Effect of ligand substitution on DNA binding ability of two new square planar copper(II)-Schiff base complexes

Article abstract of DOI:10.1016/j.molstruc.2010.04.034

Two terdentate NNO donor Schiff base ligands CH₃C(OH){double bond, long}CHC(CH₃){double bond, long}NCH₂C₅H₄N [HL¹] and C₆H₅C(OH){double bond, long}CHC(CH₃){

Full text of DOI:10.1016/j.molstruc.2010.04.034

Journal of Molecular Structure 975 (2010) 265–273
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Effect of ligand substitution on DNA binding ability of two new square
planar copper(II)–Schiff base complexes
Dipali Sadhukhan a, Aurkie Ray a, Saurabh Das a, Corrado Rizzoli b, Georgina M. Rosair c, Samiran Mitra a,
*
a Department of Chemistry, Jadavpur University, Raja S.C. Mullick Road, Kolkata 700 032, West Bengal, India
b Universitá degli Studi di Parma, Dipartimento di Chimica Generale ed Inorganica, Chimica Analitica, Chimica Fisica Viale, G.P. Usberti 17/A I-43100 Parma, Italy
c School of Engineering and Physical Sciences, Heriot Watt University, Edinburgh EH14 4AS, UK
a r t i c l e i n f o
a b s t r a c t
Two terdentate NNO donor Schiff base ligands CH3C(OH)@CHC(CH3)@NCH2C5H4N [HL1] and
C6H5C(OH)@CHC(CH3)@NCH2C5H4N [HL2] were used to synthesize two mononuclear square planar cop-
per(II) complexes, [CuL1(CF3COO)] (1) and [CuL2(CF3COO)] (2). The complexes were characterized by ele-
mental analyses, FT-IR, UV–vis spectral methods and their structures were established by single crystal
X-ray diffraction study. The biochemical activity of the complexes towards DNA binding were performed
using absorbance and fluorescence titration methods and their mode of binding with calf thymus DNA
has been established by viscosity measurement technique.
Article history:
Received 16 March 2010
Received in revised form 26 April 2010
Accepted 26 April 2010
Available online 29 April 2010
Keywords:
Copper(II) complexes
Square planarity
Bifurcated H-bonding
CT-DNA
Ó 2010 Elsevier B.V. All rights reserved.
Hypochromism
Intercalation
1. Introduction
amine with acetyl acetone and benzoyl acetone [11,12]. Those
complexes were characterized structurally, spectroscopically and
The Schiff base metal complexes are very important tools for
inorganic chemists as these are widely used to design molecular
ferromagnets, in catalysis, in biological modeling applications
and as liquid crystals [1]. There are numerous accounts in the liter-
ature describing the chemistry of metal complexes of Schiff base li-
gands containing two, three, four, five and six donor atoms [2].
Terdentate Schiff bases with NNO donor sets have been often used
to block the coordination sites of metal ions which prefer square
planar or square pyramidal geometry and to saturate the coordina-
tion number of the metal ion. A bridging ligand has been used to
synthesize dinuclear and multinuclear complexes [3–6]. Carboxyl-
ate ligands serve this purpose well [7, 8]. Sometimes when a com-
plex strongly prefers square-planar geometry the carboxylate
anion binds in terminal monodentate fashion [9,10]. Small Schiff
base ligands with N-heterocyclic aromatic ring where the hetero
atom can coordinate to the metal ion are expected to induce a good
extent of planarity to the complexes because of the rigidity in-
serted within the ligand framework by the coordinating heterocy-
cle. Earlier our research group synthesized such planar complexes
of copper(II) with terdentate NNO donor Schiff base ligands con-
taining N-heterocyclic fragment by the condensation of 2-picolyl
magnetically but the utility of the complexes in biochemical field
remained unexplored.
Planar mononuclear complexes with solubility in buffered
aqueous medium can penetrate more easily through cell mem-
brane compared to the large organic molecules and hence have
gained much importance now a day to interact with DNA and to
act as chemical nuclease [13–21]. Metal complexes of the ligands
having planar aromatic heterocyclic ring such as pyridine, bipyri-
dine, phenanthroline moieties are good metallointercalators [22–
27] and have been employed so far in binding studies with DNA.
Although a number of small inorganic complexes containing CuII,
ZnII, CoIII, RuII metal ions with varieties of ligand systems have been
reported so far [28] still there are scope to design and study new
complexes of the same or different metal ions with special modifi-
cations in the ligand system as new chemical nuclease. Aspired by
this idea, in continuation to our earlier works [11,12] we have syn-
thesized two new copper(II) complexes [1 and 2] with the terden-
tate NNO donor Schiff base ligands by the condensation of 2-
picolyl amine with acetyl acetone [HL1] and benzoyl acetone
[HL2]. We have designed our ligand systems [HL1 and HL2] so that
they differ only by a substituent (methyl in HL1 and phenyl in HL2)
and have monitored the effect imposed by the substituents on the
structure–function relationship of respective complexes [1 and 2]
during interaction with DNA by absorption and fluorescence
* Corresponding author. Tel.: +91 033 2414 6666x2779; fax: +91 033 2414 6414.
E-mail address: smitra_2002@yahoo.com (S. Mitra).
0022-2860/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2010.04.034

266
D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
titration methods. The probable modes of binding of the two com-
plexes were investigated by viscosity measurement of the DNA in
absence and in presence of the complex solutions.
In this paper we have described the crystal structures of two
new square planar copper(II)–Schiff base complexes (1 and 2)
along with spectral characterization methods and have monitored
their reverberation towards interaction with calf thymus DNA by
UV–vis, fluorescence and viscometric methods.
(HL1) (1 mmol) was added to the former. The mixture was allowed
to stir for 2 h with warming. The dark green solution was kept in
refrigerator at 16 °C. Brown plate shaped single crystals suitable
for X-ray diffraction were obtained within one day. Crystals were
isolated by filtration and were air-dried. Yield: 0.33 g (85%). Anal.
Calc. for [C13 H13 Cu F3 N2 O3]: C, 42.65; H, 3.55; N, 7.65. Found: C,
42.69; H, 3.60; N, 7.68%.
2.2.3. Synthesis of [CuL2(CF3COO)] (2)
2. Experimental
Cu(CF3COO)2.6H2O (0.397 g, 1 mmol) was dissolved in 5 ml
methanol. 10 ml yellow methanolic solution of the Schiff base
(HL2) (1 mmol) was added to the former in stirring condition.
The mixture was allowed to stir for 2 h at room temperature. The
dark green solution was kept in refrigerator at 16 °C yielding green
needle shaped crystals suitable for X-ray diffraction after a week.
Crystals were isolated by filtration and were air-dried. Yield:
0.32 g (80%). Anal. Calc. for [C18 H15 Cu F3 N2 O3]: C, 50.48; H,
3.51; N, 6.54. Found: C, 50.50; H, 3.54; N, 6.57%.
2.1. Materials
All solvents were of reagent grade and used without further
purification. 2-picolylamine, acetyl acetone and benzoyl acetone
were purchased form Aldrich Chemical Company and were used
as received. Copper trifluoroacetate hexahydrate was prepared by
treatment of copper carbonate (E. Merck, India) with 60% trifluoro-
acetic acid (E. Merck, India) followed by the slow evaporation on
the steam bath. It was then filtered through a fine glass-frit and
preserved in a CaCl2 desiccator for further use.
2.3. Physical measurements
The Fourier Transform Infrared spectra (4000–400 cmꢀ1) of the
ligands and the complexes were recorded on a Perkin–Elmer
Spectrum RX I FT-IR system with solid KBr disc. The electronic
spectra were recorded, in acetonitrile or in 5 mM Tris–50 mM NaCl
buffer (pH ꢁ 7.4), on a Perkin–Elmer Lambda 40 (UV–vis) spectro-
photometer with a 1 cm quartz cuvette in the range 200–800 nm.
Electrochemical measurements were performed on a VersaStat-
PotentioStat II cyclic voltammeter using HPLC grade acetonitrile
as solvent where tetrabutylammonium perchlorate was used as
supporting electrolyte at a scan rate 50 mV/s. Platinum and satu-
rated calomel electrode (SCE) were the working and the reference
electrodes in the process respectively. C, H, N microanalyses were
carried out with a Perkin–Elmer 2400 II elemental analyzer. Fluo-
rescence spectroscopic measurements were performed on a JOBIN
YVON Fluoro Max 3 fluoremeter using a 10 ꢂ 10 mm2 cuvette. Vis-
cosity measurements were conducted on an Ubbelodhe viscome-
ter, immersed in a thermostated water bath maintained at 298 K.
The relative viscosities of the samples were determined using the
2.2. Syntheses
2.2.1. Syntheses of the Schiff base ligands (HL1 and HL2)
The Schiff base ligand (HL1) was synthesized by the reflux con-
densation of
a 25 ml methanolic solution of 2-picolylamine
(0.51 ml, 5 mmol) with 25 ml methanolic solution of acetyl ace-
tone (0.52 ml, 5 mmol). The ligand was refluxed for 2 h when a yel-
low solution was obtained. The colour indicated the formation of
the Schiff base HL1 which was used without further purification.
HL2 was synthesized by refluxing 25 ml methanolic solution of
2-picolylamine (0.51 ml, 5 mmol) with 25 ml methanolic solution
of benzoyl acetone (0.18 gm, 5 mmol) following the same proce-
dure as HL1. The schematic representation of the syntheses of the
ligands is depicted in Scheme 1.
2.2.2. Synthesis of [CuL1(CF3COO)] (1)
Cu(CF3COO)2.6H2O (0.397 g, 1 mmol) was dissolved in 20 ml 2-
propanol. 10 ml yellow methanolic solution of the Schiff base
H3C
NH2
H3C
R
Refluxed at 40oC
for 2 h
R
N
+
N
O
N
O
O
R= Me for 1 and Ph for 2
H3C
N
H
H3C
H
R
R
Tautomerisation
HO
N
N
O
N
Enol form of the
ligand
Keto form of the ligand
Deprotonation
H3C
R
N
_O
N
Terdentate NNO donor ligand
Scheme 1. Synthesis of the Schiff base ligand and its keto-enol tautomerism.

D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
267
Table 1
equation
g
= (t ꢀ t0)/t0, where t0 is the flow time for buffer alone
Crystal structure parameters of 1 and 2.
and t is the flow time of either only DNA solution or DNA solu-
tion + different concentrations of 1 and 2.
Parameters
1
2
Empirical formula
Formula weight
Crystal system
C13 H13 Cu F3 N2 O3
365.79
Triclinic
C18 H15 Cu F3 N2 O3
427.86
Triclinic
2.4. X-ray crystallography
Space group
P-1
7.6843(4)
P-1
7.8075(18)
The X-ray diffraction experiment of 1 was carried out at 295(2) K
on a brown plate shaped single crystal (0.28 mm ꢂ 0.22 mm ꢂ
0.08 mm). The intensity data of 1 were collected with a Bruker
0
a (AÅ)
0
9.2951(5)
10.732(3)
10.785(3)
b (AÅ)
0
11.1932(6)
APEX-II CCD diffractometer using
x scan technique with Mo-Ka
c (AÅ)
a
(°)
93.1770(8)
104.7480(9)
98.9368(8)
759.98(7)
97.227(9)
108.391(7)
96.360(6)
839.8(3)
radiation (k = 0.71073 Å). Data reduction was performed with SAINT
program [29]. Empirical absorption corrections were carried out
using the SADABS program [30]. The structures were solved by di-
rect methods by using the SIR97 program [31]. All non hydrogen
atoms were refined anisotropically by full-matrix least-squares
based on F2. The H atoms were generated geometrically and were in-
cluded in the refinement in the riding model approximation. The
lattice constants were refined by least-square refinement using
8528 total reflections (1.9° < h < 25.2°), 2757 unique reflection
(Rint = 0.021). Structure solution and refinement based on 2451
b (°)
c
(°)
0
V (AÅ3)
Z
2
295
0.71073
2
Temperature (K)
100(2)
0.71073
0
kMo–K (AÅ)
a
Dc (Mg mꢀ3
)
1.599
1.481
370
1.9–25.2
8528
1.692
1.355
434
2.79–27.47°
17,576
3687
l
(mmꢀ1
)
F (0 0 0)
h range for data collection (°)
Total data
observed reflections with I > 2r(I) and 258 parameters gave final
Unique data
2757
R = 0.0275, wR = 0.0789 and S = 1.02. The fluorine atoms of the tri-
fluoromethyl group are disordered over three orientations rotated
about the C12–C13 bond, with occupancy factors of 0.53(3),
0.34(3) and 0.13(2), respectively. During the refinement the C–F
and Fꢃ ꢃ ꢃF distances were restrained to be equal and the Uij compo-
nents of the disordered F atoms were restrained to approximate
isotropic behavior. The X-ray diffraction experiment of 2 was
carried out at 100(2) K on a green needle shaped single crystal
(0.08 mm ꢂ 0.18 mm ꢂ 0.62 mm). The selected crystal was
Observed data [I > 2
r
(I)]
2451
0.0275
0.0789
3039
0.0462
0.1320
Ra
Rwb
s
1.02
0.021
0.30
1.03
0.0402
1.33
Rint
0
D
D
qmax (eꢀ. ÅAꢀ3
qmin (eꢀ. ÅAꢀ3
)
0
ꢀ0.13
ꢀ1.03
)
mounted on a Bruker CCD area diffractometer with Mo-Ka radiation
using non-linear curve fit analysis [Eqs. (3) and (4)]. All experimen-
tal points for binding isotherms were fitted by least-square
analysis;
(k = 0.71073Å). The lattice constants were refined by least-square
refinement using 17,576 total reflections (2.79° < h < 27.47°), 3687
unique reflection (Rint = 0.0402). Structure solution and refinement
h
ꢀ
ꢁ
ih
ꢀ
ꢁ
i
based on 3039 observed reflections with I > 2r(I) and 245 parame-
D
A
DA
C0 ꢀ
C0 CD
ꢀ
C0
ters gave final R = 0.0462, wR = 0.1320 and S = 1.03. Data collection
and data reduction were done with the APEX 2 suite programs
[32]. All non hydrogen atoms were refined anisotropically by full-
matrix least-squares based on F2. The H atoms were generated
geometrically and were included in the refinement in the riding
model approximation. Selected crystallographic data, experimental
conditions and relevant features of the structural refinements for
both the complexes are summarized in Table 1.
D
Amax
D
Amax
ꢀ
ꢁ
Kd
¼
ð3Þ
DA
C0
D
Amax
ꢂ
ꢃ
ꢂ
ꢃ
D
A
DA
Amax
C0
2 ꢀ ðC0 þ CD þ KdÞ
þ CD ¼ 0
ð4Þ
D
Amax
D
D
A is the change in absorbance at 313 nm for 1 and 325 nm for
2 while Amax is the same parameter when the respective com-
D
plexes are totally bound to CT-DNA. CD is the concentration of
CT-DNA and C0 is the initial concentration of each complex. A dou-
2.5. Procedure of DNA binding
ble reciprocal plot was used for the determination of
using Eq. (5) [34]:
DAmax and Kd
Binding of the complexes, 1 and 2 with calf thymus DNA (CT-
DNA) was monitored by UV–vis spectroscopy and fluorescence
spectroscopy in 50 mM tris–HCl/NaCl buffer (pH ꢁ 7.4). The con-
1
1
Amax
Kd
¼
þ
ð5Þ
centration of the complexes in solution was 10 lM. Purity of CT-
D
A
D
D
AmaxðCD ꢀ C0Þ
DNA was verified by the ratio of absorbance at 260 nm and
280 nm. The A260/A280 value was between 1.8 and 1.9 indicating
no further deproteinization of the DNA was required. Concentra-
tion of the DNA solution was obtained by dividing absorbance at
This approach is based on the assumption that absorbance is
linearly proportional to the concentration of the complexes. The
concentration of the complexes was 10 M and CT-DNA concentra-
l
260 nm by the molar extinction coefficient e
260 = 6600 Mꢀ1 cmꢀ1
tion was kept around 12–15-fold greater than that of the complex.
per base for CT-DNA [33]. Therefore, in all experiments the DNA
concentration has been expressed in terms of base.
For DNA binding studies the following compound-DNA equilib-
rium was considered
3. Results and discussion
3.1. Fourier transform IR spectroscopy
L þ D ꢀ LD
ð1Þ
The IR spectra of 1 and 2 were analyzed in comparison with
their free ligand HL1 and HL2 in the region 4000–400 cmꢀ1. A
strong sharp absorption band around 1607 cmꢀ1 in the spectrum
of HL1 and around 1654 cmꢀ1 in the spectrum of HL2 can be as-
signed to the C@N group which initially confirms the formation
Kd ¼ ½Lꢄ½Dꢄ=½LDꢄ
ð2Þ
L represents the complexes used in the experiments while D
represents CT-DNA. The apparent dissociation constant (Kd = 1/Kapp
where Kapp is the apparent binding constant) was determined

268
D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
of the Schiff bases. In the complexes, 1 and 2 those bands are
shifted to 1574 and 1593 cmꢀ1 respectively upon complexation
with the metal, which can be attributed to the coordination of
the nitrogen atom of the imino group to the metal ion [35]. The li-
gands HL1 and HL2 show well defined bands at 3410 and
3404 cmꢀ1 respectively due to OAH stretching which disappear
in the complexes and a CAO absorption band at 1057 cmꢀ1 in both
the spectra of 1 and 2 appears indicating that the ligands have
undergone deprotonation on complexation. The low energy
pyridine ring in plane and out of plane vibrations observed in the
spectrum of HL1 at 604 and 628 cmꢀ1 and of HL2 at 610 and
620 cmꢀ1 respectively whereas the corresponding bands for 1
and 2 are shifted to higher frequencies which is a good indication
of the coordination of the heterocyclic nitrogen [36]. The typical
3.4. X-ray crystal structures of 1 and 2
The structures of 1 and 2 have been confirmed by single crystal
X-ray diffraction study. The ORTEP diagrams of the complexes are
shown in Figs. 1 and 2 and selected bond distances and angles are
given in Table 2. The structures of both 1 and 2 consist of mononu-
clear units of [CuL1(CF3COO)] and [CuL2(CF3COO)] where
HL1@CH3C(OH)@CHC(CH3)@NCH2C5H4N and HL2@C6H5C(OH)@
CHC(CH3)@NCH2C5H4N. In both complexes the copper ions have
distorted square-planar geometry and are bound to the donor
atoms N1, N2, O1 and O2. Out of the four coordination sites three
positions are occupied by two nitrogen (pyridyl and imino) atoms
and a deprotonated enolato oxygen atom of the corresponding tri-
dentate NNO donor Schiff base (L1)ꢀ and (L2)ꢀ. The fourth position
of the square plane is occupied by one of the oxygen atoms (O2) of
a trifluoroacetate ion coordinated to the central copper atom in ter-
minal fashion. The Cu1AO2 distance is 1.9653(15) Å in 1 and
1.961(2) Å in 2 and matches well with the reported data of other
copper(II)-monodentate trifluoroacetato complexes [40, 41]. The
geometry around the Cu1 square plane suffers distortion which
is evident from the chelate bite angles made by the NNO donor
set of the respective ligands in both the complexes. The
N1ACu1AN2 angles [83.52(7)° in 1 and 84.17(10)° in 2] undergo
compression whereas the O1ACu1AN2 angles [94.80(7)° in 1 and
94.44(10)° in 2] undergo expansion from their ideal 90° value
trifluoroacetate vibrations
1612 cmꢀ1 for 2 as well as the
1 and 2 are observed in the spectra of the complexes. The differ-
ences between asym(COO) and sym(COO),
(259 and 256 cmꢀ1
m
asym(COO) at 1615 cmꢀ1 for 1 and
sym(COO) at 1356 cmꢀ1 for both
m
m
m
Dm
in 1 and 2 respectively) are much greater than 164 cmꢀ1 observed
in free trifluoroacetate ion indicating the presence of a deproto-
nated carboxylate group coordinated to the metal centers in mono-
dentate terminal fashion [36, 37]. Sharp bands appearing at 458
and 412 cmꢀ1 in the spectra of 1 and 2 respectively correspond
to mCu–N indicating the metal-nitrogen coordination.
3.2. UV–vis spectroscopy
UV–vis spectra of the free ligands and the complexes were re-
corded at 300 K in HPLC grade acetonitrile solution. The electronic
spectral data for 1 and 2 in acetonitrile solvent are in good agree-
ment with their geometry. The UV–vis spectra of the free ligands
exhibit two charge transfer (CT) bands at 312, 270 nm in HL1 and
at 331, 248 nm in HL2 which can be attributed to n ? pꢅ and
p
?
pꢅ transitions within the Schiff base ligands. Three distinct
CT bands appeared in the spectra of the complexes at 325, 282,
234 nm in 1 and 361, 305, 244 nm in 2 which indicate that the li-
gand centered CT bands may have undergone large bathochromic
shifts in the spectra of the complexes appearing at 325, 282 nm
in 1 and 361, 305 nm in 2. The band at 234 nm in 1 and 244 nm
in 2 can be considered as L ? M charge transfer transitions as those
were absent in the spectra of the respective free ligands. In the
spectra of the complexes, much weaker, less well defined broad
bands are found at the lower energy regions associated with d–d
transitions. 1 and 2 are assumed to have a square-planar geometry
with a d–d transition at 595 and 600 nm respectively, similar to
those observed for structurally well characterized square-planar
copper(II) complexes [38].
Fig. 1. An ORTEP drawing of 1 with displacement ellipsoids drawn at the 50%
probability level. Only the major component of disorder affecting the trifluromethyl
group is shown.
3.3. Cyclic voltammetry
The electrochemical behavior of 1 and 2 were studied in HPLC
grade acetonitrile medium with tetrabutylammonium perchlorate
as supporting electrolyte at a scan rate of 50 mV secꢀ1. 1 showed a
reductive response at 0.452 V vs SCE, which is assigned to CuII ?
CuI reduction with no oxidative response for CuI ? CuII oxidation
during anodic scan suggesting that the CuI species formed during
reduction is too unstable to get oxidized and may have undergone
decomposition [39]. Unlike 1, 2 showed a reductive response at
+0.389 V and an oxidative response at +0.555 V for CuII ? CuI ?
CuII cycle. This reduction and oxidation are quasi reversible, char-
acterized by a peak to peak separation (DEp) 166 mV which
remain unchanged upon changing the scan rate suggesting quasi
reversible redox behavior of 2. A distinct ligand oxidation peak
appeared at +1.275 V and +1.248 V respectively for HL1 and HL2.
Fig. 2. An ORTEP drawing of 2 with displacement ellipsoids drawn at the 50%
probability level.

D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
269
Table 2
0
Selected bond lengths (AÅ) and bond angles (°) for 1 and 2.
0
1
2
Bond lengths (AÅ)
Cu1AO1
Cu1AO2
Cu1AN1
Cu1AN2
1.8905(15)
1.9653(15)
1.9912(17)
1.9141(17)
1.879(2)
1.961(2)
1.963(3)
1.919(2)
Bond angles (°)
O1ACu1AO2
O1ACu1AN1
O1ACu1AN2
O2ACu1AN1
O2ACu1AN2
N1ACu1AN2
88.91(7)
176.85(7)
94.80(7)
92.96(7)
174.66(7)
83.52(7)
86.29(9)
176.39(8)
94.44(10)
95.37(9)
175.31(8)
84.17(10)
Fig. 3. The dimeric unit of 1 formed by intermolecular CAHꢃ ꢃ ꢃO hydrogen bonds
(dashed lines). Intramolecular H bonds are also shown. Only the major component
of disorder is shown. Symmetry code: (i) 2ꢀx, ꢀy, 2ꢀz.
to the 2-picolyl amine part of (L1)ꢀ along with one aromatic proton
(H4) of the pyridyl moiety are close enough to O3 to act as H-bond
donor. In both complexes a similar intramolecular CAHꢃ ꢃ ꢃO H-
bond involving the O2 oxygen atom and the aromatic H1 hydrogen
atom is observed.
[11] and this behavior is expected because the compressed and ex-
panded bite angles are enclosed by a five membered and a six
membered chelate rings respectively. The transoid angles [ranging
from 174.66(7)° to 176.85(7)° in
1 and from 175.31(8)° to
176.39(8)° in 2] also deviate from the ideal value for a0 square pla-
nar system. The Cu1AN1 bond length of 1 [1.9912(7) ÅA] is slightly
0
longer than that of 2 [1.963(3) ÅA]. This behavior can be explained
by the fact that in 2 the N1ACu1AN2 bond angle [84.17(10)°] is
closer to the ideal value (90°) than the angle in 1 [83.52(7)°] which
is responsible for better overlap of the Cu-d and N-p orbitals in 2
resulting in shorter Cu1AN1 bond distance. The N2ACu1AO1 bond
angles in both 1 and 2 have quite similar values, the Cu1AN2 bond
lengths [1.9141(17) ÅA in 1 and 1.919(2) ÅA in 2] are comparable
indicating same extent of dp–pp orbital overlap of the Cu1AN2
bonds.
3.5. Results of DNA binding
3.5.1. Absorbance spectroscopic study to determine the binding
constants of 1 and 2
The ability of the complexes (1 and 2) to bind with calf thymus
DNA (CT-DNA) was investigated by UV–vis absorption spectros-
copy at physiological pH (7.4). Prior to titration of the complexes
with CT-DNA their stability was checked in 50 mM tris–HCl/NaCl
buffer (pH 7.4) at room temperature for 24 h which was then mon-
itored by UV–vis spectroscopy. No liberation of ligands was ob-
served under the above-mentioned conditions indicating that the
complexes are very stable in the said medium. In tris-buffer 1
0
0
Earlier Mitra and his group have synthesized a few mononu-
clear copper(II) complexes with the same Schiff base ligand system
of 1 [11] and 2 [12] with similar structural features but different
coordinating non chelating anions. Both complexes 1 and 2 are al-
most planar. The deviation of the Cu1 atom above the mean square
and
2 showed charge transfer bands at 313 and 325 nm
respectively.
0
0
plane [0.0108(2) ÅA in 1 and 0.0120(4) ÅA in 2] is much smaller than
those observed in the copper(II) complexes reported earlier [11–
12]. Moreover the angles between the planes defined by Cu1, N1,
N2 and Cu1, N2, O1 are 2.68(6)° and 3.34(8)° respectively, in 1
and 2, much smaller than that in similar copper(II) complexes
[11] indicating much less tetrahedral distortion for 1 and 2 from
perfect square-planar geometry.
Centrosymmetrically-related mononuclear units of 1 and 2 are
interconnected by H-bonding interactions to form dimers. The H-
bonding data are presented in Table 3. The H-bonded dimeric
structures of 1 and 2 are shown in Figs. 3 and 4 respectively. Crys-
tal packing diagrams for the two complexes are depicted in Figs. 5
and 6 respectively. Though the intermolecular H-bond acceptors in
both the complexes are provided by the same oxygen atom (O3) of
the trifluoroacetate anion, the donor hydrogens are different due to
the difference in spatial orientation of the mononuclear units
resulting in bifurcated H-bonding in 1. In 2 the methylene proton
(H8) belonging to the benzoyl acetone part of (L2)ꢀ is the sole H-
bond donor whereas in 1 the methylene proton (H6B) belonging
In order to measure the strength of binding of the complexes to
CT-DNA 10
lM solutions of 1 and 2 in tris-buffer were titrated with
increasing concentration of CT-DNA (0–140
l
M) and the change in
absorbance intensities at 313 nm for 1 and 325 nm for 2 were ob-
served. In both cases the bands showed appreciable degree of hyp-
ochromism (Figs. 7 and 8) but in case of 1 saturation was reached
much earlier than 2. From the spectral changes the apparent bind-
ing constants (Kapp) were calculated by regression analyses using
Eqs. (4) and (5) (under Section 2.5.) for both complexes 1 and 2
as shown in Figs. 9 and 10 respectively. The Kapp values obtained
from UV–vis and fluorescence studies by applying different meth-
ods of analysis are summarized in Table 4. It is therefore clear from
the DNA binding constants of the complexes that 1 (Kapp = 2.61 ꢂ
104 Mꢀ1) has a greater affinity for binding to calf thymus DNA in
comparison to 2 (Kapp = 1.54 ꢂ 104 Mꢀ1).
Table 3
H-bond parameters of 1 and 2.
0
0
0
DAHꢃ ꢃ ꢃA
<(DHA)°
d(DAH) ÅA
d(Hꢃ ꢃ ꢃA) ÅA
d(Dꢃ ꢃ ꢃA) ÅA
1
2
C1AH1ꢃ ꢃ ꢃO2
C4AH4ꢃ ꢃ ꢃO3
C6AH6Bꢃ ꢃ ꢃO3
0.93
0.93
0.97
2.55
2.59
2.54
3.051(3)
3.312(3)
3.316(3)
114
134
137
Fig. 4. The dimeric unit of 2 formed by intermolecular CAHꢃ ꢃ ꢃO hydrogen bonds
(dashed lines). Intramolecular H bonds are also shown. Symmetry code: (i) 1ꢀx,
2ꢀy, ꢀz.
C1AH1ꢃ ꢃ ꢃO2
C8AH8ꢃ ꢃ ꢃO3
0.95
0.95
2.58
2.43
3.087(4)
3.302(4)
114
152

270
D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
Fig. 5. The crystal packing of 1 showing intermolecular hydrogen-bonding interactions as dashed lines. Only the major component of disorder is shown.
Fig. 6. The crystal packing of 2 showing intermolecular hydrogen-bonding interactions as dashed lines.
Fig. 8. Change in absorption spectra of 2 on titration with increasing CT-DNA
concentration.
Fig. 7. Change in absorption spectra of 1 on titration with increasing CT-DNA
concentration.
spectra of 1 in presence of different amount of calf thymus DNA
was recorded following an excitation at 220 nm. The emission
spectrum exhibits a maximum at 405 nm. From Fig. S1 (provided
as a supplementary data), it is clear that fluorescence emission
intensity decreased gradually with increasing amount of calf thy-
mus DNA showing that fluorescence of 1 got sufficiently quenched
upon binding to DNA.
3.5.2. Fluorescence spectroscopic study
In order to see whether 1 and 2 interact with calf thymus DNA,
solution of the complexes at physiological pH (7.4) was titrated
with increasing amount of CT-DNA keeping concentration of 1
and 2 unchanged during such titration. For this purpose separate
solutions were made containing constant concentration of com-
plexes and different concentrations of CT-DNA. The fluorescence
The binding isotherms were analyzed using non-linear curve
fitting following equilibrium (1) [42]. Apparent dissociation

D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
271
Fig. 9. Plot of non-linear variation of
D
A/
D
Amax with increasing concentration of
Fig. 11. Plot of non-linear variation of
D
F/D
Fmax with increasing concentration of
DNA using fluorescence titration method for 1. The inset shows the double
reciprocal plot of interaction of 1 with CT-DNA.
DNA using UV–vis titration method for 1. The inset shows the double reciprocal plot
of interaction of 1 with CT-DNA.
1. Double reciprocal plot (Fig. 11 inset) was used for the determina-
tion of
DFmax and Kd using the following equation:
1
1
Fmax
Kd
¼
þ
ð8Þ
D
F
D
D
FmaxðCD ꢀ C0Þ
This approach is based on the assumption that fluorescence is
linearly proportional to the concentration of 1, the concentration
of which was 10 M while CT-DNA concentration toward the end
l
of the titration was around 20-fold greater than that of the com-
plex in solution.
Fig. 11, which shows the binding isotherms of 1 with CT-DNA
was used to calculate the binding constant [Eqs. (6) and (7)] as de-
scribed above. The apparent binding constant obtained was
0.74 ꢂ 104 Mꢀ1. The apparent binding constant obtained from dou-
ble reciprocal plot following Eq. (8) [Fig. 11 inset] was
0.80 ꢂ 104 Mꢀ1. The values indicate that the apparent binding con-
stants obtained from two different methods of analysis are similar.
2, which has a kmax at 325 nm was excited at 340 nm using a
fluorescence spectrophotometer. The emission spectra which were
recorded from 500 nm to 700 nm did not show any characteristic
peak that could be used to monitor any possible interaction with
CT-DNA, for which reason interaction of 2 could not be followed
by fluorescence spectroscopy.
Fig. 10. Plot of non-linear variation of
DNA using UV–vis titration method for 2. The inset shows the double reciprocal plot
of interaction of 2 with CT-DNA.
DA/DAmax with increasing concentration of
constant (Kd = 1/Kapp, where Kapp is the apparent binding constant)
was determined using non-linear curve fit analysis [Eqs. (6) and
(7)] based on such equilibrium. All experimental points for binding
isotherms were fitted according to least-square analysis;
h
ꢀ
ꢁ
ih
ꢀ
ꢁ
i
3.5.3. Viscosity measurement
D
Fmax
F
DF
C0 ꢀ
C0 CD
ꢀ
C0
D
D
Fmax
Optical or photophysical investigations are necessary but not
sufficient to establish the mode of binding between metal com-
plexes and DNA. In absence of crystallographic information, hydro-
dynamic methods are proved to be least ambiguous to support a
complex-DNA binding model [43] as these measurements are very
much sensitive to length change (i.e. viscosity and sedimentation).
When a substance intercalates between DNA base pairs it unwinds
the DNA helix and hence increases the overall DNA contour length
resulting in significant increase in the viscosity of DNA solution. In
contrast, non classical interaction that bend (or kink) the DNA
helix reduce its effective length and concomitantly its viscosity.
ꢀ
ꢁ
Kd
¼
ð6Þ
D
F
C0
D
Fmax
ꢂ
ꢃ
ꢂ
ꢃ
D
F
DF
Fmax
C0
2 ꢀ ðC0 þ CD þ KdÞ
þ CD ¼ 0
ð7Þ
D
Fmax
D
where
D
F is the change in fluorescence emission intensity of 1 at
Fmax
405 nm (kex = 220 nm) for each point of the titration curve,
is the same parameter when 1 is totally bound to CT-DNA, CD is
the concentration of CT-DNA and C0 is the initial concentration of
D
Table 4
The Kapp values obtained from UV–vis and fluorescence studies by applying different methods of analysis.
Methods of analysis
UV–vis
Fluorescence
Double reciprocal fit
Non-linear fit
Double reciprocal fit
Non-linear fit
1
2
2.61 ꢂ 104 Mꢀ1
2.65 ꢂ 104 Mꢀ1
0.80 ꢂ 104 Mꢀ1
0.74 ꢂ 104 Mꢀ1
1.54 ꢂ 104 Mꢀ1
1.51 ꢂ 104 Mꢀ1
–
–

272
D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
in the ligand fragment have shown much difference in their H-
bonding features as well as in their biochemical activities.
Acknowledgements
D. Sadhukhan is thankful to University Grants Commission,
New Delhi, Government of India and A. Ray acknowledges Council
of Scientific and Industrial Research, New Delhi, Govt. of India for
providing them financial support to carry out the work. The
authors are also thankful to Mr. Partha Sarathi Guin of Saha Insti-
tute of Nuclear Physics, Kolkata, India for helping in the fluores-
cence study.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2010.04.034.
CCDC numbers for 1 and 2 are 754,875 and 664,113. These data
can be obtained free of charge at www.ccdc.cam.ac.uk (or from
Cambridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB2 1EZ, UK; email: deposit@ccdc.cam.ac.uk). The supple-
mentary data also contains Fig. S1 (plot of fluorescence intensity
vs wave length for 1).
Fig. 12. The plot showing the effect of increasing concentrations of 1 and 2 on the
viscosity of DNA.
Moreover binding of a molecule to the surface or groove of the DNA
helix produces no significant change in DNA solution-viscosity
[13]. The effects of 1 and 2 on the viscosity of CT-DNA are shown
in Fig. 12. The relative viscosity of the DNA solution has increased
gradually on addition of increasing concentration of 1 but with the
increasing concentration of 2 it has not shown any significant
change. Thus 1 is found to show intercalative mode of binding like
that of the proven intercalator ethidium bromide. In contrast, 2
being bulkier than 1 is a less effective intercalator and may have
bound exclusively to the DNA grooves like that of netropsin, dista-
mycin etc. or to the surface of the DNA helix causing less pro-
nounced change in DNA solution-viscosity.
References
[1] C.O. Rodriguez de Barbarin, N.A. Bailey, D.E. Fenton, Q.-Y. He, J. Chem. Soc.,
Dalton Trans. (1997) 161.
[2] Y.Y. Chen, D.E. Chu, B.D. McKinney, L.J. Willis, S.E. Cummings, Inorg. Chem. 20
(1981) 1885.
[3] J. Ribas, M. Monfort, I. Resino, X. Solans, P. Rabu, F. Maingot, M. Drillan, Angew.
Chem., Int. Ed. Engl. 35 (1996) 2520.
[4] R. Vicente, A. Escuer, E. Penalba, X. Solans, M. Font-Badia, J. Chem. Soc., Dalton
Trans. (1994) 3005.
[5] R. Vicente, A. Escuer, J. Ribas, X. Solans, J. Chem. Soc., Dalton Trans. (1994) 259.
[6] J. Ribas, C. Diaz, X. Solans, M. Font-Bardia, J. Chem. Soc., Dalton Trans. (1997)
35.
[7] R.J. Doedens, Prog. Inorg. Chem. 21 (1976) 209.
[8] S.J. Rettig, R.C. Thompson, J. Trotter, S. Xia, Inorg. Chem. 38 (1999) 360 (and
references therein).
[9] H.V. Huynh, D.L. Van, F.E. Hahn, T.S.A. Hor, J. Organomet. Chem. 689 (2004)
1766.
[10] Y.-Y. Wang, Q. Shi, Q.-Z. Shi, Y.-C. Gao, Z.-Y. Zhou, Polyhedron 18 (1999) 2009.
[11] R. Karmakar, C. Roy Choudhury, A.S. Batsanov, S.R. Batten, S. Mitra, Struct.
Chem. 16 (2005) 535.
[12] A. Ray, G. Pilet, C.J. Gomez-Garcia, S. Mitra, Polyhedron 28 (2009) 511.
[13] S. Banerjee, S. Mondal, W. Chakraborty, S. Sen, R. Gachhui, R.J. Butcher, A.M.Z.
Slawin, C. Mandal, S. Mitra, Polyhedron 28 (2009) 2785.
[14] L. Li, K.D. Karlin, S.E. Rokita, J. Am. Chem. Soc. 127 (2005) 520.
[15] J.-H. Li, J.-T. Wang, P. Hu, L.-Y. Zhang, Z.-N. Chen, Z.-W. Mao, L.-N. Ji,
Polyhedron 27 (2008) 1898.
3.5.4. Discussions
From the DNA binding studies it is evident that 1 binds with
DNA more strongly compared to 2. Moreover 1 is a good intercala-
tor whereas 2 shows groove binding characteristics. Such behavior
of the complexes can be interpreted highlighting the difference in
substituents on the ligands used in the two complexes. The pres-
ence of a methyl group at the C9 position in 1 which in 2 is phenyl
(Figs. 1 and 2) may be thought to be a source of difference in reac-
tivity. The presence of a phenyl ring should promote the possibility
of intercalation [44] but here the situation is reverse. The methyl
group at the C9 position in 1 has induced more planarity within
the complex compared to the phenyl group attached to the same
position in 2 which is evident from the X-ray crystallography data.
The relative displacement of the C9 atom holding the methyl and
[16] D. Kong, J. Reibenspies, J. Mao, A. Clearfield, A.E. Martell, Inorg. Chim. Acta 342
(2003) 1568.
the phenyl groups in 1 and 2 respectively from the mean plane
[17] L.M. Rossi, A. Neves, R. Hörner, H. Terenzi, B. Szpoganicz, J. Sugai, Inorg. Chim.
Acta 337 (2002) 366.
[18] K.J. Humphreys, K.D. Karlin, S.E. Rokita, J. Am. Chem. Soc. 124 (2002) 6009.
[19] G.N. De Iuliis, G.A. Lawrance, S. Fieuw-Makaroff, Inorg. Chem. Commun. 3
(2000) 307.
0
passing through N1, N2, O1 and O2 is 0.093 ÅA in 1 whereas
0
0.225 ÅA in 2. Thus comparatively smaller deviation of 1 from pla-
narity compared to 2 makes it a good DNA intercalator.
[20] L. Zhu, O. dos Santos, C.W. Koo, M. Rybstein, L. Pape, J.W. Canary, Inorg. Chem.
42 (2003) 7912.
4. Conclusion
[21] S.-P. Tang, L. Hou, Z.-W. Mao, L.-N. Ji, Polyhedron 28 (2009) 586.
[22] M.C. Prabhakara, B. Basavaraju, H.S. Bhojya Naik, Bioinorg. Chem. Appl. 2007
(2007) 36497.
Though the complexes, 1 and 2 are isostructural their H-bond-
ing patterns are quite different due to the difference in spatial ori-
entation of two centrosymmetrically-related mononuclear units.
From the spectroscopic titration results it is evident that 1 has
greater DNA binding ability than 2, probably due to the presence
of the methyl substituent attached to C9 of HL1 which is less bulk-
ier, less electron rich and more planar than the phenyl substituent
attached to C9 of HL2. The phenyl group reduces the ability of 2 to
bind with DNA bases though the reverse should be expected. From
the viscosity measurement it is clear that 1 binds with DNA by
intercalative method whereas 2 shows surface/groove binding.
Thus the complexes obtained by varying only an alkyl substituent
[23] Y. Wang, G. Lin, J. Hong, T. Lu, L. Li, N. Okabe, M. Odoko, Inorg. Chim. Acta 362
(2009) 377.
[24] A.M. Thomas, A.D. Naik, M. Nethaji, A.R. Chakravarty, Inorg. Chim. Acta 357
(2004) 2315.
[25] J.D. Ranford, P.J. Sadler, D.A. Tocher, J. Chem. Soc., Dalton Trans. (1993) 3393.
[26] M.J. Plater, M.R. St, J. Foreman, J.M.S. Skakle, R.A. Howie, Inorg. Chim. Acta 332
(2002) 135.
[27] C.A. Johns, G.M. Golzar Hossain, K.M. Abdul Malik, S. Zahir Haider, U.K.
Rowzatur Romman, Polyhedron 20 (2001) 721.
[28] S. Banerjee, S. Mondal, S. Sen, S. Das, D.L. Hughes, C. Rizzoli, C. Desplanches, C.
Mandal, S. Mitra, Dalton Trans. (2009) 6849 (and references therein).
[29] SAINT V 4.035 Software for the CCD Detector System, Siemens Analytical
Instruments Division, Madison, WI, 1995.
[30] SADABS, Program for Absorption Corrections using Siemens CCD based on the
Method of Bob Blessing, Acta Crystallogr. A51 (1995) 33.

D. Sadhukhan et al. / Journal of Molecular Structure 975 (2010) 265–273
273
[31] A. Altomare, M.C. Burla, M. Camalli, G. Gascarano, G. Giacovazzo, A.G.G.
Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr. 32 (1999) 115.
[32] Apex2 Version 2.1: Area Detector Control and Integration Software, v. 5.629
and 6.45, Bruker Analytical X-Ray Instruments, Inc., Madison, WI, USA, 2003,
2006.
[37] M.T.H. Tarafder, A. Kasbollah, K.A. Crouse, A.M. Ali, B.M. Yamin, H.-K. Fun,
Polyhedron 20 (2001) 236.
[38] A.B.P. Lever, Inorganic Electronic Spectroscopy, Elsevier, Amsterdam, 1984. p.
553.
[39] S. Basak, S. Sen, S. Mitra, C. Marschner, W.S. Sheldrick, Struct. Chem. 19 (2008)
115.
[40] T. Nakamura, H. Higuchi, K. Izutsu, Bull. Chem. Soc. Jpn. 62 (1989) 3089.
[41] R.L. Paul, S.M. Couchman, J.C. Jeffery, J.A. McCleverty, Z.R. Reeves, M.D. Ward, J.
Chem. Soc., Dalton Trans. (2000) 845.
[33] M.E. Reichmann, S.A. Rice, C.A. Thomas, P. Doty, J. Am. Chem. Soc. 76 (1954)
3047.
[34] S. Roy, R. Banerjee, M. Sarkar, J. Inorg. Biochem. 100 (2006) 1320.
[35] M. Dolaz, M. Tümer, M. Dig˘rak, Transition Met. Chem. 29 (2004) 528.
[36] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, fifth ed., Theory and Applications in Inorganic Chemistry, John
Wiley and Sons Inc., New York, 1997.
[42] P.S. Guin, S. Das, P.C. Mandal, J. Inorg. Biochem. 103 (2009) 1702.
[43] V.G. Vaidyanathan, B.U. Nair, J. Inorg. Biochem. 93 (2003) 271.
[44] M. Cusumano, A. Giannetto, J. Inorg. Biochem. 65 (1997) 137.