JOURNAL OF

Inorganic
Biochemistry
Journal of Inorganic Biochemistry 101 (2007) 434–443
www.elsevier.com/locate/jinorgbio

Cobalt complexes of terpyridine ligand: Crystal structure

and photocleavage of DNA
Ramasamy Indumathy a, Srinivasan Radhika a, Mookandi Kanthimathi a,
T. Weyhermuller b, Balachandran Unni Nair a,*
a
Chemical Laboratory, Central Leather Research Institute, Adyar, Chennai 600 020, India
b
Max-Planck Institut Fur Bioanorganische Chemie, D-45413 Mulheim an der Ruhr, Germany

Received 7 August 2006; received in revised form 6 November 2006; accepted 6 November 2006
Available online 15 November 2006

Abstract

Two new cobalt complexes, [Co(pytpy)2](ClO4)2, 1, and [Co(pytpy)2](ClO4)3, 2 where pytpy = pyridine terpyridine, have been synthe-
sized and characterized. Single-crystal X-ray structure of both the complexes has been resolved. The structure shows the complexes to be
a monomeric cobalt(II) and cobalt(III) species with two pytpy ligands coordinated to the metal ion to give a six coordinate complex.
Both cobalt(II) and cobalt(III) complexes crystallize in meridional configuration. The interaction of these complexes with calf thymus
DNA has been explored by using absorption, emission spectral, electrochemical studies and viscosity measurements. From the experi-
mental results the DNA binding constants of 1 and 2 are found to be (1.97 ± 0.15) · 104 M1 and (2.7 ± 0.20) · 104 M1 respectively.
The ratio of DNA binding constants of 1 and 2 have been estimated to be 0.82 from electrochemical studies, which is in close agreement
with the value of 0.73 obtained from spectral studies. The observed changes in viscosity of DNA in the presence of increasing amount of
complexes 1 and 2 suggest intercalating binding of these complexes to DNA. Results of DNA cleaving experiments reveal that complex 2
efficiently cleaves DNA under photolytic conditions while complex 1 does not cleave DNA under similar conditions.
 2006 Elsevier Inc. All rights reserved.

Keywords: DNA interaction; Cobalt complexes; Photocleavage

1. Introduction binding properties of metal complexes gains importance

Recent years have seen a growing interest in the binding
of small molecules to DNA and DNA cleaving with metal
complexes [1–3]. The interaction of transition metal com-
plexes with DNA has been extensively studied in order to
develop novel probes of DNA structure [4,5] and DNA
mediated electron transfer reactions [6]. Metal ion coordi-
nation to nucleic acids is not only required for charge neu-
tralization, it is also essential for the biological function of
nucleic acids [7]. Metal ions are used in purifying nucleic
acids [8] and in probing the structure and biochemistry of
nucleic acids [9]. The precise understanding of the DNA
*
Corresponding author. Tel.: +91 44 2441 1630; fax: +91 44 2491 1589.
E-mail address: bunair@clri.info (B. Unni Nair).
because of therapeutic approaches; study of nucleic acid
conformations and in the development of new tools for
nanotechnology [10–12]. Prominent among the various
metal complexes employed by our group so far in the stud-
ies of DNA interaction, are those of metallo-intercalators
which incorporate either pyridine, phenanthroline or a
modified phenanthroline moiety as ligands [13–16]. In these
complexes, the ligands or metal ion may be varied in an
easily controlled way to facilitate the individual applica-
tion. Polypyridyl transition metal complexes can bind to
DNA by non-covalent interactions such as external surface
binding, groove binding for large molecules and intercala-
tion for planar molecules or compounds containing a ring
system [17]. All the studies reveal that modification of the
metal or ligands would lead to subtle or substantial
changes in the binding modes, location of binding and

0162-0134/$ - see front matter  2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jinorgbio.2006.11.002
 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443
affinity of DNA binding. This gives valuable information
to explore the various site-specific DNA probes and poten-
tial chemotherapeutical agents [18]. The interaction of
transition metal polypyridyl and Schiff base ligand com-
plexes with DNA has been extensively studied by us [19–
21] because, their general photoactivity make them suitable
candidates as probes for DNA secondary structure,
photocleavers and antitumour drugs [22–24]. Cobalt com-
plexes have gained importance because of their application
as potential hypoxia-activated prodrug [25–27]. We chose
to concentrate our studies on complexes of cobalt, which
have the same interesting characteristics and DNA cleaving
properties as ruthenium complexes, but have not received
as much attention as the ruthenium(II) systems [28,29].
Among different methodologies adopted for the photoclea-
vage of DNA, the one based on irradiation with light of
longer or shorter wavelength has gained importance for
their potential use in photodynamic therapy (PDT) of
cancer [30–33]. We have designed new cobalt complexes
derived from a tridentate ligand, pyridine terpyridine with
the aim to involve the metal based d–d and/or charge trans-
fer band(s) in the photoexcitation process.
Substituted pyridines including terpyridines are promi-
nent building blocks in both organic and inorganic supra-
molecular chemistry [34,35] with their p-stacking ability,
directional H-bonding and coordination properties. The
coordination chemistry of 2,2 0 :6 0 ,200 -terpyridine (terpy)
and its derivatives has been intensively explored due to
the array of interesting electronic, photonic, magnetic,
reactive and structural properties shown by the transition
metal complexes of this family of ligands [36]. The lumines-
cence properties of pyridine terpyridine ligands stem from
the conjugated aromatic cores [37]. The present work is
in continuation of our interest in defining and evaluating
DNA binding properties of pyridine based ligands, which
would help in the design of newer drugs and develop new
selective and efficient DNA recognition and cleaving
agents. We aim at exploring the design of new metal com-
plexes, which possess more potent DNA binding affinities
and DNA cleaving ability. In this paper, we report the syn-
thesis and characterization of two new cobalt complexes
derived from terpyridine ligand, as shown below:
2+ / 3+
N N
N N Co N N
N N
1 and 2 ( 1 is dipositive and 2 is tripositive complex)
A series of physical methods like absorption, fluores-
cence, thermal denaturation studies,cyclic voltammetry
and viscosity measurements have been used to probe the
435
interactions of the cobalt complexes with CT DNA. The
photochemical cleavage of DNA by complex 2 is also
demonstrated.
2. Experimental section
2.1. Materials
The materials used in this investigation such as 2-acetyl
pyridine, pyridine-4-carboxaldehyde were purchased from
Aldrich Chemicals and used as received. Other materials
like sodium hydroxide, ammonium acetate, sodium per-
chlorate, potassium permanganate and solvents like aceto-
nitrile, dimethyl sulphoxide, diethyl ether, perchloric acid
and hydrochloric acid were of reagent grade. The ligand,
pyridine–terpyridine was prepared using published proce-
dures [38]. The purity of the ligand was examined by mea-
suring the melting point and compared with the literature
value. CT DNA was purchased from Bangalore Genie
(India). Agarose (molecular biology grade), and ethidium
bromide (EB) were from Sigma. Tris (hydroxymethyl)
aminomethane–HCl (Tris–HCl) buffer was prepared using
deionized and sonicated triple distilled water. All the exper-
iments regarding the binding and cleavage of DNA using
complexes 1 and 2 were carried out in Tris buffer (pH,
7.2). A solution of CT DNA in the buffer gave a ratio of
UV absorbance at 260 and 280 nm of about 1.8–1.9:1, indi-
cating that the DNA was sufficiently free from protein [39].
The DNA concentration per nucleotide was determined by
absorption spectroscopy using molar absorption coefficient
6600 M1 cm1 at 260 nm [40].
2.2. Physical measurements
Elemental analyses were performed using a Heraeus
CHN-O-rapid analyzer. UV–visible spectra of the com-
plexes were recorded on a Perkin–Elmer Lambda 35
double beam spectrophotometer at 25 C. The emission
spectra were recorded on a Hitachi 650-40 spectrofluori-
meter. The electrospray ionization (ESI) mass spectra of
the complexes were recorded with a Micro mass Quattro
II triple quadrupole mass spectrometer. The infrared spec-
trum of the complex was recorded on a Perkin–Elmer
FTIR spectrometer. Cyclic voltammetry was performed
on an EG and G PAR 173 potentiostat/galvanostat ana-
lyzer. Tetrabutyl ammonium perchlorate (TBAP) was used
as the supporting electrolyte. The sample in dried DMSO
was purged with nitrogen prior to measurement. A stan-
dard three electrode system comprising of glassy carbon
as a working electrode, platinum electrode as an auxillary
electrode and a saturated calomel as reference electrode
(SCE) was used. Cyclic voltammetric investigations on
complexes 1 and 2 have been carried out in 10 mM Tris
buffer, 50 mM NaCl (pH 7.2). 1H NMR was recorded in
DMSO – d6 solution using JEOL – 500 MHz spectrometer.
Chemical shifts (d) are given in ppm. (Caution. Perchlorate
436 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443

salts of these complexes are potentially explosive and 2.5. X-ray crystallographic data collection and refinement
should be handled in small quantities with care.) of the structures of complex (1) and (2)

2.3. Synthesis of bis(pyridine–terpyridine) cobalt(II) A red single crystal of 1 and a yellow translucent single
perchlorate, [Co(pytpy)2](ClO4)2, (1) crystal of 2, having parallelepiped shapes and dimensions
of 0.07 · 0.06 · 0.03 mm3 and 0.11 · 0.03 · 0.03 mm3
The cobalt(II) complex was prepared by a direct reac- respectively were coated with perfluoropolyether. The crys-
tion of cobalt(II) chloride and the ligand in 1:2 ratio of tals were mounted in the nitrogen cold stream (100 K) of a
metal to ligand, using methanol as the reaction medium. Nonius Kappa-CCD diffractometer equipped with a Mo-
The ligand, pytpy (0.62 g, 2 mmol) was dissolved in meth- target rotating-anode X-ray source and a graphite mono-
anol and the solution was heated to boiling. A boiling, chromator (Mo-Ka, k = 0.71073Å). A total of 771 images
aqueous solution of cobaltous chloride (0.23 g, 1 mmol) were taken, 90 s each and stepwise rotation of 1.0 in
was transferred to the ligand solution. The resulting solu- omega and phi for compound 1 and a total of 1654 images
tion turned to reddish brown on the addition of cobalt(II) were taken, 35 s each and stepwise rotation of 0.5 in
solution. The reaction mixture was heated for 10 min. An omega and phi for compound 2. Final cell constants were
aqueous solution of sodium perchlorate (5 mL of 0.5 M.) obtained from a least squares fit of all integrated reflec-
was added. On cooling, dark reddish brown precipitate set- tions. Crystallographic data of the compounds are listed
tled down at the bottom of the reaction vessel. The crude in Table 1. The Siemens ShelXTL [41] software package
product thus obtained was filtered and the precipitate dis- was used for solution and artwork of the structure,
solved in 0.05 M HClO4 at 50 C. The slurry was filtered ShelXL97 [42] was used for the refinement. The structure
and the filtrate cooled to 5 C. The desired complex crystal- was readily solved by Patterson method and subsequent
lized as brown micro crystals. (Yield 60%.) The authentic- difference Fourier techniques. All non-hydrogen atoms
ity of the cobalt(II) complex was ascertained from were anisotropically refined. Hydrogen atoms attached to
elemental analysis. (Found C, 54.41; H, 2.69; N, 12.72; carbon atoms were placed at calculated positions and
Co 6.78, [Co(C20H14N4)2] (ClO4)2, 1 requires C, 54.61; H, refined as riding atoms with isotropic displacement param-
3.19; N, 12.74; Co, 6.81%.) M/z, 339.5. IR (KBr phase, eters. For compound 1, hydrogen atoms of solvent water
cm1): 1613 s, 735 w, 409 w, 3011 s, 1095 br (s, strong; molecules O(100) and O(105) were localized from the dif-
w, weak; br, broad). Diffraction quality crystals were ference map and were refined with restrained O–H dis-
obtained as nitrate salt by recrystallising the complex tances and displacement parameters. The nitrate anion
from acetonitrile in the presence of few drops of conc. containing N(60), the acetonitrile and two water solvent
nitric acid. molecules, namely O(90) and O(95) were found to be disor-
dered over two sites. Two split positions with restrained
2.4. Preparation of bis(pyridine–terpyridine) cobalt(III)
perchlorate, [Co(pytpy)2](ClO4)3, (2) Table 1
Crystallographic data for 1 Æ MeCN Æ 4H2O and 2 Æ 2.5H2O
The cobalt(II) complex prepared as above was dissolved Parameters 1 2
in methanol and 5 mL of dilute perchloric acid was added.
Chem formula C42H39CoN11O10 C40H35Cl5CoN8O22.5
The solution was taken in a 250 ml beaker and chlorine gas Fw 916.77 1223.94
was bubbled through the solution for 15 min. The solution Space group P-1, No. 2 P-1, No. 2
turned deep yellow indicating the oxidation of cobalt(II) to a, Å 10.7034(6) 10.8231(5)
cobalt(III). After the completion of oxidation, the solution b, Å 12.8391(7) 18.5198(9)
c, Å 15.4751(8) 25.307(2)
was concentrated by heating over a water bath. The
a, deg 100.651(4) 76.869(4)
required cobalt(III) complex separated as yellow crystals b, deg 95.462(4) 88.281(4)
on the addition of ether with yield of 75%. (Found C, c, deg 98.627(4) 76.948(4)
48.98; H, 2.43; N, 11.42; Co, 5.95, [Co(C20H14N4)2] V, Å 2049.7(2) 4811.1(5)
(ClO4)3, 2 requires C, 49.03; H, 2.86; N, 11.44 and Co, Z 2 4
T, K 100(2) 100(2)
6.03%). The sample was recrystallized from acetonitrile–
q calcd, g cm3 1.485 1.690
methanol mixture by slow evaporation of the solvent. Sin- Refl. collected/2Hmax 37895/58.0 47387/50.0
gle crystal suitable for X-ray analysis was obtained by ether Unique refl./I > 2r(I) 10 882/8678 16 577/11387
diffusion into acetonitrile solution of the complex. (Found No. of params/restr. 621/20 1478/919
C, 38.98; H, 2.78; N, 9.24; Co, 4.78, [Co(C20H15N4)2] l, cm1/k, Å 4.94/0.71073 7.29/0.71073
R1a/goodness of fit b 0.0554/1.054 0.0679/1.021
(ClO4)5 Æ 2.5 H2O requires C, 39.21; H, 2.86; N, 9.15 and
wR2c (I > 2r (I)) 0.1355 0.1629
Co, 4.82%). IR (KBr phase, cm1): 3010 s, 1615 s, 1092 Residual density, eÅ3 +0.78/0.94 +2.82/1.15
br, 740 w, 412 w (s, strong; w, weak; br, broad). 1H a P P
Observation criterion: I > 2r(I). R1 = jjF0j  jFcjj/ jF0j.
NMR (DMSO – d6) d 9.90 (s, 2H); d 9.21 (d, 2H); d 9.12 b P 2 2 2 1=2
GooF ¼ ½ ½wðF 0  F c Þ =ðn  pÞ .
P P
(d, 2H); d 8.61 (d, 2H); d 8.40 (t, 2H); d 7.59 (d, 2H); d c
wR2 ¼ ½ ½wðF 20  F 2c Þ2 = ½wðF 20 Þ2 1=2 where w ¼ 1=r2 ðF 20 Þ þ ðaP Þ2 þ
2 2
7.49 (t, 2H) (s, singlet; d, doublet; t, triplet). bP ; P ¼ ðF 0 þ 2F c Þ=3.
 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443
bond lengths and angles were refined by giving equal ther-
mal displacement parameters to corresponding atoms.
CCDC reference number 605985; see http://www.ccdc.
cam.ac.uk/deposit for crystallographic data in CIF or
other electronic format. For compound 2, the protonated
pyridine functions of the ligands were treated the same
way after density peaks for H positions were localized from
the difference map and hydrogen bonds to water molecules
and perchlorate anions indicated their protonation. Hydro-
gen position for the water molecules could not be reliably
determined. Most of the perchlorate anions were found
to be disordered. Anions containing Cl(21) and Cl(22)
are disordered next to a crystallographic inversion center
and an occupation factor of 0.5 was therefore assigned
for the two sites. Anions containing Cl(23)/Cl(53) and
Cl(25)/Cl(55) were pair wise split with occupation ratios
of 0.67/0.33 and 0.73/0.27 respectively. Cl–O and O–O dis-
tances were restrained to be equal within errors using the
SADI instruction in ShelXL. EADP was used to define
equal thermal displacement parameters to the correspond-
ing disorderd atoms. A disordered pyridine ring was also
split. The split components containing N(152) and
N(162) were refined with an occupation ratio of 0.59/6.41
and equal displacement parameters for corresponding
atoms. CCDC reference number 295743; see http://
www.ccdc.cam.ac.uk/deposit for crystallographic data in
CIF or other electronic format.
2.6. DNA binding and cleavage experiments
Concentrated stock solutions of metal complex was pre-
pared by dissolving the complex in DMSO:acetonitrile
(1:100) and diluted suitably with buffer to required con-
centrations for all the experiments. Absorption spectral
titration experiments were carried out with both the com-
plexes by maintaining constant concentration of the com-
plex (20 lM) while varying the nucleic acid concentration
(20–300 lM). Equal solution of DNA was added to both
complex solution and reference solution to eliminate the
absorbance of DNA itself. The intrinsic binding constant
was estimated using the following equation [43].
½DNA=ðea  ef Þ ¼ ½DNA=ðeb  ef Þ þ 1=K b ðeb  ef Þ;
where, ea, ef, and eb correspond to Aobsd/[Co], the extinc-
tion coefficient for the free cobalt complex, and the extinc-
tion coefficient for the cobalt complex in the fully bound
form, respectively. A plot of [DNA]/(ea  ef) vs [DNA],
gives Kb, the intrinsic binding constant as the ratio of the
slope to the intercept. For emission intensity measure-
ments, the tris-buffer was used as a blank to make prelimin-
ary adjustments. The excitation wavelength was fixed at
500 nm and the emission intensity was measured at
602 nm. DNA was pretreated with ethidium bromide in
the ratio [DNA]/[EB] = 2.5 for 30 min at room tempera-
ture. The effect on emission intensity was measured by
varying the concentration of complexes from 0 to 30 lM.
Viscosity measurements were carried out using an Ubber-
437
hold viscometer, immersed in a thermostated water-bath
maintained at 25 ± 1 C. DNA samples approximately
200 base pairs in average length were prepared by sonica-
tion to minimize complexities arising from DNA flexibility.
Data are presented as (g/g0)1/3 vs the concentration of co-
balt complex, [44] where g is the viscosity of DNA in the
presence of complex and g0 is the viscosity of DNA alone.
Viscosity values were calculated from the observed flow
time of DNA-containing solutions (t > 100 s) corrected
for the flow time of buffer alone (t0), using equation
g = (t  t0)/t0 [45]. For the gel electrophoresis experiments,
supercoiled pBR 322 DNA (100 lM) was treated with 50
and 100 lM complex 1 and complex 2 independently in
50 lM tris buffer, pH 7.2 and the solutions were irradiated
with a UV lamp (350 nm, 10 W). One set of solution also
contained 0.005 mL of 100 lM sodium azide. The samples
were analyzed by electrophoresis for 2.5 h at 40 V on a
0.8% agarose gel in tris–Boric acid–EDTA buffer, pH 7.2.
The gel was visualized by photographing the fluorescence
of intercalated ethidium bromide under a UV illuminator.
3. Results and discussion
3.1. Synthesis and characterization of the complexes
Complex 1 was prepared by direct reaction of cobalt(II)
chloride and the ligand in 1:2 molar ratio of metal to ligand,
using methanol as the reaction medium. The corresponding
cobalt(III) complex 2 was obtained by oxidation of the
cobalt(II) complex by passing chlorine gas. The complexes
were synthesized in good yield and isolated as perchlorate
salts. Fig. 1 depicts the electronic spectra of (1) cobalt(II)
complex and (2) cobalt(III) complex in the spectral range
of 200–500 nm. The electronic spectra of complexes 1 and
2 are dominated by ligand centered transitions, as observed
in the case of polypyridyl complexes of cobalt. It is observed
that the free ligand exhibits bands at 240, 277 and 317 nm.
10
2
10 ε (M cm )
-1
1
-1
5
-4
0
250 300 350 400
Wavelength (nm)
Fig. 1. Absorption spectra (1) of complex 1 and (2) 2 in acetonitrile in the
spectral range 200–400 nm.
438 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443

Fig. 2. ORTEP drawing of [Co(pytpy)2]3+, complex 2. Hydrogen atoms are omitted for clarity.

These are due to the aromatic nature of terpyridine ligand. complex 1 the two pyridine nitrogens are not protonated
In case of the cobalt(II) complex, the 277 nm band of the and hence crystallized as a dication. Both complexes 1
ligand splits into two centering at 277 and 283 nm. In the and 2 belong to triclinic system. Unlike in the case of com-
case of the cobalt(III) complex this band is observed at plex 1 the unit cell of complex 2 consists of two crystallo-
281 nm. The 317 nm band of ligand is shifted to 329 nm graphically independent cations. The perspective view of
for cobalt(II) and in case of cobalt(III), besides the band complex 2 is shown in Fig. 2. In both the complexes the
at 317 nm, two additional bands are observed at 346 and two tridentate ligands are coordinated to the central metal
361 nm. The NMR spectrum of the complex 2 showed well ion to give a CoN6 type six coordinate complex with merid-
resolved peaks due to low spin, diamagnetic nature of the ional configuration. The important bond lengths and bond
cobalt(III) complex. Complex 1 and 2 show reversible elec-
trochemical wave in DMSO attributable to Co(II)/Co(III)
couple with E1/2 0.159 and 0.164 V respectively vs SCE. Table 3
The redox potentials of cobalt complexes coordinated Important bond distances and bond angles of complex 2
with bpy or phen have been reported to be in the range of Bond distances (Å)
0.037–0.173 V [46]. Co(1)–N(42) 1.857(4) Co(2)–N(142) 1.848(5)
Co(1)–N(12) 1.860(4) Co(2)–N(112) 1.857(4)
Co(1)–N(31) 1.939(4) Co(2)–N(131) 1.934(5)
3.2. Description of the structure Co(1)–N(48) 1.941(4) Co(2)–N(118) 1.938(5)
Co(1)–N(18) 1.943(4) Co(2)–N(101) 1.941(5)
The complexes 1 and 2 have been characterized from a Co(1)–N(1) 1.946(4) Co(2)–N(148) 1.945(4)
single-crystal X-ray diffraction study. The complex 2 crys- C(13)–N(18) 1.373(7) N(101)–C(106) 1.372(7)
C(17)–N(18) 1.333(7) N(101)–C(102) 1.346(7)
tallized as [Co(LH)2](ClO4)5(LH = pytpy) due to proton- C(49)–C(50) 1.379(9) C(107)–C(108) 1.377(8)
ation of the two pyridine nitrogens whereas in the case of C(49)–C(54) 1.408(8) C(108)–C(109) 1.406(8)
C(51)–N(52) 1.333(9) C(117)–N(118) 1.332(7)
N(52)–C(53) 1.329(9) N(122)–C(123) 1.335(8)
Table 2
Important bond distances and bond angles of complex 1 Bond angles ()
N(42)–Co(1)–N(12) 179.30(19) N(142)–Co(2)–N(112) 179.3(2)
Bond distances (Å)
N(42)–Co(1)–N(31) 82.12(18) N(142)–Co(2)–N(131) 82.8(2)
Co–N(42) 1.8932 (0.0018)
N(12)–Co(1)–N(31) 97.66(18) N(112)–Co(2)–N(131) 97.5(2)
Co–N(12) 1.9009 (0.0018)
N(42)–Co(1)–N(48) 82.62(18) N(142)–Co(2)–N(118) 96.61(19)
Co–N(31) 2.0640 (0.0021)
N(12)–Co(1)–N(48) 97.60(18) N(112)–Co(2)–N(118) 82.79(19)
Co–N(48) 2.0653 (0.0021)
N(31)–Co(1)–N(48) 164.72(18) N(131)–Co(2)–N(118) 90.85(19)
Co–N(18) 2.0827 (0.0020)
N(42)–Co(1)–N(18) 96.63(19) N(142)–Co(2)–N(101) 98.0(2)
Co–N(1) 2.0832 (0.0019)
N(12)–Co(1)–N(18) 82.70(19) N(112)–Co(2)–N(101) 82.65(19)
Bond angles () N(31)–Co(1)–N(18) 88.36(18) N(131)–Co(2)–N(101) 90.9(2)
Co–N42–N12 179.52 (0.08) N(48)–Co(1)–N(18) 92.67(18) N(118)–Co(2)–N(101) 165.43(19)
Co–N31–N42 80.28 (0.08) N(42)–Co(1)–N(1) 98.38(19) N(142)–Co(2)–N(148) 82.97(19)
Co–N31–N12 99.24 (0.08) N(12)–Co(1)–N(1) 82.29(18) N(112)–Co(2)–N(148) 96.72(19)
Co–N48–N42 79.73 (0.08) N(31)–Co(1)–N(1) 94.85(18) N(131)–Co(2)–N(148) 165.7(2)
Co–N48–N12 100.75 (0.08) N(48)–Co(1)–N(1) 88.09(18) N(118)–Co(2)–N(148) 91.68(18)
Co–N48–M31 160.00 (0.08) N(18)–Co(1)–N(1) 164.95(18) N(101)–Co(2)–N(148) 90.21(18)
Co–N18–N42 99.88 (0.08) C(50)–C(49)–C(54) 118.7(5) C(150)–C(149)–C(154) 120.0(8)
Co–N18–N12 80.04 (0.08) C(50)–C(49)–C(39) 121.5(5) C(150)–C(149)–C(139) 128.4(10)
Co–N18–N31 89.13 (0.08) C(54)–C(49)–C(39) 119.8(5) C(154)–C(149)–C(139) 111.7(9)
Co–N1–N48 91.24 (0.08) C(49)–C(50)–C(51) 120.1(6) C(151)–C(150)–C(149) 120.6(12)
Co–N1–N18 159.90 (0.07) N(52)–C(51)–C(50) 119.1(7) N(152)–C(151)–C(150) 118.4(11)
Co–N1–N42 100.10 (0.07) C(53)–N(52)–C(51) 123.1(6) C(153)–N(152)–C(151) 124.0(9)
Co–N1–N12 79.96 (0.08) N(52)–C(53)–C(54) 120.1(6) N(152)–C(153)–C(154) 121.2(12)
Co–N1–N31 92.08 (0.08) C(53)–C(54)–C(49) 118.9(6) C(153)–C(154)–C(149) 115.8(12)
 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443 439

angles for the complexes 1 and 2 are given in Tables 2 and 3
respectively. From Table 2 it can be seen that the central
nitrogen of the two pyridine rings N12 and N48 are closer
to the Co(II) ion compared to other four nitrogens. The
other two nitrogens of each ligand are almost equidistant
from the central metal ion. However, it is interesting to
note that the bond distance between the Co(II) and two ter-
minal nitrogens of one ligand (N1 and N18) are longer
than that between Co(II) and the two terminal nitrogens
of the other ligand (N31 and N48) by around 0.017 A.
From Table 3, it is observed that the bond lengths of com-
plex 2 are typical of those observed for low spin cobalt(III)
complexes. The central nitrogen of the two terpyridine
moieties N12 and N42 are closer to the central metal ion
compared to the other two nitrogens of each ligand by
almost 0.08 Å, as observed in the case of complex 1. On
the other hand, the other two nitrogens of each ligand
are almost equidistant from the central metal ion, the
Co(1)–N(1), Co(1)–N(18), Co(1)–N(31) and Co(1)–N(48)
bond lengths being 1.946(4), 1.943(4), 1.939(4) and
1.941(4) respectively. It is to be noted that in the case of
complex 1 these four bonds are not of equal length. There
is considerable deviation in the N(12)–Co(1)–N(18).
N(12)–Co(1)–N(1), N(42)–Co(1)–N(48) and N(42)–Co(1)–
N(31) bond angles from the value expected for a perfect
octahedral coordination (complex 2). All these bond angles
are around 82.70. Only the N(42)–Co(1)–N(12) angle is
closer to the ideal value of 180.
4. DNA binding experiments
4.1. Electronic absorption spectra
Absorption spectroscopy is one of the most useful tech-
niques to study the binding of any drug to DNA [47]. Inter-
calative binding of complex with DNA generally results in
hypochromism and bathochromism. The extent of hypo-
chromism generally indicates the intercalative binding
strength. The electronic spectra of both complexes 1 and
2 showed hypochromism in the presence of DNA. The elec-
tronic spectra of the complex 2 in presence and absence of
CT DNA are illustrated in Fig. 3. As expected, the spectra
clearly show a decrease in absorption intensity at 280, 317,
346 and 360 nm. The hypochromism observed for all the
bands are accompanied by a small red shift of around
5 nm. Since the bands at 317, 346 and 360 nm are not so
intense, the band at 280 nm arising from ligand-based tran-
sitions was monitored for absorption titration experiment
with CT DNA. If the complex interacts with DNA accord-
ing to intercalation model, the chromophore would bury

Fig. 3. Spectral changes of complex 2 in Tris-buffer (pH 7.2) in the absence and in the presence of CT DNA. Arrows indicate the direction of change upon
increasing DNA concentration. Inset: Plot of [DNA]/(ea  ef) vs [DNA] · 106 M.
440 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443
itself in the stack of DNA bases, leading to spectral
changes caused by the interaction of the electronic states
of the complex with that of the DNA bases. After the com-
plexes intercalate with the base pairs of DNA, the p* orbi-
tal of the intercalated ligands of the complexes could
couple with p orbitals of the base pairs, thus, decreasing
the p–p* transition energies. The intercalative binding of
complex 2 to DNA is not surprising since the ligand pos-
sesses planar pyridine ring, which can easily intercalate
between DNA base pairs. Phen complexes of cobalt(III)
too have been shown to bind DNA intercalatively [48].
From the [DNA]/(ea  ef) vs [DNA] plots, the binding con-
stants, Kb for complexes 1 and 2 were estimated to be
(1.97 ± 0.15) · 104 M1 and (2.7 ± 0.2) · 104 M1 respec-
tively. It is of interest to note that cobalt(II) complex of
another tridentate ligand bzimpy shows surface binding
to DNA with a binding constant of (1.6 ± 0.3) · 105 M1
[49]. The Kb for complex 2 reported here is much smaller
Fig. 4. Emission spectrum of EB–DNA ([DNA]/[EB] = 2.5) in the
absence and presence of various concentrations of complex 2 (0–30 lM).
Arrows indicate the direction of change upon increasing concentration
of 2.
compared to that reported for classical intercalators like
ethidium bromide. It is not surprising, since in complex 2
one can expect only the pyridine ring to intercalate unlike
in the case of ethidium bromide, which has an extended
ring structure and hence can have a better p overlap with
DNA bases. Even though the peripheral pyridine ring is
co-planar with the terpy unit in the crystalline form of
the complex (as seen from the crystal structure of 1 and
2), in solution because of free rotation of pyridine ring with
respect to the terpy unit the whole ligand is unable to
intercalate.
4.2. Competitive binding between ethidium bromide (EB)
and complexes 1 and 2
EB is one of the most sensitive fluorescence probes that
can bind DNA. The interaction pattern of EB with DNA
belongs to the intercalation model, as reported in the liter-
ature [50]. The fluorescence intensity of EB tends to
increase after interacting with DNA due to its burial in
the hydrophobic region of DNA. If complex 1 and 2 inter-
calate into DNA, it will lead to decrease in the binding sites
of DNA available for EB, and hence quenching of fluores-
cence intensity of the EB–DNA mixture. Groove binding
of the drug also under some circumstances may lead to
quenching of EB emission. Fig. 4 shows the emission inten-
sity of EB intercalated to DNA in the presence of increas-
ing amounts of complex 2. Addition of complex 2 leads to
gradual decrease in the fluorescence intensity. The fluores-
cence intensity finally approaches that of EB in the absence
of DNA. Since the DNA binding constant of complex 2 is
much smaller than that of EB, concentration of complex 2
needed to displace EB from its DNA binding site is large.
Similar kind of spectral changes were observed for complex
1 also. Although these results do not rule out groove bind-
ing of complexes 1 and 2, because of the planarity of the
ligand one can expect intercalation of the ligand in between
the base pairs displacing the EB from binding sites.
4.3. Cyclic voltammetry
The cyclic voltammogram of 0.5 mM complex 2 in the
absence (a) and presence (b) of 2.5 mM DNA is shown in
Fig. 5. The complex has been found to show an almost
reversible electrochemical wave in the buffer with E1/2 of
0.164 V vs SCE. Presence of 2.5 mM DNA causes a consid-
erable decrease in the voltammetric current of the redox
wave with a slight shift in E1/2 (E1/2 = 0.159 V) to less neg-
ative potential (Fig. 5b). This electrochemical behavior is
similar to that observed for [Co(phen)2(IP)]3+ [18]. The
drop of the voltammetric currents in the presence of
DNA may be attributed to slow diffusion of the metal com-
plex bound to CT DNA. This in turn indicates the extent
of binding affinity of the complex to DNA. The net shift
in E1/2 can be used to estimate the ratio of equilibrium con-
stants for the binding of the 3+ and 2+ ions to DNA.
Scheme 1 represents a Nernstian electron transfer to a
 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443
Fig. 5. Cyclic voltammogram of 0.5 mM complex 2 in the absence (a) and presence of 2.5 mM DNA (b).
- Ef0
M3+ + e M2+
K3+
K2+
M3+-DNA + e
-
M2+ - DNA
Scheme 1.
system in which both the oxidized and reduced forms are
associated with a third species in solution (DNA).
In this scheme M3+ and M2+ represent the oxidized and
reduced forms of the metal complex. E0f and E0b are the for-
mal potentials of the 3+/2+ couple, in the free and bound
forms respectively. K3+ and K2+ are the corresponding
equilibrium constants for binding of the 3+ and 2+ species
to DNA. The Nernst equation for the reversible redox
reactions of the free and bound species and the correspond-
ing equilibrium constants for binding of each oxidation
state to DNA for an one electron redox process is given
[46] as follows:
E0b  E0f ¼ 0:059 logðK2þ =K3þ Þ
Substituting the experimental values in the above equation
for the redox reaction of 2 in presence and absence of
DNA, the ratio of equilibrium constants for the binding
of the 2+ and 3+ ions to DNA was estimated to be 0.82.
This value is in reasonably good agreement with the K2+/
K3+ value of 0.73 obtained from spectroscopic titration.
The value of K2+/K3+ indicates that the complex in its
+3 oxidation state has slightly higher affinity to DNA com-
441
pared to the complex in its +2 oxidation state. This is as
expected since both complexes have same ligand which is
involved in intercalation. A marginal increase in the affinity
of complex 2 for DNA, evident from their binding constant
values may be attributed to the increased electrostatic
charge on complex 2.
Eb0
4.4. Viscosity measurements
Further clarification on the interaction between two
complexes and DNA was carried out by viscosity measure-
ments. Intercalation of the complexes to DNA is known to
cause a significant increase in the viscosity of a DNA solu-
tion due to the separation of the base pairs at the interca-
lation site and, hence, an increase in the overall DNA
molecular length. In order to confirm the binding mode
of the complex with DNA, the effect of complex 1 and 2
on the viscosity of CT DNA solution was studied. The
effect of viscosity of rod like DNA (200 lM) in presence
of complexes 1 and 2 (20–200 lM) are shown in Fig. 6.
As seen from the figure, the viscosity of the mixture
increases with increasing addition of the complex, till the
ratio of complex to DNA reaches 0.65 suggesting both
the complexes assume intercalative mode of binding by
the pytpy ligand as reported for [Co(phen)2(imp)]3+ [51].
This result also parallels the hypochromism, spectral red
shift and quenching of emission intensity of EB in fluores-
cence measurements.
4.5. Photocleavage of pBR 322 DNA
The DNA cleaving ability of both the complexes were
analyzed by determining the ability of the complexes to
442 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443

1.9 Table 4
Intensities of Form I and Form II
1.8 Lanes
1 2 3 4 5 6 7
1.7
% Form I 65 61 36 14 11 80 74
1.6 % Form II 35 39 64 86 89 20 26
0 1/3

1.5
( η /η )

cleavage of the plasmid, DNA incubated with the cobal-

1.4
1.3
1.2
1.1
0.0 0.2 0.4 0.6 0.8
1/R (R=[DNA]/[complex])
Fig. 6. The change in viscosity of CT DNA (200 lM) in presence of
increasing amounts of complex 1 (d) and complex 2 (j).
convert the supercoiled DNA (SC) to nicked circular (NC)
and linear form. Fig. 7 shows the gel electrophoretic sepa-
ration of plasmid pBR 322 DNA after incubation with the
two complexes. It is seen from the gel that neither irradia-
tion of DNA at 350 nm without cobalt complexes 1 or 2
(lane 1), nor incubation of cobalt(III) and cobalt(II) com-
plexes with DNA in dark (lane 2 and 6) yield significant
strand scission. On the other hand, when plasmid DNA
is irradiated for 30 min in the presence of 50 lM and
100 lM of cobalt(III) complex cleavage of supercoiled
Form I to open circular Form II is observed in a concentra-
tion dependent manner (lane 3 and 4). The intensities of the
Form I and Form II in the gel has been quantified and the
percentage of the two forms of DNA in different lanes are
given in Table 4. It can be clearly seen from (Table 3 and
Fig. 7) that when DNA is photolysed in the presence of
100 lM of cobalt(III) complex almost all the supercoiled
Form I gets cleaved to nicked Form II. Irradiation of
DNA in the presence of the cobalt(II) analogue on the
other hand did not bring about any appreciable DNA
cleavage as evident from lane 7 of Fig. 7. In order to estab-
lish the reactive species responsible for the photoactivated
t(III) complex was irradiated in the presence of singlet oxy-
gen scavenger sodium azide. Lane 5 of Fig. 7 clearly shows
that azide had no marked influence on the photocleavage
of DNA in the presence of the cobalt(III) complex. This
clearly rules out any role for singlet oxygen in the photoc-
leavage of DNA. Hence it is clear that the photoexcited
1.0 state of the cobalt(III) complex is responsible for the
observed DNA cleavage. Photoexcited states of cobalt(III)
complexes have previously been shown to bring about pro-
tein cleavage [52] and hence it is not surprising that com-
plex 2 is able to bring about DNA cleavage under
photolytic conditions.
5. Conclusion
Two cobalt complexes derived from pyridine terpyridine
have been synthesized and characterized spectroscopically.
The X-ray crystal structures of both the complexes have
been solved. Both cobalt(II) and cobalt(III) complexes
show CoN6 coordination with meridional configuration.
Both these complexes bind to CT DNA intercalatively indi-
cating that the DNA binding mode is dictated by the coor-
dinated ligand. The intrinsic DNA binding constant of
cobalt(II) and cobalt(III) complexes were found to be
1.97 ± 0.15 · 104 M1 and 2.7 ± 0.20 · 104 M1 respec-
tively indicating cobalt(III) complex is stronger binder
when compared to the cobalt(II) complex. This clearly
indicates that electrostatic considerations also play signifi-
cant role in the binding of these complexes to DNA. The
DNA cleavage studies show that the cobalt(III) complex
has the ability to cleave DNA under photolytic condition
while the cobalt(II) analogue does not have any DNA
cleaving ability.
6. Abbreviations

bzimpy 2,6- bis(benzimidazol-2-yl)pyridine

1 2 3 4 5 6 7 CT DNA calf thymus DNA
EB ethidium bromide
Form II ESI eionization
DMSO dimethyl sulphoxide
Form I IP imidazo[4,5-f][1,10]phenanthroline
phen phenanthroline
Fig. 7. Gel electrophoresis diagram showing the cleavage of SC plasmid pytpy pyridine terpyridine
DNA (150 lg) in Tris buffer (pH 7.2) containing DMSO:CH3CN (1:9 v/v).
SCE saturated calomel electrode
Lane 1, DNA control, lane 2, DNA + 50 lM 2 in dark, Lane 3,
DNA + 50 lM 2 in light, Lane 4, DNA + 100 lM 2 in light, Lane 5, TBE tris–boric acid–EDTA
DNA + 100 lM 2 + 0.005 mL of 100 lM sodium azide in light, Lane 6, TBAP tetrabutyl ammonium perchlorate
DNA + 100 lM 1 in dark, Lane 7, DNA + 100 lM 1 in light. terpy terpyridine
 R. Indumathy et al. / Journal of Inorganic Biochemistry 101 (2007) 434–443 443

Tris tris(hydroxymethyl)methylamine [24] K. Naing, M. Takashami, M. Tanicuchi, A. Yamagishi, Inorg. Chem.

34 (1995) 350–356.
SC supercoiled [25] D.C. Ware, P.J. Brothers, G.R. Clark, W.A. Denny, B.D. Palmer,
W.R. Wilson, Dalton Trans. (2000) 925–932.
[26] P.J. Blower, J.R. Dilworth, R.I. Maurer, G.D. Mullen, C.A.
Appendix A. Supplementary data Reynolds, Y. Zheng, J. Inorg. Biochem. 85 (2001) 15–22.
[27] P.D. Bonnitcha, M.D. Hall, C.K. Underwood, G.J. Foran, M.
Zhang, P.J. Beale, T.W. Hambley, J. Inorg. Biochem. 100 (2006)
Supplementary data associated with this article can 963–971.
be found, in the online version, at doi:10.1016/j.jinorgbio. [28] L. Jin, P. Yang, Polyhedron 16 (1997) 3395–3398.
2006.11.002. [29] S. Arounaguiri, B.G. Maiya, Inorg. Chem. 35 (1996) 4267–4270.
[30] R. Bonnett, Chem. Soc. Rev. 24 (1995) 19–33.
References [31] R. Ackroyd, C. Kelty, N. Brown, M. Reed, Photochem. Photobiol.
74 (2001) 656–669.
[32] M.C. De Rosa, R.J. Crutchley, Coord. Chem. Rev. 233/4 (2002) 351–
[1] M.A. J-Hakeen, S.S. Sommer, Anal. Biochem. 163 (1987) 433–439.
371.
[2] C.V. Kumar, H. Emma, J. Chem. Soc., Chem. Commun. (1992) 470–
[33] J.L. Sessler, G. Hemmi, T.D. Mody, T. Murai, A. Burrell, S.W.
472.
Young, Acc. Chem. Res. 27 (1994) 43–50.
[3] D.Z. Yang, J.T. Strode, H.P. Spielmann, A.J. Wang, T.G. Burake,
[34] R.K.R. Jetti, A. Nagia, F. Xue, T.C.W. Mak, J. Chem. Soc. Chem
J. Am. Chem. Soc. 120 (1998) 2979–2980.
Commun. (2001) 919–920.
[4] R.E. Hemin, J.A. Yao, J.K. Barton, Inorg. Chem. 38 (1999) 174–189.
[35] Z.C. Watson, N. Bampos, J.K.M. Sanders, New. J. Chem. (1998)
[5] N.Y. Sardesai, K. Zimmermann, J.K. Barton, J. Am. Chem. Soc. 116
1135–1138.
(1994) 7502–7508.
[36] E.C. Constable, C.E. Housecroft, M. Neuburger, D. Phillips, P.R.
[6] J.K. Barton, A.L. Raphael, Proc. Natl. Acad. Sci. USA (1985) 6460–
Raithby, E. Schofield, E. Sparr, D.A. Tocher, M. Zehnder, Y.
6464.
Zimmermann, J. Chem. Soc. Dalton Trans. (2000) 2219–2228.
[7] Y.G. Gae, M. Sriram, A.H. Wang, Nucleic Acids Res. 21 (1993)
[37] R. Buchner, C.T. Cunningham, J.S. Field, R.J. Haines, D.R.
4093–4101.
McMillin, G.C. Summerton, J. Chem. Soc. Dalton Trans. (1999)
[8] J.K. Barton, S.J. Lippard, in: T.G. Siro (Ed.), Metal Ions in Biology,
711–717.
vol. 1, Wiley, New York, 1980, p. 31.
[38] G.W.V. Cave, C.L. Raston, J.Chem. Soc., Perkin Trans. I (2001)
[9] R.M.K. Dale, D.C. Livingston, D.C. Ward, Proc. Natl. Acad. Sci.
3258–3264.
USA 70 (1973) 2238–2242.
[39] J. Murmur, J. Mol. Biol. 3 (1961) 208–218.
[10] C.J. Burrows, S.E. Rokita, Acc. Chem. Res. 27 (1994) 295–301.
[40] M.F. Reichmann, S.A. Rice, C.A. Thomas, P. Doty, J. Am. Chem.
[11] A.M. Pyle, J.K. Barton, S.J. Lippard (Eds.), Progress in Inorganic
Soc. 76 (1954) 3047–3053.
Chemistry, vol. 38, Wiley, New York, 1990, pp. 413–475.
[41] ShelXTL V. 5, Siemens Analytical X-Ray Instruments, Inc.,
[12] A.R. Banerjee, J.A. Jaeger, D.H. Turner, Biochemistry 32 (1993) 153–
Madison, Wisconsin, USA, 1994.
163.
[42] ShelXL97, G.M. Sheldrick, University of Göttingen, 1997.
[13] V.G. Vaidyanathan, B.U. Nair, J. Inorg. Biochem. 95 (2003) 334–342.
[43] A.M. Pyle, J.P. Rehmann, R. Meshoyrer, C.V. Kumar, N.J. Turro,
[14] R. Vijayalakshmi, M. Kanthimathi, R. Parthasarthi, B.U. Nair,
J.K. Barton, J. Am. Chem. Soc. 111 (1989) 3051–3058.
Bull. Chem. Soc. Jpn. 78 (2005) 270–276.
[44] B. Chaires, N. Dattaguota, D.M. Crothers, Biochemistry 21 (1982)
[15] V.G. Vaidyanathan, B.U. Nair, J. Inorg. Biochem. 93 (2003) 271–276.
3933–3940.
[16] V. Uma, M. Kanthimathi, T. Weyhermuller, B.U. Nair, J. Inorg.
[45] S. Satyanarayana, J.C. Daborusak, J.B. Chaires, Biochemistry 32
Biochem. 99 (2005) 2299–2307.
(1993) 2573–2584.
[17] Q.L. Zhang, J.G. Liu, H. Xu, H. Li, J.Z. Liu, H. Zhou, L.H. Qu,
[46] M.T. Carter, M. Rodriguez, A.J. Bard, J. Am. Chem. Soc. 111 (1989)
L.N. Ji, Polyhedron 20 (2001) 3049–3055.
8901–8911.
[18] Q.L. Zhang, J.G. Liu, H. Chao, G.Q. Zue, L.N. Ji, J. Inorg. Biochem.
[47] T.M. Kelly, A.B. Tossi, D.J. McConnell, T.C. Strekas, Nucleic Acids
83 (2001) 49–55.
Res. 13 (1985) 6017–6034.
[19] V. Uma, M. Kanthimathi, J. Subramanian, B.U. Nair, Biochem.
[48] Q.L. Zhang, J.G. Liu, J. Liu, G.Q. Xue, H. Li, J.Z. Liu, L.H. Zhou,
Biophys. Acta 1760 (2006) 814–819.
H. Qu, L.N. Ji, J. Inorg. Biochem. 85 (2001) 291–296.
[20] V.G. Vaidyanathan, B.U. Nair, Dalton Trans. (2005) 2842–2848.
[49] V.G. Vaidyanathan, B.U. Nair, J. Inorg. Biochem. 94 (2003) 121–126.
[21] R. Vijayalakshmi, M. Kanthimathi, R. Parthasarathi, B.U. Nair,
[50] M.J. Waring, J. Mol. Biol. 13 (1965) 269–282.
Bioorg. Medicinal Chem. 14 (2006) 3300–3306.
[51] P.T. Selvi, M. Palaniandavar, Inorg. Chim. Acta 337 (2002) 420–428.
[22] M.J. Clarke, Coord. Chem. Rev. 236 (2003) 209–233, and references
[52] H.Y. Shrivastava, M. Kanthimathi, B.U. Nair, Biochim. Biophys.
therein.
Res. Commun. 265 (1999) 311–314.
[23] F.L. Liu, K.A. Meadows, D.R. McMillin, J. Am. Chem. Soc. 115
(1993) 6699–6704.