Synthesis and Evaluation of the Cytotoxic Activity of Water-Soluble Cationic Organometallic Complexes of the Type [Pt(η1-C2H4OMe)(L)(Phen)]+ (L = NH3, DMSO; Phen = 1,10-Phenanthroline)

Starting from the [PtCl(η1-C2H4OMe)(phen)] (phen = 1,10-phenanthroline, 1) platinum(II) precursor, we synthesized and characterized by multinuclear NMR new [Pt(η1-C2H4OMe)(L)(phen)]+ (L = NH3, 2; DMSO, 3) complexes. These organometallic species, potentially able to interact with cell membrane organic cation transporters (OCT), violating some of the classical rules for antitumor activity of cisplatin analogues, were evaluated for their cytotoxicity. Interestingly, despite both complexes 2 and 3 resulting in greater cell uptake than cisplatin in selected tumor cell lines, only 3 showed comparable or higher antitumor activity. General low cytotoxicity of complex 2 in the tested cell lines (SH-SY5Y, SK-OV-3, Hep-G2, Caco-2, HeLa, MCF-7, MG-63, ZL-65) appeared to depend on its stability towards solvolysis in neutral water, as assessed by NMR monitoring. Differently, the [Pt(η1-C2H4OMe)(DMSO)(phen)]+ (3) complex was easily hydrolyzed in neutral water, resulting in a comparable or higher cytotoxicity in cancer cells with respect to cisplatin. Further, both IC50 values and the uptake profiles of the active complex appeared quite different in the used cell lines, suggesting the occurrence of diversified biological effects. Nevertheless, further studies on the metabolism of complex 3 should be performed before planning its possible use in tissue- and tumor-specific drug design.


Introduction
Cisplatin was synthesized for the first time in 1845, together with its trans isomer, by the Italian chemist Michele Peyrone. Its antitumor properties were highlighted only after 1960, thanks to the work of Barnett Rosenberg et al. [1,2]. Interestingly, a long time was required for its approval as an anticancer drug (in 1979) due to its relevant side effects, notwithstanding the generally observed increase of life expectancy or the complete recovery of oncological patients [3,4]. Another critical aspect of cisplatin was identified in its intrinsic low solubility, strongly limiting the possible pharmacological application with resistant tumors, probably requiring higher dosages. In any case, due to the progresses made in cisplatin therapeutic protocols, to date it remains one of the most widely used anticancer drugs [4,5]. For this reason, research on new platinum-based antitumor drugs has been mainly focused on the development of new complexes violating the classical rules early defined for antitumor activity of platinum compounds (i.e., cis-isomers, presence of two medium labile ligands and two inert N-donors different from tertiary amines) [4][5][6][7][8][9][10]. About cisplatin mechanism of action, it is well known that passive diffusion is not the only mechanism of cellular uptake, as several experiments showed that cisplatin is also imported by cell membrane carriers normally involved in the transport of different substrates. In this context, the two cis chlorido ligands of cisplatin, characterized by a medium lability, seem necessary to achieve the equilibrium formation of the antitumor active cationic solvento species cis-[PtCl(NH 3 ) 2 (OH 2 )] + , cis-[Pt(NH 3 ) 2 (OH 2 ) 2 ] 2+ , and cis-[Pt(NH 3 ) 2 (OH)(OH 2 )] + . The peculiar ability of cisplatin to produce such cationic species once inside the cytoplasm, due to the relatively low chloride concentration with respect to blood plasma (extracellular plasma [Cl − ] > 100 mM; cell cytoplasm [Cl − ] ≈ 23 mM; cell nucleus ≈ 4 mM) [4,5,11] is one of the serendipity aspects that have determined the success of this drug. In fact, the formation of reactive charged platinum solvento species inside the blood plasma seems related to the severity of side effects observed elsewhere in the body [12,13]. Indeed, the unselective binding of such reactive hydrolysis products to serum proteins, etc., is considered responsible for the observed cisplatin toxicity and side effects. Moreover, the reactivity of a platinum drug in the blood serum is a key factor determining how much of the administered drug actually reaches the cell membranes to begin uptake via passive diffusion or active transport (involving hCtr1, OCT, etc.) [14,15]. Furthermore, the hydrolyzed drug, when formed inside the cell cytoplasm, constitutes the essential active species for the desired antitumor activity. These reactive forms seem capable to interact with suitable nucleophilic targets present in cells, such as nucleic acids, proteins, thiols, phospholipids, cytoskeleton, etc. Among them, DNA is considered the main pharmacological target, as its functionality results strongly altered by interaction with cisplatin [4,[14][15][16]. Since its approval, to solve some of the problems occurred in clinical applications of cisplatin, many attempts have been made to develop new different antitumor active platinum-based drugs with a greater spectrum of efficacy, a different mechanism of action, lower side effects, and enhanced water solubility [4,5,[17][18][19]. In particular, recent studies on the action mechanism of cisplatin and oxaliplatin showed that cell membrane organic cation transporters (OCT) can be involved in the uptake of platinum drugs in the form of cationic solvento species [20][21][22][23][24][25]. This was also confirmed by recent studies on new cationic Pt(II)-complexes showing a relevant antitumor activity [5,19,26,27].
It should be also underlined that there is a limited number of studies on platinum organometallic compounds tested as antitumor drugs [28][29][30][31][32][33][34][35]. In this regard, the Zeise's anion, [PtCl 3 (η 2 -C 2 H 4 )] − , showing an extreme lability of the trans to olefin chlorido ligand and a peculiar reactivity of the coordinated ethene, resulted to be a suitable precursor for the synthesis of new antitumor organometallic complexes [19]. In fact, it allows the formation of platinum(II) organometallic complexes by the promotion of coordination in the platinum(II) sphere of specific carrier ligands and the exploitation of the intrinsic π-acid character of platinum-coordinated η 2 -ethene [36][37][38][39]. We already explored the synthetic possibilities offered by the peculiar reactivity of the Zeise's anion in strongly basic oxydrilated solvents, demonstrating that in these conditions, anionic species of the type trans-[PtCl 2 (η 2 -C 2 H 4 )(OR)] − (R = H, alkyl) are easily formed. Interestingly, the last product shows a quite different reactivity with respect to that of the Zeise's anion, allowing the synthesis of [PtCl(η 1 -C 2 H 4 OR)(N-N)] (N-N = dinitrogen ligand) complexes. These include, besides the here considered [PtCl(η 1 -C 2 H 4 OMe)(phen)] (phen = 1,10-phenanthroline, 1), also other complexes bearing even more hindered dinitrogen ligands [39].
The aim of this work was to synthesize and characterize new cationic Pt(II) complexes able to cross the cell membrane and exhibiting antitumor activity with mechanisms potentially different from that of cisplatin. In this way, we planned to change and possibly improve the selectivity and efficacy toward tumor cells, with eventual reduction of side effects. In particular, we focused on Pt(II)-phen derivatives, which can also be synthesized in high yields, Ref. [39] targeting new water-soluble cationic coordination compounds of the type [Pt(η 1 -C 2 H 4 OR)(L)(phen)] + (L = medium labile or not labile substituent; R = alkyl). The specific interest toward this type of complexes relies on their potential ability to interact with organic cation transporters [20][21][22][23][24][25] and to intercalate DNA and other nucleic acids, as generally observed in the case of phen derivatives [40,41]. These complexes were obtained and tested as new potential Pt-based antitumor drugs, as described in the present work.

Reagents and Methods
Commercially available reagents and solvents were used as received, without further purification. Zeise's salt, K[PtCl 3 (η 2 -C 2 H 4 )]·H 2 O, was synthesized with a previously reported procedure, as indicated in ref. [39]. Elemental analyses were performed with a CHN Eurovector EA 3011. NMR spectra were recorded with Bruker Avance III 400 and 600 NMR spectrometers, equipped with probes for inverse detection and with z gradient for gradient-accelerated NMR spectroscopy. 1

Synthesis of Pt(II) Complexes
[PtCl(η 1 -C 2 H 4 OMe)(phen)] (1). Na (81 mg, 3.5 mmol) was dissolved in 3.5 mL of MeOH (−20 • C, ice-NaCl bath). After evolution of H 2 , the temperature was increased to 0 • C (ice bath), then the Zeise's salt, K[PtCl 3 (η 2 -C 2 H 4 )]·H 2 O, (100 mg, 0.26 mmol) was added under magnetic stirring. A white precipitate of KCl was immediately formed, and the color of the solution changed from bright to pale yellow. After about 2 min from dissolution, 1,10-phenanthroline (47 mg, 0.26 mmol) was added to the reaction mixture, maintaining a vigorous stirring. After a few more minutes, the formation of a yellow precipitate, poorly soluble in cold methanol, started. The reaction was then followed by 1 H NMR spectroscopy, by analyzing 10 µL aliquots of the rection suspension after dissolution in an NMR tube containing 500 µL of CD 3 OD. In this way, the 1 H NMR signals of the final product 1, the free phen ligand, and the trans-[PtCl 2 (η 2 -C 2 H 4 )(OMe)] − intermediate species, could be observed. The three specie's relative amount changed with time in favor of complex 1, which was the only remaining precipitate at the end of the reaction, after about 4 h. The final product was first isolated by filtration, then washed with abundant water (to eliminate excess base and formed NaCl), and finally dried under vacuum. Complex 1 yield 115 mg, 95% referred to platinum. Anal. Calcd. for C 15

Cell Cultures
The human cancer cell lines used for the evaluation of the Pt compounds cytotoxicity (see Table 1) were cultured in different culture media, as reported below:

Cytotoxicity Assays
We evaluated the IC 50 with 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenoltetrazolium bromide (MTT) and sulforhodamine B (SRB) assays. The SRB assay and the conversion of MTT by cells were both used as an indicator of cell number, as described previously [42]. Cells, at 70-80% confluency, were trypsinized (0.25% trypsin with 1 mM EDTA), washed, and resuspended in growth medium. Then, 100 µL of a cell suspension (8 × 10 4 cells/mL) was added to each well of a 96-well plate. After overnight incubation, cells were treated with different Pt compounds for 12-72 h. The percentage cell survival was calculated as the absorbance ratio between treated and untreated cells (medium-treated control cell). Viable cells were also counted by the Trypan blue exclusion assay and light microscopy. The data presented are means standard deviation (S.D.) from eight replicate wells per microliter plate, repeated three times.

Analysis by ICP-AES (Inductively Coupled Plasma Atomic Emission Spectroscopy)
For the determination of platinum concentration, each sample (consisting of cellular pellet) was previously treated with 0.5 mL of 67% super-pure nitric acid at room temperature for 24 h, to obtain a clear solution. Acid digestion was required to destroy the complex organic matrix and obtain the complete mineralization of the samples, essential for subsequent determinations. The samples were then diluted to a final volume of 5 mL, in order to obtain a suitable concentration of the acid used in the mineralization process and avoid damage to the system. Each sample was filtered before analysis (0.45 µm) to prevent the entry of any remaining suspension in the measuring instrument. The platinum concentration in the analyzed samples was determined by a Thermo iCAP 6000 spectrometer. The spectrophotometer was calibrated with a calibration line consisting of four points, each corresponding to a concentration of the element: 1 µg/L, 10 µg/L, 100 µg/L, 1000 µg/L.

Results and Discussion
3.1. PtCl(η 1 -C 2 H 4 OMe)(phen)] (1), phen = 1,10-phenanthroline We obtained the solid [PtCl(η 1 -C 2 H 4 OMe)(phen)] (1) complex as a precipitate, by following a previously reported synthetic procedure [39,43]. The Zeise's salt, K[PtCl 3 (η 2 -C 2 H 4 )]·H 2 O, was dissolved in strongly basic methanol ([NaOMe] ≈ 1M) before addition of the stoichiometric amount of phen. In this way, the neutral [PtCl 2 (η 2 -C 2 H 4 )(phen)] pentacoordinate complex, normally produced when the phen ligand is added to a methanol solution of the Zeise's salt [44], was not observed. The absence of pentacoordinate species in the presence of excess NaOMe was due to both its high reactivity in the adopted experimental conditions and the preliminary formation of the trans-[PtCl 2 (η 2 -C 2 H 4 )(OMe)] − anionic complex which is characterized by a lower reactivity with respect to N-donors [39]. This reaction mechanism (Scheme 1) was also confirmed in the present case by 1 H NMR monitoring of complex 1 formation reaction.  In this work, we evaluated the possibility to synthesize water-soluble phenanthroline cationic complexes suitable for exhibiting antitumor activity, as previously observed in the case of similar [PtCl(η 2 -C 2 H 4 )(N-N)] + , N-N = (R,R) and (S,S)-1,2-diaminocyclohexane, cationic species [19]. These latter compounds were also suggested as possible prodrugs introducing the active Pt(1R,2R-diaminocyclohexane) moiety in the cells. Unfortunately, due to the extremely high tendency to lose the η 2 -ethene ligand in the presence of nucleophilic agents (as the simple Cl − ligand), the analogous [PtCl(η 2 -C 2 H 4 )(phen)] + complex results unsuitable for a slow release of the Pt(phen) moiety in physiological conditions. On the other hand, also the direct use of [PtCl(η 1 -C 2 H 4 OMe)(phen)] (1), as precursor of the olefin cationic species [39,43], for antitumor activity evaluations, was prevented due to its extremely low water solubility. Therefore, we decided to synthesize and test new types of water-soluble cationic species of the type [Pt(η 1 -C 2 H 4 OMe)(L)(phen)] + obtained starting from complex 1, after chlorido substitution with a neutral L ligand. Such new species are expected to be water-soluble prodrugs, possible sources of Pt(L)(phen) and Pt(phen) active moieties, and suitable for antitumor activity evaluation. For this reason, in the present study we focused on the synthesis and biological activity evaluation of two [Pt(η 1 -C 2 H 4 OMe)(L)(phen)]Cl complexes as prototypes of those containing, besides the chelated phen and the η 1 -C 2 H 4 OMe, a further nitrogen (L = NH 3 ) or sulfur (L = DMSO) ligand in the Pt(II) coordination sphere. The observed δ( 195 Pt) NMR signal exhibited by complex 2 corresponds to a chemical shift about 80 ppm lower, with respect to that of complex 1, as expected on the basis of the slightly higher NMR shielding ability of a coordinated ammonia with respect to a chlorido ligand [45]. The bidimensional [ 1 H, 195 Pt]-HETCOR spectrum, highlighting the cross peaks, accounts for the NMR couplings between phenanthroline H9 (close to the N10 which is in trans to NH 3 ), coordinated NH 3 , σ-bonded C α H 2 , C β H 2 , and the central 195 Pt (evidenced in red, in the schematic representation of the molecular structure reported in Figure 1
A 195 Pt NMR chemical shift decrease of about 800 ppm was observed, going from complex 1 to 3. Consistently, a similar 195 Pt NMR chemical shift decrease was observed, going from the cis-[PtCl 2 {1,4-(i-Pr) 2   As observed by 1 H NMR monitoring in neutral D 2 O solution, in sharp contrast with respect to 2, complex 3 slowly underwent ethene and methanol release and further DMSO dissociation when dissolved in water, even in the presence of DMSO ( Figure S1). As a consequence, besides the hydrolysis of the σ-C 2 H 4 OMe moiety, with release of ethane and methanol, the formation of hydrolytic products was also observed. In particular, NMR spectral data and literature evidence of the high acidity of Pt coordinated water in these systems, suggests the formation of [Pt(DMSO)(OH)(phen)] + and [{Pt(µ-OH)(phen)} 2 ] 2+ complexes with an asymmetric and a symmetric set of phenanthroline 1 H NMR signals, respectively. Both the new [Pt(DMSO)(OH)(phen)] + and the already known [{Pt(µ-OH)(phen)} 2 ] 2+ complexes [47] were spectroscopically identified ( Figure S1), in the latter case also by 195 Pt NMR spectroscopy ( Figure S2). A mechanism for the observed hydrolysis process is reported in Scheme 2. During hydrolysis, the formation of slight amounts of the insoluble [PtCl 2 (phen)] complex, separating as a precipitate, was also observed. Therefore, for further biological studies, complex 3 was prepared in and used from DMSO/H 2 O = 90:10 stock solutions where it demonstrated a long-term stability (over 2 weeks at room temperature). It should be noted that, despite the observed decomposition of 3 in D 2 O, a small amount of water in the presence of excess DMSO disfavored complex 3 decomposition by chlorido reentering in the Pt coordination sphere and increased complex solubility in the stock solution. Complex 3 resulted indefinitely stable when stocked at −20 • C in DMSO/H 2 O = 90:10 mother solutions at a [3] ≈10 mM. For all cell lines, viable cell number values obtained after [Pt(η 1 -C 2 H 4 OMe)(DMSO)(phen)] + treatment were corrected for a "blank" value ('blank' was the mean optical density of the background control wells containing the incubation medium and 0.725% DMSO, representing the highest DMSO quantity present in complex 3 solution). Furthermore, cell viability after 0.725% DMSO treatments of SK-OV-3, Hep-G2, and SH-SY5Y cell lines did not change significantly over 0-72 h incubation times and is shown in Figure S3.
Of the two new organometallic Pt(II)-complexes containing 1,10-phenantroline, only complex 3 proved to be more effective than cisplatin in many of the cell lines used. In fact, it was possible to calculate the IC 50 for compound 2 only in HeLa cells and after 72 h of incubation (190.9 ± 8.2 µM), as shown in Figure 3; in all the other cell lines and for all the incubation times tested, it was not possible to determine an IC 50 value. It is generally known that the presence of DMSO causes a reduction of cytotoxicity or complete inactivation of most of the tested antitumor drugs [46][47][48][49]. Interestingly, we found that in the case of complex 3, when administered to tumor cells, the presence of DMSO conferred a level of cytotoxicity comparable to that of cisplatin, rather than inactivating the molecule. This was probably due to the hydrolysis mechanism involving first the loss of the η 1 -C 2 H 4 OMe moiety and further the generation of active hydrolytic species exhibiting a cytotoxicity comparable to that of cisplatin. Conversely, [Pt(η 1 -C 2 H 4 OMe)(DMSO)(phen)] + (3) showed higher cytotoxic effects than cisplatin in six of the eight cell lines treated, with an IC 50 evaluable already at 12 h of incubation ( Figure 3 and Table S1). IC 50 at 12 h ranged between 56.9 ± 7.7 µM (SH-SY5Y cells) and 139.1 ± 6.8 µM (ZL-55 cells), and IC 50 at 24 h ranged between 45.8 ± 7.8 µM (SH-SY5Y cells) and 178.5 ± 9.3 µM (Hep-G2 cells) (Table S1). Compared to cisplatin, a drastic reduction of cell viability after treatment with 3 was observed mainly in neuroblastoma (SH-SY5Y); no significant differences between cisplatin and complex 3 were observed in Hep-G2 cells. Furthermore, an intermediate sensitivity to 3 was shown by SK-OV-3 cells (Figure 3 and Table S1). The comparisons between the mortality curves of cisplatin,  As it can be seen, in SH-SY5Y cells, complex 3 was significantly more cytotoxic than cisplatin up to 24 h of treatment (IC 50 between 56.9 ± 7.7 µM and 45.8 ± 7.8 µM), and at much longer times (72 h) the effects were the same (Figure 4). In SK-OV-3 cells, complex 3 caused a halving of the cell population rather early (12-24 h) (IC 50 between 92.1 ± 7.8 µM and 62.8 ± 6.6 µM). By contrast, the IC 50 was evaluable with cisplatin starting from 24 h of treatment (IC 50 130 ± 8.9 µM). However, also in this cell line, no significant differences were observed between 3 and cisplatin at very long incubation times (72 h) (IC 50 37.5 ± 3.2 µM and 41.7 ± 4.6 µM, respectively) ( Figure 5). With regard to Hep-G2 cells, there were no differences between the effects of 3 and those of cisplatin at any of the times tested and for any concentration in the 0.1-200 µM range ( Figure 6).
Then, we established whether the different sensitivity of the three cell lines mentioned above (Hep-G2, SH-SY5Y, and SK-OV-3) depended on a different intracellular accumulation of the Pt(II) compounds. In accordance with what has been found for cisplatin [50] and for [Pt(O,O -acac)(γacac)(DMS)] [42,51,52], the here considered new Pt(II) compounds seemed to enter cells via a passive or facilitated passive transport very quickly, and the cellular accumulation of 3 was higher than that of both cisplatin and 2 ( Figure 7 and Table S2). Thus, the high cytotoxicity of the [Pt(η 1 -C 2 H 4 OMe)(DMSO)(phen)] + complex may be related to a high intracellular uptake with respect to cisplatin, both in SH-SY5Y (about 4.5-fold) ( Figure 7A) and SK-OV-3 cells (about 18-fold) ( Figure 7B). However, no correlation between the cytotoxicity and the intracellular uptake of the [Pt(η 1 -C 2 H 4 OMe)(NH 3 )(phen)] + complex was found; in fact, this compound did not decrease cell viability, although its intracellular uptake was greater than that of cisplatin in both SH-SY5Y (about four-fold) ( Figure 7C) and SK-OV-3 cells (about three-fold) ( Figure 7B). In addition, although complex 2 accumulated considerably, it showed no significant cytotoxic effects in Hep-G2 cells ( Figure 6). Therefore, these Pt(II)-based compounds exhibited very different uptake profiles and cytotoxicity levels in the three considered cell lines, suggesting diversified biological effects, thus making them potentially tissue-and tumor-specific drugs.

Conclusions
Several strategies have been developed to solve some of the problems related to the clinical application of cisplatin, since its approval. Greater spectrum of efficacy, lower side effects, and enhanced water solubility have been constantly targeted by the researchers in the field [1][2][3][4]16]. In this work, specific platinum drugs in the form of cationic solvento species and, therefore, possible substrates for organic cation transporters (OCT) were studied. In particular, two novel water-soluble organometallic cationic species, [Pt(η 1 -C 2 H 4 OMe)(NH 3 )(phen)] + (2) and [Pt(η 1 -C 2 H 4 OMe)(DMSO)(phen)] + (3), were obtained from [PtCl(η 1 -C 2 H 4 OMe)(phen)] by clorido substitution, using the Zeise's anion, [PtCl 3 (η 2 -C 2 H 4 )] − as starting material. The new cationic complexes (2, 3) were spectroscopically characterized, and their cellular uptake and cytotoxicity evaluated, in comparison with cisplatin, for a range of tumor cell lines. The Pt(II)-based drugs exhibited very different uptake profiles and cytotoxicity levels in the considered cell lines, suggesting diversified biological effects. The DMSO complex (3) showed general higher cytotoxicity, with respect to the amino species 2, exhibiting also lower IC 50 values with respect to cisplatin in almost all cell lines starting from the first hours of treatment, particularly in neuroblastoma and adenocarcinoma cell lines. It is also worth pointing out that cellular uptake did not appear to be directly related to cytotoxicity as (a) there was a significant uptake of the DMSO complex 3 in hepatocarcinoma cells without this compound being more effective than cisplatin; (b) the uptake of the NH 3 complex was elevated in liver cells without demonstrating important cytotoxic effects; (c) the uptake of the new complexes was always greater than that of cisplatin and strain-dependent in the investigated cells, even if their cytotoxicity was not always greater than that of cisplatin. The different behavior of the two [Pt(η 1 -C 2 H 4 OMe)(L)(phen)] + complexes could be related to the much higher lability of the DMSO derivative with respect to that with NH 3 , with consequent easier formation of active solvento species containing the Pt(phen) moiety. This aspect differentiates complex 3 from cisplatin where the presence of DMSO can inhibit the formation of the antitumor active aquo species, with a consequent general reduction of the observed cytotoxicity [46,49,53]. More in-depth studies will be conducted in order to determine the mechanism of action of the [Pt(η 1 -C 2 H 4 OMe)(DMSO)(phen)] + (3) complex and its metabolic alterations in the cell lines more sensitive to its antiproliferative activity. Nevertheless, the new Pt(II)based compounds exhibited specific uptake profiles and different cytotoxicity levels in the considered cell lines, suggesting the occurrence of diversified biological effects. Therefore, a potential use of the obtained species as models for tissue-and tumor-specific drugs design is also envisaged.