p-Cymene Complexes of Ruthenium(II) as Antitumor Agents

In this work, the cytotoxic behavior of six ruthenium(II) complexes of stoichiometry [(η6-p-cymene)RuCl2L] (I-VI), L = 4-cyanopyridine (I), 2-aminophenol (II), 4-aminophenol (III), pyridazine (IV), and [(η6-p-cymene)RuClL2]PF6; L = cyanopyridine (V), L = 2-aminophenol(VI) towards three cell lines was studied. Two of them, HeLa and MCF-7, are human carcinogenic cells from cervical carcinoma and human breast cancer, respectively. A comparison with healthy cells was carried out with BGM cells which are monkey epithelial cells of renal origin. The behavior of complex II exhibits selectivity towards healthy cells, which is a promising feature for use in cancer treatment since it might reduce the side effects of most current therapies.

Furthermore, ruthenium compounds have low levels of toxicity and can be tolerated in vivo. Their advantages over platinum-based complexes include their various oxidation states, reaction mechanism, and different ligand substitution kinetics, thereby making them suitable for use in biological applications. Several studies have focused attention on the interaction between ruthenium complexes and their biological targets [32]. For example, Weiss et al. [33] developed a series of organometallic ruthenium(II)-arene complexes that exerted antimetastatic activity and lowered primary tumor growth. They demonstrated that the prototype compound, [Ru(ɳ 6 -p-cymene)Cl2(pta)], where pta = 1,3,5-triaza-7-phosphaadamantane (RAPTA-C), reduces the expansion of primary tumors in preclinical models for colorectal and ovarian carcinomas The organoruthenium compounds formed from Ru(II)(η 6 -p-cymene) chloride moieties and oxicam-based ligands have also been studied [34]. The aim of the mentioned work was to combine the anti-inflammatory properties of oxicams, a versatile family of heterocyclic compounds, and the anticancer activity of Ru(II)(arene) Scheme 2. Complexes [(η 6 -p-cymene)Ru(CN)X] 0/+ (X=Cl, py or 4-NMe 2 py) containing a cyclometalated 2-ppy or 1-ppz with a non-coordinated CHO [24].
Furthermore, ruthenium compounds have low levels of toxicity and can be tolerated in vivo. Their advantages over platinum-based complexes include their various oxidation states, reaction mechanism, and different ligand substitution kinetics, thereby making them suitable for use in biological applications. Several studies have focused attention on the interaction between ruthenium complexes and their biological targets [32]. For example, Weiss et al. [33] developed a series of organometallic ruthenium(II)-arene complexes that exerted antimetastatic activity and lowered primary tumor growth. They demonstrated that the prototype compound, [Ru(η 6 -p-cymene)Cl 2 (pta)], where pta = 1,3,5-triaza-7-phosphaadamantane (RAPTA-C), reduces the expansion of primary tumors in preclinical models for colorectal and ovarian carcinomas The organoruthenium compounds formed from Ru(II)(η 6 -p-cymene) chloride moieties and oxicam-based ligands have also been studied [34]. The aim of the mentioned work was to combine the anti-inflammatory properties of oxicams, a versatile family of heterocyclic compounds, and the anticancer activity of Ru(II)(arene) complexes. By means of in vitro assays, it was established that the complexes were active against the colon carcinoma HCT116 and breast cancer MDA MB 231 cancer cell lines. The cytotoxicity was found to be strongly dependent on the lipophilicity of the compound, as the most lipophilic compound was the most active Molecules 2020, 25, 5063 3 of 12 in HCT116 cells. Moreover, the ruthenium(II) p-cymene complexes of naphthoquinone derivatives [Ru(II)(η 6 -p-cymene)(Lap)(PTA)](PF 6 ), and [Ru(II)(η 6 -p-cymene)(Jug)(PTA)](PF 6 ), 4 (Lap: lapachol, Plum: plumbagin, Law: lawsone, Jug: juglone, PTA: 1,3,5-triaza-7-phosphaadamantane) showed in vitro antiproliferative activity against human melanoma A375, liver hepatocellular carcinoma HepG-2, breast MCF-7, colon adenocarcinoma LoVo, ovary A2780 and colon carcinoma HCT-8 cancer cell lines under hypoxic conditions [35].
Ru(II) complexes are known to enter cells through multiple mechanisms, such as passive diffusion, active transport, and endocytosis [37]. However, it was noted that most nanostructured ruthenium complexes enter cells by endocytosis [38], although, the changes in ligands and hydrophobicity can modulate uptake and cellular localization. On the other hand, most Ru(II) complexes are known to have high selectivity for binding to DNA [39][40][41][42] and can also bind to DNA via interaction with aromatic ligands.
In this context, we studied the synthesis of complexes similar to those previously described by us, containing the Ru(η 6 -p-cymene) fragment, starting from the dimer [Ru(η 6 -p-cymene)(µ-Cl)Cl]2 [43] and different ligands in order to investigate the possible antitumor activity of these complexes. The compounds were characterized by C, H and N elemental analysis, infrared spectroscopy, proton nuclear magnetic resonance and high-resolution mass spectrometry. The cell lines chosen to evaluate the cytotoxic activity of the synthesized compounds were HeLa, MCF-7, and BGM cells, human carcinogenic cells from cervical carcinoma and human breast cancer and monkey epithelial cells of renal origin, respectively. The cytotoxicity was determined using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay.

Results and Discussion
The compounds used as anticancer agents are summarized in the following scheme (Scheme 3). From the p-cymene dimer, by the addition of the corresponding ligands in dichloromethane, the complexes I, II, III and IV were achieved. Compounds V and VI were obtained from I and II, the corresponding ligand, NH4PF6, and a small amount of water. It is worth mentioning that the yields of the neutral compounds III and IV are higher than those of the ionic ones, V, but mainly VI, probably because the presence of water makes the reaction medium more ionic, dissolving and dragging part of the NH4PF6.
All the isolated ruthenium compounds are air-stable solids and gave satisfactory partial elemental analyses and their colors, yields, mass spectrometry and decomposition points are listed in Table 1. All these data are consistent with the proposed formula. From the p-cymene dimer, by the addition of the corresponding ligands in dichloromethane, the complexes I, II, III and IV were achieved. Compounds V and VI were obtained from I and II, the corresponding ligand, NH 4 PF 6, and a small amount of water. It is worth mentioning that the yields of the neutral compounds III and IV are higher than those of the ionic ones, V, but mainly VI, probably because the presence of water makes the reaction medium more ionic, dissolving and dragging part of the NH 4 PF 6.
All the isolated ruthenium compounds are air-stable solids and gave satisfactory partial elemental analyses and their colors, yields, mass spectrometry and decomposition points are listed in Table 1. All these data are consistent with the proposed formula. Infrared spectroscopy confirmed that the ligand acted as a monodentate even when the reaction conditions or proportions where changed. Thus, in all cases, ruthenium(II) prefers to be coordinated by pyridine or amine nitrogen acting as a monodentate ligand. This behavior has been observed previously [25].
The ruthenium(II) compounds used as antitumor agents are of two types: neutral and cationic. In the case of the cationic, the stabilizing anion was PF 6 − , which is easily distinguished both by infrared spectroscopy (due to the presence of two bands at, approximately, 840 and 560 cm −1 , see Table 2) and by mass spectrometry (due to the presence of the peak corresponding to the anion at 145). In all the cases, the signals corresponding to ν (Ru-Cl) appear in the 265-300 range. The new compounds were also characterized by NMR-1 H and Cosy 1 H-1 H. The observed signals in Table 3 are consistent with the predicted values. The 1H NMR spectra of complexes III, IV and V are attached in Figures S1 to S3 of the Supporting Information as well as the 1H-1H COSY of IV ( Figure  S4). The solvent used in each case can be observed. The mass spectra of III, IV, and V are also attached in Figures S5 to S7. Decompose into solution 208 Infrared spectroscopy confirmed that the ligand acted as a monodentate even when the reaction conditions or proportions where changed. Thus, in all cases, ruthenium(II) prefers to be coordinated by pyridine or amine nitrogen acting as a monodentate ligand. This behavior has been observed previously [25].
The ruthenium(II) compounds used as antitumor agents are of two types: neutral and cationic. In the case of the cationic, the stabilizing anion was PF6 -, which is easily distinguished both by infrared spectroscopy (due to the presence of two bands at, approximately, 840 and 560 cm −1 , see Table 2) and by mass spectrometry (due to the presence of the peak corresponding to the anion at 145). In all the cases, the signals corresponding to ν (Ru-Cl) appear in the 265-300 range.  The new compounds were also characterized by NMR-1 H and Cosy 1 H-1 H. The observed signals in Table 3 are consistent with the predicted values. The 1H NMR spectra of complexes III, IV and V are attached in Figures S1 to S3 of the Supporting Information as well as the 1H-1H COSY of IV ( Figure S4). The solvent used in each case can be observed. The mass spectra of III, IV, and V are also attached in Figures S5 to S7.  Infrared spectroscopy confirmed that the ligand acted as a monodentate even when the reaction conditions or proportions where changed. Thus, in all cases, ruthenium(II) prefers to be coordinated by pyridine or amine nitrogen acting as a monodentate ligand. This behavior has been observed previously [25].
The ruthenium(II) compounds used as antitumor agents are of two types: neutral and cationic. In the case of the cationic, the stabilizing anion was PF6 -, which is easily distinguished both by infrared spectroscopy (due to the presence of two bands at, approximately, 840 and 560 cm −1 , see Table 2) and by mass spectrometry (due to the presence of the peak corresponding to the anion at 145). In all the cases, the signals corresponding to ν (Ru-Cl) appear in the 265-300 range.  The new compounds were also characterized by NMR-1 H and Cosy 1 H-1 H. The observed signals in Table 3 are consistent with the predicted values. The 1H NMR spectra of complexes III, IV and V are attached in Figures S1 to S3 of the Supporting Information as well as the 1H-1H COSY of IV ( Figure S4). The solvent used in each case can be observed. The mass spectra of III, IV, and V are also attached in Figures S5 to S7.

VI
Decompose into solution.
The assignment of the aromatic protons of complex IV was done by COSY 1 H-1 H. In any case, the presence of isomers in the anionic species was observed.

Cytotoxicity of the Complexes and the Ligand
In vitro cytotoxicity tests of the synthesized ruthenium(II) compounds were carried out to establish their potential anticancer activities and to select the most active in this respect but, at the same time, the least harmful for healthy cells. A colorimetric MTT assay, that assesses cell metabolic

VI
Decompose into solution.
The assignment of the aromatic protons of complex IV was done by COSY 1 H-1 H. In any case, the presence of isomers in the anionic species was observed.

Cytotoxicity of the Complexes and the Ligand
In vitro cytotoxicity tests of the synthesized ruthenium(II) compounds were carried out to establish their potential anticancer activities and to select the most active in this respect but, at the same time, the least harmful for healthy cells. A colorimetric MTT assay, that assesses cell metabolic activity, was used to determine cytotoxicity. This test is cost-effective, convenient and rapid [44]. The cytotoxicity of the complexes was seen to be strongly influenced by the chosen cell lines and by the structural features such as the ligand used. The IC 50 values of HeLa, MCF-7 and BGM cells after exposure to a series of ruthenium(II) compounds for 48 h were calculated using a dose-response model, which was obtained from sigmoidal fitting of dose-response curves, as stated in the Experimental Section. The calculated IC 50 values of the compounds and free ligands are shown in Tables 4 and 5, respectively. The IC 50 values of the ruthenium(II) compounds synthesized were compared in Table 4 with the values of cisplatin obtained from the literature [45,46] for the same cell lines. All complexes (except complex VI with Hela) were found to be less toxic than cisplatin with the cancer cell lines. Nevertheless, cisplatin is much more aggressive towards the BGM healthy cells. The differential selectivity of an anticancer drug towards cancer cells versus normal cells increases the likelihood of tumor-specific cytotoxicity, reducing the side effects in patients. The corresponding dose-response curves of Complexes II, III and VI are represented in Figure 1.   The results showed that ligands 4-cyanopyridine, pyridazine and 4-aminophenol do not contribute to the cytotoxic behavior of ruthenium (II) compounds (IC50 > 250 μM) on any of the three cell lines studied. The ligand 2-aminophenol only presents a cytotoxic effect against MCF-7 cells, but much lower than the ruthenium (II) complexes. The cytotoxic potency of Complexes I, IV and V is also negligible (IC50 > 250 μM) on the three cell lines studied. However, Complexes II, III and VI (see Figure 1) were cytotoxic in the case of at least two cell lines. Complexes II and VI (both with the ligand L = 2-aminophenol) were the most cytotoxic towards cancer cells both being more aggressive to MCF-7 breast cancer cells than towards cervical cancer cells HeLa. Of these two complexes, Complex II must be considered the best choice for cancer therapy because it was not cytotoxic towards the The results showed that ligands 4-cyanopyridine, pyridazine and 4-aminophenol do not contribute to the cytotoxic behavior of ruthenium (II) compounds (IC 50 > 250 µM) on any of the three cell lines studied. The ligand 2-aminophenol only presents a cytotoxic effect against MCF-7 cells, but much lower than the ruthenium (II) complexes. The cytotoxic potency of Complexes I, IV and V is also negligible (IC 50 > 250 µM) on the three cell lines studied. However, Complexes II, III and VI (see Figure 1) were cytotoxic in the case of at least two cell lines. Complexes II and VI (both with the ligand L = 2-aminophenol) were the most cytotoxic towards cancer cells both being more aggressive to MCF-7 breast cancer cells than towards cervical cancer cells HeLa. Of these two complexes, Complex II must be considered the best choice for cancer therapy because it was not cytotoxic towards the healthy cells used in this study (BGM), when IC 50 values higher than 250 µM were reached. By contrast, Complex VI, while it had the stronger cytotoxic effect against tumor cells, showed lower selectivity between the tumor and healthy cells, with a cytotoxicity of 95 µM in BGM cells. Finally, Complex III (with the ligand 4-aminophenol) was more cytotoxic towards healthy cells than cancer cells, which rules out its suitability for the purpose of this work. Therefore, it seems that position 2 in the ligand is preferred for cytotoxic behavior. The above described behavior of Complex II is highly promising because its selectivity in the face of tumor cells is higher than most of the tumor used in current cancer therapies.

Materials and Methods
The solvents were dried by conventional methods. The ligands were commercial grade chemicals and [{(p-cymene)RuCl 2 } 2 ] were prepared by published methods [23]. 1

H NMR spectra were recorded
Molecules 2020, 25, 5063 8 of 12 on a Bruker Avance 200, 300 and 400 MHz instrument. IR spectra were recovered on a 100 FTIR Spectrometer as nujol mulls. The C, H and N analyses were obtained with a LECO CNHS-932 elemental microanalyzer. Thermal decomposition studies were carried out on a TGA-DTA TA Instruments. High resolution (HR)-ESI-MS spectrometry was obtained using a MS TOF Agilent Model 6220 spectrometer.

Synthesis of the Complexes
3.1.1. Complexes I, II, III and IV These were prepared according to the following procedure. To a dichloromethane (15 mL) solution of [{(p-cymene)RuCl 2 } 2 ] (0.4902 mmol), the appropriate ligand (0.9804 mmol for I, II, III and IV) was added. The resulting suspension was stirred for 1 h and was concentrated. The solid obtained was separated by filtration, and repeatedly washed with diethyl ether. Complex III, as mentioned, was synthesized according to reference [29].

Complexes V and VI
These were prepared according to the following procedure. To an ethanol (7 mL) solution of the respective compound II and IV, respectively, (0.6098 mmol) the corresponding ligand (0.6098 mmol), and NH 4 PF 6 (0.6098 mmol) respective ligand (0.6098 mmol) were added. The mixture was stirred for 5 min and then, water (1.5 mL), was added. The solution darkened immediately and was stirred for 1 h. Partial evaporation of the solvent and the subsequent addition of diethyl ether caused the formation of a precipitate, which was filtered off and air-dried. These complexes were recrystallized from ethanol-diethyl ether.

Cell lines and Culture Media
Human cervical cancer cells (HeLa), human breast cancer cells (MCF-7) and green monkey kidney epithelial cells (BGM) were acquired from the American Type Culture Collection (ATCC, USA). The reason to choose human cervical cancer cells (HeLa) and human breast cancer cells (MCF-7) was to use a cell culture model that closely represented the human in vivo situation. The green monkey kidney epithelial cells (BGM) were chosen to compare cancer cells with healthy cells. Cell lines were maintained in Dulbecco's Modified Eagle Medium (DMEM) with a low glucose content (1 g/L) supplemented with 10% (v/v) fetal bovine serum (FBS), 1 mM glutamax, 1% antibiotics (penicillin-streptomycin) and 1 mM pyruvate. In all cases, the cells were maintained at 37 • C in 5% CO 2 atmosphere of 95% humidity. Cells were sub-cultured and the medium was changed once a week. In all cases, 0.25% trypsin, 0.25 mM ethylenediaminetetraacetic acid (EDTA) was used. Before and after the experiments, all cell lines were mycoplasma-free, as determined by the Hoechts DNA stain method [47].

Cytotoxicity Assay
A total of 5 × 10 3 cells/well (200 µL of the culture medium described above) were seeded into a 96-well plate and incubated at 37 • C in a 5% CO 2 and 95% humidity atmosphere for 24 h. A solution of each compound was prepared at a final concentration of 250 µM in DMSO (<1%). Successive 1:1 dilutions were performed, obtaining a total of sixteen solutions of concentrations ranging from 250 µM to 0.00762 µM, all of them in culture medium. Finally, a 200 µL aliquot of each of these last sixteen solutions was added to the wells. Cells were incubated at 37 • C for 48 h. The medium was then removed from the wells and 200 µL of MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, 1 mg/mL final concentration) was added. After 4 h incubation in identical conditions, MTT was removed and 100 µL of dimethyl sulfoxide (DMSO) added. The absorbance at 560 nm was measured and recorded in a Fluostar Omega spectrophotometer.
Absorbances at each compound concentration were translated into inhibition percentages, I%, according to Equation (1): where A T and A C are the absorbance of treated and control cells, respectively. IC 50 values were obtained from a three-parameter fitting of the semi-logarithmic curves (I% as a function of the logarithm of the compound concentration) according to Equation (2): where Imax is the maximum inhibition observed, IC 50 is the compound concentration at which 50% of the cell population is death, C is the compound concentration at what the inhibition I% is observed and n is the slope of the curve at the IC 50 value. The fitting was performed using GraphPad Prism v.8 software. All compounds were tested in three independent sets with triplicate points. The in vitro studies were performed in SACE (Support Service for Experimental Sciences, University of Murcia, Murcia, Spain) with biosecurity Level 2.

Conclusions
Of the six complexes analyzed in this work, two had been described previously and the other four were specifically designed and synthesized and fully characterized for this study. The cytotoxicity of all the new compounds was evaluated against three cell lines by means of an MTT assay. The results showed that Complexes I, IV and V (with ligands 4-cyanopyridine and pyridazine) had no cytotoxic effect against HeLa, MCF-7 or BGM cells, when they used IC 50 concentrations lower than 250 µM. By contrast, Complexes II, III and VI (with the ligand L = aminophenol) were cytotoxic against at least two cell lines. However, while Complexes II and VI, (both with the ligand 2-aminophenol) were more cytotoxic against cancer cells, while Complex III (with the ligand 4-aminophenol) was more aggressive against healthy cells. All these findings justify further studies into the use of ruthenium compounds as promising anticancer agents due to their unique and versatile biochemical properties, that serve as alternatives to cisplatin and its derivatives.
Supplementary Materials: The supplementary materials are available online. Figure S1: 1H NMR spectrum of complex III, Figure S2: 1H NMR spectrum of complex IV, Figure S3: 1H NMR spectrum of complex V, Figure S4: The 1H-1H COSY of complex IV, Figure S5: Mass spectrum of complex III, Figure S6: Mass spectrum of complex IV, Figure