AA-3 .
The amplified PCR products were then separated by
electrophoresis at 80 V, on a 2% agarose gel in TBE 1× buffer
and the bands were visualized by UV illumination com-
pared to molecular weight ladder.
2.4. Effect of Cu–Ad on mitochondrial complexes I and II
Mitochondria were isolated from HepG2 cells as
described from isolated rat liver cells [23]. Mitochondria
were frozen (−80 ◦ C) and thawed three times before using
them in activity assay. The effect of Cu–Ad on respira-
tory chain complexes I and II, the main entry sites of the
reducing equivalents NADH and FADH2 , respectively, was
investigated. Activities of the rotenone sensitive NADH-UQ-
reductase (complex I) and antimycin sensitive succinate
dehydrogenase (complex II) were assayed spectrophoto-
metrically [24]. The rate of absorbance change (Abs) of
NADH oxidation at 340 nm and K3 Fe(CN)6 reduction at
400 nm, respectively, by mitochondria (30 ␮g) was moni-
tored with time (30 min). Mitochondria was pre-incubated
for 5 min with Cu–Ad (0.5–4.5 nmol) prior to initiation of
the reaction. The absorbance change was recorded and was
compared to a vehicle control at a final concentration of
1%.
3. Results
The synthesis of various copper–adenine complexes
(1–6) was achieved starting with different copper salts,
ligands and solvents. The complexes’ composition was con-
firmed by elemental analysis, infrared spectroscopy and
thermal analysis. A summary of the characteristic infrared
spectral bands for the various complexes 1–6 are listed in
Table 1. Thermal analysis of complexes 1–6 using TG and
DSC mode are presented in Table 2 showing the %mass loss
(theoretical and experimental) and the enthalpy change
(H) of the process.
Complexes 1, 2, 4 and 5 (15–125 ␮M) were qualitatively
screened initially for their cytotoxic effect on different cell
lines in culture, using the trypan blue exclusion test (data
not shown). Complexes 3 and 6 were not included because
of their relative insolubility in aqueous medium; they used
to come out of solution even at low concentration. Among
the six complexes, only complex 1, Cu–Ad, demonstrated
cytotoxic effects. Hence in this study we have investigated
the in vitro biological effects of Cu–Ad only.

Table 1
Infrared spectral data (KBr) for adenine and its copper complexes (1–6)

Band assignment Adenine 1 2 3 4 5 6

(OH) , NH2 3327 3332 3390 3356 3398 3415

NH2 , 2ıNH2 3294 3186 3267 3290 3171 3300 3338
(C8–H) , (C2–H) , NH2 3118 3112 3080 3095 3111 3130 3200
ıNH2 1668 1652 1653 1650 1657 1666 1653
(C O) 1582 1578 – – 1583
(C4–C5) , (C8–N9) + ı(C8–H) 1558 1558 1556 1558 1558 1575 1544
ı(N1–H) 1489 1489 1474 1489 1480 1489 1489
ı(C2–H) , (C8–N9) + ı(C8–H) 1450 1458 1451 1458 1458 1452 1464
(N1–C6N6) 1420 1437 1415 1398 1408 1404 1425
(C5–N7–C8) 1335 1335 1337 1335 1348 1346 1344
(N9–C8) + (N3–C2) + ı(C–H) 1308 1317 1300 1305 1319 1312 1308
ring , ı(C–H) 1252 1278 1256 1272 1246 – 1281
ı(C8–H) + (N7–C8) 1232 1211 1236 1236 1215 1215 1209
ı(C8–N7) + ı(C6–N6–H) 1157 1145 1157 1151 1147 1172 1153
(C2–N3) 1125 1138 1126 1118 1110 –
NH2 1024 1020 1020 1020 1018 1083 1031
NH2 + N1–C6 939 928 934 935 937 929 949
ı(N1–C2–N3) + (C5–N7) + (N9–H) 870 866 800 860 – – 858
NH2 ring deformation 640 654 656 657 684 665 638
o.o.p. deformation N9–H 621 617 619 611 611 594 623
R +  (N9–H) 666 680 680 670 672 665 671
WNH2 + ıNH2 542 575 540 547 547 550 554
ı(C O) – 557 561 – – 579
 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96 89

Table 2
Thermal analysis data: thermogravimetry (TG) and differential scan calorimetry (DSC) for the complexes (1–6)

Complex no. and formula T (◦ C) Experimental loss Theoretical loss Expected fragment Enthalpy change,
(%mass) (%mass) H (J/g) (process)

(1) Cu2 (adenine)4 Cl4 ·2EtOH 71.76 10.96 10.21 2EtOH 251.74 (endo)
233.01 16.01 15.76 4Cl 9.31 (exo)
339.64 19.22 18.65 4N2 CH 123.47 (endo)

(2) Cu(acac)2 (adenine)·EtOH 100.35 10.93 10.38 EtOH 89.06 (endo)

207.53 42 44.7 2acac 207.25 (endo)
260–400 12.92 13.32 N2 CH2 , NH2 (endo)

(3) Cu(acac)(adenine)·H2 O 186.52 29.29 31.38 Acac 168.2 (endo)

229.4 5.64 5.7 H2 O 8.2 (endo)
316.38 12.92 13.3 N2 CH2 12.87 (endo)
394.59 120.9 (endo)

(4) Cu(adenine)2 Cl2 ·2H2 O 69.53 6.5 8.17 2H2 O 106.89 (endo)
216 7.85 8.06 Cl 11.24 (endo)
248.53 7.89 8.06 Cl 8.91 (exo)
275.76 35.96 31.3 2NH2 24.94 (endo)
378.99 2C2 N2 H 204.57 (endo)

(5) Cu(adenine)2 (BF4 )2 ·2EtOH 112.49 6.08 7.68 EtOH 18.5 (endo)
255.06 17.4 17.36 EtOH, N2 CH2 , NH2 131.82 (endo)
387.41 19.37 21.5 N2 CH2 , BF4 172.61 (endo)
450–750 42.87 45.54 BF4 , C4 N3 H3 , C4 N2 H (endo)

(6) Cu(acetate)(adenine)(SCN)·0.5CH3 OH 79.14 9.19 6.82 0.5CH3 OH 160.3 (endo)

258.46 16.41 17.48 Acetate 124.68 (endo)
331.97 14.13 17.19 SCN 92.01 (endo)
410.22 11.82 12.15 N2 CH 62.28 (endo)

Cu–Ad complex came out as a powder from the reaction
mixture. Attempts to crystallize complex 1 from solvent
systems failed which limited its structural characterization
by X-ray and mass spectroscopy. Spectral analysis of Cu–Ad
(2.2 ␮mol) in Nujol revealed (data not included) the follow-
ing max at 288 nm (due to ␲–␲* of adenine ligand), 395 nm
(shoulder) and 580 nm (due to d–d transition). Nujol how-
ever interacted with DNA thus could not be used to monitor
the spectral changes occurring upon binding to Cu–Ad.
Complex 1 was stable under all biological assay condi-
tions; thermal analysis findings have shown (Fig. 1) a mass
loss corresponding to 2EtOH molecules at 75 ◦ C; 4Cl groups
at 233 ◦ C, and 4N CH–N fragments of adenine at 339 ◦ C
confirming their presence in the Cu–Ad complex (Table 2).
The high temperature needed for adenine fragmentation is
indicative of the relative stability of copper coordination to
adenine in the complex Cu–Ad.
3.1. In vitro biological effects of Cu–Ad complex
3.1.1. Cytotoxicity of Cu–Ad
The effect of varying Cu–Ad concentration (15–125 ␮M)
on different cell lines in culture was evaluated using trypan
blue exclusion test. Our results (Fig. 2a) show a significant
(p < 0.005) increase in cell death with increasing concentra-
tion of Cu–Ad. A maximal cell death of: 96% with HepG2,

Fig. 1. Thermal analysis of Cu–Ad complex using TG and DSC mode and showing %mass loss and the enthalpy change of the process.
90 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96

Fig. 3. Representative UV-absorption spectra of Cu–Ad (yellow tracing),

Cu–Ad + DNA (blue tracing) and DNA (red tracing). Genomic DNA (800 ng)
was incubated with 10 nmol of Cu–Ad for 15 min at 37 ◦ C in 1 M Tris buffer
pH 7.0. Absorption spectrum (200–400 nm) was monitored using Shi-
madzu spectrophotometer. (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of the article.)

pH 7.0, in the presence and absence of DNA (1620 ng) then

scanned between 200 and 800 nm. Cu–Ad alone exhibited
the following max at 261, 343 and 590 nm; whereas when
mixed with DNA we observed an increase in the absorbance
at 261 nm, a shift in the absorption from 343 to 402 nm and
Fig. 2. Effect of Cu–Ad on different cell lines in culture. Cells cultured
under optimal conditions of humidity and using appropriate media were a small shift from 590 to 587 nm (data not shown).
treated for 24 h with Cu–Ad (15–125 ␮M). %cell death was compared to a
control, considered as 100% treated with vehicle. The effect of Cu–Ad was 3.1.2. Cu–Ad inhibits DNA binding to transcription factor
estimated using: (a) trypan blue exclusion assay compared to a control
Nuclear extracts expressing the rat GATA proteins
treated with vehicle and (b) MTT assay performed only on HepG2 cells.
were incubated with 32 P-radio-labeled double-stranded
oligonucleotide containing GATA consensus binding motifs.
84% with C2C12, 74% with NIH3T3, 69% with MCF7, 60% The addition of 5, 10 and 25 nmol of Cu–Ad inhibited com-
with MCF7-Bcl2, 28% with A549, was obtained with final pletely GATA-5 binding to DNA (Fig. 4) whereas a similar
concentration of 125 ␮M Cu–Ad. Estimated IC50 for the dif- effect was obtained with CuCl2 , neither adenine nor DMSO
ferent cell lines were: 45 ␮M HepG2; 73 ␮M C2C12; 103 ␮M exerted any on transcription.
NIH3T3; 108 ␮M MCF7; and 115 ␮M MCF7-BcL2, indicating
that HepG2 cells are the most sensitive, while A549 cells are 3.1.3. Cu–Ad inhibits Taq polymerase activity in vitro
the least sensitive. The toxicity of Cu–Ad on HepG2 cells was The effect of Cu–Ad on replication was assessed
further verified using the MTT assay. A decrease (Fig. 2b) in indirectly via its effect on amplification by Taq polymerase-
HepG2 viability was obtained with increasing Cu–Ad con-
centration with an estimated IC50 of 40 ␮M. Neither free
adenine nor copper chloride salt, when added to HepG2
cells, at a final concentration of 30, 100 and 200 ␮M, exerted
any effect (data not shown) on cell viability indicating the
effect is due to intact Cu–Ad complex.

3.1.1.1. Cu–Ad binds genomic DNA. The interaction of the

copper–adenine complex with genomic DNA was investi-
gated by monitoring the changes in the absorption intensity
of free Cu–Ad complex compared to that of DNA pre-
incubated with Cu–Ad. The UV spectra of Cu–Ad showed a
single peak at 260 nm characteristic of the aromatic ade-
nine base. Incubation of Cu–Ad complex (10 nmol) with
genomic DNA (800 ng/ml) isolated from human blood,
decreased the absorption intensity by 30%; the yellow and
blue tracings in Fig. 3 are for Cu–Ad complex alone and
Cu–Ad complex with DNA, respectively. No other peaks at
higher wavelength were observed at the Cu–Ad concentra-
tions (10 and 50 nmol) used in our study. To examine if spec-
Fig. 4. Representative figure of the effect of Cu–Ad on gel shift assay. Dif-
tral changes, at higher wavelength, occur upon the binding
ferent concentrations of Cu–Ad were added to nuclear extracts treated
of Cu–Ad to DNA, high amount of Cu–Ad (2.2 ␮mol) were with GATA-5 32 P-labeled DNA probe, then separated, exposed and com-
adsorbed on a filter paper saturated with 1 M Tris buffer, pared to CuCl2 , adenine and DMSO (vehicle) as described in Section 2.
 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96 91

Fig. 5. Representative figure of the effect of Cu–Ad on Taq polymerase-catalyzed reaction. The effect of Cu–Ad on the amplification of exon-11 of the ATP7B
(Wilson gene) and exon-1 of the TbX20 gene was investigated and compared to CuCl2 , adenine and DMSO.

catalyzed reactions in vitro. The amplification of exon-11

of the Wilson gene (WND, ATP7B) and exon-1 of the
TbX20 gene was examined. The complex was added to the
PCR reaction with the appropriate forward and reverse
primers for each of the amplified exons. A concentration-
dependant decrease in the intensity of the amplified bands
in both WND (54%) and TbX20 (68%) genes (Fig. 5) was
observed with 0.5 nmol of Cu–Ad complex. Similar results
were obtained when Cu–Ad complex was subjected to
the PCR program, heating–cooling cycles and prior to
adding it to the Taq polymerase amplification reaction
indicating thus its stability under the reaction conditions.
On the other hand CuCl2 (0.5 and 1 nmol) inhibited the Taq
polymerase reaction whereas adenine and DMSO had no
effect.

3.1.4. Effect of Cu–Ad on mitochondrial complexes

Copper–adenine effect on mitochondrial function
was assessed via testing its effect on the activity of
both complexes I and II, the main entry site of NADH
and FADH2 , respectively. Our results (Fig. 6a) show a
concentration-dependent decrease in the activity of com-
plex I reaching a maximal inhibition (74.8%) at 4.0 nmol
Cu–Ad, with an estimated IC50 of 2.8 nmol. On the other
hand, Cu–Ad (1–4 nmol) exerted no effect on complex II
activity (Fig. 6a). Control enzyme activities of 104 ± 10.3
and 94.8 ± 8.4 abs/(mg/h) (considered as 100%) were for
complex I (NADH-UQ-reductase) and complex II (succinate
dehydrogenase), respectively, treated with the vehicle of
Cu–Ad.

3.1.5. Cu–Ad increases ROS generation and decreases

intracellular ATP level
Inhibition of complex I by Cu–Ad will decrease
electron transfer through the electron transport chain;
consequently will decrease ATP level and increase ROS gen-
eration. We examined the level of ATP in Cu–Ad-treated
HepG2 cells. Our results show 80% decrease in ATP level at Fig. 6. The effect of Cu–Ad on (a) NADH-UQ-reductase () and
50 ␮M Cu–Ad (Fig. 6b). succinate dehydrogenase () activities in mitochondria (30 ␮g) iso-
Similarly Cu–Ad decreased NBT reduction into for- lated from HepG2 cells. Activities were compared to a control
value (100%) of 104 ± 10.3 abs/(mg h) for NADH-UQ-reductase and
mazan. Our results (Fig. 6c) show a decrease in NBT
94.8 ± 8.4 abs/(mg h) for succinate dehydrogenase treated with vehicle.
reduction with increasing concentration of Cu–Ad, indicat- (b) ATP bioluminescence assay as per supplier instructions and (c) NBT
ing an increase in ROS generation. A 60% increase in ROS reduction into formazan. Results are average ± S.E.M. of three determina-
was obtained at 45 ␮M Cu–Ad. tions from three different experiments.
92 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96

4. Discussion

In this study the toxicity and biochemical targets of

a copper coordination complex synthesized with adenine
as ligand, were assessed in vitro. The compound Cu–Ad
induced cell death in many cell lines and bound genomic
DNA. In addition it inhibited Taq polymerase-catalyzed
reactions, the binding of transcription factor to DNA and
mitochondrial NADH oxidation in vitro. Cu–Ad increased
also ROS generation and decreased ATP level that may be
one of the possible underlying mechanisms of its cytotox-
icity. These findings identify Cu–Ad as a compound with
several biochemical targets and potential anti-cancerous
activity.
Copper complexes 1–6 were synthesized (Scheme 1)
from different ligands and solvents and were initially quali-
tatively (data not shown) screened for their cytotoxic effect
on different cell lines in culture. Out of the six complexes,
only complex 1 (Cu–Ad), induced cell death in all the tested
cell lines, except for A549 (Fig. 2a). The most sensitive were
HepG2 cells showing a dose-dependant effect (Fig. 2b).
Elemental analysis, comparative IR spectra and thermal
analysis profile of complexes 2–6 verified the structural
components of Cu–Ad.
Several bands appeared in the IR spectra of adenine
and its copper complexes (Table 1). Bands in the region
3500–3000 cm−1 could be attributed to stretching vibra-
tion for –NH group of adenine and OH group of water,
ethanol or methanol which are present in the lattice of
the complexes [25,26]. The peak at 3390 cm−1 could be
attributed to OH of water in complex 3 and at 3415 cm−1
to OH of methanol in complex 6. The carbonyl stretch of Fig. 7. Crystal structure of Cu(acac)2 (adenine)·EtOH compound (2).

acetylacetone is clearly observed at 1582 and 1578 cm−1 in

complexes 2 and 3, respectively, while the carbonyl stretch
of the acetate appeared at 1583 cm−1 in complex 6.
The NH2 stretches of adenine at 3290–3118 cm−1 were
slightly shifted upon adenine complexation with metal
(Table 1). The ıNH2 mode of free adenine at 1668 cm−1
undergoes shifts to about 1650 cm−1 as expected for cor-
responding N-bounded adenine complexes [25–27]. Thus
NH2 group (exocyclic NH2 nitrogen) is coordinated to the
metal ion in the complexes 1–4 and 6. Adenine coordi-
nate through ring nitrogen with appreciable shifts and
occasional splitting of C C, C N and ring vibrations of
the ligand (1605–1300 cm−1 ) [28–30]. The NH region,
2900–2500 cm−1 , suffered considerable changes in the
metal adenine complexes resulting in two weak maxima in
this region [26]. The 1232 cm−1 of adenine due to N7–C8
shift to lower wavelength upon complexation in the case
of complex 1 and 4–6 indicating the binding of adenine is
through ring nitrogen [25,29].
Various coordinated sites have been observed for ade-
nine in copper complexes as indicated by X-ray studies:
among the four nitrogens N(1), N(3), N(7) and N(9) of ade-
nine, the N(9) is the most basic hence bears a proton ren-
dering it the most preferred metal binding site. An example
is the crystal structure of [Cu(tren)(adeninato)]ClO4
where N(9) of adeninate anion is bound to Cu at
the axial position {tren = Tris(2-aminoethyl)amine} [31].
The same mode of binding is observed in the crystal
structure of [Cu(tren)(adeninato)]Cl·2H2 O [32]. Adenine
cation also coordinates through ring nitrogen N(9)
in [Cu(adenineH)2 Cl2 ]2+ [33]. On the other hand, the
crystal structure of [Cu(MOBIDA)(adenineH)(H2 O)]. H2 O
reveals the selective formation of rare Cu–N(3) adenine-H
bound, (MOBIDA = N-(p-methoxybenzyl)-imidiacetato(2-))
ligand [33]. In a recent X-ray study the structure of
Cu(acetylacetone)2 (adenine)·EtOH shows that the complex
is square pyramidal with acetylacetone ligands occupying
the equatorial positions and adenine in the axial position
coordinating through its N(7) nitrogen [34,35] (Fig. 7).
Further evidence on the formula of adenine–metal
complexes comes from thermal analysis (Table 2). The
%mass loss with expected fragments and involved heat dur-
ing thermal analysis TG–DSC of various complexes (1–6)
are listed in Table 2. Thermal analysis of free adenine
shows the presence of two decomposition steps: one cen-
tered at 383 ◦ C and the second at 667 ◦ C, with associated
endothermic heat of 892 and 376 J/g due to loss of C4 N3 H3
and N2 CH2 , respectively. Coordinated adenine although
showed two degradation steps in all the studied complexes
at similar temperature regions 250–400 ◦ C and 400–800 ◦ C,
the %mass loss shifts to lower values in the first step and to
larger values in the second step. The binding of N-adenine
to copper introduces some sort of rigidity to the ring struc-
ture that degrades in the second step with increased %mass
loss. While N CH–NH of five-member ring of adenine is
lost in the first step in most cases, TG–DSC curve for com-
plex 1 is shown in Fig. 1. In addition thermal analysis of
 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96
Fig. 8. Proposed structure of Cu–Ad (1) with formula [Cu(adenine)2 Cl]2 Cl2 ·2EtOH when L = Cl− , and [Cu(adenine)2 (EtOH)]2 Cl4 when L = ethanol.
copper bis(acetylacetone) shows a sharp endothermic peak
at 200 ◦ C (H = 465 J/g). The same pattern is observed in
complexes 2 and 3 which clearly confirm the presence of
acetyl acetone in both.
Previous studies have investigated the interaction of
Cu(II) with adenine in solution. These studies reported
that copper adeninate complexes were stable involving N3
and N9 as binding sites simultaneously [36]. The binding
constants for copper and related metal complexes have
indicated the high stability of these complexes [36–38].
In this study the prepared Cu–Ad complex remained sta-
ble in DMSO. No decomposition or change in its spectral
properties or thermal data was obtained even after a
long time exposure to atmosphere. The crystal structure
(Fig. 7) of complex 2, [34] has shed some light on ade-
nine binding to copper in Cu–Ad; both complexes 1 and
2 were prepared similarly; with one exception the addi-
tion of ammonium chloride to Cu–Ad. A dimeric structure
is proposed for Cu–Ad with four bridging adenine and
chloride or ethanol as axial ligand (L) suggesting the fol-
lowing formulae, respectively [Cu(adenine)2 Cl]2 Cl2 ·2EtOH
or [Cu(adenine)2 (EtOH)]2 Cl4 (Fig. 8). Previous investiga-
tors have reported with related complexes [39,40] similar
dimeric structures involving bridging of adenine through
its N3 and N9 with the axial sites occupied by water or
chloride ion.
Of the entire complexes (1–6) only Cu–Ad complex
significantly decreased cell viability in a concentration-
dependant manner. HepG2 and C2C12 were the most
sensitive whereas A549 was the least sensitive (Fig. 2a).
The obtained effects indicate that Cu–Ad is bioavailable,
capable of overcoming the physical barriers and entering
the cells. Prolonged treatment (data not shown) of HepG2
with Cu–Ad for 36 and 48 h resulted in less cell death when
compared to 24 h, which may be attributed to proliferation
of HepG2, instability of the Cu–Ad or possible metabolism
of Cu–Ad into less toxic compound. Our experiments do
not provide direct evidence, as to whether Cu–Ad gets bio-
93
transformed with time. However no effect on cell viability
was obtained when cells were treated with either CuCl2
or free adenine; thus indirectly ruling out the dissociation
of Cu–Ad into its primary components. We hence opted to
further investigate the target Cu–Ad in vitro.
Coordination metal complexes among which is copper,
have been designed and used in the treatment of many
diseases including cancer. Of the many suggested mech-
anisms is the possible interaction with DNA and disruption
of its tertiary structure. This may influence the recognition
of DNA by specific factors or proteins leading ultimately to
inhibition of DNA synthesis and or transcription [2].
In this study we show (Fig. 3) a decrease in the
absorbance of Cu–Ad compared to that pre-incubated with
genomic DNA, suggesting binding of the complex to the
DNA. On the other hand the observed increase in the
absorbance of the Cu–Ad–DNA (blue tracing) compared to
that of genomic DNA (red tracing) may result from dis-
ruption of the double-stranded DNA into single strand by
Cu–Ad. A shift in the structure of DNA from double stranded
into single is known to increase the absorbance. Alterna-
tively, possible base pairing, favored by ␲–␲ interactions,
between the adenine moieties of the Cu–Ad complex and
the thymine bases of the genomic DNA would facilitate
the separation of the double-stranded DNA favoring single-
stranded chains, subsequently increasing absorbance [18].
Molecules that modify nucleic acids have received con-
siderable attention because of their potential application
as chemotherapeutic agents. Transition metal complexes
endowed with redox properties and DNA affinity have
been developed as chemical nucleases. DNA cleavers such
as iron–bleomycin, Mn(III)–porphyrin, nickel, Co, Rh, and
Ru phenanthroline/bipyridine and Cu(phen)2 2+ complexes
were able to mediate oxidative damage to nucleobases/or to
deoxyribose moiety [41]. Among the many possible mech-
anisms of metal carcinogenesis, is the direct or indirect
binding of the metal ion to DNA. The direct binding to
DNA involves the binding of the metal to either phosphates
94 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96
and/or nucleobases. The coordination of the divalent metal
ion to N7 of guanine in DNA promotes the proton switch
of the hydrogen from guanine to cytosine, and increases
the probability of mispairing hence mutations [42]; similar
situation may be occurring with the Cu–Ad complex.
The ability of copper II complexes to interact with DNA
and induce strand scission was reported [43]. Copper-
adenylated polymer templates resulted in the complete
conversion of supercoiled into nicked DNA. The above
finding prompted us to investigate the in vitro effect of
Cu–Ad complex on gene amplification by Taq polymerase
(indirectly replication) and on the binding of transcrip-
tion factors to DNA probes (indirectly transcription). Cu–Ad
complex inhibited both.
Taq polymerase is an enzyme that catalyzes the in
vitro amplification of gene/exon by polymerizing deoxynu-
cleotides. Cu–Ad inhibited significantly the amplification
of exon-11 of the Wilson gene, a copper ATPase encoding
gene. To rule out the effect is because it is a copper-related
gene, exon-1 of TbX20 gene was amplified in the presence
of Cu–Ad (Fig. 5). Again the Taq polymerase activity was
significantly inhibited by Cu–Ad. Neither adenine nor the
vehicle (DMSO) demonstrated any effect. CuCl2 however
exhibited a similar effect, suggesting inhibition is proba-
bly mediated by the copper. When subjecting the Cu–Ad
alone to the PCR-programmed cycle of heating and cool-
ing prior to adding it to the Taq polymerase amplification
reaction, we obtained the same result (data not shown)
indicating that Cu–Ad complex is stable under these con-
ditions. In addition thermal analysis data have shown the
presence of two-step decomposition of the adenine at tem-
perature greater than 200 ◦ C. Currently our data do not
indicate if this inhibition is the result of a direct inhibition of
Cu–Ad on the Taq polymerase or is an effect of the Cu–Ad
on DNA. Cu(II) complexes were reported to cleave inacti-
vated peptide bonds and the thermostable enzyme Taq DNA
polymerase at ␮M concentration under mild pH and tem-
perature conditions [42,44]. Strand breaks and modified
bases were induced in vitro by copper ion-mediated reduc-
tion of hydrogen peroxide in the presence of ascorbate
[45]. Copper and copper(I)-metalated nucleobase polymers
induced DNA cleavage and promoted hydrolysis of non-
natural and natural phosphate ester substrates [46,47]. In
addition Bruston et al. reported hydrogen peroxide dis-
proportionation into oxygen and water by the catalytic
activity of copper–adenine complexes [48]. DNA cleavage
and complete conversion of supercoiled plasmid DNA into
linearized DNA was reported to occur by the catalytic activ-
ity endowed to copper complexes in the presence of a
co-oxidant [46].
In a previous study phenanthroline Cu(I) complex inhib-
ited transcription by binding to a site generated by the
interaction of the RNA polymerase with the promoter that
may partly be attributed to its potent cytotoxicity [49]. We
hereby show inhibition (Fig. 4) of the binding of the tran-
scription factor GATA-5 to DNA probe (harboring GATA-5
consensus site) by Cu–Ad. Likewise neither adenine nor
DMSO had any effect whereas CuCl2 did.
Among transition metal ions, copper-based artificial
systems have received considerable focus because of their
dual mode of action in modifying nucleic acids either
through oxidative scission or via hydrolytic pathway.
Induced DNA strand scission has been reported to occur
by reactive radical species generated in the presence of
transition metal ion catalysts and redox [50]. Of all essen-
tial metallo elements, copper has the highest bonding
affinity for nucleic acids. Nucleic acid phosphate and base
donor sites yield Cu chelates that are much more sta-
ble than Cu chelates of most amino acids and near the
stability of protein chelates. A higher physiological concen-
tration accounts for inhibition of ribonuclease syntheses:
initiation, nucleotides condensations and release of newly
formed mRNA [51].
Contrary to Cu–Ad or complex 1, monomeric complexes
such as complex 2 (Fig. 7), complex 4 and complex 5 did
not exert any cytotoxic effect. The proposed dimeric struc-
ture of Cu–Ad in which bridging of adenine to two copper
atoms occurs via N3 and N9, would leave behind N1 and
N6 available for H-bonding with thymine base of DNA.
This substitutes the normal hydrogen bonding that exists
between adenosine and thymine in double-stranded DNA,
contributing thus to some of the cytotoxic effects imparted
by Cu–Ad [52].
Mitochondria are one of the subcellular targets of metal-
based toxicity drugs. For instance severe morphological
alterations were reported to occur following cisplatin treat-
ment to isolated or inside intact kidney cells. These include
mitochondrial swelling, collapse of membrane potential,
inhibition of ADP stimulated respiration and increased for-
mation of reactive oxygen species [15,53]. Mitochondria
were also suggested to be involved in the maintenance
of malignant phenotype by unknown mechanisms related
to ATP production [54]. In tumorogenesis the role of
mitochondria may involve the transfer and insertion of
mitochondrial DNA into nuclear DNA or altering the expres-
sion of mt-DNA encoded proteins.
Cu–Ad inhibited NADH-UQ-reductase activity in iso-
lated HepG2 mitochondria but had no effect on succinate
dehydrogenase activity (Fig. 6a). NADH-UQ-reductase or
complex I of the respiratory chain complex, is the main
entry site of the reducing equivalent NADH whose oxi-
dation is coupled to ATP production. Cu–Ad (2.8 nmol)
inhibited 50% of the rotenone sensitive NADH oxidation of
the ETC, which is in line with reported toxic concentration
of the copper-based antineoplastic drugs, the casiopeinas
[15]. Inhibition of NADH oxidation would increase oxy-
gen radical generation, deplete intracellular ATP level
and ultimately lead to cell death. Alternatively inhibition
of complex I would increase oxygen radical generation
favoring oxidative stress and leading to DNA damage by
modifying bases or deoxyriboses [18]. In this study we show
an increase in ROS generation and a decrease in ATP pro-
duction (Fig. 6b and c) which may be one of the mechanisms
underlying the demonstrated toxicity of Cu–Ad.
In summary, we report in this paper the synthesis of
a copper–adenine complex, a compound with multi in
vitro biochemical effects. The Cu–Ad complex was relatively
toxic and was found to inhibit Taq polymerase-catalyzed
reaction, the binding of transcription factors to DNA probes
and mitochondrial NADH oxidase activity. The different
multi-biochemical targets and effects of Cu–Ad identify it
as a potential anti-cancerous compound. Design of different
 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96
structural analogues of Cu–Ad will allow better evalua-
tion of structure–function relationship and increases target
specificity and potency.
References
[1] A.Y. Louie, T.J. Meade, Metal complexes as enzymes inhibitors, Chem.
Rev. 99 (1999) 2711–2734.
[2] R.N. Patel, N. Singh, K. Shukla, U.K. Chauhan, S. Chakraborty, J. Niclos-
Gutierrez, A. Castineiras, X-ray, spectral and biological (antimicrobial
and superoxide dismutase) studies of oxalate bridged Cu-II com-
plexes with penta methyldiethylenetriamine as capping ligand, J.
Inorg. Biochem. 98 (2004) 231–323.
[3] A.H. Fairlamb, G.B. Henderson, A. Cerami, Trypanothione is the pri-
mary target for arsenical drugs against African trypanosomes, Proc.
Natl. Acad. Sci. A86 (1989) 2607–2611.
[4] Z. Balcarova, J. Kasparakova, A. Zakovska, O. Novakova, M.F. Sivo, G.
Natile, V. Brabeck, DNA interactions of a novel platinum drug, cis-
[PtCl(NH3 )2 (N7-Acyclovir)]+ , Mol. Pharmacol. 53 (1998) 846–855.
[5] R.L. LaFemina, Requirement of active human immunodeficiency
virus type 1 integrase enzyme for productive infection of human
T-lymphoid cells, J. Virol. 66 (1992) 7414–7419.
[6] P.S. Moore, C.J. Jones, N. Mahmood, I.G. Evans, M. Goff, R. Cooper,
A.J. Hay, Anti-(human immunodeficiency virus)activity of polyoxo-
tungstates and their inhibition of human immunodeficiency virus
reverse transcriptase, Biochem. J. 307 (1995) 129–134.
[7] O. Rixe, W. Ortuzar, M. Alvarez, R. Parker, E. Reed, K. Paull, T. Fojo,
Oxaliplatin, tetraplatin, cisplatin and carboplatin: spectrum of activ-
ity in drug-resistant cell lines and in the cell lines of the National
Cancer Institute’s Anticancer drug screen panel, Biochem. Pharmacol.
52 (1996) 1855–1865.
[8] R. Bakhtiar, E.-I. Ochiai, Pharmacological applications of inorganic
complexes, Gen. Pharmacol. 32 (1999) 525–540.
[9] J.H. Murray, M.M. Harding, Organometallic anticancer agents: the
effect of the central metal and halide ligands on the interaction of
metallocene dihalides Cp2Mx2 with nucleic acids constituents, J.
Med. Chem. 37 (1994) 1936–1941.
[10] J.S. Thayer, Organometallic Compounds and Living Organisms (1984),
Monographs (Organometallic Chem. Ser.), Elsevier Science and Tech-
nology Books, San-Diego, CA, USA, 1984.
[11] V.A. Shlykhovenko, V.A. Milinevskaya, I.L. Zagorujko, Y.V. Yanish, V.V.
Kozak, V.N. Kokozey, A. Schernova, Exp. Oncol. 19 (1997) 348–358.
[12] R. Gopalakrishna, Z.H. Chen, U. Gundimeda, Seleno compounds
induce a redox modulation of protein kinase C in the cell, com-
partmentally independent from cytosolic glutathione: its role in
inhibition of tumor promotion, Arch. Biochem. Biophys. 348 (1997)
37–48.
[13] R. Gopalakrishna, U. Gundimeda, Z.H. Chen, Cancer preventive seleno
compounds induce a specific redox modification of cysteine-rich
regions in Ca(2+)-dependant isoenzymes of protein kinase-c, Arch.
Biochem. Biophys. 348 (1997) 25–36.
[14] E. Monti, Interaction of Cu-II anthracyclin complexes with protein
kinase-C. Spectromagnetic assessment of inhibitory effect, Inorg.
Chim. Acta 205 (1993) 181–184.
[15] A. Marin-Hernandez, I. Gracia-Mora, L. Ruiz-Ramirez, R. Moreno-
Sanchez, Toxic effects of copper based antineoplastic drugs
(casiopeinas) on mitochondrial function, Biochem. Pharmacol. 65
(2003) 1979–1989.
[16] R. Martinez, L. Chacon-Garcia, The search of DNA-intercalators as
antitumoral drugs: what it worked and what did not work, Curr. Med.
Chem. 12 (2005) 127–151.
[17] X. Shui, M.E. Peek, L.A. Lipscomb, A.P. Wilkinson, L.D. Williams, M.
Gao, C. Ogata, B.P. Roques, C. Garbay-Jaureguiberry, Effects of cationic
charge on three-dimensional structures of intercalative complexes:
structure of a bis-intercalated DNA complex solved by mad phasing,
Curr. Med. Chem. 7 (2000) 59–71.
[18] T. Theophanides, J. Anastassopoulou, Copper and carcinogenesis, Crit.
Rev. Oncol. Hematol. 42 (2002) 57–64.
[19] T.F. Kagawa, B.H. Geierstanger, A.H. Wang, P.S. Ho, Covalent modifi-
cation of guanine bases in double-stranded DNA. The 1.2- a z-DNA
structure of d(cgcgcg) in the presence of CuCl2 , J. Biol. Chem. 266
(1991) 20175–20184.
[20] R. Mital, G.M. Shah, T.S. Srivastava, R.K. Bhattacharya, The effect of
some new platinum(II) and palladium(II) coordination complexes on
rat hepatic nuclear transcription in vitro, Life Sci. 50 (1992) 781–790.
[21] K. Maskos, The interaction of metal ions with nucleic acids. NMR
study of the copper(II) interaction with inosine derivatives, Acta
Biochim. Pol. 28 (1981) 317–335.
95
[22] S.Y. Paik, K.H. Koh, S.M. Beak, S.H. Paek, J.A. Kim, The essential oils
from Zabthoxylum Schinifoliukm pericap induce apoptosis of HepG2
human hepatoma cells through increased production of reactive oxy-
gen species, Biol. Pharm. Bull. 28 (5) (2005) 802–807.
[23] P. Gellerfors, B.D. Nelson, A rapid method for the isolation of intact
mitochondria from isolated rat liver cells, Anal. Biochem. 93 (1979)
200–203.
[24] J. Usta, S. Kreydiyyeh, K. Bajakian, H. Nakkash-Chmaisse, In vitro
effect of eugenol and cinnamaldehyde on membrane potential and
respiratory chain complexes in isolated rat liver mitochondria, Food
Chem. Toxicol. 40 (7) (2002) 935–940.
[25] I. Somasundaram, M. Palaniandavar, Models for enzyme–copper–
nucleic acid interaction: interaction of some biomimetic copper com-
plexes derived from salicylaldehyde, glycine and ␣-alanine with
adenine and adenosine, Indian J. Chem. 32A (1993) 495–499.
[26] R. Singh, S. Tyagi, S. Singh, S.M. Singh, U.P. Singh, Solid, solution and
antitumor activity studies of mixed ligand complexes with adenine-
5-bromouracilbase pair, Synth. React. Inorg. Met. Org. Chem. 32 (5)
(2002) 853–859.
[27] M.S. Masoud, A.A. Soayed, A.E. Ali, Complexing properties of nucleic-
acid constituents adenine and guanine complexes, Spectrochim. Acta
60 (2004) 1907–1915.
[28] R. Savoie, J.J. Jutier, L. Prizant, A.L. Beauchamp, Raman and infrared
spectra of methylmercury complexes of adenine, Spectrochim. Acta
A: Mol. Spectrosc. 38 (1982) 561–566.
[29] A.N. Speca, C.M. Mikulski, F.J. Iaconianni, L.L. Pytlewski, N.M.
Karayanni, Characterization studies of adenine complexes with 3d
metal perchlorates, J. Inorg. Nucl. Chem. 43 (11) (1981) 2771.
[30] A.N. Speca, L.L. Pytlewski, C.M. Mikulski, N.M. Karayanni, Zinc(II)
complexes of adenine, Inorg. Chim. Acta 66 (1982) 153–158.
[31] M.A. Salam, K. Aoki, Metal ion interactions with nucleobases in the
tripodal Tris(2-aminoethyl)amine (tren) ligand system, Inorg. Chim.
Acta 314 (2001) 71–78.
[32] A. Marzotto, A. Ciccarese, D.A. Clemente, G. Valle, Co-ordination
chemistry of adenine (hade): synthesis and characterization of
[Cu(tren)(nucleobase)]X2 , J. Chem. Soc., Dalton Trans. (1995)
1461–1467.
[33] D.B. Brown, J.W. hall, H.M. Helis, E.G. Walton, D.J. Hodgson, W.E. Hal-
field, The preparation and magnetic and structural examination of
monomeric, dimeric and polymeric adenine complexes of copper(II),
Inorg. Chem. 16 (11) (1997) 2675–2682.
[34] M.T. Zaworotko, H.H. Hammud, G. Mc-Manus, A.T. Kabbani, M.S.
Masoud, A.M. Ghannoum, deposited at Cambridge Crystallographic
Data Center # CCDC-281077, 2005.
[35] T.A. George, H.H. Hammud, S. Isber, Structure, thermal analysis
and magnetism of the doubly bridged linear polymer [Ni(4-NH2 -
C6 H4 CO2 )2 (CH3 OH)2 ]·CH3 OH, Polyhedron 25 (2006) 2721–2728.
[36] R. Tauler, M.J.A. Rainer, B.M. Rode, The interaction of H+ , Zn(II) and
Cu(II) with adenine and 9-methyladenine, Inorg. Chim. Acta 123
(1986) 75–82.
[37] S. Satyanarayana, K. Venugopal Reddy, Potentiometric determina-
tion of the stability constants of ternary complexes of biologically
important metal ions, Indian J. Chem. 28A (1989) 630–632.
[38] H. Demir, M. Pekin, A.K. CuCu, E. Dolen, H.Y. Aboul-Enein, Potentio-
metric studies of mixed complexes of cobalt(II) and copper(II) with
l-asparagine and adenine, Toxicol. Environ. Chem. 71 (1999) 357–367.
[39] A. Terzis, A.L. Beauchamp, R. Rivest, Crystal and molecular struc-
ture of tetra-u-adenine-diaqua-dicopper(II) perchlorate dihydrate,
[Cu2 (C5 H5 N5 )4 (H2 O)2 ](ClO4 )4 ·2H2 O, Inorg. Chem. 12 (5) (1973)
1166–1170.
[40] P. De Meester, A.C. Skapski, Crystal structure of dichlorotetra-p-
adenine-dicopper(II) chloride hexahydrate, J. Chem. Soc. (1971)
2167–2169.
[41] M. Pitie, C.J. Burrows, B. Meunier, Mechanisms of DNA cleavage by
copper complexes of 3-clip-phen and of its conjugate with a dis-
tamycin analogue, Nucleic Acids Res. 28 (2000) 4856–4864.
[42] P.A. Ochoa, M.I. Rodriguez-Tapiador, S.S. Alexandre, C. Pastor,
F. Zamora, Structural models for the interaction of Cd(II) with
DNA: trans-[Cd(9-rgh-N7)2 (H2 O)4 ]2+ , J. Inorg. Biochem. 99 (2005)
1540–1547.
[43] S. Verma, S.G. Srivatsan, C.A. Claussen, E.C. Long, DNA strand scission
by a Cu-I adenylated polymeric template: preliminary mechanis-
tic and recycling studies, Bioorg. Med. Chem. Lett. 13 (2003) 2501–
2504.
[44] M.C. De-Oliveira, M. Scarpellini, A. Neves, H. Terenzi, A.J. Bortoluzzi,
B. Szpoganics, A. Greatti, A.S. Mangrich, E.M. de Souza, P.M. Fernan-
dez, M.R. Soares, Hydrolytic protein cleavage mediated by unusual
mononuclear copper(II) complexes: X-ray structures and solution
studies, Inorg. Chem. 44 (2005) 921–929.
96 H.H. Hammud et al. / Chemico-Biological Interactions 173 (2008) 84–96
[45] H. Rodriguez, R. Drouin, G.P. Holmquist, T.R. O’Connor, S. Boiteux,
J. Laval, J.H. Doroshow, S.A. Akman, Mapping of copper/hydrogen
peroxide-induced DNA damage at nucleotide resolution in human
genomic DNA by ligation-mediated polymerase chain reaction, J. Biol.
Chem. 270 (1995) 17633–17640.
[46] S.G. Srivatsan, M. Parvez, S. Verma, Modeling of prebiotic catalysis
with adenylated polymeric templates: crystal structure studies and
kinetic characterization of template assisted phosphate ester hydrol-
ysis, Chem. Eur. J. 8 (22) (2002) 5184–5191.
[47] S.G. Srivatsan, M. Parvez, S. Verma, Adenine–copper coordi-
nation polymer as an oxidative nucleozyme: implications for
simple prebiotic catalytic units, J. Inorg. Biochem. 97 (2003) 340–
344.
[48] F. Bruston, J. Verbne, L. Grajcar, B. Drahi, R. Calvayrac, M.-H. Baron,
M.-C. Maurel, Copper–adenine catalyst for O2 production from H2 O2 ,
Biochem. Biophys. Res. Commun. 263 (1999) 672–677.
[49] A. Mazumder, D.M. Perrin, K.J. Watson, D.S. Sigman, A transcription
inhibitor specific for wound DNA in RNA polymerase promoter open
complexes, Proc. Natl. Acad. Sci. 90 (1993) 8140–8144.
[50] S.G. Srivatsan, Modeling prebiotic catalysis with nucleic acid-like
polymers and its implications for the proposed RNA world, Pure Appl.
Chem. 76 (2004) 2085–2099.
[51] J.R. Sorenson, Cu, Fe, Mn, and Zn chelates offer a medicinal chemistry
approach to overcoming radiation injury, Curr. Med. Chem. 9 (2002)
639–662.
[52] M.E. Prater, D.J. Mindiola, X. Ouyang, K.R. Dunbar, A quadruply-
bonded dirhenium complex bridged by 2 N1 /N6 adenate ligands,
Inorg. Chem. Commun. 1 (1998) 475–477.
[53] J.A. Gordon, V.H. Gattone, Mitochondrial alterations in cisplatin-
induced acute renal failure, Am. J. Physiol. 250 (1986) F991–F998.
[54] J.S. Penta, F.M. Johnson, J.T. Wachsman, W.C. Copeland, Mitochondrial
DNA in human malignancy, Mutat. Res. 488 (2001) 119–133.