DNA Binding and Antitumor Activity of α-Diimineplatinum(II) and Palladium(II) Dithiocarbamate Complexes

The two water-soluble designed platinum(II) complex, [Pt(Oct-dtc)(bpy)]NO3 (Oct-dtc = Octyldithiocarbamate and bpy = 2,2′ -bipyridine) and palladium(II) complex, [Pd(Oct-dtc)(bpy)]NO3, have been synthesized and characterized by elemental analyses, molar conductivity measurements, IR, 1H NMR, and electronic spectra studies. Studies of antitumor activity of these complexes against human cell tumor lines (K562) have been carried out. They show Ic50 values lower than that of cisplatin. The complexes have been investigated for their interaction with calf thymus DNA (CT-DNA) by utilizing the electronic absorption spectroscopy, fluorescence spectra, and ethidium bromide displacement and gel filtration techniques. Both of these water-soluble complexes bound cooperatively and intercalatively to the CT-DNA at very low concentrations. Several binding and thermodynamic parameters are also described.


Introduction
In order to reduce the toxicity of cisplatin, that is, the wellknown anticancer drug, and modulate its activity, a new strategy is the design of novel metal complexes containing N and S donor ligands [1][2][3]. This interest has probably initiated from detoxicant properties of sulfur-containing ligands against heavy metal intoxication [4]. As an example is the use of sodium diethyldithiocarbamate (DDTC) in the treatment of patients with acute poisoning of nickel carbonyl, arsenic, and thallium [5]. A further interest in this chelating ligand has arisen due to its highly selective use to protect a variety of animal species from renal, gastrointestinal, and bone marrow toxicity, induced by cisplatin without inhibition of cisplatin's antitumor effect [6][7][8][9]. In addition, diethyldithiocarbamate has remarkable property of reversing platinum binding to macromolecules responsible for host toxicity. However, it does not interfere with the tumoricidal Pt-DNA interaction in the tumor cell. Thus, the protective action of the dithiocarbamate against the toxicity of cisplatin seems to be the formation of its stable platinum dithiocarbamate complexes [7,10].
It has also known that dithiocarbamate complexes of [M(NN)(SS)] type, with diimine (NN) as a π-electron acceptor and dithiocarbamate (SS) as an electron donor, exhibit intramolecular mixed metal ligand to ligand charge transfer bands [30,31]. This band appears in a region where CT-DNA has no absorption and thus has been widely used for the absorption spectrophotometric binding studies.
In this work, we have chosen a bioactive octyldithiocarbamate ligand whose structure not only resembles a great number of antibacterial, antiviral, anthelmintic, and insecticidal agent, but also can be used as inhibitor of cisplatin-induced nephrotoxicity [32] while as diimine we have employed 2,2 -bipyridine (bpy) whose structure adopts a planar conformation when it chelates to Pt(II) or Pd(II) center [33], and thus can intercalate in CT-DNA. These complexes have been tested against human cell tumor line K562. In order to confirm the mode of binding of these complexes to CT-DNA, detailed interaction studies of them with CT-DNA are attempted.

Materials and Methods.
Octylamine and carbon disulfide were purchased from Aldrich (England). Palladium(II) chloride was obtained from Fluka (Switzerland). Potassium tetrachloroplatinate, 2,2 -bipyridine, highly polymerized calf thymus DNA sodium salt, and Tris-HCl buffer were obtained from Merck (Germany). Other chemicals used were of analytical reagent or higher purity grade. [Pt(bpy)Cl 2 ] and [Pd(bpy)Cl 2 ] were prepared by the procedures described in the literature [34]. Solvents used were of reagent grade and purified before use by the standard methods. All the experiments involving interaction of the complexes with CT-DNA were performed in Tris-HCl buffer (20 mM) of pH = 7.0 medium containing 20 mM NaCl. Monitoring the ratio of the absorbance at 260 to that of 280 nm checks purity of CT-DNA. The solution gave a ratio of >1.8 at A 260 /A 280 , indicating that CT-DNA is sufficiently free from protein [35,36]. The CT-DNA concentration per nucleotide was determined by absorption spectroscopy using the known molar extinction coefficient value of 6600 M −1 cm-1 at 260 nm [37].
Electronic absorption spectra of the title metal complexes were measured on a J AS.CO UV/Vis-7850 recording spectrophotometer. Infrared spectra of the metal complexes were recorded on a J AS.CO -460 Plus FT-IR spectrophotometer in the range of 4000-400 cm −1 in KBr pellets. Microchemical analysis of carbon, hydrogen, and nitrogen for the complexes were carried out on a Herause CHNO-RAPID elemental analyzer. 1 H NMR spectra were recorded on a Brucker DRX-500 Avance spectrometer at 500 MHz in DMSO-d 6 using tetramethylsilane as internal reference. Fluorescence intensity measurements were carried out on a Hitachi spectrofluorimeter model MPF-4. Melting point was measured on a Unimelt capillary melting point apparatus and reported uncorrected. Conductivity measurements of the above platinum and palladium complexes were carried out on a Systronics conductivity bridge 305, using a conductivity cell of cell constant 1.0. Doubly distilled water was used as solvent all along. (Oct-dtcNa). This ligand was synthesized by the method as   [20,38].

Cytotoxic Studies.
The procedure for cytotoxic studies of the [Pt(bpy)(Oct-dtc)]NO 3 and [Pd(bpy)(Oct-dtc)]NO 3 was similar to that reported earlier [15]. Here, also 1 × 10 4 cells per mL of K562 chronic myelogenous leukemia were used in Tris-HCl buffer solution of PH 7.0.

Metal Complexes-DNA-Binding Studies.
[Pt/Pd(bpy) (Oct-dtc)]NO 3 complexes were interacted with calf thymus DNA in Tris-HCl buffer (20 mM, PH = 7.0) containing 20 mM sodium chloride. The procedure followed to determine binding and thermodynamic parameters were similar to what was reported earlier [16]. The stock solutions of the complexes (0.5 mmol/L) and CT-DNA (4 mg/mL) were made in the same buffer. The DNA-metal complex solutions were incubated at 300 K and 310 K, and, then, the spectrophotometric reading at λ max of the complexes (321 nm for Pt(II) complex and 305 nm for pd(II) complex), where CT-DNA has no absorption, were measured. Using trial and error method, the incubation time for solutions of DNA-metal complexes at 300 K and 310 K was found to be 6 h. No further changes were observed in the absorbance reading after longer incubation. All the experiments repeated to get the constant results.

Results and Discussion
The compounds correspond to the composition Oct-dtcNa, [Pd(bpy)(Oct-dtc)]NO 3 and [Pt(bpy)(Oct-dtc)]NO 3 , where bpy = 2,2 -bipyridine, and Oct-dtc = octyldithiocarbamato ligands were prepared and characterized by chemical analysis, conductance measurements, ultraviolet-visible, infrared and, 1 H NMR spectroscopic methods. These analytical data of the complexes are given in experimental section and the proposed structure in Scheme 1. Cytotoxicity and detailed calf thymus DNA-binding studies of these water-soluble complexes have been studied.

Cytotoxic Measurement of the Metal Complexes.
[Pt/Pd (bpy)(Oct-dtc)]NO 3 complexes were screened for their antitumor activities against K562 chronic myelogenous leukemia cells [15]. These cells were maintained in RPMI 1640 medium supplemented with 10% FCS in a humidified incubator (310 K and 5% CO 2 ). The cells were then grown in RPMI medium supplemented with L-glutamine (2 mM), streptomycin and penicillin (5 μg/mL), and 10% heatinactivated fetal calf serum, at 310 K under a 5% CO 2 /95% air atmosphere. In this study, the harvested cells were seeded into 96-well plate (1 × 10 4 cell/mL) with various concentrations of metal complexes ranging from 0 to 0.25 mM and incubated for 24 h [42]. The 50% cytotoxic concentrations (Ic 50 ) of the Pt(II) and Pd(II) complexes were determined to be 0.0017 and 0.007 mM, respectively ( Figure 1 for pt(II) complex and the inset for Pd(II) complex). As shown in Figure 1, cell growing after 24 h was significantly reduced in the presence of various concentrations of the metal complexes. Furthermore, Ic 50 value of cisplatin under the same experimental conditions was determined. This value (0.154 mM) is much higher as compared to the Ic 50 value of the above two complexes. However, the Ic 50 values of these complexes are comparable with those of our analogous Pt(II) and Pd(II) dithiocarbamate complexes reported earlier [15][16][17]. This procedure for growth inhibition studies of the metal complexes established that the cell DNA is the target biomolecule for these complexes [39].

CT-DNA Denaturation Data and Evaluation of Thermodynamic Parameters.
Absorption spectroscopy is one of the most useful techniques to study the binding of any drug to DNA [43][44][45][46]. The procedure followed was similar to that reported earlier [47,48]. These experiments were carried out separately at two temperatures of 300 K and 310 K in Tris-HCl buffer medium. The absorbance at 260 nm was     [16] with CT-DNA. It is noticeable that absorbance of DNA at 260 nm should increase in presence of increasing amount of each metal complex (denaturant agent). This is true for palladium complex (Figure 2). However, the opposite trend is observed for the platinum analog. Keeping in mind that platinum complexes react about 10 5 -10 6 times slower than palladium complexes [49,50], the decrease in the absorbance at 260 nm with the increase of the amount of Pt(II) complex added to CT-DNA may be due to (i) a possibility that interaction between CT-DNA and the metal complex causes the double helix of CT-DNA to become more straight leading to stacking; this stacking may cause conformational changes leading to a sort of denaturation or (ii) each strand after denaturation gets associated in a more stacked structure and (iii) metal complex slips into the helix and masks the hydrophobic bases leading to a decrease in absorbance. As will be seen in the later part of this paper, the [Pt(Octdtc)(bpy)]NO 3 and [Pd(Oct-dtc)(bpy)]NO 3 complexes can bind CT-DNA taking the mode of intercalation. This mode of binding supports the above three hypotheses.
Using the CT-DNA denaturation plots ( Figure 2) and Pace method [50], the value of K, that is, unfolding equilibrium constant and ΔG • , unfolding free energy of CT-DNA, at two temperatures of 300 K and at 310 K in the presence of Pt(II) and Pd(II) complexes have been calculated. In this method, Pace had assumed two-state mechanism, nature and denature, and then calculated unfolding free energy of DNA, that is, (ΔG • ) by using (1): where A obs is absorbance readings in transition region and A N and A D are absorbance readings of nature and denatured conformation of DNA, respectively. A straight line is observed when the values of ΔG • are plotted against the concentrations of each metal complex in the transition region at 300 K and at 310 K. These plots are shown in Figure 3 for Pt(II) complex and the inset for Pd(II) complex. The equation for these lines can be written as follow [51]: Here, the values of ΔG • (H2O) for each curve are measured from the intercept on ordinate of the plots and it is conformational stability of DNA in the absence of metal complex. m (the slope of each curve in the same plots) is a measure of the metal complex ability to denature CT-DNA and is summarized in Table 1. The values of m for the above complexes are much higher than those of Pt(II) and Pd(II) complexes reported earlier [15][16][17], which indicate the higher ability of this Pt(II) and Pd(II) complexes to denature CT-DNA. As we know, the higher the values of ΔG • , the larger the conformational stability of CT-DNA. However, the values of ΔG • (Table 1) are decreased by increasing the temperature. This is as expected because, in general, the decrease in ΔG • (H2O) value is the main reason for the decrease in CT-DNA stability [52]. Molar enthalpy  of CT-DNA denaturation in absence of Pt(II) and Pd(II) complexes, ΔH • (H2O) , is another important thermodynamic parameter. To find this, we calculated the molar enthalpy of CT-DNA denaturation in presence of the metal complexes, ΔH • conformation or ΔH • denaturation , in the range of two temperatures using Gibbs-Helmholtz equation [53]. On plotting the values of these enthalpies versus the concentration of metal complexes, straight lines will be obtained which are shown in Figure 4 for [Pt(Oct-dtc)(bpy)]NO 3 and the inset for [Pd(Oct-dtc)(bpy)]NO 3 complexes. Intrapolation of these lines (intercept on ordinate, i.e., absence of metal complexes) gives the values of ΔH • (H2O) ( Table 1). These plots show that in the range of 300-310 K the changes in the enthalpies in presence of Pt(II) and Pd(II) complexes are ascending. These observations indicate that, on increasing the concentration of Pt(II) and Pd(II) complexes, the stability of CT-DNA is increased. Moreover, the entropy of CT-DNA unfolding by Pt(II) and Pd(II) complexes ΔS • (H2O) has been calculated using equation ΔG = ΔH − TΔS and the data are given in Table 1. These data show that the metal-DNA complexes are more disordered than those of native CT-DNA, because the entropy changes are positive for Pt(II)-or Pd(II)-DNA complexes in the denaturation processes of CT-DNA ( Table 1). The above thermodynamic parameters agree well with those we have reported for [Pt/Pd(bpy)(Et-dtc)]NO 3 [15] and [Pt/Pd(bpy)(Bu-dtc)]NO 3 [16] complexes.     Figure 6). These plots are curvilinear concave downwards, suggesting cooperative binding [54]. Similar cooperativity in binding of analogous complexes with CT-DNA has also been observed [15,55]. On substituting the above experimental data (ν and [L] f ) in Hill equation,

UV-Vis Spectral Studies and Evaluation of Binding
we get a series of equations with unknown binding parameters n, K, and g. Using Eureka software, the theoretical values of these parameters have been deduced. The results are shown in Table 2. These results are comparable with those of 2,2 -bipyridine-platinum and -palladium complexes of dithiocarbamates as reported earlier [15,16] [58] and [Ru (dmb) 2 (NMIP)] 2 5.46 × 10 3 M −1 [59]. Obviously, these thermodynamical characteristics suggest that the two complexes intercalatively bind to CT-DNA, involving a strong stacking interaction between the aromatic chromophore and the base pairs of the DNA [60]. Similar results were obtained Bioinorganic Chemistry and Applications   Figure 9 for Pt(II) complex and the inset for Pd(II) complex. This plot shows that the peak obtained for the two wavelengths are not resolved and suggests that CT-DNA has not separated from the metal complexes. Thus, it implies that the binding between CT-DNA and the metal complexes is not reversible under such circumstances. This is due to the fact that if the interaction between CT-DNA and metal complexes was weak, the CT-DNA should have come out of the column separately [63].

Conclusions
The above work describes the synthesis, characterization, and cytotoxic studies of a novel and water-soluble platinum(II) and palladium(II) complexes possessing a planar aromatic ligand for intercalation and a dithiocarbamate ligand to avoid its renal damage. Ic 50 values of these complexes are much lower than those of cisplatin. Furthermore, detailed interaction studies of the complexes with CT-DNA have been carried out. They cooperatively bind to DNA through intercalation mode and unexpectedly denature DNA at extremely low concentration. Determinations of several binding and thermodynamic parameters have also been attempted. They 8 Bioinorganic Chemistry and Applications  may be helpful to understand the mechanism of the interaction between these complex and nucleic acid that must be quite different from that of cisplatin.

Competitive Binding between Ethidium Bromide (EB) and Pt(II)/Pd(II) Complexes for CT-DNA.
No fluorescence was observed for aqueous solution of above Pt(II) and Pd(II) complexes alone or in the presence of calf thymus DNA. So, the binding of Pt(II) and Pd(II) complexes with CT-DNA cannot be directly presented in the emission spectra. It has been studied by competitive EB binding experiments [64,65]. The fluorescence of EB is greatly enhanced upon intercalation to DNA. The effect of platinum(II)-2,2bipyridine and palladium(II)-2,2 -bipyridine complexes of octyldithiocarbamate on the fluorescence intensity of DNA-EB complexes was studied to get the mode of their binding to CT-DNA. Figure 10 shows the effect of Pt(II) and Pd(II) complexes ( [66,67]. The number of EB molecules intercalated to CT-DNA in presence of different concentrations of the Pt(II) complex and Pd(II) complex was calculated using Scatchard analysis [68]. In this experiment, the wavelengths of excitation and emission were set at 540 nm and 700 nm, respectively, with both having 0.5 nm slit widths. Solutions of CT-DNA, EB, and metal complexes were prepared in Tris-HCl buffer of pH 7.0. In this mediumsolutions of Pt(II) complex and Pd(II) complex were interacted with CT-DNA by incubating them at 300 K and 310 K for 6 h; appropriate amount of EB was added to them and further incubated at room temperature (300 K) for 6 h and finally processed for fluorescence spectral measurement. Saturation curves of fluorescence intensity for [Pt/Pd(Octdtc)(bpy)] + -DNA systems at different r f values (0.55, 1.18, and 1.85 for Pt(II) complex and 0.2, 0.58 and 0.92 for Pd(II) complex) in presence of increasing concentrations of EB (2, 4,. . ., 20 μM) were obtained. The fluorescence Scatchard plots obtained for binding of EB to CT-DNA in absence ( ) and presence (♦, Δ, and •) of various concentration of Pt(II) and Pd(II) (inset) complexes were shown in Figure 11. This figure shows that the complexes inhibit competitively the EB binding to CT-DNA (type A behavior) [68], where number of binding sites n (intercept on the abscissa) remains constant and the slope of the graphs that is K app (apparent association constant) decreases with increasing the concentrations of Pt(II) and Pd(II) complexes (Table 3). This implies that the [Pt/Pd(Oct-dtc)(bpy)]NO 3 complexes are intercalating in CT-DNA and thereby competing for intercalation sites occupied by EB. The values of K app and n, the number of binding sites per nucleotide, are listed in Table 3. These data suggested that the interaction of the [Pd(Oct-dtc)(bpy)]NO 3 complex with CT-DNA was stronger than that of [Pt(Octdtc)(bpy)]NO 3 complex, which were consistent with the above absorption spectral results. Compare their K app values with those of other known CT-DNA-intercalative complexes which possess analogical structure; the Pd(II) complexes in our paper have similar or stronger affinities with CT-DNA [17]. Similar modes of binding seem to be involved in other complexes [52,69].