The replacement of chlorides by cytosines in copper(II) complexes containing guanidine derivatives

Article abstract of DOI:10.1016/j.ica.2010.11.030

Two copper(II) chloride complexes of amidino-O-methylurea (L¹), [Cu(L¹)Cl₂] (1), and (N-benzyl)-amidino-O-methylurea (L²), [Cu(L²)Cl₂] (2), were prepared and characterized by elemental anal

Full text of DOI:10.1016/j.ica.2010.11.030

Inorganica Chimica Acta 366 (2011) 357–365
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
The replacement of chlorides by cytosines in copper(II) complexes
containing guanidine derivatives
⇑
Atittaya Meenongwa, Unchulee Chaveerach , Khatcharin Siriwong
Department of Chemistry and Center of Excellence for Innovation in Chemistry, Faculty of Science, Khon Kaen University, Khon Kaen 40002, Thailand
a r t i c l e i n f o
a b s t r a c t
Two copper(II) chloride complexes of amidino-O-methylurea (L1), [Cu(L1)Cl2] (1), and (N-benzyl)-ami-
dino-O-methylurea (L2), [Cu(L2)Cl2] (2), were prepared and characterized by elemental analysis, infrared,
diffuse reflectance, electron spin resonance and electrospray ionization mass spectra. Their cytosine bind-
ing abilities has been studied and found that two cytosine molecules are able to coordinate with the cop-
per centers by replacing the chloride ligands to yield the bifunctional binding adducts [Cu(L1)(cyt)2]Cl2
(1c) and [Cu(L2)(cyt)2]Cl2 (2c), respectively. The shift of the C@O band of cytosine in both cytosine-bound
products to higher energy suggested that the N(3)–cytosine atom coordinates to the copper center. The
large blue shifts of the d–d absorbance maxima and the nine superhyperfine splitting from the CuN4
chromophore were also observed in their electronic and EPR spectra. Their thermal decompositions have
also supported the interaction of cytosine with complexes 1 and 2. Density functional calculations have
also been performed and revealed that square planar coordination geometry is more stable for both 1c
and 2c. The binding energy of 1c is found to be ꢀ20% lower than that of 2c, indicative of the higher bind-
ing potential of 1c.
Article history:
Received 15 July 2010
Received in revised form 10 September 2010
Accepted 18 November 2010
Available online 1 December 2010
Keywords:
Copper(II)
Cytosine binding
Amidino-O-alkylurea
Spectroscopic analysis
Theoretical calculation
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
important role in all living systems as an essential trace element
[9], has currently received intense interest prior to its structural
The success of an anticancer drug cisplatin, [PtCl2(NH3)2], has
led to the development in a large number of metal-based chemo-
therapy drugs. Cisplatin is a small molecule containing a square
planar PtN2Cl2 unit. Its biological activity is believed that one of
the chloride ligands is slowly displaced by an aqua ligand yielding
[PtCl(H2O)(NH3)2]+ which is subsequently reacted with a basic site
in DNA. Finally, cross-linking of the two DNA bases, particularly
guanines, takes place by means of displacement of the remaining
chloride ligand [1,2], thereby interfering with mitosis and then
undergoing apoptosis. Hence, the lability of two chlorides is of
importance in the anticancer activity. Although cisplatin offers a
good result in treatment of cancer patients, several side-effects
such as nephrotoxicity, neurotoxicity, nausea and vomiting are
found [3]. As a result, there is a need for searching new agents
which do not exhibit cross-resistance and which are less toxic.
Cisplatin analogs containing the PtN2Cl2 chromophore have
been synthesized by several research groups [4–8]. However, coor-
dination compounds of other transition metals, such as palla-
dium(II), gold(III), ruthenium(II), copper(II) and etc., have also
been studied in terms of the interactions with nucleic acids and
their cytotoxicity. Particularly, copper(II) ion, which plays an
diversity, redox property and DNA cleaving ability. For example,
a series of copper(II) compounds containing N,N-bidentate benz-
imidazole derivatives with the CuN2Cl2 chromophore have been
prepared and showed their biological activities toward cancer cell
lines [10]. Moreover, three CuN2Cl2 complexes with guanidinium/
ammonium groups were synthesized and investigated their cleav-
age efficiency [11]. These activities have guided us to prepare cop-
per(II) complexes containing a CuN2Cl2 chromophore analog to
cisplatin and then to study their interactions with DNA nucleo-
bases. Several studies have mainly focused on the direct interac-
tion of metal salts, where M = Cu, Rh, Cd, Mn or Co, with the
cytosine nucleobase and its derivative 1-methylcytosine [12–19],
but not the direct nucleobase replacement of the anionic ligands
on metal complexes like the cisplatin system. However, Reedijk
and coworkers have studied on interactions of the octahedral
ruthenium(II) complexes [cis-Ru(bpy)2Cl2] [20] and [a-Ru(azpy)2-
(NO3)2] [21] (bpy = 2,20-bipyridine and azpy = 2-(phenylazo)pyri-
dine) with 9-ethylguanine, 9-methylhypoxanthine and guanosine,
which replace only one of the two anionic ligands to yield the
monofunctional binding adducts.
Towards achieving our goal of finding a new choice of metal
complexes based upon a cheaper alternative copper(II) ion show-
ing an efficient binding potential to nucleobase, we firstly studied
on the interaction of the two copper(II) chloride complexes,
⇑
Corresponding author. Tel.: +66 43 202222x12243; fax: +66 43 202373.
E-mail address: sunchul@kku.ac.th (U. Chaveerach).
0020-1693/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2010.11.030

358
A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
Scheme 1. Ligands in the present work. L1 = amidino-O-methylurea and L2 = (N-methylphenyl)-amindino-O-methylurea.
2. Experimental
2.1. Materials and methods
Copper(II) chloride dihydrate, cyanoguanidine and cytosine
were purchased from Sigma–Aldrich Chemical Company Ltd. and
used without further purification. The elemental analysis was
determined using a Perkin–Elmer PE-2400II CHNS/O elemental
analyzer. The infrared spectra were obtained in the range 4000–
400 cmꢁ1 as KBr pressed pellets on a Perkin–Elmer spectrum one
FT-IR spectrophotometer. Diffuse reflectance spectra were ob-
tained on a Shimadzu 3101 UV–Vis–NIR scanning spectrophotom-
eter. Positive electrospray ionization (ESI+) mass spectral
measurements were undertaken using a Q-TOF mass spectrometer.
Thermogravimetric curves were obtained on a Perkin–Elmer Pyris
Scheme 2. Cytosine molecule with atomic numberings.
Diamond TG/DTA thermal analyzer with
a heating rate of
[Cu(L1)Cl2] (1) and [Cu(L2)Cl2] (2), where L1 = amidino-O-methylu-
rea and L2 = (N-benzyl)-amidino-O-methylurea (Scheme 1), toward
cytosine (cyt) molecules (Scheme 2). In 2003, the copper(II) chlo-
ride complex consisting of two L2 ligands, [Cu(L2)2]Cl2, has been
structurally studied and reported [22] that the copper(II) center
are coordinated by two symmetry-related N,N-bidentate L2 ligands
in a trans configuration to give a square planar geometry. In the
present work, compounds 1 and 2 contain only a single molecule
of L1 and L2 ligands, respectively, which only differ in the N-substi-
tution sidearm groups, NH2 for 1 and NH2CH2Ph for 2 (Scheme 1).
This will give us an opportunity to examine the pendent-arm effect
on their cytosine-binding abilities. Besides, other advantages of
this complex system are that (i) complexes 1 and 2 are similar to
cisplatin in that they contain the same MN2Cl2 (M = Cu or Pt) chro-
mophore; (ii) the L1 and L2 ligands are an N,N-bidentate ligand
which will chelate to a copper center and give a more stable com-
plex; (iii) this ligand system also results a compound in cis-isomer,
just like cisplatin; and (iv) hence, one site of complexes 1 and 2 is
blocked by the L1 and L2 ligands and another site is situated by two
labile groups (Clꢁ) which will be easily replaced by nucleobases
such as cytosine. It was found that the predominant metal binding
site of cytosine nucleobase is the nitrogen atom N(3) (Scheme 2)
[11,12,14,15,17,23,24]. In addition, the secondary interaction be-
tween copper and the O(2) exocyclic carbonyl group of cytosine
also exists in a few cases [16,25,26].
5 °C minꢁ1 from 30 to 1000 °C under 100 mL minꢁ1 N2 flow. Elec-
tron spin resonance spectra in the frozen DMSO solution at 77 K
were recorded by a RE-2X electron spin resonance spectrometer
operating at
m = 9.16 GHz (X-band).
2.2. Copper(II) complex preparations
Copper(II) complexes [Cu(L1)Cl2] (1) and [Cu(L2)Cl2] (2) were
prepared by the slightly modified procedure as previously reported
[27]. A 1:1 molar ratio of CuCl2ꢂ2H2O (0.3410 g; 2 mmol) and
cyanoguanidine for 1 (0.1682 g; 2 mmol) or (N-benzyl)-cyanogua-
nidine for 2 (0.3480 g; 2 mmol) was used instead. The preparation
reactions were shown below.
NH2
[Cu(L1)Cl2], 1
NCN
C
NH2
MeOH
reflux
or
or
CuCl2 2H2O +
NHCH2(C6H5)
[Cu(L2)Cl2], 2
NCN
C
Herein, our attempts were to prepare two copper(II) complexes
of [Cu(L1)Cl2] (1) and [Cu(L2)Cl2] (2) and investigate their interac-
tions toward cytosine nucleobases. Spectroscopic methods (infra-
red, electrospray ionization mass, diffuse reflectance and electron
spin resonance), elemental analysis and thermogravimetric analy-
sis were utilized to give some evidences for the cytosine replace-
ment by the two chlorides in both compounds. Additionally,
density functional calculations for the cytosine-bound complexes
have been carried out to obtain the binding energies and predict
theoretically their coordination geometries.
NH2
2.2.1. [Cu(L1)Cl2], 1
Yield: 0.4920 g; 1.96 mmol; 98.2%. M.p. (°C): 228.4–229.2. Anal.
Calc. for C3H8N4OCuCl2: C, 14.37; H, 3.19; N, 22.36. Found: C,
14.02; H, 3.14; N, 22.63%. IR (KBr/cmꢁ1): 3373s, br; 3343s, br;
3259s, br; 3180s, br; 1692s; 1656s; 1547s; 1508m; 1430m;
1343m; 1197m; 1165m; 1117w; 948w; 760m; 732m; 688w;

A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
359
647w; 568w; 518m; 453w. Diffuse reflectance (kmax/cmꢁ1):
15 049. ESI+ mass spec (m/z): 179 [Cu(L1)ꢁH+]+; 117 [(L1)H]+.
all calculations to describe the valence orbitals. The density func-
tional semi-core pseudo-potentials (DSPP) [29] were also utilized
to treat core electrons. The exchange–correlation contribution to
the total electronic energy was treated in a generalized-gradient
approximation (GGA-PW91) function. It was previously reported
that these functions have been successfully used to predict geom-
etry and energy for the anticancer drug oxaliplatin [30,31]. The
2.2.2. [Cu(L2)Cl2], 2
Yield: 0.4147 g; 1.22 mmol; 60.9%. M.p. (°C): 209.9–210.3. Anal.
Calc. for C10H14N4OCuCl2: C, 35.24; H, 4.11; N, 16.45. Found: C,
35.40; H, 4.25; N, 16.73%. IR (KBr/cmꢁ1): 3375m, br; 3301s, br;
3225m, br; 3144m, br; 2961w; 1677s; 1556s; 1525m; 1442m;
1385m; 1356m; 1326m; 1226m; 1211m; 1169m; 1144m;
1089w; 1055w; 930w; 897w; 774w; 734m; 712m; 696m; 610w;
535w. Diffuse reflectance (kmax/cmꢁ1): 16 176. ESI+ mass spec
(m/z): 268 [Cu(L2)ꢁH+]+; 207 [(L2)H]+.
binding energy (
DEbinding) including relaxation can be expressed as
D
Ebinding ¼ ECuðLÞðcytÞ ꢁ ðECuðLÞ þ 2Ecyt
Þ
2
where ECuðLÞðcytÞ , ECuðLÞ and Ecyt are the zero-point corrected energies
of the optimiz2ed structure for [Cu(L)(cyt)2]2+, [Cu(L)]2+, and cyto-
sine, respectively. L is either the ligand L1 or L2.
2.3. Cytosine binding to copper(II) complexes
3. Results and discussion
Cytosine (0.0888 g; 0.8 mmol) was added into 25 mL of the
warm ethanol–water (1:2 v/v) solution of complex 1 (0.1002 g;
0.4 mmol) or complex 2 (0.1362 g; 0.4 mmol). The dark blue solu-
tion was stirred under reflux condition for 24 h. A pink or purple
solid, respectively, was obtained by solvent reduction under vac-
uum. The binding reactions were shown below.
3.1. General aspects
The pale blue [Cu(L1)Cl2] (1) and deep blue [Cu(L2)Cl2] (2) com-
plexes were generated in a good yield by methanolysis reaction of
cyanoguanidine and (N-benzyl)-cyanoguanidine, respectively, in
the presence of copper(II) chloride in a 1:1 molar ratio. Their ele-
mental analyses were in agreement with the desired compounds.
[Cu(L1)(cyt)2]Cl2, 1c
or
[Cu(L1)Cl2], 1
or
EtOH/H2O
reflux
+
2(C4H5N3O)
(cyt)
[Cu(L2)(cyt)2]Cl2, 2c
[Cu(L2)Cl2], 2
2.3.1. [Cu(L1)(cyt)2)]Cl2, 1c
Yield: 0.1705 g; 0.36 mmol; 90.2%. M.p. (°C): 181.2–184.2. Anal.
Calc. for C11H18N10O3CuCl2: C, 27.94; H, 3.81; N, 29.63. Found: C,
27.98; H, 3.82; N, 29.63%. IR (KBr/cmꢁ1): 3370s, br; 3305s, br;
3193s, br; 2996s; 1746m; 16721s; 1614s; 1497m; 1447m;
1370m; 1298w; 1236m; 1218m; 1146w; 1126w; 1012w; 947w;
843w; 807m; 784m; 760m; 599m; 579w; 547w; 431w. Diffuse
reflectance (kmax/cmꢁ1): 19 327. ESI+ mass spec (m/z): 400
[Cu(L1)(cyt)2ꢁH+]+; 289 [Cu(L1)(cyt)ꢁH+]+; 118 [(L1)+2H+]; 112
[(cyt)+H+]. TGA Found (calc): [Cu(cyt)]Cl+ (50–450 °C) 45.2 (44.4);
CuCl+ (450–1100 °C) 22.4 (20.9).
2.3.2. [Cu(L2)(cyt)2)]Cl2, 2c
Yield: 0.1922 g; 0.34 mmol; 85.4%. M.p. (°C): 186.8–187.4. Anal.
Calc. for C18H24N10O3CuCl2: C, 38.40; H, 4.27; N, 24.89. Found: C,
38.49; H, 4.25; N, 24.88%. IR (KBr/cmꢁ1): 3294s, br; 3220s;
3120s, br; 2974s; 1738m; 1667s; 1591m; 1496m; 1457m;
1365w; 1297w; 1219m; 1156w; 1030w; 952w; 904w; 800m;
780m; 760m; 731m; 700m; 600m; 578m; 544w; 475w. Diffuse
reflectance (kmax/cmꢁ1): 19 153. ESI+ mass spec (m/z): 490
[Cu(L2)(cyt)2ꢁH+]+; 379 [Cu(L2)(cyt)ꢁH+]+; 268 [Cu(L2)ꢁH+]+; 207
[(L2)H]+. TGA Found (calc): [Cu(L2)(cyt)2]2+ (60–230 °C) 87.9
(87.4); [Cu(cyt)]2+ (230–425 °C) 31.7 (31.0).
2.4. Computational details
In order to investigate the binding competence between cyto-
sine bases and the two copper(II) complexes (1 and 2) and to pre-
dict the geometry of cytosine-bound complexes (1c and 2c), the
first principle density functional (DF) calculations were used. The
structures of cytosine, [Cu(L1)]2+, [Cu(L2)]2+, [Cu(L1)(cyt)2]2+ and
[Cu(L2)(cyt)2]2+ were optimized without restriction using Dmol3
module of Material Studio 4.3 package [28] at the National Nano-
technology Center, Thailand. The numerical basis sets of double
zeta quality plus polarization functions (DNP) were employed in
Fig. 1. Overlayered FT-IR spectra of the free complexes 1 and 2 (top), the bound
complexes 1c and 2c (middle) and cytosine (bottom).

360
A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
The molecular ion peak of the cationic species, [Cu(L1)ꢁH+]+ for 1
and [Cu(L2)ꢁH+]+ for 2, was appeared at m/z = 179 and 268, respec-
tively, in their ESI+ mass spectra. Additionally, the protonated li-
gands were observed at m/z = 117 for [(L1)H]+ and 207 for
[(L2)H]+. Melting points were observed in the narrow temperature
range of 228.4–229.2 for 1 and 209.9–210.3 °C for 2.
Subsequently, the binding experiments were carried out in a
1:2 molar ratio of the obtained complexes and cytosine nucleo-
bases. Colors of the resulting products were changed to pink (1c)
and purple (2c). Their melting properties were significantly chan-
ged to 181.2–184.2 °C for 1c and 186.8–187.4 °C for 2c. These mod-
ifications preliminarily point out that environment on copper(II)
centers of the bound compounds totally differ from that of the free
compounds. Thus it is suggested that the interaction between com-
pounds 1 or 2 and cytosine bases has taken place. To gain further
evidences on the cytosine binding ability of 1 and 2, various spec-
troscopic methods (infrared, mass, diffuse reflectance, electron
spin resonance), thermal analysis and density functional calcula-
tions were undertaken and deeply described in the following
topics.
2), cytosine and the bound complexes (1c and 2c) are illustrated in
Fig. 1. Band assignments are listed in Table 1.
Infrared spectra of 1c and 2c revealed the presence of all func-
tional groups from both cytosine and the corresponding free com-
plexes. The strong and broad infrared absorption bands in the
range 3370–3120 cmꢁ1 are assigned to asymmetric and symmetric
NH2 stretching vibrations of both cytosine and the free complexes.
The broadening CH stretching bands of complexes 1c and 2c were
also observed in the range of 2996–2974 cmꢁ1. The N–H bending
vibrations also appeared in the range of 1672–1447 cmꢁ1. The
bands assigned to C@C and C@N stretching vibrations were shifted
from 1614–1539 cmꢁ1 for the free cytosine to higher frequency at
1672–1591 cmꢁ1 for the bound complexes since the conjugated
p-
electrons in the cytosine ring were redistributed. This evidence
preliminarily confirms the coordination of the N(3) cytosine to
the copper(II) center of 1 and 2 [32]. In addition, the bands related
to
found in their infrared spectra in the lower frequency than
1500 cmꢁ1
m-ring and m(C–N) single bonded stretching frequencies were
.
It is noticed in the region of 1750–1730 cmꢁ1 that the additional
new peak at 1746 and 1738 cmꢁ1 corresponding to the C@O
stretching vibration of cytosine was observed in compounds 1c
and 2c, respectively. The C@O band of the free cytosine at
ꢀ1665 cmꢁ1 was shifted to higher frequency in both compounds,
thus eliminating the exocyclic O(2) coordination of cytosine. The
shift of infrared bands in the complexes is therefore a more and
strong evidence to suggest that cytosine molecules coordinate to
the copper center of 1 and 2.
3.2. Analysis of infrared spectra
Vibrational spectrophotometry is a basis method to determine
the functional groups of compounds. Herein, it can be used to mon-
itor the interaction between the starting copper(II) complexes and
cytosine nucleobases. Infrared spectra of the free complexes (1 and
Table 1
Assignment of infrared vibrational modes for cytosine, the free complexes (1 and 2) and the bound complexes (1c and 2c)
a
Cytosine
3382s, br
Wave number (cmꢁ1
)
Assignment
Free complex
Bound complex
1
2
1c
2c
3373s, br
3343s, br
3259s, br
3180s, br
2984w
3375m, br
3301s, br
3225m, br
3144m, br
2961w
3370s, br
3305s, br
3294s, br
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
m
c
m
m
m
as(NH)
as(NH)
s(NH)
s(NH)
(CH)
(C@O)
(C@C),
(C@C),
(ring),
(ring),
(ring),
(ring), d(CH)
(ring), d(CH)
(ring), d(CH)
(ring), d(CH)
(ring), d(CH)
(ring),
(ring),
(ring),
(ring),
(ring),
(ring),
(ring),
(ring),
(ring),
(NH2)
3220s
3120s, br
2974s
1738m
1667s
1591m
3172s, br
2927s
1665s
3193s, br
2996s
1746m
1672s
1692s
1656s
1547s
1508m
1430m
1343m
1677s
m
m
(C@N), d(NH2)
(C@N), d(NH2)
1614s, sh
1539s
1504m
1466m
1364m
1614s
1556s
m
m
m
(C@N), d(NH2)
(C–NH2), d(NH2)
(C–N), d(NH2), d(CH)
1525m
1442m
1385m
1356m
1326m
1497m
1447m
1370m
1496m
1457m
1365w
1277m
1237m
1298w
1236m
1297w
1219m
1226m
1211m
1169m
1144m
1089w
1055w
1197m
1165m
1218m
q(NH2),
q(NH2),
q(NH2),
q(NH2),
q(NH2)
q(NH2)
q(NH2)
q(NH2)
q(NH2)
m(C–O)
m(C–O)
m(C–O)
m(C–O)
1155w
1146w
1126w
1012w
1156w
1030w
1117w
1010w
994w
965w
948m
930w
897w
947w
843w
807m
784m
760m
952w
904w
800m
780m
760m
731m
700m
816m
792m
774w
734m
712m
696m
(ring), d(ring)
(ring), d(ring)
(ring), d(ring)
760m
732m
688w
647w
700w
NH2 wagging
m
m
(ring), d(ring)
(ring), d(ring)
601w
549m
512m
610w
535w
599m
579m
547w
431w
600m
578m
544w
475w
568w
518m
453w
NH2 wagging
NH2 wagging
NH2 wagging
a
Abbreviation: s, strong; m, medium; w, weak; br, broad; sh, shoulder;
bending.
m
as, asymmetric stretching; ms, symmetric stretching; d, bending; q, libration, c, out-of-plane

A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
361
3.3. Mass spectra
at 118 and [(L2)H]+ at 207 were also found. In the case of 1c, the
protonated cytosine species, [(cyt)+H+], was also found at
m/z = 112. The obtaining results are therefore supported that two
cytosine-bound complexes have been achieved. The possible frag-
mentation pathways for both complexes are schematically sum-
marized as followed:
Mass spectroscopy is so useful that it can provide information
about the relative molecular mass of a compound. Furthermore,
it is frequently possible to determine which ligands or groups of
atoms are bonded together from the way the compound breaks
up or fragments during the experiment. Mass spectra of the bound
complexes 1c and 2c appeared the molecular ion peak of the cat-
ionic species at m/z = 400 for [Cu(L1)(cyt)2ꢁH+]+ and 490 for
[Cu(L2)(cyt)2ꢁH+]+, respectively. Furthermore, the fragment ions
of a single cytosine-containing species, [Cu(L1)(cyt)ꢁH+]+ and
[Cu(L2)(cyt)ꢁH+]+, were observed at m/z = 289 and 379, respec-
tively. For 2c, the m/z ratio at 268 has been assigned to the
[Cu(L2)ꢁH+]+ fragment. Finally, the protonated ligands [(L1)+2H+]
− 2Cl−
− H+
− cyt
[Cu(L1/L2)(cyt)2]Cl2
[Cu(L1/L2)(cyt)2 – H+]+
[Cu(L1/L2)(cyt) – H+]+
m/z = 400 for 1c;
490 for 2c
m/z = 289 for 1c;
379 for 2c
− cyt
For 1c: [(L1)+2H+]
m/z = 118
+
[(cyt)+H+]
112
− Cu2+
H+
[Cu(L1/L2) – H+]+
For 2c: [(L2)H]+
m/z = 207
m/z = not found for 1c;
268 for 2c
3.4. Diffuse reflectance spectra
Electronic spectra in solid state samples, obtained by diffuse
reflectance spectrophotometry, are employed to ascertain the
binding of complexes with the nitrogenous base cytosine. The
covalent binding interaction between the copper(II) coordination
compounds and cytosine DNA bases can be observed through the
shifts in the d–d absorbance maxima of the copper(II) complexes
in the visible range. The reflectance spectra of the free and bound
complexes are shown in Fig. 2. Their maximum absorptions and
their differences are collected in Table 2.
The copper(II) complexes 1 and 2 with blue in color showed a
single broad d–d absorption band at kmax = 664.5 nm
(15 049 cmꢁ1) and 618.2 nm (16 176 cmꢁ1), respectively, indica-
tive of approximate square pyramidal geometry [33]. In 1986,
M.J. Begley et al. reported the copper(II) dimer of amidino-O-eth-
ylurea (aOeu), [Cu(aOeu)Cl2]2 [34], which is rather similar to our
systems. The [Cu(aOeu)Cl2]2 dimer appears as blue crystals with
the absorption maxima at 670 nm (14 900 cmꢁ1) and reveals a
square pyramid. According to these reports, it is supported that
compounds 1 and 2 should possibly be a dimer and have been pro-
posed to adopt a square pyramidal coordination with the formula
of [Cu(L1)Cl2]2 and [Cu(L2)Cl2]2, respectively, as shown in Fig. 3.
Upon cytosine binding to the complexes 1 and 2, the colors of
solid products were changed to pink for 1c and purple for 2c, pre-
liminarily suggesting that the bound complexes adopt the differ-
ent coordination geometries and their copper centers have
different environment from the initial complexes. Their reflec-
tance spectra revealed that the d–d absorption peak of the bound
complexes 1c and 2c showed the kmax value of 517.4 nm
(19 327 cmꢁ1) and 522.1 nm (19 153 cmꢁ1), respectively, recom-
mending square-planar CuN4 complexes [35–38]. Their maximum
absorptions were significantly shifted to shorter wavelength
Fig. 2. Comparison of diffuse reflectance spectra of (a) complexes 1 (ꢂ ꢂ ꢂ ꢂ) and 1c (—)
and (b) complexes 2 (ꢂ ꢂ ꢂ ꢂ) and 2c (—).
Table 2
Effect of cytosine on the electronic absorption bands and the EPR spectral parameters in a frozen DMSO solution at 77 K of the free complexes (1 and 2) and the bound complexes
(1c and 2c).
Complex
Color
Electronic spectral data, nm (cmꢁ1
)
EPR spectral data
Gb
A|| (G)
g||/A|| (cm)
AN (G)
a
kmax
D
kmax
g||
g\
1
1c
2
Light blue
Pink
Deep blue
Purple
664.5 (15 049)
517.4 (19 327)
618.2 (16 175)
522.1 (19 153)
ꢁ147.1 (4278)
ꢁ96.1 (2978)
2.28
2.25
2.27
2.26
2.06
2.04
2.06
2.05
4.81
6.57
4.64
5.40
168
181
158
185
136
124
144
122
15
15
2c
a
D
kmax = kbound–kfree
.
b
G = (g|| ꢁ2.0023)/(g\ ꢁ2.0023).

362
A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
Fig. 3. Proposed structures of the free complexes 1 (R = H) and 2 (R = CH2C6H5).
(hypsochromic or blue shift), thus resulting in the wavelength
differences
(
D
k = kbound–kfree
)
of ꢁ147.1 nm (4278 cmꢁ1
) and
ꢁ96.1 nm (2978 cmꢁ1) (Table 2). The large shift is caused by coor-
dination of cytosine to copper(II) center instead of the labile
groups (Clꢁ) in the initial complexes. This evidence helps us to
further confirm the occurrence of the replacement of the two
chloride ligands in 1 and 2 by two N(3)–cytosine molecules.
3.5. Electron spin resonance spectra
To gain further verification of the cytosine-binding ability of
compounds 1 and 2, EPR spectral studies was carried out in the fro-
zen DMSO solution at 77 K. The spin Hamiltonian parameters pro-
vide the information on the stereochemistry of the copper(II)
center and are significantly related to the type of donor atoms.
Hence, this method can be utilized to monitor the interaction be-
tween copper(II) complexes and cytosine nucleobases. The X-band
EPR spectra of the free and bound complexes are shown in Fig. 4.
The EPR parameters are summarized in Table 2.
Fig. 4. X-band EPR spectra of a frozen DMSO solution of the starting complexes (1
and 2) and the cytosine-bound complexes (1c and 2c) at 77 K.
The obtained EPR spectra of all compounds are the typical
anisotropic spectrum of monomeric copper(II) complexes where
the four hyperfine lines in the g|| region arising from the interaction
of the S = 1/2 electron spin with the I = 3/2 copper nucleus are ob-
served. The g tensor values are used to examine the electronic
ground state of the copper center. The two initial and two bound
complexes presented the trend of g|| > g\ > 2.0 and the G value
higher than 4.0 suggesting that the unpaired electron of the cop-
124 cm for 1c and 122 cm for 2c appear to be nearly a typical pla-
nar complex.
The EPR results are strongly confirmed that cytosine nucleo-
bases interact with complexes 1 and 2 through the substitution
reaction of the N(3) cytosine atoms at the two chloride ligands in
agreement with the large shift to higher energy in the electronic
absorption bands in the bound complexes (1c and 2c).
per(II) ion locates in the ground state dx2
orbital of an axial
ꢁy2
3.6. Thermal analysis
square-based geometry [39,40]. Noticeably, the minor signals at
low field were observed in the spectra of the starting complexes
(1 and 2), possibly due to the interaction between the complexes
and solvent (DMSO) molecules.
Thermogravimetric analysis is based on the change in mass of a
sample as the temperature is varied, providing qualitative or quan-
titative information. Thermal decomposition processes of the
bound complexes (1c and 2c) are summarized in Table 3. Both
complexes show the similar patterns with two steps of mass losses.
In the initial step of thermal decomposition of 1c occurred in
the temperature range from 50 to 450 °C which is attributed to
the liberation of one chloride anion, the L1 ligand and one cytosine
molecule. An obvious weight loss of 54.8% (calc. 55.6%) appeared in
this process with an endothermic peak at ꢀ310 °C. An intermediate
with proposed composition [Cu(cyt)]Cl+ was formed. Its thermal
decomposition in the last step between 450 and 1100 °C was the
release of one remaining cytosine molecule with the found mass
loss of 22.8% (calc. 23.5%) along with a sharp endothermic peak
at 777 °C. Two steps of its thermal decomposition are demon-
strated below:
EPR profiles of complexes 1c and 2c are considerably different
from those of complexes 1 and 2 in the g\ region. Nine superhyper-
fine splitting lines (2nIN + 1) with the coupling value of 15 G found
in the EPR spectra of 1c and 2c arise from four nitrogen donor
atoms around the copper center. On the other hand, complexes 1
and 2 showed no nitrogen superhyperfine splitting in their spectra.
This result is an important information to strongly confirm that the
replacement of two chloride leaving groups in 1 and 2 by two
nitrogen atoms from two cytosine molecules was occurred, thus
resulting in changes of the chromophore from CuN2Cl2 (1 and 2)
to CuN4 (1c and 2c). Additionally, a degree of distortion (g||/A||)
can tell us the tetrahedral distortion of a square planar geometry
and then the actual stereochemistry of the copper(II) complexes.
The g||/A|| ratio in the range of 110–120 cm are typical for planar
complexes, while the range of 130–150 cm is a characteristic of
slight to moderate distortion and the range of 180–250 cm indi-
cates considerable distortion [41,42]. The g||/A|| values of 1 and 2
are 136 and 144 cm, respectively, suggesting a slight to moderate
distortion. The much lower values of the bound complexes of
- L1 - cyt - Cl
[Cu(L1)(cyt)2]Cl2
[Cu(cyt)]Cl+
[Cu(cyt)]Cl+
CuCl+
50-450οC
- cyt
450-1100οC

A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
363
Table 3
Characteristic parameters of thermal decomposition for the cytosine bound complexes.
Complex
Step
Lost species
Decomp. range (°C)
DTA peaka (°C)
Weight loss (%)
Found
Calculated
[Cu(L1)(cyt)2]Cl2, 1c
[Cu(L2)(cyt)2]Cl2, 2c
1
2
1
2
Clꢁ + L1 + cyt
cyt
50–450
450–1100
60–230
310
777
148
296
54.8
22.8
12.1
56.2
55.6
23.5
12.6
56.4
2Clꢁ
L2 + cyt
230–425
a
All showed endothermic peaks.
According to the thermal decomposition of both bound com-
plexes, it is strongly recommended that two cytosine molecules
can coordinate to the copper center by replacing the two chloride
ligands of the complexes [Cu(L1/L2)Cl2] as desired.
For 2c, two steps of thermal decomposition appeared in the
temperature range of 60–230 and 230–420 °C for the first and sec-
ond processes, respectively. The first stage revealed the 12.1% loss
of weight corresponding to the elimination of two chloride anions
(calc. 12.6%) and confirming by an endothermic peak at 148 °C. The
remaining species were proposed as [Cu(L2)(cyt)2]2+. Then, the re-
moval of the L2 ligand and one cytosine molecule with the mass
loss found 56.2% (calc. 56.4%) was manifested in the final disinte-
gration stage corresponding to its DTA profile showing an endo-
thermic peak at 296 °C. Two steps of its thermal decomposition
are concluded as followed:
3.7. Computational analysis
Due to the unsuccessful efforts to obtain suitable single crystals
of 1c and 2c for structural analysis, molecular modeling studies
were undertaken to gain a better understanding in the cytosine
binding behavior and to investigate their coordination geometries.
The theoretical geometries for both compounds were calculated
by the DFT method and optimized without restriction by using
Dmol3 module. Geometry optimizations were performed on two
possible 4-coordinate conformers, square planar and tetrahedral
-2Cl
60-230οC
[Cu(L2)(cyt)2]Cl2
[Cu(L2)(cyt)2]2+
[Cu(L2)(cyt)2]2+
[Cu(cyt)]2+
- L2 - cyt
230-425οC
forms, for the cationic species [Cu(L1)(cyt)2]2+ and [Cu(L2)(cyt)2]2+
.
Fig. 5. Optimized geometries of (a) [Cu(L1)(cyt)2]2+ and (b) [Cu(L2)(cyt)2]2+ at the different views.

364
A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
Each conformer was treated by placing the oxygen–O(2) atom of
the two coordinated cytosine molecules in two orientations, at
the same and opposite directions. Data of energy minimization
for both conformers have shown that the strain energies of square
planar structure are less than those of tetrahedral coordination in
all cases. This result agrees well with the diffuse reflectance and
EPR data of 1c and 2c. The optimized geometries are illustrated
in Fig. 5 and the selected geometric parameters are collected in
Table 4.
The optimized structure of [Cu(L1)(cyt)2]2+ providing the lowest
energy is a square planar coordination geometry where the O(2)
atom of two cytosine molecules orientated in the opposite direc-
tion (Fig. 5a). The Cu–N bond lengths for Cu–L1 and Cu–(cyt)2 are
ꢀ1.98 and ꢀ2.05 Å, respectively. In the case of [Cu(L2)(cyt)2]2+, it
was found the distortion in its optimized structure at the N(3)
and N(30) positions to be approximately 3° above and below the
N(1)–Cu–N(2) plane (Fig. 5b). Therefore, the geometry optimiza-
tion of [Cu(L2)(cyt)2]2+ reveals a slightly distorted square planar
with the two O(2)–cytosine atom in the opposite positions. Bond
lengths between copper and nitrogen donors of L2 and cytosine
are in the range of 1.97–2.00 and 2.05–2.07 Å, respectively. Addi-
tionally, it is found that the Cu–O(2) and Cu–O(20) distances are
found to be the same length (2.78 Å) in compound 1c. For com-
pound 2c, they are 2.82 and 2.81 Å, respectively. It is indicated that
the two axial O(2)–cytosine atoms in both compounds are too far
away to form coordinate covalent bonds with the copper center.
In order to investigate the binding interactions between cyto-
sine bases and the present complexes, the total energy with
zero-point vibrational energy (ZPVE) for each optimized structure
Table 4
Selected bond lengths (Å) and angles (°) for the optimized geometry of [Cu(L1)
(cyt)2]2+ and [Cu(L2)(cyt)2]2+
.
[Cu(L1)(cyt)2]2+
[Cu(L2)(cyt)2]2+
Bond lengths
Cu–N(1)
Cu–N(2)
Cu–N(3)
Cu–N(30)
1.981
1.978
2.046
2.052
1.977
1.997
2.061
2.052
Bond angles
N(1)–Cu–N(2)
N(3)–Cu–N(30)
N(1)–Cu–N(3)
N(2)–Cu–N(30)
N(1)–Cu–N(30)
N(2)–Cu–N(3)
88.1
90.9
90.4
90.6
178.6
178.5
88.3
91.1
90.0
90.6
176.0
177.3
was utilized to calculate the binding energy (DEbinding). As the
results, the ZPVE corrected energies of [Cu(L1)(cyt)2]2+, [Cu(L2)
(cyt)2]2+
,
[Cu(L1)]2+
,
[Cu(L2)]2+ and cytosine are ꢁ894402.6,
ꢁ1063985.9, ꢁ398684.8, ꢁ568298.7 and ꢁ247781.9 kcal/mol,
respectively. The calculated
D
Ebinding values are ꢁ154.0 and
ꢁ123.4 kcal/mol for [Cu(L1)(cyt)2]2+ and [Cu(L2)(cyt)2]2+, respec-
tively. These obtained binding energies have pointed that the
two cytosine molecules are able to coordinate with the complex.
The higher binding potential is found for [Cu(L1)(cyt)2]2+. It can
be clearly seen from a plot of the highest occupied molecular
orbitals or HOMOs (Fig. 6) that the HOMOs of the cytosine-bound
complexes are contributed by the orbitals from copper(II), ligand
and cytosine, strongly confirming the occurrence of the cytosine-
binding to the complex.
4. Conclusions
Copper(II) chloride complexes of N,N-bidentate L1 and L2 ligands
were prepared and studied on their cytosine-binding abilities.
Interactions of complexes 1 and 2 with two cytosine bases were
investigated by several basic techniques. Results exhibit a strong
evidence to confirm the capability of both complexes to coordinate
with two N(3)–cytosine molecules by replacing two chloride li-
gands, yielding the bifunctional adducts 1c and 2c. The theoretical
calculations have shown their optimized structures as a square pla-
nar geometry. The ꢀ20% lower binding energy for 1c than 2c indi-
cates the stronger binding potential to cytosine. The reason may
arise from the fact that compounds 1 and 2 contain the different
N-substitution groups on the ligand, NH2 and NHCH2(C6H5), respec-
tively. The phenyl moiety could be possible to sterically interfere
the incoming cytosines to replace the chlorides on the opposite site.
However, this factor would be a little impact on the coordination
geometry. Therefore, the optimized structure of 2c tends to be
slightly distorted from the perfect square planar geometry, while
that of 1c is a square planar. Here we have shown that copper(II)
complexes with bidentate ligands like the L1 and L2 ligands are good
candidates for binding to cytosines and worthwhile further devel-
oping to be a DNA-binding agent.
Acknowledgments
The authors are grateful to the Center of Excellence for
Innovation in Chemistry (PERCH-CIC), Khon Kaen University, the
Fig. 6. Plots of the HOMOs for (a) [Cu(L1)(cyt)2]2+ and (b) [Cu(L2)(cyt)2]2+
.

A. Meenongwa et al. / Inorganica Chimica Acta 366 (2011) 357–365
365
[17] G. De Munno, M. Medaglia, D. Armentano, J. Anastassopoulou, T.
Theophanides, J. Chem. Soc., Dalton Trans. (2000) 1625.
[18] P.T. Selvi, M. Murali, M. Palaniandavar, M. Kockerling, G. Henkel, Inorg. Chim.
Acta 340 (2002) 139.
[19] S. Kalyanaraman, V. Krishnakumar, K. Ganesan, Spectrochim. Acta 66A (2007)
1340.
[20] P.M. van Vliet, J.G. Haasnoot, J. Reedijk, Inorg. Chem. 33 (1994) 1934.
[21] A.C.G. Hotze, A.H. Velders, F. Ugozzoli, M.B. Cingi, A.M.M. Lanfredi, J.G.
Haasnoot, J. Reedijk, Inorg. Chem. 39 (2000) 3838.
Development and Promotion of Science and Technology Talents
Project (to A.M.), and the Office of the Higher Education Commis-
sion under National Research University Project for their financial
support. The computer resource was supported by the National
Nanotechnology Center (NANOTECH).
[22] U. Suksangpanya, A.J. Blake, P. Hubberstey, D.J. Parker, S.J. Teat, C.L. Wilson,
CrystEngComm 5 (2003) 10.
References
[23] O. Jardetzsky, P. Pappas, N.G. Wade, J. Am. Chem. Soc. 85 (1963) 1657.
[24] M. Sundralingam, J.A. Carrabine, J. Mol. Biol. 61 (1971) 287.
[25] S. Shirotake, T. Sakaguchi, Chem. Pharm. Bull. 25 (1977) 3232.
[26] D.J. Szalda, L.G. Marzilli, T.J. Kistenmacher, Inorg. Chem. 14 (1975) 2076.
[27] U. Chaveerach, A. Meenongwa, Y. Trongpanich, C. Soikum, P. Chaveerach,
Polyhedron 29 (2010) 731.
[1] A. Eastman, in: B. Lippert (Ed.), Cisplatin: Chemistry and Biochemistry of a
Leading Anticancer Drug, VHCA and Wiley-VCH, Zurich and Germany, 1999, p.
111.
[2] T.W. Hambley, J. Chem. Soc., Dalton Trans. (2001) 2711.
[3] H.M. Pinedo, J.H. Schornagel (Eds.), Platinum and Other Metal Co-ordination
Compounds in Cancer Chemotherapy, vol. 2, Plenum, New York, 1996.
[4] M.J. Bloemink, H. Engelking, S. Karentzopoulos, B. Krebs, J. Reedijk, Inorg.
Chem. 35 (1996) 619.
[28] Material Studio 4.3, Accelrys Software Inc., San Diego, CA, 2008.
[29] B. Delley, Phys. Rev. B. 66 (2002) 155125/1.
[30] P. Tyagi, P. Gahlot, R. Kakkar, Polyhedron 27 (2008) 3567.
[31] R. Kakkar, R. Grover, P. Gahlot, Polyhedron 25 (2006) 759.
[32] R. Wagner, W. von Philipsborn, Helv. Chim. Acta 53 (1970) 299.
[33] A.B.P. Lever, Inorganic Electronic Spectroscopy, Elsevier, New York, 1984.
[34] M.J. Begley, P. Hubberstey, C.H.M. Moore, J. Chem. Res. (S) (1986) 172.
[35] L. Fabbrizzi, P. Paoletti, A.B.P. Lever, Inorg. Chem. 15 (1976) 1502.
[36] A.W. Addison, M. Carpenter, L.K.M. Lau, M. Wicholas, Inorg. Chem. 17 (1978)
1545.
[37] A.W. Addison, M. Palaniandavar, W.L. Driessen, F. Paap, J. Reedijk, Inorg. Chim.
Acta 142 (1988) 95.
[38] T. Pandiyan, M. Palaniandavar, M. Lakshminarayanan, H. Manohar, J. Chem.
Soc., Dalton Trans. (1992) 3377.
[39] I.M. Procter, B.J. Hathaway, P. Nicholas, J. Chem. Soc., Section A (1968) 1678.
[40] B.J. Hathaway, D.E. Billing, Coord. Chem. Rev. 5 (1970) 143.
[41] A.W. Addison, Spectroscopic and redox trends from model copper complexes,
in: K.D. Karlin, J. Zubieta (Eds.), Copper Coordination Chemistry: Biological
and Inorganic Perspectives, Adenine Press, New York, 1983.
[42] V.V. Pavlischuk, Theor. Exp. Chem. 31 (1995) 1.
[5] J.A. Navarro, J.M. Salas, M.A. Romero, R. Vilaplana, F. Gonzalez-Vilchex, R.
Faure, J. Med. Chem. 41 (1998) 332.
[6] M.P.M. Marques, T. Girao, M.C. Pedroso de Lima, A. Gameiro, E. Pereira, P.
Garcia, Biochim. Biophys. Acta 1589 (2002) 63.
[7] Y. Jung, S.J. Lippard, Chem. Rev. 107 (2007) 1387.
[8] I. Eryazici, C.N. Moorefield, G.R. Newkome, Chem. Rev. 108 (2008) 1834.
[9] M.C. Linder, M. Hazeqh-Azam, Am. J. Clin. Nutr. 63 (1996) 797s.
[10] F. Saczewski, E. Dziemidowicz-Borys, P.J. Bednarski, R. Grunert, M. Gdaniec, P.
Tabin, J. Inorg. Biochem. 100 (2006) 1389.
[11] J. He, P. Hu, Y.-J. Wang, M.-L. Tong, H. Sun, Z.-W. Mao, L.-N. Ji, Dalton Trans.
(2008) 3207.
[12] C.M. Mikulski, C.J. Lee, T.B. Tran, Inorg. Chim. Acta 136 (1987) L13.
[13] D. Tran Qui, E. Palacios, Acta Crystallogr., Sect. C 46 (1990) 1220.
[14] G. De Munno, S. Mauro, T. Pizzino, D. Viterbo, J. Chem. Soc., Dalton Trans.
(1993) 1113.
[15] A. Panfil, A. Terrón, J.J. Fiol, M. Quiros, Polyhedron 13 (1994) 2513.
[16] M. Palaniandavar, I. Somasundaram, M. Lakshminarayanan, H. Manohar, J.
Chem. Soc., Dalton Trans. (1996) 1333.