Journal of Inorganic Biochemistry 103 (2009) 1221–1227

Contents lists available at ScienceDirect

Journal of Inorganic Biochemistry

journal homepage: www.elsevier .com/locate / j inorgbio

Chemical, structural and biological studies of cis-[Pt(3-Acpy)2Cl2]

Tian Tian a, Ilpo Mutikainen b, Gilles P. van Wezel a, Nuria Aliaga-Alcalde a,c, Jan Reedijk a,*

a Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, P.O. Box 9502, 2300 RA Leiden, The Netherlandsb Laboratory of Inorganic Chemistry, Department of Chemistry, University of Helsinki, P.O. Box 55 (A.I.Virtasen aukio 1), 00014 Helsinki, Finlandc ICREA and Universitat de Barcelona, Facultat de Química, Martí i Franquès, 1-11, 08028 Barcelona, Spain

a r t i c l e i n f o

Article history:Received 6 March 2009Received in revised form 29 June 2009Accepted 29 June 2009Available online 3 July 2009

Keywords:Platinum compound3-AcetylpyridineCircular dichroismCytotoxicityAntibacterial studies and DNA studies

0162-0134/$ - see front matter � 2009 Elsevier Inc. Adoi:10.1016/j.jinorgbio.2009.06.009

* Corresponding author.E-mail address: reedijk@chem.leidenuniv.nl (J. Ree

a b s t r a c t

Recent developments in the field of platinum anticancer drugs have revealed that compounds containingderivates of pyridine may exhibit highly cytotoxic activity against a variety of tumor cells, with AMD473(cis-PtCl2(NH3)(2-methylpyridine)) as one of the most relevant examples. Following these advances, thispaper describes the synthesis, characterization and X-ray structure of the square-planar compound cis-[Pt(3-Acpy)2Cl2] (1, Acpy stands for acetylpyridine), where the coordination of 3-acetylpyridine takesplaces through the pyridine nitrogen of the ligand. The structural arrangement of this compound is highlypeculiar and it is the first example with two of these 3-acetylpyridine molecules in a cis disposition. Inaddition, the anticancer and antibacterial activities of this compound together with studies of DNA bind-ing are also described in detail, with selective activity of compound 1 against A2780R cells. cis-[Pt(3-Acpy)2Cl2] apparently coordinates to the DNA double helix upon exchange of at least one of the Cl� ionswith the media and shows very interesting bacteriolytic and bacteriostatic activity against Escherichia coliand Streptomyces, respectively.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction

Since the early nineteen eighties exhaustive efforts have beenmade in developing analogs of cisplatin following the archetypeof structure-activity relationships (SARs) [1] to broaden the spec-trum of activities of platinum-based pharmaceuticals as well asto improve their therapeutic properties [2,3]. In general, however,it was found that direct structural analogs of this drug did neitheralter the biomedical response nor improve the efficacy at the bio-logical level, mainly due to their similar array of adducts formedwith DNA [4,5].

No more than two decades ago, Hollis and Farrell first reportedexamples of platinum-based anticancer agents that defy the classicrules of SARs [6,7] and later many of other platinum compoundsthat disregard these principles have been found to act as potentialdrugs and shown to be even active in resistant-cancer cells to cis-platin. Currently, the scientific community proceeds with a num-ber of different approaches in the design of novel platinum-baseddrugs that differ from those of the existing cisplatin analogs, thisway: oxidation state +4 (produgs) [8], trans isomers [6], chargedPt(II) compounds [9], and mixed-metal clusters [10] to interact ina classical, coordination or non-covalent way to DNA are some ofthe tactics used nowadays in the battle against cancer [11].

ll rights reserved.

dijk).

Recently, a new class of compounds has been synthesized bysubstituting at least one of the NH3 groups for a sterically moredemanding ligand, usually a heterocyclic amine like pyridine[12]. Even though this adjustment may not seem a priori overlydrastic, it has been shown that such bulky ligands have the poten-tial to overcome resistance. One of the most successful examples ofthis family of compounds is AMD473 (cis-[PtCl2(NH3)(2-methyl-pyridine)]) that entered clinical trials and has shown to be partic-ularly helpful in the treatment of ovarian cancer resistant tocarboplatin [12,13].

The pyridine ring is a basic, structural fragment in numerouscompounds with primary applications, playing a key role in severalbiological processes. Currently, many synthetic pyridine deriva-tives function as therapeutic agents and agrochemicals [14–16].In this group, acetylpyridine derivatives have been found to be veryversatile molecules; for instance, an interesting class of bioactivecompounds derived from 2-acetylpyridine have been fully studieddue to their recorded antibacterial, antimalarial, antiviral, and anti-neoplastic activities [17,18].

Similarly, 3-acetylpyridine acts as a precursor in the synthesisof some analgesics and as a cofactor for depleting detoxification in-side the cells [19,20]. In addition, there are several structures in theliterature where this molecule functions as a ligand attached to 3dand 4f metals of different architectures [21].

Despite all of the above, not much is known about the coordina-tion chemistry of 3-acetylpyridine and Pt and the biological effectsof the coordination compounds. Motivated by this synergy and its

1222 T. Tian et al. / Journal of Inorganic Biochemistry 103 (2009) 1221–1227

potential bioactivity we became interested in the development ofantitumor and/or antibacterial drugs using both Pt and 3-acetyl-pyridine (3-Acpy). In an attempt to further explore this hypothesis,a new compound, cis-[Pt(3- Acpy)2Cl2] (1), was synthesized andcharacterized. Meischer et al. [22] mentioned this complex in aprevious report, however no synthetic description or further char-acterization was shown. Our investigations on this compound havebeen directed towards the antitumor activity in vitro and the bind-ing studies with DNA. The present work aims to contribute to theunderstanding of the structural and reactivity characteristics ofcompound 1.

2. Materials and methods

2.2. Chemistry

K2PtCl4. was obtained from a loan scheme with Johnson andMatthey and 3-acetylpyridine was purchased from Aldrich andused without further purification. The synthetic procedure wasperformed in the absence of light and the final product was keptaway from light.

Infrared spectra were recorded from 4000 to 300 cm�1on Per-kin–Elmer Paragon 1000 FTIR spectrometer equipped with a Gold-en Gate ATR device, using the reflectance technique. 1H NMR and195Pt NMR experiments were performed on a Bruker DPX 300 spec-trometer, with measurements carried out at room temperature ind6MSO. Elemental analysis of C, H, N and Pt analysis were carriedout on a Perkin–Elmer 2400 series II analyzer. Electrospray ioniza-tion-mass spectra (ESI-MS) were achieved from dmso and recordedon a Thermo Finnigan AQA apparatus.

2.3. cis-3-Acetylpyridinedichloridoplatinum (1)

3-Acetylpyridine (20.3 ll, 0.18 mmol) was added dropwise to aslurry of 25.3 mg (0.06 mmol) of K2PtCl4 dissolved in 3.0 ml of H2O.The mixture was stirred for 24 h in a complete darkness. Theresulting light yellow precipitate was filtered, washed with ice-cold H2O, air-dried and protected from the light. Yellow crystalswere obtained by re-crystallization from dichloromethane anddiethyl ether. Yield: 46%; IR (m/cm�1): 3056w, 1694vs, 1603m,1574m, 1431m, 1358m, 1264s, 1197m, 1064m, 1029w, 965m,831m, 807s, 692vs, 602s, 590s, 334s, 345s, 319m; EA: forPtCl2C14H14N2O2, Anal. (Calcd)% C 32.80 (33.08),% H 2.16 (2.78),%N 5.25 (5.51); 1H NMR (d6-dmso, d/ppm): 9.37 (singlet (s), 1H,H4), 9.01 (s, 1H, H3), 8.44 (s, 1H, H1), 7.62 (s, 1H, H2), 2.26 (m,3H, CH3) (Fig. 1); 195Pt NMR (d6-dmso, d/ppm): �2844 ppm; ESI-

∗

∗

3-acetylpyridine

Compound 1

10.0 3.09.0 7.0 6.0 5.0 4.0 2.08.0 0.0 1.0

Fig. 1. 1H NMR of cis-[Pt(3-Acpy)Cl2] recorded in dmso-d6. � – Solvent, N – watercontained in the deuterated solvent and d – impurity. The dotted lines highlight theshift between the peaks of the free (top) and coordinated ligand (bottom).

MS: m/z: 507.71 (Pt(3-Acpy)2Cl2) [M-1e�]+, 550.67 (Pt(3-Ac-py)2Cl(dmso))+ [M]+.

2.4. Cytotoxicity studies

The A2780 and A2780R cells (human ovarian cancer cells sensi-tive and resistant to cisplatin, respectively) were generously pro-vided by Dr. J.M. Perez (Universidad Autónoma de Madrid,Spain). The cells were grown as monolayers in Dulbeccos modifiedEagles Medium supplemented with 10% fetal calf serum (Gibco,Paisley, Scotland), penicillin (100 units/ml: Dufecha, The Nether-lands) and streptomycin (100 lg/ml: Dufecha, The Netherlands).The L1210 and L1210R cell lines (mouse leukemia cells sensitiveand resistant, in that order) were cultured in McCoys 5a mediumsupplemented with 10% fetal calf serum (Gibco, Paisley, Scotland),penicillin (100 units/ml: Dufecha, The Netherlands) and strepto-mycin (100 lg/ml: Dufecha, The Netherlands). The rest of cell linesdescribed in this paper were studied by Pharmachemie: MCF7 (hu-man breast carcinogenic cell line), EVSA-T (human breast carcino-genic cell line), WIDR (human colon carcinogenic cell line), IGROV(human ovarian carcinogenic cell line), M19 (human melanomacarcinogenic cell line), A498 (human lung carcinogenic cell line)and H226 (human renal carcinogenic cell line).

For the cell growth assay, cells were pre-cultured in 96 multi-well plates (1500–2000 cells/well) for 48 h at 37 �C in a 5% CO2-containing incubator and subsequently treated with 45 ll of thesolution of the compound in triplicate. The stock solution of thesamples were made in dmso and later diluted with full mediumin such a way that the total amount of dmso in solution did not ex-ceeded more than 1% in each cell plate (the compound was addedin different concentrations from 0.1 lM to 90 lM). After 48 h incu-bation, 50 ll of a 5 mg/ml MTT ((3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) solution in PBS (phosphate bufferedsaline) was added to each well, and allowed to develop at 37 �C for2 h. The medium was then carefully removed and 100 ll of dmsowas added to each well. The absorbance of the resulting purplesolutions was measured at 590 nm using a Biorad 550 microplatereader. The cytotoxicity is mainly determined by the IC50 valuesof each compound. The IC50 values are drug concentrations that in-hibit cell growth for 50% with respect to control. The lower the IC50

value, the better antitumor activity can be expected. The numberswere determined graphically using Graphpad Prism analysissoftware.

2.5. Thermal denaturation of DNA

To prepare platinum adducts with double-stranded calf-thymusDNA (CT-DNA), a platinum stock solution was prepared by dissolv-ing 3 mg platinum compound (compound 1) with 3-acetylpyridinein 6 ml Tris buffer pH = 7.2. To a mixture of a fixed amount of10�4 M CT-DNA in 10 mM Tris buffer were added predeterminedquantities of platinum stock solution to achieve different R values.R = [DNA]/[Pt-compound]; it is a ratio of DNA concentration innucleotide phosphate units per metal). Together, they were incu-bated for 24 h at 37 �C after which 150 ll samples will be removedto the cuvettes with 500 ll paraffin being added last. The solutionin the cuvettes was heated from 20 �C to 95 �C within 2 h in orderto achieve the thermal hydrolysis. Temperature absorbance pro-files were automatically recorded on a Varian Cary 300 Bio instru-ment, equipped with a Cary temperature controller attached to aUV–vis spectrophotometer. The equipment contains a cell blockwhich is electrically heated, the temperature being programmedto change at a constant rate of about l �C/min. The concentrationof the solutions of nucleic acid was determined spectro-photomet-rically at the wavelength of the maximum of absorption, at260 nm.

Table 1Crystal data and structure refinement for cis-[Pt(3-Acpy)2Cl2].

Compound cis-[Pt(3-Acpy)2Cl2]

Empirical formula C14H14N2O2PtCl2

Formula weight 508.26Temperature (K) 173(2)Wavelength (Å) 0.71073Crystal system, space group Monoclinic, P21/nUnit cell dimensionsa (Å) 7.9160(16)b (Å) 15.852(3)c (Å) 12.380(3)a (�) 90b (�) 94.11(3)c (�) 90Volume (Å3) 1549.5(5)Z, Dcalc (Mg/m3) 4, 2.179Absorption coefficient (mm�1) 9.404F(000) 960Crystal size (mm) 0.40 � 0.06 � 0.06h range for data collection (�) 2.96 � 27.00Limiting indices �10 6 h 6 10, �20 6 k 6 20, �15 6 l 6 15Reflections collected/unique [Rint] 24,541/3375 [0.0452]Completeness to h = 27.00 (%) 99.6Absorption correction Semi-empirical from equivalentsMax. and min. transmission 0.6023 and 0.1169Refinement method Full-matrix least-squares on F2

Data/restraints/parameters 3375/0/192Goodness-of-fit on F2 1.160Final R indices [I > 2r(I)] R1 = 0.0199, wR2 = 0.0491R indices (all data) R1 = 0.0257, wR2 = 0.0547Largest diff. peak and hole (e �3) 0.715 and �2.067

T. Tian et al. / Journal of Inorganic Biochemistry 103 (2009) 1221–1227 1223

2.6. Circular dichroism

Solutions were made by mixing CT-DNA and the compound 1 atdifferent R ratios (defined above) varying the concentration of Ptcompound, meanwhile the sample of CT-DNA was kept constantat a concentration of 10�4 M. The used R values were in a rangeof 5, 2, 1 and 0.7. They were incubated for 24 h at 37 �C. Circulardichroism (CD) measurements were obtained on a JASCO J-815CD spectrometer. Cell path-lengths, time constants, scanningspeeds, and sensitivity settings were varied to obtain the highestpossible signal-to-noise ratio. In some cases, to increase the sig-nal-to-noise ratio, multiple scanning was done using the data pro-cessor, which accompanies the JASCO J-815. Calibration of thespectrometer was done before each scan with a standard solutionof Tris buffer as background. The average of five scans was usedfrom 400 nm to 200 nm in the determination of each spectrum.Absorption spectra were obtained on Cary 17 and Cary 18 scanningspectrophotometers. All scans in CD and absorbance measure-ments were done at 37.0 ± 0.4 �C. The temperature was controlledby the use of a JASCO PTC-423S. A thermocouple (Yellow SpringsInstrument Co.) was placed in the cell for calibration of the temper-ature in the solution. Difference spectra were obtained using 1 cmtandem cuvettes (Pyrocell Mfg. Co.) placed in a thermostable cellholder designed for a Cary 17. All solutions were accurately madeup using volumetric pipettes.

2.7. UV–vis titration

UV–vis titration experiments were performed in order to clearlydiscern the behavior of the Pt compound and CT-DNA at differentratios and also to support previous CD analysis. Tris buffer (1 MTris–HCl pH = 7.2, 1.5 M NaCl) was used in all the solutions andvariations on the amount of platinum compound vs. a constantconcentration of DNA were accomplished (R ratio from 5 to 0.3).Data was collected on Varian Cary 50 UV–vis Spectrophotometerwith new Fiber Optic Dip Probe accessory. The Cary 50 utilizes avery long-life Xenon source lamp with wavelength range between190 nm and 1100 nm as well as photometric range of ±3.3 A.

2.8. Antibacterial studies

Antibacterial activity of compound 1 and cisplatin were testedat different concentrations using the Gram-negative bacteriumEscherichia coli and the Gram-positive bacterium Streptomyces coe-licolor as indicator strains (Table 4). 100 ll suspensions of E. coliJM109 or Streptomyces coelicolor M145 containing approximately106 cfu (colony forming units) were mixed with 3 ml of soft agar(LB with 0.6% bactoagar) and plated on LB agar plates (for E. coli)or minimal medium agar plates (for Streptomyces), respectively[23]. Sterile filter discs containing 10 ll sample solutions werepositioned on the plates. The plates were incubated at 30 �C for16–24 h. Antimicrobial activity was assessed as the zone of growthinhibition around the filter discs. For a detailed description see Royet al. [24].

2.9. X-ray crystal structure determination details

Intensity data for single crystal was collected at 173 K using MoKa radiation (k = 0.71073 Å) on a Nonius KappaCCD diffractometer.The crystal was mounted using oil-drop method [25]. Crystal andrefinement data for the compound is in Table 1. The intensity datawas corrected for Lorentz and polarization effects and for absorp-tion (multiscan absorption correction [26]). The structures wassolved by direct methods [27]. The programs EvalCCD [28], SHEL-XS97 [27] and SHELXL97 [29] were used for data reduction, struc-ture solution and refinement, respectively. All non-hydrogen

atoms were refined with anisotropic displacement parameters.All hydrogen atoms were placed at calculated positions and wererefined riding on the parent atoms.

3. Results and discussion

3.1. Synthesis and spectroscopic characterization of cis-[Pt(3-Acpy)2Cl2]

Compound 1 was readily obtained by reacting K2PtCl4 and theligand, 3-acetylpyridine, in water in a 1:3 ratio. Earlier attemptsto perform the same reaction in DMF led to the formation of oils,complicating purification and resulting in very low yields. Instead,with H2O as a solvent a light yellow precipitate was formed and itscrystallization was possible by layering with CH2Cl2/diethyl ether.Modification of the metal-to-ligand ratio affected the yield, and theratio of 1:3 was found to be optimal. Clear indications of coordina-tion of the ligand with the metal were observed studying the infra-red spectrum, where the mpy shifted from 692 cm�1 in the freeligand to 702 cm�1 in the final compound. Furthermore, the mC@O

stretching moved slightly from 1683 to 1694 cm�1, and a new, in-tense and sharp band was observed at 334 cm�1 and assigned tothe mPt–Cl stretch [30]. The spectroscopic characterization in solu-tion was performed using dmso-d6 as a solvent. The 1H NMR spec-trum of the compound in deuterated dmso solution shows fourpeaks within the aromatic region (Fig. 1) at 7.6, 8.4, 9.0 and9.4 ppm all shifted from those found for the free 3-acetylpyridine(7.5, 8.3, 8.8 and 9.1 ppm, respectively).

The methyl group signals are overlapped by the deuterated sol-vent in both, compound 1 and the free ligand. In principle, thesmall number of peaks observed can be explained by the formationof a solution species of high symmetry containing one, two or evenfour molecules of 3-acetylpyridine; moreover, in the case of two 3-acetylpyridine the compound could be cis or trans. In fact 195PtNMR performed in dmso-d6 provided more insight about the plat-inum coordination. A single peak at �2844 ppm was obtained

Table 2Selected bond lengths (Å) and angles (�) for cis-[Pt(3-Acpy)2Cl2].

Pt(1)–N(21) 2.018(3) N(21)–Pt(1)–N(11) 90.73(12)Pt(1)–N(11) 2.033(3) N(21)–Pt(1)–Cl(2) 87.81(8)Pt(1)–Cl(1) 2.2988(9) N(11)–Pt(1)–Cl(2) 178.31(8)Pt(1)–Cl(2) 2.2868(9) Cl(2)–Pt(1)–Cl(1) 90.13(4)N(11)–C(12) 1.351(4) C(16)–N(11)–C(12) 118.5(3)N(21)–C(22) 1.346(4) C(16)–N(11)–Pt(1) 120.9(2)C(17)–O(18) 1.206(5) N(11)–C(12)–C(13) 122.3(3)C(12)––C(13) 1.386(5) N(11)–C(12)–H(12A) 118.9(3)C(13)–C(17) 1.510(5) C(14)–C(13)–C(17) 120.2(3)C(17)–C(19) 1.479(6) O(18)–C(17)–C(13) 118.6(4)C(19)–H(19A) 0.9800(5) H(19A)–C(19)–H(19B) 109.5(3)

Fig. 3. Ball and stick view and used atomic numbering of cis-[Pt(3-Acpy)2Cl2].

1224 T. Tian et al. / Journal of Inorganic Biochemistry 103 (2009) 1221–1227

which corresponds to a platinum surrounded by two nitrogen, onechloride anion and one sulfur atoms [31].

This way, the possibilities of one or four 3-Acpy ligands coordi-nated to the Pt were excluded. In a parallel way, ESI+ (Fig. 2) wasfound a very powerful tool that supported the existence of a com-pound with two chlorides as ligands and two 3-Acpy ligands, aswell as [Pt(3-Acpy)2Cl(dmso)]+ species where a chloride anionwas substituted for a molecule of solvent, as the previous 195PtNMR indicated. Elemental analysis of the powdered sample alsoconfirmed the ratio 1:2 (M:L) and finally the crystallographic dataallowed the identification of the cis isomer (vide infra).

3.2. Structure description from X-ray diffraction

cis-[Pt(3-Acpy)2Cl2] (1) crystallizes in the monoclinic spacegroup P21/n with an asymmetric unit composed by four of thesemolecules. X-ray crystallographic data and details of the refine-ment of the structure are listed in Table 1, and selected bondlengths and angles are given in Table 2. Compound 1 exhibits asquare-planar configuration with angles close to the ideal valuesof 90� and 180�, as Fig. 3 shows. Pt–Cl bond lengths range from2.287 to 2.298 Å, close to the expected values and Pt–N bondlengths (2.018 and 2.033 Å) are also comparable to those of relatedstructures on the literature [24,32]. One of the most significant fea-tures of this structure is the orientation of the 3-acetylpyridinerings with respect to the platinum square plane. Both aromatic li-gands are symmetrically not-related and tilted by angles of 36.6�and 51.2�, respectively. In addition, the two acetyl groups are notcoplanar with the pyridine ring with angles of 6.32� and 9.96�.

The arrangement of these coordination entities in the crystaldoes not allow standard hydrogen bonds; however, the overallpacking appears to be stabilized by weak C–H� � �O hydrogen bondstogether with weak C–H� � �Cl interactions [33,34]. The details ofthis are given in the supporting information (Figs. S1–S3).

A view along the b-axis of the crystal lattice is shown in Fig. S3.Compound 1 stacks in columns where the platinum atoms define azigzag arrangement along the piles. However, this arrangementdoes not involve close Pt� � �Pt interactions. The stacks show inter-molecular distances of 3.645 and 4.557 Å periodically repeatedand both longer than the upper distance limit of �3.5 Å that usu-ally indicates d2

z ðPtÞ � dz2ðPtÞ interactions. Generally, mononuclear

256.83 507.71

550.67

549.66

550.67

552.68

m/z250 300 350 400 450 500 550 595

Fig. 2. ESI+ spectrum of compound 1. This result shows clearly the isotope featuresof the platinum compound, as well as the cationic mass of two species: [Pt(3-Acpy)2Cl2]+ (507) and [Pt(3-Acpy)2Cl(dmso)]+ (550), respectively.

platinum species with metal–metal and ligand–ligand interactionsexhibit unusual colors [35]. The absence of color in the crystal cor-roborates the lack of d2

z interactions.A CSD search has shown that this is only the second crystal

structure incorporating 3-acetylpyridine and Pt together. Tessierand coworkers only recently communicated the structure of a Ptcompound with two molecules of 3-acetylpyridine, though in atrans mode and with two additional iodide ligand (trans-[Pt(3-Ac-py)2I2]) [36]. Comparing both structures some patterns were foundsimilar, as for example, the absence of standard hydrogen bondsand the non-coplanar arrangement of the acetyl groups and thepyridine rings.

It should be emphasized that this compound is the first cis-Plat-inum-(3-Acpy) species crystallographically characterized in the lit-erature. Previous attempts to synthesize cis-[Pt(3-Acpy)2I2]resulted in the trans-isomer and biological activity studies werenot made available [36]. However, work performed by Meischerand coworkers [22] as early as 1976 described the anticancer prop-erties of 46 cis-Platinum-amine derivates against L1210 leukemiamice cells, also mentioning cis-[Pt(3-Acpy)2Cl2] among these com-pounds, albeit without presenting supplementary structural char-acterization. In 1979 Fazakerley [21] described the NMR studiesof cis-[Pt(NH3)2(3-Acpy)2]X2 species (X = Cl� and ClO�4 ), but nocrystal structure was reported. So, the present study was devel-oped with the aim of portraying a comprehensive picture of boththe detailed characterization of this novel compound and alsothe evaluation of its activity in vitro.

3.3. Biological assays and biophysical studies

3.3.2. Growth inhibition assayThe in vitro cytotoxicity of cis-[Pt(3-Acpy)2Cl2] was initially

tested using the available cell lines (A2780, L1210/0 and their cor-responding cisplatin resistant analogs A2780R and L1210/2) andsubsequently in a number of additional cell lines (A498, EVSA-T,H226, IGROV, M19, MCF-7 and WIDR) by Pharmachemie (Haarlem,The Netherlands). IC50 values (compound concentration in lMwhich induces 50% cell death) of 1 were always compared with

240 260 280 300 320 340-10

-5

0

5

10

CD

/ m

deg

Wavelength / nm

R=5.0 R=2.0 R=1.0 R=0.7 Free DNA

Fig. 4. CD graph of compound 1 and DNA at different R values in the range of 230–260 nm. The arrows indicate the decrease of R values.

T. Tian et al. / Journal of Inorganic Biochemistry 103 (2009) 1221–1227 1225

those of cisplatin for all cell lines and data are summarized in Table3. Growth inhibition by the compound was determined using theMTT-based assay at Leiden University and the SRB cell viability testfor the rest.

The lack of activity of cis-[Pt(3-Acpy)2Cl2] in the murine leuke-mia cell line was previously reported in vivo [22]. The present workshows similar results in vitro for this L1210/0 cell line and its resis-tant analog, L1210/2. However, compound 1 exhibits a similaractivity as cisplatin towards A2780R (Fig. S4); but its RF (resistantfactor), defined as IC50(A2780R)/IC50(A2780) is lower (1.3 for 1 vs.4.25 for cisplatin). For the rest of the cell lines, the results in vitroindicate that compound 1 has a relatively low cytotoxic activity.

3.3.3. Thermal denaturation of DNAB-type DNA of calf thymus (CT-DNA) with a size of approxi-

mately 2000 base pairs was employed to study the nature of theinteractions of cis-[Pt(3-Acpy)2Cl2] with DNA. The thermal processof denaturalization of the resulting adducts (CT-DNA-1) started at20 �C and finished at 95 �C, heating the samples up at intervals of1 deg/min. The melting temperature values (Tm, where more than50% of the sample is denaturalized) could be precisely achievedfrom the first derivative of the thermal spectra. Comparativeblanks of free CT-DNA in the absence of compound 1 were also per-formed at similar conditions. In addition, different R ratios werestudied to measure the effects of Pt concentration on thermal sta-bility. In these experiments, the concentration of CT-DNA was keptconstant varying only the concentration of compound 1. The final Rvalues used were 0.3, 0.4, 0.5, 0.7, 1.0, 2.0 and 5.0 (the lower the Rvalues, the higher the concentration of platinum compound insolution). All samples were incubated and equilibrated at 37 �Cfor 24 h previous to the measurements.

As a result, the blanks (0.1 mM of free CT-DNA in 5 mM Tris buf-fer pH 7.2) denatured at around 70 ± 2 �C and all the Tm valuesfound for the adducts were lower than this value. In addition, itwas observed that the higher the concentration of Pt in the solu-tion (low Rs) the lower the observed Tm. In fact, the melting tem-peratures of at R = 0.3 and 0.4 were the lowest (50 �C and 53 �C,respectively). These experiments showed that the thermal stabilityof the studied DNA samples decreased more than 15� after theaddition of an excess of cis-[Pt(3-Acpy)2Cl2]. Related to these re-sults, the literature [37,38] shows that the formation of cross-linksbetween Pt compounds and DNA may reduce the double-helicalstability. In particular, cisplatin exhibits a similar behavior due tothe formation of intra-strand cross-links that destabilize double-helical DNA [38].

3.3.4. Circular dichroismCircular dichroism (CD) is a useful technique for the assessment

of DNA-binding mode and type of affinity of ligand–DNA interac-tions [39–41]. With this aim in mind, different concentrations ofcompound 1 were incubated with calf-thymus DNA (constant con-centration =10�4 M) for 24 h in a Tris buffer of pH = 7.2 and the va-lue of R was varied (vide supra) and their spectra recorded from220 nm to 300 nm.

Fig. 4 shows the CD experiments of free CT-DNA and [CT-DNA-1] solutions with R = 0.7, 1, 2 and 5, respectively. As it was ex-pected, the curve found for the free DNA show a positive band at

Table 3IC50 values of compound 1 and cisplatin of a variety of cell lines.

Compounds A498 EVSA-T H226 IGROV M19

1 66 17 81 21 52Cisplatin 8 1.4 11 0.6 2

The error associated to these numbers is up to 5%.

278 nm and a negative band at 241 nm. In principle, the presenceof compound 1 in solution induced changes in the ellipticity ofthese positive and negative bands exhibiting hyperchromic andhypochromic shifts, respectively, within increasing the R values.These results agree with modifications in the secondary structureof DNA caused by 1. The interpretation of these CD spectra may im-ply mainly cross-links binding modes of 1 with DNA; however,additional interactions could be also involved (e.g. monofunctionalbinding); in general, changes in the negative band are interpretedas reduced repulsive electrostatic forces within the DNA molecule,due to the presence of the cross-linked Pt–DNA fragments, whilethe changes in the positive band are related to the unstacking ofbase pairs at the adduct sites [39]. The complexity of these resultsagrees with the UV–vis titration studies described below.

3.3.5. UV–vis titrationUV–vis spectra of several solutions containing CT-DNA (10�4 M)

and compound 1 at different ratios were prepared with Tris bufferat pH = 7.2 and measured in the wavelength interval of 200–400 nm.

As Fig. 5 shows, an absorption band appearing at 258 nm (solidline) is typical for free DNA. At lower concentration of compound 1in the solution (R > 1; R = [DNA]/[Compound]) a smooth increase isobserved in the intensity of this absorbance band; however, athigher concentrations of compound 1 (R = 0.3, 0.7 and 1), a hypo-chromic shift of the band at 258 nm was observed and a distinc-tive, new band appears at around 225 nm. Variations in intensityand shift of the absorption bands indicated major changes in theDNA’s conformation consistent with the CD data shown previously.

3.4. Bacterial studies

The antimicrobial activity of compound 1 relative to that of cis-platin was assessed using E. coli and S. coelicolor as indicator strains(Table 4). As described in the Methods section, the antimicrobialactivity was measured as the zone of growth inhibition effected

MCF-7 WIDR A2780 A2780R L1210/0 L1210/2

57 22 20 26 >100 >1002 3 4 17 4 25

200 250 300 350 4000.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

Free DNA R=0.3 R=0.7 R=1.0 R=2.0 R=5.0

Abs

orba

nce

Wavelenght / nm

Fig. 5. UV–vis spectra of compound 1 and DNA at different R values in the range of200–400 nm. The arrows indicate the growing of R values.

Table 4Antibacterial data for compound 1 and cisplatin.

Compounds Concentration(mg/ml)

Solvent Ø E. coli (mm) Ø S. coelicolor (mm)

1 16.5 dmso 9 8a

Cisplatin 1.7 H2O 10 0Blank – dmso 0 0

a Growth inhibition (bacteriostatic activity). Zones were measured as mm out-side the filter discs (5 mm). dmso without compound was used as the control.

1226 T. Tian et al. / Journal of Inorganic Biochemistry 103 (2009) 1221–1227

by the compound in a filter disc assay [24]. cis-[Pt(3-acetylpyri-dine)2Cl2] presented bactericidal activity against E. coli, with aclearing zone of 15 mm (9 mm outside the filter disc) where bacte-ria are killed by the complex. Interestingly, against S. coelicolor anopaque zone was formed, indicating that while growth was inhib-ited, the cells did not die, suggesting compound 1 has bacterio-static activity against Gram-positive bacteria. As a control,cisplatin was also tested; while it exhibited a significantly higheractivity against the Gram-negative E. coli cells (same zone withtenfold lower concentration, Table 4), it had no effect in Gram-po-sitive bacteria.

The protective mechanism of bacteria against outside agentshas a complex interpretation and is a matter of exhaustive investi-gation nowadays. Nevertheless, the present experiment shows thatcisplatin and compound 1 act in different ways against bacteria,being the former more active against Gram-negative bacteria butunable to act against Gram-positive, meanwhile cis-[Pt(3-acetyl-pyridine)2Cl2] seems to be less aggressive in its interaction withGram-negative but more effective against the growth of Gram-po-sitive. This study exposes the in vitro differences between bothcompounds as well as the rest of biological experiments describedabove.

4. Concluding remarks

The research on the synthesis and characterization of the newplatinum complex, cis-[Pt(3-Acpy)2Cl2] (1), has resulted in a prom-ising new compound. It is, to the best of our knowledge, the firstcompound that encloses two 3-Acpy molecules in a cis dispositionand also the first reported together with both extensive physicaland biological analyses are presented.

Detailed characterization of the solid compound was obtainedby using X-ray crystallography, which provided the final atomic

arrangement of this compound in the solid state. Analyses in solu-tion, i.e. 195Pt NMR and MS-ESI+, agreed with these results and re-vealed the existence of new species, cis-[Pt(3-Acpy)2Cl(dmso)]+,due to the exchange of a Cl� ion in 1 with a molecule of solvent.

In addition, CD and UV–vis titration studies clearly show thatcis-[Pt(3-Acpy)2Cl2] interacts with DNA, most likely by coordinat-ing to the double helix following on an exchange of one of theCl� ions with the media (already observed in solution in the char-acterization part), although this process may involve additional in-tra- and interstrand coordination at different ratios that have notbeen analyzed yet. Further studies of viscosity will be performedto analyze in detail the nature of these interactions. In vitro cyto-toxicity studies have shown that the title compound is activeagainst A2780 and A2780R cells, but shows relatively little activityagainst other cell lines. Moreover, this compound presents inter-esting bactericidal and bacteriostatic activity against the Gram-negative bacterium E. coli and the Gram-positive S. coelicolor,respectively, thus expanding the number of different platinumcompounds with antibacterial activity. The structural differencesbetween compound 1 and AMD473 (nature and number of the pyr-idine derivatives) are clear in the low biological activity of theformer.

At present, detailed studies on related, derivative pyridines areongoing to complete the information on platinum compounds con-taining acetylpyridine ligands; our goal is to obtain an accurate li-brary of these complexes to fully understand the effects ofsubstituents in the pyridine groups.

Acknowledgements

The authors gratefully acknowledge Dr. J.M. Perez (UniversidadAutónoma de Madrid, Spain) for providing A2780 and A2789R celllines. Dr. N.A.A. thanks to ICREA. The authors wish to thank John-son amd Matthey (Reading, UK) for their generous loan of K2PtCl4.Continuous support from The Netherlands Research Organisation(NWO) and its chemical council (CW) is also gratefullyacknowledged.

Appendix A. Supplementary material

Description of the packing of the title compound, including 3Figures, as well as IC50 graph for compound 1 using the A2780Rcell line. Crystallographic data are deposited as a CIF File, CCDC719925. These data can be obtained free of charge from The Cam-bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/da-ta_request/cif. Supplementary data associated with this articlecan be found, in the online version, at doi:10.1016/j.jinorgbio.2009.06.009.

References

[1] N. Farrell, Transition Metal Complexes as Drugs and Chemotherapeutic Agents,Dordrecht, The Netherlands, 1989.

[2] Y.W. Jung, S.J. Lippard, Chem. Rev. 107 (2007) 1387–1407.[3] J. Reedijk, Proc. Natl. Acad. Sci. USA 100 (2003) 3611–3616.[4] R.J. Knox, F. Friedlos, D.A. Lydall, J.J. Roberts, Cancer Res. 46 (1986) 1972–1979.[5] B.J. Monk, D.S. Alberts, R.A. Burger, P.T. Fanta, A.V. Hallum, K.D. Hatch, S.E.

Salmon, Gynecol. Oncol. 71 (1998) 308–312.[6] N. Farrell, T.T.B. Ha, J.P. Souchard, F.L. Wimmer, S. Cros, N.P. Johnson, J. Med.

Chem. 32 (1989) 2240–2241.[7] L.S. Hollis, A.R. Amundsen, E.W. Stern, J. Med. Chem. 32 (1989) 128–136.[8] M.D. Hall, H.R. Mellor, R. Callaghan, T.W. Hambley, J. Med. Chem. 50 (2007)

3403–3411.[9] E.G. Talman, D.P. Myers, J. Reedijk, Inorg. Chim. Acta 240 (1995) 25–28.

[10] M. Milkevitch, H. Storrie, E. Brauns, K.J. Brewer, B.W. Shirley, Inorg. Chem. 36(1997) 4534–4538.

[11] T.W. Hambley, Dalton Trans. (2007) 4929–4937.[12] Y. Chen, Z.J. Guo, S. Parsons, P.J. Sadler, Chem. Eur. J. 4 (1998) 672–676.[13] J. Holford, F. Raynaud, B.A. Murrer, K. Grimaldi, J.A. Hartley, M. Abrams, L.R.

Kelland, Anti-cancer Drug. Des. 13 (1998) 1–18.

T. Tian et al. / Journal of Inorganic Biochemistry 103 (2009) 1221–1227 1227

[14] D. Chopra, T.P. Mohan, K.S. Rao, T.N.G. Row, Acta Crystallogr. E. 60 (2004)O2415–O2417.

[15] B.H. Kim, C.H. Yin, Q.X. Guo, E.A. Bach, H. Lee, C. Sandoval, S. Jayabose, A.Ulaczyk-Lesanko, D.G. Hall, G.H. Baeg, Mol. Cancer Ther. 7 (2008) 2672–2680.

[16] J.A. Turner, M.R. Dick, T.M. Bargar, G.M. Garvin, T.L. Siddall, ACS Symp. Ser. 892(2005) 119–129.

[17] K. Onodera, N.C. Kasuga, T. Takashima, A. Hara, A. Amano, H. Murakami, K.Nomiya, Dalton Trans. (2007) 3646–3652.

[18] K. Singh, M.S. Barwa, P. Tyagi, Eur. J. Med. Chem. 42 (2007) 394–402.[19] N.A. Abdel-Latif, N.M. Sabry, A.M. Mohamed, M.M. Abdulla, Monatsh. Chem.

138 (2007) 715–724.[20] I.I. Budihardjo, D.L. Walker, P.A. Svingen, C.A. Buckwalter, S. Desnoyers, S.

Eckdahl, G.M. Shah, G.G. Poirier, J.M. Reid, M.M. Ames, S.H. Kaufmann, Clin.Cancer Res. 4 (1998) 117–130.

[21] G.V. Fazakerley, K.R. Koch, Inorg. Chim. Acta 36 (1979) 13–25.[22] S.J. Meischen, G.R. Gale, L.M. Lake, C.J. Frangakis, M.G. Rosenblum, E.M. Walker,

L.M. Atkins, A.B. Smith, J. Natl. Cancer Inst. 57 (1976) 841–845.[23] S. Rigali, F. Titgemeyer, S. Barends, S. Mulder, A.W. Thomae, D.A. Hopwood, G.P.

van Wezel, EMBO Rep. 9 (2008) 670–675.[24] S. Roy, K.D. Hagen, P.U. Maheswari, M. Lutz, A.L. Spek, J. Reedijk, G.P. van

Wezel, ChemMedChem 3 (2008) 1427–1434.[25] T. Kottke, D. Stalke, J. Appl. Crystallogr. 26 (1993) 615–619.[26] G.M. Sheldrick, in: Program for Empirical Absorption Correction, University of

Göttingen, Göttingen, Germany, 1996.[27] G.M. Sheldrick, SHELXS-97, Program for Crystal Structure Solution, University

of Göttingen, Germany, 1997.[28] A.J.M. Duisenberg, L.M.J. Kroon-Batenburg, A.M.M. Schreurs, J. Appl.

Crystallogr. (2003) 220–229.

[29] G.M. Sheldrick, SHELXL-97. Program for Crystal Structure Refinement,University of Göttingen, Germany, 1997.

[30] P. Marques-Gallego, H. den Dulk, J. Brouwer, H. Kooijman, A.L. Spek, O.Roubeau, S.J. Teat, J. Reedijk, Inorg. Chem. 47 (2008) 11171–11179.

[31] M.D. Reily, L.G. Marzilli, J. Am. Chem. Soc. 108 (1986) 6785–6793.[32] S. Moradell, J. Lorenzo, A. Rovira, S. van Zutphen, F.X. Aviles, V. Moreno, R. de

Llorens, M.A. Martinez, J. Reedijk, A. Llobet, J. Inorg. Biochem. 98 (2004) 1933–1946.

[33] M.R. Pressprich, R.D. Willett, R.J. Terjeson, R. Winter, G.L. Gard, Inorg. Chem. 29(1990) 3058–3060.

[34] F.T. Vieira, G.M. de Lima, J.L. Wardell, S. Wardell, K. Krambrock, A. Alcantara, J.Organomet. Chem. 693 (2008) 1986–1990.

[35] W.R. Caseri, H.D. Chanzy, K. Feldman, M. Fontana, P. Smith, T.A. Tervoort, J.G.P.Goossens, E.W. Meijer, A.P.H.J. Schenning, I.R. Dolbnya, M.G. Debije, M.P. deHaas, J.M. Warman, A.M. van de Craats, R.H. Friend, H. Sirringhaus, N.Stutzmann, Adv. Mater. 15 (2003) 125–132.

[36] C. Tessier, F.D. Rochon, Acta Crystallogr. Sec. C – Cryst. Struct. Commun. 62(2006) M620–M622.

[37] Y.Z. Chen, Y.L. Zhang, E.W. Prohofsky, Phys. Rev. E. 55 (1997) 5843–5848.[38] A. Kagemoto, H. Takagi, K. Naruse, Y. Baba, Thermochim. Acta 190 (1991) 191–

201.[39] N.C. Garbett, P.A. Ragazzon, J.B. Chaires, Nat. Protoc. 2 (2007) 3166–3172.[40] J. Ruiz, J. Lorenzo, C. Vicente, G. Lopez, J.M. Lopez-De-Luzuriaga, M. Monge, F.X.

Aviles, D. Bautista, V. Moreno, A. Laguna, Inorg. Chem. 47 (2008) 6990–7001.[41] J. Ruiz, M.D. Villa, N. Cutillas, G. Lopez, C. de Haro, D. Bautista, V. Moreno, L.

Valencia, Inorg. Chem. 47 (2008) 4490–4505.