Synthesis, Characterization, Cytotoxic Activity, and Interactions with CT-DNA and BSA of Cationic Ruthenium(II) Complexes Containing Dppm and Quinoline Carboxylates

The complexes cis-[Ru(quin)(dppm)2]PF6 and cis-[Ru(kynu)(dppm)2]PF6 (quin = quinaldate; kynu = kynurenate; dppm = bis(diphenylphosphino)methane) were prepared and characterized by elemental analysis, electronic, FTIR, 1H, and 31P{1H} NMR spectroscopies. Characterization data were consistent with a cis arrangement for the dppm ligands and a bidentate coordination through carboxylate oxygens of the quin and kynu anions. These complexes were not able to intercalate CT-DNA as shown by circular dichroism spectroscopy. On the other hand, bovine serum albumin (BSA) binding constants and thermodynamic parameters suggest spontaneous interactions with this protein by hydrogen bonds and van der Waals forces. Cytotoxicity assays were carried out on a panel of human cancer cell lines including HepG2, MCF-7, and MO59J and one normal cell line GM07492A. In general, the new ruthenium(II) complexes displayed a moderate to high cytotoxicity in all the assayed cell lines with IC50 ranging from 10.1 to 36 µM and were more cytotoxic than the precursor cis-[RuCl2(dppm)2]. The cis-[Ru(quin)(dppm)2]PF6 were two to three times more active than the reference metallodrug cisplatin in the MCF-7 and MO59J cell lines.

Due to this background of promising biological results in complexes containing the cis-[Ru(P-P) 2 ] unit our current strategy consists in evaluating other derivatives with different chelating moiety replacing the chlorido ligands in the search for new cytotoxic agents against tumor cells. In the current work quinoline carboxylates were chosen as coligands in order to generate cationic complexes and also to explore the possible different coordination modes of the ligands.
In this work the synthesis and characterization of two new derivatives with formula cis-[Ru(quin)(dppm) 2 ]PF 6 (1) and cis-[Ru(kynu)(dppm) 2 ]PF 6 (2) are reported. The interaction of these complexes with CT-DNA and BSA was investigated by circular dichroism (CD) and fluorescence spectroscopies, respectively. Besides, preliminary in vitro tests of cytotoxic activities against a variety of human cell lines are presented and discussed.

Instrumentation.
Elemental analyses were performed on a Perkin Elmer 2400 Series II CHNS/O microanalyser. Molar conductivities of freshly prepared 1.0 × 10 −3 mol⋅dm −3 methanol solutions were measured using a Digimed DM-31 conductivity meter. IR spectra were recorded on a Shimadzu FTIR-Prestige 21 spectrophotometer, using KBr pellets. UVvis spectroscopy was recorded on a Femto model 800 XI spectrophotometer using cuvettes of 1 cm path length. 1 H and 31 P{ 1 H} NMR experiments were performed on a Bruker Avance III HD 400 MHz (9.4 T) at 298 K. Spectra were recorded in CDCl 3 with TMS and 85% H 3 PO 4 as external references, respectively, for 1 H and 31 P{ 1 H}.

Synthesis.
The precursor cis-[RuCl 2 (dppm) 2 ] (0.103 mmol; 100 mg) was solubilized in 100 mL of methanol and the mixture was heated during 20 minutes. Quinaldic acid (0.103 mmol; 17.8 mg) or kynurenic acid (0.103 mmol; 19.5 mg), respectively, for synthesis of complexes 1 and 2, was solubilized in 10 mL of methanol and deprotonated with triethylamine (0.132 mmol; 0.014 ml) and this solution was added dropwise on the precursor solution. After this processes, the mixture was stirred and refluxed for 48 h. The final solution was concentrated to ca. 5 mL and an aqueous solution of NH 4 PF 6 (0.150 mmol; 24.4 mg) was added for the precipitation of a yellow solid. The solid was filtered off and washed with water (3 × 5 mL) and diethyl ether (3 × 5 mL) and dried under reduced pressure.

Circular Dichroism (CD) Experiments.
CD spectra were recorded on a spectropolarimeter JASCO J-720 between 540 and 240 nm in a continuous scanning mode (200 nm⋅min −1 ). The final data are expressed in molar ellipticity (millidegrees). All of the CD spectra were generated and represented averages of three scans. Stock solutions of each complex were freshly prepared in DMSO prior to use. An appropriate volume of each solution was added to the samples of a freshly prepared solution of CT-DNA (50 M) in Trizma buffer to achieve molar ratios ranging from 0.05 to 0.5 DNA⋅drug −1 . The samples were incubated at 37 ∘ C for 18 h.

BSA-Binding Experiments.
The protein interaction with complexes 1 and 2 was examined in 96-well plates used for fluorescence assays on a fluorimeter Synergy H1. BSA (2.0 mol⋅L −1 ) was prepared in Trizma buffer at pH = 7.4 (4.5 mM Trizma HCl, 0.5 mM Trizma base and 50 mM NaCl).
The inner-filter effect for ruthenium complexes was corrected by using where corr and obs are the corrected and measured fluorescence intensity of protein, respectively. em and ex are the absorption values of the system at the excitation wavelength and emission wavelength of the complex, respectively. Complexes were dissolved in sterile DMSO. For fluorescence measurements, the BSA concentration was kept constant in all samples, while the complex concentration was increased from 3.125 to 150 M, and quenching of the emission intensity of the tryptophan residues of BSA at 320 nm (excitation wavelength 280 nm) was monitored at different temperatures (300 and 310 K). The experiments were carried out in triplicate and analyzed using the classical Stern-Volmer equation where 0 and are the fluorescence intensities in the absence and presence of quencher, respectively, [ ] is the quencher concentration, and sv is the Stern-Volmer quenching constant, which can be written as where is the biomolecular quenching rate constant and is the average lifetime of the fluorophore in the absence of quencher (6.2 × 10 −9 s) [20]. Therefore, (2) was applied to determine sv by linear regression of a plot of 0 / versus [ ].
The binding constant ( ) and number of complexes bound to BSA ( ) were determined by plotting the double log graph of the fluorescence data using The thermodynamic parameters were calculated from the van't Hoff equation: where is analogous to the Stern-Volmer quenching constant, sv is at the corresponding temperature (the temperatures used were 300 and 310 K), and is the gas constant, from which the Δ and Δ of the reaction can be determined from the linear relationship between ln and the reciprocal absolute temperature. Furthermore, the change in free energy (Δ ) was calculated from the following equation:

Spectroscopical
Characterization. The diamagnetic and monoelectrolytes compounds 1-2 exhibited satisfactory microanalytical (C, H, and N) data. The 1 H NMR spectra showed two broad signals due to the CH 2 group of the dppm ligand close to 4.10 and 4.70 ppm [21]. The hydrogens of the phenyl groups (H o , H m and H p ) were observed as several multiplets between 6.26 and 7.86 ppm [21]. could be assigned to the kynu ligand [22]. Other hydrogens of the quin and kynu ligands were obscured by the signals corresponding to the hydrogens of the phenyl groups. The total number of hydrogens and proportion between the ligands dppm and quin or kynu was confirmed by the integral values for both complexes. In the 31 P{ 1 H} NMR spectra these complexes displayed a pair of triplets resonance signals, corresponding to two transpositioned phosphorus atoms and two phosphorus atoms transpositioned to oxygen atoms from carboxylate groups of the quinaldate and kynurenate anions, respectively. Triplets are observed at 9.70 and −12.2 ppm and at 9.94 and −12.1 ppm, respectively, for 1 and 2, with the splitting pattern typical of an AX pattern [16]. These signals are downfield shifted when compared with the triplets signals for the cis-[RuCl 2 (dppm) 2 ] that are observed at 0.64 and −25.3 ppm with PP = 36 Hz [17].

Circular Dichroism (CD) Experiments.
The CD spectral technique is very sensitive for diagnosing changes in the secondary structure of DNA, resulting from drug-DNA interactions [25]. A typical CD spectrum of CT-DNA shows a maximum at 275 nm, due to the base-stacking and a minimum at 248 nm attributed to the right-handed helicity, characteristic of the B conformation [26]. Thus simple groove binding and electrostatic interaction of small molecules show less or no perturbation on the base-stacking and helicity bands, while intercalation enhances the intensities of both the bands stabilizing the right-handed B conformation of CT-DNA as observed for the classical intercalator methylene blue [27]. To determine if the ruthenium(II) complexes cause changes in DNA, CD spectra of CT-DNA with increasing concentrations of 1 and 2 were acquired, up to molar ratio drug⋅DNA −1 (Ri) = 0.4. As shown in Figure 2 significant changes were not observed indicating that these compounds were not able to intercalate DNA [28].

BSA-Binding Experiments.
Serum albumin is the most abundant protein in plasma and is involved in the transport of metal ions and metal complexes with drugs through the blood stream. Binding to these proteins to complexes may lead to loss or conformational change in the protein subunit and provide paths for drug transportation. Bovine serum albumin, BSA (containing two tryptophans, Trp-134 and Trp-212) is the most extensively studied serum albumin, due to its structural homology with human serum albumin, HAS (one Trp-214). The BSA solution exhibits a strong fluorescence emission with a peak at 340 nm, due to the tryptophan residues, when excited at 280 nm [29]. Fluorescence quenching of BSA can occur by different mechanisms, usually classified as either dynamic or static quenching, which can be distinguished by their differing on temperature, viscosity, and lifetime measurements [30]. A dynamic quenching refers to the collisional process between the fluorophore and the quencher (in this case, ruthenium complexes) during the transient existence of the excited state. Dynamic quenching depends on diffusion, as higher temperatures result in high diffusion coefficient, and consequently, the constant quenching must also increase. In contrast, for the static quenching, an increase in temperature results in lower extinction values of the constants due to a fluorophore and quencher complex formation in the ground state [31].
The Stern-Volmer equation (2) has been used to understand the nature of the quenching mechanism of BSA in the presence of complexes 1 and 2 at different temperatures [31]. Figure 3 shows the quenching of the BSA in the presence of the different concentrations of the complexes 1 and 2.
In general, these complexes showed no significant variation sv with increasing temperature. However, the results of have values greater than the maximum possible for a dynamic mechanism (2.0 × 10 10 L⋅mol −1 ⋅s −1 ) [32] and in both cases were of the order 10 13 M −1 ⋅s −1 (Table 1), which is 1000-fold higher than the maximum value possible for diffusion controlled quenching of various kinds of quencher to biopolymer. It is suggested that the suppression of BSA Bioinorganic Chemistry and Applications with these complexes is a static quenching mechanism. The decreasing of with increasing temperature is in accordance with sv dependence on temperature.
In both cases, the obtained values of n indicate that the proportion between BSA-complex is equal 1 : 1, indicating that there is only one binding site in the BSA for each ruthenium complex, similar or equal to those reported before for other metal complexes [33][34][35]. Furthermore, values of confirmed the moderated interaction force between complex-BSA [36][37][38] and the temperature is not significant. Thus, these complexes can be stored and carried by protein in the body. The interaction forces between drugs and biomolecules may include van der Waals interaction, hydrogen bonds, and electrostatic and hydrophobic interactions. The thermodynamic parameters ΔG (free energy change), ΔH (enthalpy change), and ΔS (entropy change) were calculated to evaluate the intermolecular forces involving the molecules of complex and BSA. The values for ΔH > 0 and ΔS > 0 imply the involvement of hydrophobic forces in protein binding, ΔH < 0 and ΔS < 0 correspond to van der Waals and hydrogen bonding interactions, and ΔH < 0 and ΔS > 0 suggest an electrostatic force [39]. Thermodynamic parameters (ΔH and ΔS) were calculated from the van't Hoff plots, (5); ΔG was estimated from (6). All the results are shown in Table 1. Table 1: Stern-Volmer quenching constant ( sv , L⋅mol −1 ), biomolecular quenching rate constant ( , L⋅mol −1 ⋅s −1 ), binding constant ( , L⋅mol −1 ), the number of binding sites ( ), and ΔG 0 (KJ⋅mol −1 ), ΔH 0 (KJ⋅mol −1 ), and ΔS 0 (J⋅mol −1 ⋅K) values for the complex-BSA system at different temperatures.  The negative values of free energy, ΔG, suggested that the interaction process was spontaneous; the values entropy ΔS and enthalpy ΔH negative indicated that the hydrogen bonds and van der Waals forces are the more important interactions in the reaction [38].

Cytotoxicity
Assays. Ruthenium compounds 1 and 2, cis-[RuCl 2 (dppm) 2 ] and cisplatin were evaluated for their capability of inhibiting tumor cell growth in vitro using three human cell lines, HepG2, MCF-7, and MO59J. One nontumorigenic cell line (GM07492A) was also assayed in the same conditions in order to verify the selectivity of these compounds. The resulting concentration-effect curves obtained with continuous exposure for 24 h are depicted in Figure 4. A more convenient comparison of the cytotoxic potency (expressed as IC 50 values) is listed in Table 2.
Overall sensitivities of the three tumor cell lines ( Figures  4(a)-4(c)) and of the normal cell (Figure 4(d)) are more pronounced for complex 1, followed by complex 2, and the less active is the precursor cis-[RuCl 2 (dppm) 2 ].
Complexes 1 and 2 have showed, in general, a moderate cytotoxicity against all the human tumor cell lines assayed with IC 50 values ranging from 10.1 to 36 M, except for complex 2 in MCF-7 cells that showed a very low cytotoxicity as shown in Table 2. Complex 1 displayed higher activity than 2 in all the cell lines assayed with IC 50 close to 10 M. This nonselective activity of complex 1 was not observed for complex 2 that was almost inactive in MCF-7 cells. The selectivity index (SI) was very close to 1 (or smaller) for all the cell lines assayed indicating a lack of selectivity for both complexes.
In the same experimental conditions the precursor complex cis-[RuCl 2 (dppm) 2 ] was less active than the complexes 1 and 2 by factors ranging from 1.4 to 15.8. A similar increased in activity was also observed against the normal cell line GM07492A. This lack of selectivity was also observed for the cis-[RuCl 2 (dppm) 2 ] for all the cell lines assayed and for cisplatin in the MCF-7 and MO59J cell lines. These data clearly indicate that exchanging two chlorido ligands by a bidentate quinoline carboxylate group turns the unity cis-[Ru(dppm) 2 ] complex more cytotoxic, probably due to its higher solubility and disponibility in the culture medium. Interestingly, the presence of one OH-group in complex 2 led to a significant decrease in cytotoxic activity, indicating that this may be a way to modulate the cytotoxic potency of this type of complex.

Conclusion
In this investigation two new ruthenium(II) complexes containing dppm and the quinaldate and kynurenate anions with formula cis-[Ru(quin)(dppm) 2 ]PF 6 and cis-[Ru(kynu)(dppm) 2 ]PF 6 were synthesized and characterized by elemental analysis and spectroscopic methods. Characterization data are in agreement with a cis geometry and a chelated coordination, through the carboxylate group, for the quinaldate and kynurenate ligands. Utilizing circular dichroism spectroscopy showed that these complexes lack the ability to intercalate DNA. On the other hand BSA-binding constants and thermodynamic parameters suggest spontaneous interactions with this protein by hydrogen bonds and van der  Waals forces. The in vitro cytotoxicity activity assays indicate a moderate to high cytotoxicity against a panel of human tumor cell lines; however, these complexes lacks selectivity. Interestingly, modulation of cytotoxic potency can probably be done by exchanging the substituents of the quinolone carboxylate ligand and will be explored in future work.

Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this paper.