Polyhedron 49 (2013) 167–176

Contents lists available at SciVerse ScienceDirect

Polyhedron

journal homepage: www.elsevier .com/locate /poly

DNA binding, nuclease and superoxide scavenging activity studies onmononuclear cobalt complexes derived from tridentate ligands

Kaushik Ghosh ⇑, Varun Mohan, Pramod Kumar, Udai P. SinghDepartment of Chemistry, Indian Institute of Technology Roorkee, Roorkee 247667, Uttarakhand, India

a r t i c l e i n f o

Article history:Received 16 May 2012Accepted 16 September 2012Available online 26 September 2012

Keywords:Cobalt complexCrystal structureElectrochemistrySOD activityDNA bindingNuclease activity

0277-5387/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.poly.2012.09.025

⇑ Corresponding author. Fax: +91 1332 273560.E-mail address: ghoshfcy@iitr.ernet.in (K. Ghosh).

a b s t r a c t

A new family of mononuclear cobalt complexes was synthesized using tridentate ligands Pyimpy(Pyimpy = 1-phenyl-1-(pyridin-2-yl)-2-(pyridin-2-ylmethylene)hydrazine) and PampH (PampH =N0-phenyl-N0-(pyridin-2-yl)picolinohydrazide). The complexes [Co(Pyimpy)Cl2] (1a), [Co(Pyimpy)2](ClO4)2

(1b), [Co(Pamp)Cl2] (2a) and [Co(Pamp)2](ClO4) (2b) were characterized by elemental analysis, UV–Vis,IR, NMR, ESI-MS spectral studies and conductivity measurements. Molecular structure of 1a was deter-mined by X-ray crystallography. Redox properties of the complexes were examined by cyclic voltamme-try. Superoxide dismutase (SOD) activity of the complexes was assayed by xanthine/xanthine oxidase/nitroblue tetrazolium assay. We investigated the DNA binding properties of these complexes by absorp-tion spectral, fluorescence quenching and circular dichroism spectral studies. Nuclease activity and theirmechanism were also investigated.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

In the recent years there has been considerable interest in theinteraction of transition metal complexes with DNA and nucleaseactivity studies because of their applications in probing structuralvariations in nucleic acids [1], identifying binding sites of DNA li-gands [2], designing artificial nucleases [3–5] and serving as che-motherapeutic agent in cancer research [6]. In this regard, amongthe first row transition elements, copper and iron got much atten-tion. Copper possesses high affinity for nucleobases and exhibitsbiologically accessible redox properties [7]. Iron-bleomycin causesDNA damage and clinically used for cancer treatment [8,9]. Severaliron complexes are known to exhibit DNA interaction and nucleaseactivity [10–13]. On the other-hand cobalt, a constituent of vitaminB12 and one of the essential elements in biosystem [14] receivedless attention [15–17]. Cobalt polypyridyl complexes were investi-gated for DNA interaction and nuclease activity studies [18–28]however, reports for oxidative cleavages are few [20,29–33].

In the present study, we have explored the coordination chem-istry of cobalt because it possesses several interesting properties.First, usual oxidation states of +2 and +3 in cobalt complexes giverise to kinetically labile and kinetically inert complexes, respec-tively [34–36]. Second, cobalt(III) is a strong Lewis acid and exhib-its ligand exchange properties. Third, Co(III) has important role incatalytic hydrolysis of phosphate esters [37]. Hettich and cowork-ers reported that Co(III) complexes are known to be most effective

ll rights reserved.

catalysts for hydrolytic cleavage of amides, esters and phosphates[38]. Fourth, cobalt ion is present in natural nucleases [36,39–41]or sometimes required for the activity of nucleases [37,42–45].There are reports that substitution of zinc by cobalt in several nuc-leases may result the enhancement of the enzyme activity [41].Fifth, DNA cleaving properties of cobalt complexes are sometimessimilar to ruthenium complexes [15]. Moreover, antiproliferativeactivity of the cobalt complexes was determined in MCF-7 andMDA-MB-231 human breast cancer cells and very promising re-sults were obtained [46].

Hence, in the present report we describe the synthesis andcharacterization of cobalt complexes [Co(Pyimpy)Cl2] (1a),[Co(Pyimpy)2](ClO4)2 (1b), [Co(Pamp)Cl2] (2a), [Co(Pamp)2](ClO4)(2b) derived from the tridentate ligands Pyimpy and PampH (Hstands for dissociable proton) which are depicted in Scheme 1[47]. Both Pyimpy and PampH have two of the three donor centersin common which are pyridine nitrogens, however the third donoris an iminium nitrogen for Pyimpy whereas carboxamido nitrogenfor PampH. The complexes were characterized by UV–Vis, IR, andNMR spectral studies. Molecular structure of 1a was determinedby X-ray crystallography. Redox property of the complexes wasscrutinized and superoxide dismutase activity was examined byxanthine/xanthine oxidase/nitroblue tetrazolium assay. DNA inter-action was investigated by UV–Vis, fluorescence and circulardichroism spectral studies. We have also investigated the nucleaseactivity of our complexes and results of mechanistic studies will bedescribed in this report. Recently we have reported that the car-boxamido functionality plays a crucial role to impart the DNAcleaving property to the manganese complexes [47]. Moreover zinc

N

NN N

N

NHN N

Pyimpy PampH

O

Scheme 1. Schematic drawing of ligands.

168 K. Ghosh et al. / Polyhedron 49 (2013) 167–176

complexes derived from Pyimpy were unable to perform DNAcleavage activity [48] whereas the copper complexes with thesame ligand were found to be efficient in DNA strand scission[49,50]. Necessity of a redox-active metal centre for superoxidescavenging as well as nuclease activity of these complexes willbe scrutinized in the light of our reported results.

2. Experimental

2.1. Materials

Analytical grade reagents Phenylhydrazine, 2-mercaptoethanol,hydrogenperoxide (S.D. Fine, Mumbai, India), sodium azide (SigmaAldrich, Steinheim, Germany), picolinic acid (Wilson Laboratories,Mumbai, India), 1-hydroxybenzotriazol, sodium perchloratemonohydrate (Himedia Laboratories Pvt. Ltd., Mumbai, India),dicyclohexylcarbodiimide (SRL, Mumbai, India), ethylenediamine-tetraacetic acid, cobalt chloride hexahydrate (Merck Ltd., Mumbai,India), pyridine-2-aldehyde, sodium hydride and 2-chloropyridine(Acros organics, USA) were used as obtained. The supercoiledpBR322 DNA and CT DNA were purchased from Bangalore Genei(India) and stored at 4 �C. Agarose (molecular biology grade) andethidium bromide were obtained from Sigma Aldrich. Xanthine, ni-tro blue tetrazolium (NBT) and catalase were obtained from Hime-dia and xanthine oxidase (XO) from bovine milk was purchasedfrom Sigma. Tris(hydroxymethyl)aminomethane–HCl (Tris–HCl)buffer and phosphate buffer were prepared in deionised water. Sol-vents used for spectroscopic studies were HPLC grade and purifiedby standard procedures before use.

2.2. Methods and instrumentation

Elemental analysis was carried out on Elemental model VarioEL-III. Infrared spectra were recorded as KBr pellets on a NicoletNEXUS Aligent 1100 FT-IR Spectrometer, using 50 scans and werereported in cm�1. Electronic spectra were recorded in methanoland phosphate buffer solution with Evolution 600, Thermo scien-tific, UV–Vis spectrophotometer using cuvettes of 1 cm path length.Fluorescence spectra were recorded by Varian Cary (Eclipse) fluo-rescence spectrophotometer. Circular dichroism (CD) spectra wererecorded on Chirascan circular dichroism spectrometer, Appliedphotophysics, UK. Molar conductivities were determined in DMFat 10�3 M at 25 �C with a Systronics 304 conductometer. 1H and13C spectra were recorded on Bruker AVANCE, 500.13 MHz spec-trometer, chemical shifts for 1H NMR spectra are related to internalstandard Me4Si for all residual protium in the deuterated solvents.The electrospray ionization (ESI) mass spectra of the complexeswere recorded as acetonitrile solutions in positive or negative ionmode with a Bruker MicroTOF-QII mass spectrometer. Magneticsusceptibilities were determined at 294 K with vibrating sampleMagnetometer model 155, using nickel as a standard. In order toinvestigate the redox behavior of the synthesized complexes, cyclicvoltammetric studies were performed on a CH-600 electroanalyser

in acetonitrile with 0.1 M tetrabutylammonium perchlorate (TBAP)as supporting electrolyte. The working electrode, reference elec-trode and auxiliary electrode were glassy carbon, Ag/AgCl and Ptwire electrode, respectively. The concentration of the compoundswas in the order of 10�3 M. The ferrocene/ferrocenium couple oc-curs at E1/2 = +0.42 (75) V versus Ag/AgCl under the same experi-mental conditions.

2.3. Synthesis of ligands and cobalt complexes

The ligands Pyimpy and PampH were synthesized by the re-ported procedures [47]. Complex [Co(Pyimpy)Cl2] (1a) was synthe-sized by the reaction of CoCl2�6H2O and ligand Pyimpy in 1:1 ratio.On the other hand, the same reaction when carried out with metalto ligand ratio 1:2 afforded [Co(Pyimpy)2](ClO4)2 (1b). Similarly[Co(Pamp)Cl2] (2a) and [Co(Pamp)2](ClO4) (2b) were synthesizedby the reaction of CoCl2�6H2O and ligand PampH in 1:1 and in1:2 ratio, respectively.

Caution: Perchlorate salts of metal complexes with organic li-gands are potentially explosive. Only small amounts of materialsshould be prepared and these should be handled with great care.

2.3.1. Synthesis of [Co(Pyimpy)Cl2] (1a)A batch of CoCl2�6H2O (238 mg, 1 mmol) was dissolved in 5 mL

of tetrahydrofuran. After stirring for 5 min, a batch of ligand (Pyim-py) (274 mg, 1 mmol) dissolved in 15 mL of tetrahydrofuran wasadded dropwise. The color of the solution changed from blue togreen and resulted into the precipitation of a green solid. After3 h of stirring, the solid was filtered and washed thoroughly withexcess of tetrahydrofuran and dried in vacuo. The complex wasrecrystallised in methanol:tetrahydrofuran mixture. Yield: 81%.Anal. Calc. for C17H14Cl2CoN4: C, 50.52; H, 3.49; N, 13.86. Found:C, 50.29; H, 3.62; N, 13.93. Selected IR data (KBr, mmax/cm�1):1596, mC@N; leff (297 K): 4.60 lB. KM/X�1 cm2 mol�1 (DMF): 17(neutral). UV–Vis [CH3OH, kmax/nm (e/M�1 cm�1)]: 230 (12360),249 (10340), 278 (8040), 359 (13160). ESI-MS (acetonitrile,neg.): m/z 402.08 (15.9%) [M�H+].

2.3.2. Synthesis of [Co(Pyimpy)2](ClO4)2 (1b)A batch of ligand (Pyimpy) (274 mg, 1 mmol) was dissolved in

10 mL of methanol. After stirring for 10 min, a batch of CoCl2�6H2O(119 mg, 0.5 mmol) in 10 mL of methanol was added dropwise tothe above stirred solution. The color of the solution changed tored. After 30 min of stirring, a batch of sodium perchlorate mono-hydrate (154 mg, 1.1 mmol) dissolved in 5 mL of methanol wasadded and stirred for 2 h. An orange solid precipitated which wasfiltered, washed with small amount of methanol, successively withdiethyl ether and dried in vacuo. Yield: 59%. Anal. Calc. for C34H28-

Cl2CoN8O8: C, 50.64; H, 3.50; N, 13.89. Found: C, 50.45; H, 3.56; N,13.85. Selected IR data (KBr, mmax/cm�1): 1599, mC@N; 1088, 625,mClO4

�. leff (297 K): 4.54 lB. KM/X�1 cm2 mol�1 (DMF): 141 (1:2).UV–Vis [CH3OH, kmax/nm (e/M�1cm�1)]: 230 (30540), 249 sh(24540), 291 (19760), 358 (29400). ESI-MS (acetonitrile, pos.):m/z 706.11 (4.5%) [M�(ClO4)]+, m/z 303.58 (100%) [M�2(ClO4)]2+.

2.3.3. Synthesis of [Co(Pamp)Cl2] (2a)Method A: A batch of ligand (PampH) (290 mg, 1 mmol) was

dissolved in 10 mL of methanol. After stirring for 5 min, a batchof sodium hydride (24 mg, 1 mmol) was added to this solutionand stirred for 1 h. This deprotonated ligand solution was addeddropwise to a continuously stirred solution of CoCl2�6H2O(262 mg, 1.1 mmol) dissolved in 5 mL of methanol. After 4 h ofstrirring, volume was reduced to 5 mL under reduced pressure. Agreen solid separated out which was washed with very smallamount of methanol, successively with diethyl ether and dried invacuo. The complex was then recrystallized in methanol. Yield:

K. Ghosh et al. / Polyhedron 49 (2013) 167–176 169

37%. Anal. Calc. for C17H13Cl2CoN4O: C, 48.71; H, 3.13; N, 13.37.Found: C, 48.32; H, 3.34; N, 13.52. Selected IR data (KBr, mmax/cm�1):1655, mC@O;. KM/X�1 cm2 mol�1 (DMF): 30 (neutral). UV–Vis[CH3OH, kmax/nm (e/M�1cm�1)]: 394 (12900). 1H NMR (500 MHz,d/ppm, CDCl3): 8.35 (d, J = 5.5 Hz, 1 H), 8.11 (t, J = 7.5 Hz, 1 H),7.96 (dd, 2 H), 7.82 (m, 3H), 7.73 (t, J = 8 Hz, 2 H), 7.62(t, J = 7 Hz, 1 H), 7.55 (t, 1 H), 6.93 (t, J = 6.5 Hz, 1H), 6.37(d, J = 9 Hz, 1H), 13C NMR (500 MHz, d/ppm, CDCl3): 163.62,161.10, 156.61, 149.17, 145.73, 142.07, 141.10, 141.04, 130.54,130.05, 130.01, 128.43, 125.85, 119.04, 109.97.

Method B: A batch of CoCl2�6H2O (262 mg, 1.1 mmol) was dis-solved in 5 mL of methanol. After stirring for 5 min, a batch of li-gand (PampH) (290 mg, 1 mmol) dissolved in 10 mL of methanolwas added dropwise to the above stirred solution and stirring con-tinued to 4 h. A green solid separated out which was washed withsmall amount of methanol, successively with diethyl ether anddried in vacuo. The complex was then recrystallized in methanol.Yield: 54%.

2.3.4. Synthesis of [Co(Pamp)2]ClO4 (2b)Method A: A batch of ligand (PampH) (145 mg, 0.5 mmol) was

dissolved in 5 mL of acetonitrile. After stirring for 5 min, a batchof sodium hydride (12 mg, 0.5 mmol) was added to the stirredsolution and stirred for 1 h. A batch of CoCl2�6H2O (60 mg,0.25 mmol) in 5 mL of acetonitrile was added dropwise to theabove stirred solution. The color of the solution changed to red.After 30 min of stirring, the solution was filtered. A batch of so-dium perchlorate monohydrate (42 mg, 0.3 mmol) dissolved in2 mL of acetonitrile was added to the filtrate. After 4 h of stirringthe solution was evaporated to dryness in vacuo. The brown solidobtained was recrystallised in dichloromethane:methanol mixture.Yield: 82%. Anal. Calc. for C34H26ClCoN8O6: C, 55.41; H, 3.56; N,15.20. Found: C, 55.14; H, 3.79; N, 15.16. Selected IR data (KBr,mmax/cm�1): 1655, mC@O; 1096, 623, mClO4

�. KM/X�1 cm2 mol�1

(DMF): 60 (1:1). UV–Vis [CH3OH, kmax/nm (e/M�1 cm�1)]: 398(22000). 1H NMR (500 MHz, d/ppm, CDCl3): 8.34 (d, J = 6 Hz, 1H), 8.08 (t, J = 6.5 Hz, 1 H), 7.95 (t, J = 6 Hz, 2 H), 7.78 (m, 3H),7.72 (t, 2 H), 7.61 (t, 1 H), 7.52 (t, 1 H), 6.90 (t, J = 6 Hz, 1 H),6.37 (d, J = 8.5 Hz, 1 H). 13C NMR (500 MHz, d/ppm, CDCl3):163.66, 161.12, 156.66, 149.22, 145.78, 141.96, 141.14, 140.97,130.53, 129.98, 129.88, 128.41, 125.84, 119, 109.98. ESI-MS (aceto-nitrile, pos.): m/z 637.15 (100%) [M�(ClO4)]+.

Method B: A batch of ligand (PampH) (145 mg, 0.5 mmol) wasdissolved in 5 mL of acetonitrile. A batch of CoCl2�6H2O (60 mg,0.25 mmol) dissolved in 10 mL of acetonitrile was added dropwiseto the above stirred solution. The color of the solution changed tored. After 2 h of stirring a batch of sodium perchlorate monohy-drate (42 mg, 0.3 mmol) dissolved in 2 mL of acetonitrile wasadded to reaction mixture and stirring was continued for addi-tional 2 h followed by filtration of the reaction mixture. The filtratewas evaporated to dryness in vacuo. The brown solid obtained wasrecrystallised in dichloromethane:methanol mixture. Yield: 68%.

2.4. Superoxide dismutase assay

The superoxide dismutase (SOD) activities of the complexeswere determined by following the standard xanthine/xanthine oxi-dase/nitroblue tetrazolium assay [47]. The superoxide radicals(O2

��) were generated in situ using xanthine/xanthine oxidase sys-tem and detected spectrophotometrically by measuring the absor-bance at 560 nm due to the reduction of NBT (nitrobluetetrazolium). The experiment was started by the addition of2.1 mU/mL xanthine oxidase to the reaction system containing0.2 mM xanthine, 0.6 mM NBT, 1000 U/mL catalase and varyingconcentration of complexes in 0.1 M phosphate buffer (pH 7.8).The measurements were started after 10 min incubation and each

experiment was performed in duplicate. IC50 value for SOD activitywas defined as the concentration of the compound for 50% inhibi-tion of the NBT reduction by the superoxide radicals generated inthe reaction system.

2.5. DNA binding and cleavage experiments

DNA binding experiments were carried out in 0.1 M phosphatebuffer (pH 7.2) using a solution of calf thymus (CT) DNA whichgave a ratio of UV–Vis absorbance at 260 and 280 nm (A260/A280)of ca. 1.8, indicating that the CT-DNA was sufficiently protein free.The concentration of DNA solution was determined by absorbanceat 260 nm and the extinction coefficient e260 was taken 6600 cm�1

as reported in the literature [51]. Absorption titration experimentswere carried out with complex concentration of 40–100 lMvarying the CT-DNA concentration from 0–135 lM in 0.1 M phos-phate buffer (pH 7.2). The binding constants Kb were determinedaccording to the reported method [52]. Fluorescence quenchingexperiments were carried out by the successive addition of the co-balt complexes to the DNA (25 lM) solutions containing 5 lMethidium bromide (EB) in 0.1 M phosphate buffer (pH 7.2). Thesesamples were excited at 250 nm and emissions were observed be-tween 500 and 700 nm. Stern–Volmer quenching constants Ksv

were calculated according to the reported procedure [52]. Circulardichroism (CD) spectra of CT-DNA in absence and presence of thecobalt complexes were recorded with a 0.1 cm path-length cuvetteafter 10 min incubation at 25 �C. The concentration of the com-plexes and CT-DNA were 50 and 200 lM, respectively.

Cleavage of plasmid DNA was monitored by using agarose gelelectrophoresis. Supercoiled pBR322 DNA (100 ng) in Tris–boricacid–EDTA (TBE) buffer (pH 8.2) was treated with cobalt com-plexes (50–200 lM) in the presence or absence of additives. Theoxidative DNA cleavage by the complexes were studied in the pres-ence of H2O2 (200 lM, oxidizing agent) or 2-mercaptoethanol(200 lM, reducing agent) and DMSO, ethanol, NaN3, D2O, L-histi-dine (20 mM each) and catalase (10 U). The samples were incu-bated for 15 min to 2 h at 37 �C followed by the addition ofloading buffer (25% bromophenol blue and 30% glycerol). The aga-rose gel (0.8%) containing 2 lL (10 mg/mL stock) of ethidium bro-mide (EB) was prepared and the electrophoresis of the DNAcleavage products was performed on it. The gel was run at 60 Vfor 2 h in TBE buffer and the bands were identified by placingthe stained gel under an illuminated UV lamp. The fragments werephotographed by using gel documentation system (BIO RAD).

2.6. X-ray crystallography

The X-ray data collection and processing for 1a was performedon Bruker Kappa Apex-II CCD diffractometer by using graphitemonochromated Mo Ka radiation (k = 0.71073 Å) at 273 K. Crystalstructure was solved by direct methods. Structure solution, refine-ment and data output were carried out with the SHELXTL program[53]. All non-hydrogen atoms were refined anisotropically. Hydro-gen atoms were placed in geometrically calculated positions andrefined using a riding model. Images were created with theDIAMOND and MERCURY program.[54]

3. Result and discussion

3.1. Synthesis and characterization of ligands and their metalcomplexes

Complexes [Co(Pyimpy)Cl2] (1a) and [Co(Pyimpy)2](ClO4)2 (1b)were synthesized by the reaction of CoCl2�6H2O and ligand Pyimpyin 1:1 and 1:2 M ratio, respectively. Complexes [Co(Pamp)Cl2] (2a)

170 K. Ghosh et al. / Polyhedron 49 (2013) 167–176

and [Co(Pamp)2](ClO4) (2b) were synthesized by two differentmethods. Complex 2a was isolated as green solid by the reactionof deprotonated ligand with CoCl2�6H2O (1:1) in aerobic condition.The oxidation of metal centre indicated the coordination of carbox-amido nitrogen to the metal centre because amide nitrogen stabi-lizes higher oxidation state of metal ions [55] and oxygen acts asoxidizing agent during such reactions. This was also supportedby spectroscopic and electrochemical studies (vide infra). Complex[Co(Pamp)2](ClO4) (2b) was prepared by the reaction of deproto-nated ligand with CoCl2�6H2O (2:1) in aerobic conditions. Browncolored microcrystalline complex was isolated and characterized(described in Method A). It is important to note here that the reac-tion of CoCl2�6H2O and PampH also afforded complexes 2a and 2bin presence of air (described in Method B). This data indicated thatin aerobic condition cobalt ion was efficient in deprotonating theligand due to its Lewis acidity. We may or may not need a deprot-onating agent (sodium hydride here) for the deprotonation ofamide nitrogen for the synthesis of complexes 2a and 2b [55,56].All the complexes were synthesized with good yield (50–80%).The synthetic procedures of complexes are summarized inScheme 2.

In complexes 1a and 1b coordination of azomethine nitrogenwas supported by shifting of mC@N to higher frequencies by 18–21 cm�1 as compared to those of free ligands (1578 cm�1) in IRspectra [57]. Complexes 2a and 2b afforded the carbonyl stretchingfrequency mC@O nearly 1655 cm�1. The decrease in the carbonylstretching frequencies as compared to free ligand by 40 cm�1

clearly expressed the coordination of carboxamido nitrogen. Thedisappearance of the peak at 3354 cm�1 which was assigned tobe N–H stretching frequency in PampH as well as shifting of mC@O

to lower wavenumbers indicated its coordination in the deproto-nated form [58]. Complex 1b and 2b showed IR bands near 1092and 1096 cm�1, respectively, together with band near 625 cm�1.These data suggest the presence of non-coordinated perchlorateion in 1b and 2b. The absorption band in the range 358–398 nmfound in UV–Vis spectra of all complexes were of charge transfertype because of higher extinction coefficients. The molar conduc-tivity measurement for complexes 1a and 2a in DMF solution (ca.10�3 M) gave rise to neutral electrolytic behavior with conductivityvalues of 17 and 30 X�1 cm2 mol�1, respectively at 25 �C. Complex1b afforded molar conductivity values of 141 X�1 cm2 mol�1 (at25 �C) indicating bi-univalent electrolytic behavior whereas 2bafforded 60 X�1 cm2 mol�1 (at 25 �C) indicating uni-univalentelectrolytic behavior [59]. Magnetic susceptibility measurementsfor complexes 1a and 1b at room temperature suggested the pres-ence of three unpaired electrons at the cobalt centre [60]. The car-boxamido nitrogen donor stabilizes the higher oxidation states socomplexes 2a and 2b showed the Co(III) centres and these twocomplexes were found to be diamagnetic with low-spin d6 elec-tronic configuration. The structures of complexes 2a and 2b wereestablished by 1H and 13C NMR spectral studies as shown inFig. S5–S6 and Fig. S7–S8. The electro-spray ionization mass spec-trum (ESI-MS) of 1a showed a peak with m/z value of 402.08 prob-ably due to the formation of [1a�H+]� anion with loss of a proton(Fig. S9). The ESI-MS spectrum of 1b exhibited two peaks with m/z

CoCl2 6

[Co(Pyimpy)(Cl)2]

[Co(Pyimpy)2](ClO4)2

1:1

1:2

Pyimpy

NaClO4 H2O

a1

b1

Scheme 2. Schematic drawing of

values of 706.11 (relative intensity 4.5%) and 303.58 (relativeintensity 100%) which were assigned to the formation of[1b�(ClO4)]+ and [1b�2(ClO4)]2+ cations due to the loss of oneand two perchlorate anions, respectively (Fig. S10). However, incase of 2b only one peak was observed with m/z value of 637.15(relative intensity 100%) assigned to the formation of [2b�(ClO4)]+

cation by the loss of perchlorate anion present outside thecoordination sphere (Fig. S11).

3.2. Crystal structure

In order to confirm mode of coordination of the ligand Pyimpy,crystals were grown for complex 1a by the diffusion of acetone inthe methanolic solution and molecular structure was determinedby X-ray diffraction studies. A ball and stick representation of themetal coordination environment in complex 1a is displayed inFig. 1. The matrix parameters are described in Table 1 and selectedbond lengths and bond angles are listed in Table 2.

Two pyridine nitrogen donors (NPy) and one azomethine nitro-gen (NIm) was bound to the cobalt centre in 1a in meridional fash-ion (Fig. 1). The equatorial plane of the coordination sphereconsists of two pyridine nitrogen (NPy), one azomethine nitrogen(NIm) and Cl atom while the axial position is occupied by other Clatom. The ligand biting angles at the metal centre were 75.26�(N(1)�Co(1)�N(2)) and 73.67� (N(4)�Co(1)�N(2)) in complex 1aand the other two angles with Cl2 were 100.25� and 101.77�. Struc-tural index parameters calculation (s = 0.30) showed the distortedsquare pyramidal geometry [61]. Two pyridine and imine nitrogendistances from Co(II) centre were consistent with the reported datafor high-spin Co(II) complexes [62–64]. These data were consistentwith the magnetic moment data (4.60 B.M.) described earlier (videsupra). The axial Co–Cl1 distance 2.315 Å was found to be longerthan the equatorial Co�Cl2 distance 2.284 Å due to Jahn–Tellerdistortion which are consistent with the literature values [65].The Co(II) centre is 0.424 Å above the plane generated by the coor-dinated N1, N2, N4 and Cl2 atoms. The ligand Pyimpy has three six-membered rings; among them, the two pyridine rings are coplanarwith the imine function whereas the other phenyl ring is roughlyperpendicular (85.55�) to the ligand binding plane.

Non-covalent interactions like p-stacking interactions with arylhydrogen and hydrogen bonding network are important in supra-molecular chemistry and crystal engineering [66]. In complex 1a,the axial Cl atom showed hydrogen bonding interaction with thehydrogen atoms of two pyridine and one phenyl rings of neighbor-ing molecules and the distances are 2.666, 2.760 and 2.878 Å,respectively. However the equatorial Cl showed the hydrogen bod-ing interactions with the hydrogen atoms of two phenyl rings andone pyridine ring of the neighboring molecules with the bond dis-tances 2.896, 2.780 and 2.773 Å, respectively (shown in Fig. S12).Interestingly, weak p–p stacking interactions were observed be-tween the pyridine rings of two neighboring molecules affordingthe centroid to centroid distance of 3.636 Å (Fig. 2) which is lesserthan those reported by Yin and coworkers [67]. The interaction ofthe axial Cl ion with pyridine hydrogen at a distance of 2.878 Åhelps for p–p interactions. These hydrogen bonding and p–p

H2O

[Co(Pamp)(Cl)2]

[Co(Pamp)2](ClO4)

1:1

1:2

PampH

NaClO4 H2O

a2

b2

synthesis of complexes 1–2.

Fig. 1. Molecular structure of the [Co(Pyimpy)(Cl)2] (1a).

Table 1Summary of crystal data and data collection parameters for [Co(Pyimpy)(Cl)2].

Empirical formula C17H14Cl2N4CoFormula weight (g mol�1) 404.17Temperature (K) 293(2)k (Å) (Mo Ka) 0.71073Crystal system OrthorhombicSpace group P212121a (Å) 8.4008(6)b (Å) 10.0715(7)c (Å) 19.8632(12)a (�) 90.00b (�) 90.00c (�) 90.00V (Å3) 1680.6(2)Z 4qcalc (g cm�3) 1.597F(000) 820Theta range for data collection 2.05–32.85Index ranges �12 < h < 8,

�15 < k < 9,�30 < l < 27

Refinement method Full matrix least-squares on F2

Data/restraints/parameters 6115/0/218(GOF)a on F2 0.734R1

b [I > 2r(I)] 0.0364R1[all data] 0.0307wR2

c [I > 2r(I)] 0.0973wR2 [all data] 0.0774

a GOF = [P

[w(Fo2 � Fc

2)2]/M � N]1/2 (M = number of reflections, N = number ofparameters refined).

b R1 =P

||Fo| � |Fc||/P

|Fo|.c wR2 = [

P[w(Fo

2 � Fc2)2]/

P[w(Fo

2)2]]1/2.

Table 2Selected bond lengths (Å) and angles (�) for [Co(Pyimpy)(Cl)2].

Bond length (Å) Bond angles (�)

Co(1)�N(1) 2.0956(17) N(1)�Co(1)�N(4) 142.23(7)Co(1)�N(2) 2.1190(17) N(1)�Co(1)�N(2) 75.26(7)Co(1)�N(4) 2.0992(18) N(4)�Co(1)�N(2) 73.67(7)Co(1)�Cl(1) 2.3152(6) N(1)�Co(1)�Cl(2) 100.25(5)Co(1)�Cl(2) 2.2844(6) N(4)�Co(1)�Cl(2) 101.77(5)

N(1)�Co(1)�Cl(1) 99.38(5)N(4)�Co(1)�Cl(1) 103.39(5)N(2)�Co(1)�Cl(1) 93.46(5)Cl(2)�Co(1)�Cl(1) 106.20(2)N(2)�Co(1)�Cl(2) 160.32(5)

K. Ghosh et al. / Polyhedron 49 (2013) 167–176 171

interactions generated a supramolecular polymer like species.Packing diagrams of the molecule and a three-dimensionalnetwork is shown in Fig. S13.

3.3. Electrochemistry and SOD activity

The electrochemical data for all the complexes are described inTable 3 and all the voltammograms are shown in Fig. 3. Two differ-ent type of voltammograms were observed for these complexes. 1aand 1b gave rise to voltammograms of type A and 2a and 2b gaverise to voltammograms of type B, respectively. In case of 1a thequasireversible redox couples were found near E1/2 values of+0.487, �0.630, and �0.879 V versus Ag/AgCl. E1/2 values of+0.487 and �0.630 V were designated as Co3+/Co2+ and Co2+/Co1+

redox couples, respectively. The quasireversible redox couple nearE1/2 values of �0.879 V versus Ag/AgCl furnished greater currentheight compared to other two couples found for this complex.We have investigated each couple individually scanning in be-tween +0.300 and +0.650 V, �0.700 and �0.500 V and �1.100and �0.650 V, respectively. Results are shown in Fig. S14. The E1/2

value found near �0.879 V versus Ag/AgCl could probably be dueto ligand centered reduction or any other species generated inthe solution. We investigated the redox property of the ligand inacetonitrile as well as an acetonitrile solution containing ligandand ZnCl2 in 1:1 ratio separately; we were unable to find any redoxcouple in that range which could authenticate ligand centeredreduction. Hence at this point we speculate that the redox coupleis probably due to some other species generated in the solution.Similar to 1a E1/2 values of +0.459 and �0.653 V versus Ag/AgClelectrode for 1b were designated as Co3+/Co2+ and Co2+/Co1+ redoxcouples, respectively. Investigation of literature revealed that ourCo3+/Co2+ data is closer to the value reported by Slattery andcoworkers [68]. However the positive potential values are higherthan that of similar complexes reported in the literature [69–72].Cyclic voltammogram of 2a and 2b displayed single quasirevers-ible redox couple with (type B in Fig. 3) E1/2 values in the range�0.373 to �0.407 V versus the Ag/AgCl electrode. These data aresimilar to the data obtained by Mascharak and coworkers [55]and are due to the Co3+/Co2+ couple.

The electrochemical behavior of these complexes showed thatall these complexes are redox-active which prompted us to studytheir superoxide scavenging ability. Due to the better solubility,1a and 2a were examined for the superoxide scavenging activity(SOD activity) using xanthine/xanthine oxidase/nitroblue tetrazo-lium assay. The SOD activity of the complexes was monitored bythe change in intensity at 560 nm which is associated with thereduction of NBT by the superoxide radicals (O2

��) generated bythe xanthine/xanthine oxidase system. The rate of absorptionchange was determined for 1a and the IC50 value was obtainedby plotting the rate of NBT reduction (along Y-axis) versus the con-centration of test solution (along X-axis) (Fig. 4). The IC50 value for

Fig. 2. Intermolecular hydrogen bonding network in [Co(Pyimpy)(Cl)2] (1a).

Table 3Electrochemical data for the redox couples Co(II/I) and Co(III/II) at 298 Ka vs. Ag/AgCl.

Complex Co(II/I) Co(III/II)

E1/2b (V) DEp

c (mV) E1/2b (V) DEp

c (mV) E1/2b (V) DEp

c (mV)

1a �0.879 118 �0.630 67 0.487 1141b – – �0.653 130 0.459 1822a – – – – �0.373 1072b – – – – �0.407 88

a Measured in acetonitrile with 0.1 M tetrabutylammonium perchlorate (TBAP).b Data from cyclic voltammetric measurements; E1/2 is calculated as average of anodic (Epa) and cathodic (Epc) peak potentials; E1/2 = 1/2 (Epa + Epc).c DEp = Epa � Epc.

172 K. Ghosh et al. / Polyhedron 49 (2013) 167–176

1a was found to be 85 ± 3 lM whereas only 10–12% inhibitioncould be obtained with 2a even at very high concentration of thecomplex (500 lM). Similar superoxide scavenging activity by Co(II)complexes was observed by Anacona and coworkers (IC50 values1000 and 580 lM) [73,74].

3.4. DNA binding studies

3.4.1. Stability of complexes in bufferWe performed all the biological activity studies in 0.1 M phos-

phate buffer so it was necessary to check the stability of our com-plexes in these buffers. Little change in absorbance was observedafter 3 days. A small change in absorbance without any consider-able shift in wavelengths (kmax) predicted the stability of these co-balt complexes in the above buffer solution.

3.4.2. Absorption spectral studiesElectronic absorption spectroscopic techniques were used to

investigate the binding of DNA with cobalt complexes. To achievethis, the absorption spectra of complexes in the absence and pres-ence of calf thymus DNA (CT DNA) at different concentrations weremeasured. The change in absorbance with a blue shift of 20 nm forcomplex 1a is shown in Fig. 5 (absorption spectra obtained forother complexes are shown in Supporting information). The intrin-sic binding constants Kb for all our complexes complexes have beendetermined according to the reported procedures and shown in Ta-ble 4 [52]. The observed binding constants for these complexes aresmaller than those of classical intercalators and metallointercala-tors where binding constant was reported to be in the order of106–107 M�1 [75,76]. Fig. S15 shows the spectral changes for 1bwith the similar shift of the charge transfer band as observed for1a. These types of spectral changes resulting into the formationof a new band are indicative of the generation of a new species

possibly due to the covalent interaction between the complexesand DNA [48–50]. Investigation of literature revealed that cobaltcomplexes may bind covalently with DNA with N7 of guaninebases [77]. Hence we predict that a new species generated due tothe attachment of Co(II) complexes with nucleic bases of DNA dur-ing DNA interaction studies. The formation of new species was alsosupported by the reverse titration experiments where a fixed con-centration of DNA was titrated with increasing concentrations ofthe complex. The spectral changes during the reverse titrationexperiments are shown in the Fig. S16. At lower concentration ofthe complex the spectrum is characteristic of the new specieswhich is dominated by the CT band of the complex at higher con-centrations. However absorption spectra for 2a and 2b exhibitedonly hypochromism without any shift in the wavelength valueswhich indicated that these complexes interacted with DNA quitedifferently from 1a and 1b (Fig. S17 and 18) probably due to thepresence of Co(III) in 2a and 2b.

3.4.3. Ethidium bromide displacement assayWe have examined the competitive binding of ethidium bro-

mide versus our complexes with DNA using fluorescence spectralstudies to get better insight into DNA binding events. Ethidiumbromide (EB) emits intense fluorescence in the presence of DNAdue to its strong intercalation between the DNA base pairs. The en-hanced fluorescence can be quenched by the addition of the metalcomplexes to the EB-DNA mixture which competitively bind toDNA resulting in the reduction in emission intensity. The fluores-cence quenching curve of ethidium bromide bound to DNA bycomplex 1a is shown in Fig. 6 (similar curves for 1b, 2a and 2bare shown in Supporting information). The Stern–Volmer quench-ing constant Ksv for all complexes were obtained by Stern–Volmerplots which are described in Table 4. These values are lesser thanthose reported for metallointercalators [78]. The Ksv values for

(A)

(B)

-1.0 -0.5 0.0 0.5

(1a)(1b)

16μA

-0.8 -0.6 -0.4 -0.2 0.0Potential/V

(2a)(2b)

18μA

Fig. 3. Cyclic voltammograms of a 10�3 M solution of (A) 1a and 1b (B) 2a and 2b inacetonitrile in presence of 0.1 M TBAP as a supporting electrolyte, glassy-carbon asa working electrode and Ag/AgCl as reference electrode; scan rate 0.1 V/s.

0 50 100 150 200 2500

10

20

30

40

50

60

70

% In

hibi

tion

Concentration (μΜΜ)

Fig. 4. SOD activity of complex 1a in the xanthine/xanthine oxidase /nitro bluetetrazolium (NBT) assay.

300 350 400 4500.0

0.2

0.4

0.6

Ab

sorb

ance

Wavelength (nm)

Fig. 5. Absorption spectra of complexes 1a (50 lM) in 0.1 M phosphate buffer (pH7.2) in the presence of increasing amounts of DNA (0–116.5 lM). Dotted linerepresents the spectrum in the absence of DNA.

Table 4Selected absorption binding constants (Kb) and Stern–Volmer quenching constants(Ksv) for complexes 1–2.

Complex Kb (M�1) Ksv (M�1)

1a 1.15 � 104 2.98 � 104

1b 4.23 � 104 7.33 � 104

2a 5.06 � 104 5.14 � 104

2b 3.27 � 104 9.99 � 104

550 600 650 7000

200

400

600

Inte

nsi

ty

Wavelength (nm)

Fig. 6. Fluorescence emission spectra of the EB�DNA in presence of complex 1a in0.1 M phosphate buffer (pH 7.2). [EB] = 5 lM, [DNA] = 25 lM, [1a] = 0–45.7 lM,kex = 250 nm and kem = 585 nm. Dotted line represents the spectrum in the absenceof complex.

K. Ghosh et al. / Polyhedron 49 (2013) 167–176 173

these complexes indicated that the 1b and 2b showed higherbinding affinity as compared to 1a and 2a probably due to bis -complexation of the ligand.

3.4.4. Circular dichroismCD spectroscopic technique can be successfully utilized to de-

tect the conformational changes of DNA induced by its interactionwith metal complexes. The CD spectrum of CT-DNA was recordedin the range 225–300 nm in 0.1 M phosphate buffer (pH 7.2) andit has been found that there were one positive band at 278 nm

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

NC

LC

SC

Fig. 8. Gel electrophoresis separations showing the cleavage of supercoiled pBR322DNA (100 ng) by complexes 1a, 2a, 1b and 2b in presence of H2O2 (200 lM) andBME (200 lM). Samples were incubated at 37 �C for 2 h. Lane 1, DNA control; lane 2,DNA + H2O2; lane 3, DNA + BME; lane 4, DNA + 1a (50 lM); lane 5, DNA + 1a(50 lM) + H2O2; lane 6, DNA + 1a (50 lM) + BME; lane 7, DNA + 2a (50 lM); lane 8,DNA + 2a (50 lM) + H2O2; lane 9, DNA + 2a (50 lM) + BME; lane 10, DNA + 1b(50 lM); lane 11, DNA + 1b (50 lM) + H2O2; lane 12, DNA + 1b (50 lM) + BME; lane13, DNA + 2b (50 lM); lane 14, DNA + 2b (50 lM) + H2O2; lane 15, DNA + 2b(50 lM) + BME.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

NC LC SC

Fig. 9. Gel electrophoresis separations showing the cleavage of supercoiled pBR322DNA (100 ng) by complex 1a (50 lM) in presence of H2O2 (200 lM) and BME(200 lM). Samples were incubated at 37 �C for 2 h. Lane 1, DNA; lane 2,DNA + 1a + H2O2; lanes 3–8, DNA + 1a + H2O2 + DMSO, ethanol, NaN3, D2O, L-histidine (20 mM) and catalase (10 U), respectively; lane 9, DNA + 1a + BME; lanes10–15, DNA + 1a + BME + DMSO, ethanol, NaN3, D2O, L-histidine (20 mM) andcatalase (10 U) respectively.

174 K. Ghosh et al. / Polyhedron 49 (2013) 167–176

due to base stacking and one negative band at 246 nm due to helic-ity. The examination of Fig. 7 indicates that the positive bandslightly increased in intensity and the negative band decreased inintensity after the binding of 1a with DNA however no consider-able shift in kmax could be observed. A very small decrease in theintensity of positive band and a small increase in the intensity ofnegative band were observed when the DNA was incubated with2a (Fig. 7). However, the spectrum of DNA was almost unaffectedby the presence of 1b or 2b (Fig. 7). These data indicated that theDNA did not undergo any appreciable conformational change afterthe DNA binding event [50].

3.5. Nuclease activity

Investigation of literature revealed that the complexes whichcontain a redox active metal centre and interact with DNA cova-lently could exhibit excellent nuclease activity [47]. In our com-plexes we found new species generation probably via covalentinteraction with DNA and we extended our DNA interaction stud-ies by examining the nuclease activity of these complexes. Thecleavage of supercoiled pBR322 DNA (100 ng) by these complexeswas analyzed by monitoring the conversion of supercoiled form(SC) to nicked circular (NC) and linear (LC) forms and the gel elec-trophoretic separation of the three forms is shown in Fig. 8 and 9.

DNA cleavage experiments were performed in the presence ofan oxidizing agent (H2O2) and a reducing agent (2-mercap-toethanol or BME). In the absence of an oxidizing or reducing agent1a and 1b could not exhibit (may be negligible) any nuclease activ-ity (Fig. 8, lanes 4 and 10, respectively). However both the com-plexes showed excellent nuclease activity in presence of H2O2 aswell as BME (Fig. 8, lanes 5–6 and 11–12, respectively) and itwas found that the nuclease activity was more pronounced in pres-ence of H2O2 than BME. Interestingly the whole DNA got convertedto NC as well as LC form in the presence of H2O2 at a minimum of50 lM concentration of these complexes. It is clear from Fig. 8 that1a showed better nuclease activity as compared to 1b. However wedid not observe any NC or LC form of DNA in case of complexes 2aand 2b exhibiting their inefficiency to cleave DNA (Fig. 8, lanes 7–9and 13–15, respectively). Hence efficient nuclease activity wasafforded by all the Co(II) complexes in presence of oxidizing orreducing agents whereas no such activity could be observed withCo(III) complexes. In the control experiment H2O2 and 2-mercap-toethanol did not exhibit any cleavage in the absence of cobaltcomplexes (Fig. 8, lane 2, 3).

Small molecule SOD mimics prove to be potent agents forcleaving the DNA. Interestingly, in our previous report manganese

(a)

-2

-1

0

1

2

θ (m

deg

)

Wavelength (nm)

DNA 1a 1b

240 260 280 300 320

Fig. 7. Circular dichroism spectra of CT-DNA and its interaction with (a) complexes 1a aincubation at 25 �C.

complexes with ligand PampH bearing the carboxamido function-ality exhibited excellent DNA cleavage whereas those with Pyimpyshowed no nuclease activity [47]. For these complexes it was foundthat the complexes were capable of dismuting superoxide and alsocleaving DNA. This may be due to the redox potential of manga-nese complexes derived from PampH which are different fromthose complexes derived from Pyimpy. In the present report theCo(III) complex derived from PampH showed no nuclease activitywhereas Co(II) complexes of Pyimpy showed excellent nucleaseactivity which is in contradiction to the previous results [47].

In order to see the effect of incubation time on the nucleaseactivity of these complexes we have performed the gel electropho-resis experiments with 1a by varying the incubation time over15–90 min (shown in Supporting information). Interestingly, con-version of SC form to NC form increased progressively with theincrease in the incubation time (Fig. S23, lanes 4–6 and 7–9).

(b)

240 260 280 300 320

-2

-1

0

1

θ (m

deg

)

Wavelength (nm)

DNA 2a 2b

nd 1b; (b) 2a and 2b, respectively, in 0.1 M phosphate buffer (pH 7.2) after 10 min

Table 5Summary of the SOD activity and nuclease activity of the complexes.

S. No. Complex SOD activity Nuclease activity Refs.

1 [Mn(Pyimpy)2](ClO4)2 not effective no [47]2 [Mn(Me-Pyimpy)2](ClO4)2 not effective no [47]3 [Mn(Pamp)2](ClO4) �3.73 lM yes [47]4 [Mn(Phimp)2] �0.29 lM yes [52]5 [Mn(Phimp)2](ClO4) �0.39 lM yes [52]6 [Mn(N-Phimp)2](ClO4) �1.12 lM yes [52]7 [Cu(Pyimpy)(H2O)](ClO4)2 �7.41 lM yes [50]8 [Cu(Pyimpy)Cl2] �4.92 lM yes [50]9 [Cu(Phimp)(H2O)]2(ClO4)2 �11.20 lM yes [51]

10 [Cu(Phimp)2] �8.31 lM yes [51]11 [Co(Pyimpy)Cl2] �85 lM yes present work12 [Co(Pamp)Cl2] not effective no present work

K. Ghosh et al. / Polyhedron 49 (2013) 167–176 175

Furthermore, the extent of cleavage also increases with theincreasing concentration of H2O2 (Fig. S23, lanes 13–15).

3.6. Mechanism of DNA cleavage

Role of diffusible radical species can be diagnosed by monitor-ing the quenching of DNA cleavage in the presence of radical scav-engers in solution [50]. Standard radical scavengers were added tothe reaction of complex 1a with H2O2 or BME during DNA cleavageexperiments. Analysis of the data obtained from the gel electro-phoresis (Fig. 9) indicated the role of singlet oxygen in DNA cleav-age activity because significant inhibition was observed inpresence of NaN3 and L-histidine. However , we did not findenhancement of cleavage activity in presence of D2O [50]. We havealso observed, although to a lesser extent, inhibition of nucleaseactivity in presence of DMSO and EtOH which are known to be hy-droxyl radical scavengers. A similar inhibition was also observed inpresence of catalase. Hence we cannot eliminate the possible roleof hydroxyl radical as well as peroxide ion in this activity.

4. Superoxide scavenging and nuclease activity

Few highlights from our previous reports are summarized in Ta-ble 5 which are necessary to explain the results obtained in this re-port and to understand the results of our superoxide scavengingand nuclease activity studies. The manganese complex derivedfrom ligand Pyimpy did not exhibit DNA cleavage activity whereasthe complex derived from PampH showed nuclease activity [47].Manganese complexes obtained from PampH exhibited superoxidescavenging activity also. These data clearly indicated that redoxproperty of the metal centre plays essential role in oxidativeDNA cleavage activity. We have mentioned this by performing aparallel chemistry of zinc with Pyimpy where we did not findany nuclease activity [48] because of redox inactive zinc centrepresent in the complex. We would like to mention here that zincdoes not bind to PampH. However, it is important to note that re-dox activity should be in the proper range [47,52,79–81] in thescale of electrode potential. From the literature we know that theimine nitrogen stabilizes lower oxidation state whereas carboxa-mido nitrogen stabilizes higher oxidation state of the metal centre.Hence the redox property of the metal centre changes completelydue to the presence of carboxamido nitrogen or imine nitrogen.[Mn(Pyimpy)2](ClO4)2 and [Mn(Pamp)2](ClO4) gave rise to E1/2 val-ues of �1.25 and +0.26 V versus Ag/AgCl reference electrode,respectively for Mn(II)/(III) couple. On the other hand[Co(Pyimpy)Cl2] and [Co(Pamp)Cl2] gave rise to Co(II)/(III) couplewith E1/2 values +0.487 and �0.373 V versus Ag/AgCl electrode. Itis clear that the Co(II)/(III) couple for [Co(Pyimpy)Cl2] is compara-ble to that of Mn(II)/(III) couple of [Mn(Pamp)2](ClO4) and these re-dox couples are responsible for the superoxide scavenging activity.

However [Co(Pamp)Cl2] and [Mn(Pyimpy)2](ClO4)2 although exhi-bit redox properties, they do not respond to superoxide scavengingactivity because of their E1/2 values which are out of the range[47,52,79–81]. With these results we moved to examine thenuclease activity of those complexes. Indeed we found out[Co(Pyimpy)Cl2] and [Mn(Pamp)2](ClO4) are efficient in DNA cleav-age activity whereas [Co(Pamp)Cl2] and [Mn(Pyimpy)2](ClO4)2 arenot efficient in nuclease activity. Interestingly, in case of cobaltcomplexes presence of imine nitrogen was necessary for superox-ide scavenging as well as nuclease activity and these results are re-verse to those reported in our previous report on chemistry ofmanganese with the same two ligands [47]. We have also observedthat several copper complexes which are small molecule SODmimics exhibited excellent nuclease activity [51] and this reportalso supports the hypothesis.

5. Conclusions

The following are the principal findings and conclusions of ourpresent study. Cobalt(II) and cobalt(III) complexes of ligands Pyim-py and PampH, respectively, were synthesized and characterizedby spectroscopic studies. Molecular structure of [Co(Pyimpy)Cl2](1a), determined by X-ray crystallography afforded high-spinCo(II). Electrochemical studies afforded Co(II)/Co(III) and Co(I)/Co(II) redox couple for complexes [Co(Pyimpy)Cl2] (1a) and[Co(Pyimpy)2](ClO4)2 (1b). On the other hand, only Co(III)/Co(II) re-dox couple was found for complexes [Co(Pamp)Cl2] (2a) and[Co(Pamp)2](ClO4) (2b). Complex 1a afforded IC50 value of85 ± 3 lM in superoxide scavenging activity by NBT assay. How-ever we found that complex 2a exhibited very poor activity. DNAinteraction studies indicated external and/or surface binding ofall the complexes. Interestingly, 1a was efficient in DNA cleavageactivity in presence of H2O2 as well as BME. Although 1b alsoshowed this nuclease activity but the efficiency of 1a was betterthan 1b. Complexes 2a and 2b did not afford SOD activity as wellas nuclease activity. Mechanistic studies indicated possible roleof reactive oxygen species (ROS) in the DNA cleavage activity.Investigation of the results of this report and our previous observa-tions (Table 5) clearly expressed that once the complex will re-spond to superoxide scavenging activity by xanthine/xanthineoxidase/nitroblue tetrazolium assay, the complex will also be effi-cient in DNA cleavage activity in presence of H2O2 or BME.

Acknowledgements

K.G. is thankful to faculty initiation grant (Scheme B) IIT Roor-kee and DST, New Delhi SERC FAST Track project (SR/FTP/CS-44/2006) for financial support. V.M. and P.K. are thankful to MHRD,UGC and CSIR, India for financial assistance. We are thankful toDST, New Delhi for financial support in DST-FIST program for

176 K. Ghosh et al. / Polyhedron 49 (2013) 167–176

ESI-MS facility at IITR. U.P.S. is thankful to IIT Roorkee for the singlecrystal X-ray facility.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.poly.2012.09.025.

References

[1] G.C. Silver, W.C. Trogler, J. Am. Chem. Soc. 117 (1995) 3983.[2] K.E. Erkkila, D.T. Odom, J.K. Barton, Chem. Rev. 99 (1999) 2777.[3] Q. Jiang, N. Xiao, P. Shi, Y. Zhu, Z. Guo, Coord. Chem. Rev. 251 (2007) 1951.[4] F. Mancin, P. Scrimin, P. Tecilla, U. Tonellato, Chem. Commun. (2005) 2540.[5] B. Armitage, Chem. Rev. 98 (1998) 1171.[6] M.J. Clarke, F. Zhu, D.R. Frasca, Chem. Rev. 99 (1999) 2511.[7] J. Chen, X. Wang, Y. Shao, J. Zhu, Y. Zhu, Y. Li, Q. Xu, Z. Guo, Inorg. Chem. 46

(2007) 3306.[8] K.I. Ansari, J.D. Grant, G.A. Woldemariam, S. Kasiri, S.S. Mandal, Org. Biomol.

Chem. 7 (2009) 926.[9] G. Roelfes, M.E. Branum, L. Wang, L. Que Jr., B.L. Feringa, J. Am. Chem. Soc. 122

(2000) 11517.[10] M. Roy, R. Santhanagopal, A.R. Chakravarty, Dalton Trans. (2009) 1024.[11] T.A. van den Berg, B.L. Feringa, G. Roelfes, Chem. Commun. (2007) 180.[12] Q. Li, W.R. Browne, G. Roelfes, Inorg. Chem. 49 (2010) 11009.[13] A. Neves, H. Terenzi, R. Horner, A. Horn Jr., B. Szpoganicz, J. Sugai, Inorg. Chem.

Commun. 4 (2001) 388.[14] S.J. Lippard, J.M. Berg, Principles of Bioinorganic Chemistry, Panima Publishing

Corporation, New Delhi, Bangalore, First Indian reprint, 1995.[15] R. Indumathy, S. Radhika, M. Kanthimathi, T. Weyhermuller, B.U. Nair, J. Inorg.

Biochem. 101 (2007) 434.[16] V.A. Kawade, A.A. Kumbhar, A.S. Kumbhar, C. Nather, A. Erxleben, U.B.

Sonawane, R.R. Joshi, Dalton Trans. 40 (2011) 639.[17] X. Liang, X. Zou, L. Tan, W. Zhu, J. Inorg. Biochem. 104 (2010) 1259.[18] C.V. Sastri, D. Eswaramoorthy, L. Giribabu, B.G. Maiya, J. Inorg. Biochem. 94

(2003) 138.[19] S. Srinivasan, J. Annaraj, P.R. Athappan, J. Inorg. Biochem. 99 (2005) 876.[20] R. Indumathy, M. Kanthimathi, T. Weyhermuller, B.U. Nair, Polyhedron 27

(2008) 3443.[21] A.C. Barve, S. Ghosh, A.A. Kumbhar, A.S. Kumbhar, V.G. Puranik, Transition Met.

Chem. 30 (2005) 312.[22] Q.L. Zhang, J.-G. Liu, H. Chao, G.-Q. Xue, L.-N. Ji, J. Inorg. Biochem. 83 (2001) 49.[23] Q.-L. Zhang, J.-G. Liu, J. Liu, G.-Q. Xue, H. Li, J.-Z. Liu, H. Zhou, L.-H. Qu, L.-N. Ji, J.

Inorg. Biochem. 85 (2001) 291.[24] P. Nagababu, S. Satyanarayana, Polyhedron 26 (2007) 1686.[25] R. Indumathy, T. Weyhermuller, B.U. Nair, Dalton Trans. 39 (2010) 2087.[26] S. Arounaguiri, B.G. Maiya, Inorg. Chem. 35 (1996) 4267.[27] M. Roy, B. Pathak, A.K. Patra, E.D. Jemmis, M. Nethaji, A.R. Chakravarty, Inorg.

Chem. 46 (2007) 11122.[28] C. Zhou, X. Du, H. Li, Bioelectrochemistry 70 (2007) 446.[29] C.H. Ng, H.K.A. Ong, K.S. Ngai, W.T. Tan, L.P. Lim, S.G. Teoh, T.S. Chong,

Polyhedron 24 (2005) 1503.[30] L.S. Kumar, H.D. Revanasiddappa, J. Coord. Chem. 20 (2011) 699.[31] G.A. McLachlan, J.G. Muller, S.E. Rokita, C.J. Burrows, Inorg. Chim. Acta 251

(1996) 193.[32] C.-C. Cheng, W.-C.H. Fu, K.-C. Hung, P.-J. Chen, W.J. Wang, Y.-J. Chen, Nucleic

Acids Res. 31 (2003) 2227.[33] B. Sreekanth, G. Krishnamurthy, H.S.B. Naik, T.K. Vishnuvardhan, B.

Vinaykumar, N. Sharath, Nucleos. Nucleot. Nucl. 30 (2011) 83.[34] E.T. Farinas, J.D. Tan, P.K. Mascharak, Inorg. Chem. 35 (1996) 2637.[35] S.K. Gupta, P.B. Hitchcock, Y.S. Kushwah, G.S. Argal, Inorg. Chim. Acta 360

(2007) 2145.[36] C. Liu, M. Wang, T. Zhang, H. Sun, Coord. Chem. Rev. 248 (2004) 147.[37] N.H. Williams, W. Cheung, J. Chin, J. Am. Chem. Soc. 120 (1998) 8079.

[38] R. Hettich, H.-J. Schneider, J. Am. Chem. Soc. 119 (1997) 5638.[39] N.A. Desai, V. Shankar, FEMS Microbiol. Rev. 26 (2003) 457.[40] S.A. Martin, R.C. Ullrich, W.L. Meyer, Biochim. Biophys. Acta 867 (1986)

67.[41] J.A. Strickland, L.G. Marzilli, J.M. Puckett Jr., P.W. Doetsch, Biochemistry 30

(1991) 9749.[42] T. Gunnlaugsson, M. Nieuwenhuyzen, C. Nolan, Polyhedron 22 (2003) 3231.[43] E.S. Rangarajan, V. Shankar, FEMS Microbiol. Rev. 25 (2001) 583.[44] S. Palit, A. Sharma, G. Talukder, Bot. Rev. 60 (1994) 149.[45] B.L. Stoddard, Q. Rev. Biophys. 38 (2005) 49.[46] I. Ott, K. Schmidt, B. Kircher, P. Schumacher, T. Wiglenda, R. Gust, J. Med. Chem.

48 (2005) 622.[47] K. Ghosh, N. Tyagi, P. Kumar, Inorg. Chem. Commun. 13 (2010) 380.[48] K. Ghosh, P. Kumar, N. Tyagi, Inorg. Chim. Acta 375 (2011) 77.[49] K. Ghosh, P. Kumar, N. Tyagi, U.P. Singh, N. Goel, Inorg. Chem. Commun. 14

(2011) 489.[50] K. Ghosh, P. Kumar, N. Tyagi, U.P. Singh, N. Goel, A. Chakraborty, P. Roy, M.C.

Barratto, Polyhedron 30 (2011) 2667.[51] K. Ghosh, P. Kumar, N. Tyagi, U.P. Singh, V. Aggarwal, M.C. Baratto, Eur. J. Med.

Chem. 45 (2010) 3770.[52] K. Ghosh, N. Tyagi, P. Kumar, U.P. Singh, N. Goel, J. Inorg. Biochem. 104 (2010)

9.[53] G.M. Sheldrick, SHELXTL-NT 2000 version 6.12, Reference Manual, University of

Gottingen, Pergamon, New York, 1980.[54] B. Klaus, Germany DIAMOND, Version 1.2 c, University of Bonn, 1999.[55] K. Delany, S.K. Arora, P.K. Mascharak, Inorg. Chem. 27 (1988) 705.[56] T. Kajiwara, R. Sensui, T. Noguchi, A. Kamiyama, T. Ito, Inorg. Chim. Acta 337

(2002) 299.[57] A. Pui, I. Berdan, I. Morgenstern-Badarau, A. Gref, M. Perree-Fauvet, Inorg.

Chim. Acta 320 (2001) 167.[58] S. Meghdadi, M. Amirnasr, M.H. Habibi, A. Amiri, V. Ghodsi, A. Rohani, R.W.

Harrington, W. Clegg, Polyhedron 27 (2008) 2771.[59] W.J. Geary, Coord. Chem. Rev. 7 (1971) 81.[60] K. Tenza, M.J. Hanton, A.M.Z. Slawin, Organometallics 28 (2009) 4852.[61] K. Ghosh, P. Kumar, N. Tyagi, U.P. Singh, Inorg. Chem. 49 (2010) 7614.[62] C.J. Davies, J. Fawcett, R. Shutt, G.A. Solan, Dalton Trans. (2005) 2630.[63] S.H. Rahaman, H. Chowdhury, D. Bose, R. Ghosh, C.-H. Hung, B.K. Ghosh,

Polyhedron 24 (2005) 1755.[64] F. Schleife, A. Rodenstein, R. Kirmse, B. Kersting, Inorg. Chim. Acta 374 (2011)

521.[65] V. Appukuttan, L. Zhang, J.Y. Ha, D. Chandran, B.K. Bahuleyan, C.-S. Ha, I. Kim, J.

Mol. Catal. A: Chem. 325 (2010) 84.[66] G.R. Desiraju, T. Steiner, The Weak Hydrogen Bond in Structural Chemistry and

Biology, Oxford University Press, New York, 1999.[67] W.J. Shi, L. Hou, D. Li, Y.G. Yin, Inorg. Chim. Acta 360 (2007) 588.[68] J. Chambers, B. Eaves, D. Parker, R. Claxton, P.S. Ray, S.J. Slattery, Inorg. Chim.

Acta 359 (2006) 2400.[69] K.T. Potts, D.A. Usifer, A. Guadalupe, H.D. Abruna, J. Am. Chem. Soc. 109 (1987)

3961.[70] A.R. Guadalupe, D.A. Usifer, K.T. Potts, H.C. Hurrell, A.-E. Mogstad, H.T. Abruna,

J. Am. Chem. Soc. 110 (1988) 3462.[71] G.D. Storrier, K. Takada, H.D. Abruna, Inorg. Chem. 38 (1999) 559.[72] B. Maity, S. Gadadhar, T.K. Goswami, A.A. Karande, A.R. Chakravarty, Dalton

Trans. 40 (2011) 11904.[73] J.R. Anacona, M. Azocar, O. Nusetti, C. Rodriguez-Barbarin, Transition Met.

Chem. 28 (2003) 24.[74] J.R. Anacona, D. Lorono, M. Azocar, R. Atencio, J. Coord. Chem. 62 (2009) 951.[75] V.G. Vaidyanathan, B.U. Nair, J. Inorg. Biochem. 94 (2003) 121.[76] F. Arjmand, M. Aziz, Eur. J. Med. Chem. 44 (2009) 834.[77] S. Ghosh, A.C. Barve, A.A. Kumbhar, A.S. Kumbhar, V.G. Puranik, P.A. Datar, U.B.

Sonawane, R.R. Joshi, J. Inorg. Biochem. 100 (2006) 331.[78] F. Dimiza, A.N. Papadopoulos, V. Tangoulis, V. Psycharis, C.P. Raptopoulou, D.P.

Kessissoglou, G. Psomas, Dalton Trans. 39 (2010) 4517.[79] Q.-X. Li, Q.-H. Luo, Y.-Z. Li, M.C. Shen, Dalton Trans. (2004) 2329.[80] J.S. Pap, B. Kripli, T. Varadi, M. Giorgi, J. Kaizer, G. Speier, J. Inorg. Biochem. 105

(2011) 911.[81] O. Iranzo, Bioorg. Chem. 39 (2011) 73.