Square-Planar vs. Trigonal Bipyramidal Geometry in Pt(II) Complexes Containing Triazole-Based Glucose Ligands as Potential Anticancer Agents

This article describes the synthesis, characterization, and biological activity of novel square-planar cationic platinum(II) complexes containing glucoconjugated triazole ligands and a comparison with the results obtained from the corresponding five-coordinate complexes bearing the same triazole ligands. Stability in solution, reactivity with DNA and small molecules of the new compounds were evaluated by NMR, fluorescence, and UV–vis absorption spectroscopy, together with their cytotoxic action against pairs of immortalized and tumorigenic cell lines. The results show that the square-planar species exhibit greater stability than the corresponding five-coordinate ones. Furthermore, although the square-planar complexes are less cytotoxic than the latter ones, they exhibit a certain selectivity. These results simultaneously demonstrate that overall stability is a fundamental prerequisite for preserving the performance of the agents and that coordinative saturation constitutes a point in favor of their biological action.


Introduction
Organometallic complexes have been currently considered valid anticancer agents due to their unique features [1][2][3]. Platinum compounds play a prevalent role in metal-based anticancer therapies, with cisplatin, carboplatin, and oxaliplatin used all over the world in clinics despite general low selectivity and drug resistance often limiting their efficacy [4,5]. To date, the efforts of the scientific community have multiplied to prepare increasingly effective agents, aimed at reducing the numerous and unpleasant side effects of cisplatin and its close derivatives [6][7][8][9][10][11][12][13][14][15][16]. One of the most successful strategies is the introduction of bio-active molecules in the coordination environment of the metal, capable of increasing its versatility and selectivity [17,18]. Among these, sugars are an excellent example due to their biocompatibility, the possibility of modulating their chemico-physical properties, and the ability of selective recognition by cancer cells, known as the Warburg effect [19][20][21][22][23][24][25][26][27][28]. In this context, over the last few years, our research group has proposed new organometallic glycoconjugate platinum complexes as innovative anticancer agents [29][30][31][32][33].
In particular, we intensely explored five-coordinate platinum(II) complexes ( Figure 1) characterized by trigonal bipyramidal geometry, with the equatorial plane occupied by the bidentate ligand 2,9-dimethyl-1,10-phenantroline (dmphen) and ethylene, while the sugar fragment occupies one of the axial positions [34]. These species combine glycoconjugation with coordinative saturation, a prerogative that is supposed to increase the stability of the structure and preserve its integrity up to the cellular target. The biological results confirmed the expectations because 1Pt [30] and 2Pt [31] showed, in some cases, remarkable activity glycoconjugation with coordinative saturation, a prerogative that is supposed to increase the stability of the structure and preserve its integrity up to the cellular target. The biological results confirmed the expectations because 1Pt [30] and 2Pt [31] showed, in some cases, remarkable activity and selectivity towards pairs of tumor and healthy cell lines, while the entire panel of complexes 3Pt [29] displayed high activity, although poor selectivity. Since the versatility of platinum organometallic chemistry allows us to span between different coordination numbers and geometries, we were stimulated to synthesize new carbohydrate-based platinum agents, 4Pt (Figure 2), to better understand the effect of the coordination environment on cytotoxicity as well as selectivity: These compounds share with the five-coordinate 3Pt agents the oxidation state II of Pt, a positive charge, a glycoconjugate ligand, and a methyl group. On the other hand, they are four-coordinate, and the bidentate nitrogen chelate dmphen is replaced with 1,10phenanthroline (phen), which does not introduce steric hindrance in the coordination plane and, hence, is more suitable in square-planar environments. It is noteworthy that the planarity of the molecules, combined with the presence of aromatic moieties, phen, and triazo-pyridine/imidazole ligands, makes the family of 4Pt potential metallointercalating agents, capable of intercalating in DNA base pairs by π-π stacking and weak electrostatic interactions [35][36][37][38]. Representative complexes present, respectively, pyridine (4Pt-py) and imidazole (4Pt-im), decorated with a peracetylated glucose and a ligand in which the sugar is completely deprotected (4Pt-imdep). To the best of our knowledge, this is a rare example of a homogeneous comparison between two classes of Since the versatility of platinum organometallic chemistry allows us to span between different coordination numbers and geometries, we were stimulated to synthesize new carbohydrate-based platinum agents, 4Pt (Figure 2), to better understand the effect of the coordination environment on cytotoxicity as well as selectivity: glycoconjugation with coordinative saturation, a prerogative that is supposed to increase the stability of the structure and preserve its integrity up to the cellular target. The biological results confirmed the expectations because 1Pt [30] and 2Pt [31] showed, in some cases, remarkable activity and selectivity towards pairs of tumor and healthy cell lines, while the entire panel of complexes 3Pt [29] displayed high activity, although poor selectivity. Since the versatility of platinum organometallic chemistry allows us to span between different coordination numbers and geometries, we were stimulated to synthesize new carbohydrate-based platinum agents, 4Pt (Figure 2), to better understand the effect of the coordination environment on cytotoxicity as well as selectivity: These compounds share with the five-coordinate 3Pt agents the oxidation state II of Pt, a positive charge, a glycoconjugate ligand, and a methyl group. On the other hand, they are four-coordinate, and the bidentate nitrogen chelate dmphen is replaced with 1,10phenanthroline (phen), which does not introduce steric hindrance in the coordination plane and, hence, is more suitable in square-planar environments. It is noteworthy that the planarity of the molecules, combined with the presence of aromatic moieties, phen, and triazo-pyridine/imidazole ligands, makes the family of 4Pt potential metallointercalating agents, capable of intercalating in DNA base pairs by π-π stacking and weak electrostatic interactions [35][36][37][38]. Representative complexes present, respectively, pyridine (4Pt-py) and imidazole (4Pt-im), decorated with a peracetylated glucose and a ligand in which the sugar is completely deprotected (4Pt-imdep). To the best of our knowledge, this is a rare example of a homogeneous comparison between two classes of These compounds share with the five-coordinate 3Pt agents the oxidation state II of Pt, a positive charge, a glycoconjugate ligand, and a methyl group. On the other hand, they are four-coordinate, and the bidentate nitrogen chelate dmphen is replaced with 1,10-phenanthroline (phen), which does not introduce steric hindrance in the coordination plane and, hence, is more suitable in square-planar environments. It is noteworthy that the planarity of the molecules, combined with the presence of aromatic moieties, phen, and triazo-pyridine/imidazole ligands, makes the family of 4Pt potential metallo-intercalating agents, capable of intercalating in DNA base pairs by π-π stacking and weak electrostatic interactions [35][36][37][38]. Representative complexes present, respectively, pyridine (4Pt-py) and imidazole (4Pt-im), decorated with a peracetylated glucose and a ligand in which the sugar is completely deprotected (4Pt-im dep ). To the best of our knowledge, this is a rare example of a homogeneous comparison between two classes of Pt-based anticancer agents sharing an oxidation state and the nature of the ligands in two different molecular geometries [39].
In this paper, we describe the synthesis and spectroscopic characterization of the new compounds, their in-solution stability and reactivity with DNA, and their cytotoxic activity in comparison with 3Pt analogs.

Synthesis and Characterization of Complexes of Type 4Pt
The sugar ligands py, im, and im dep were obtained through a click chemistry reaction between peracetylated 1-azido-β-D-glucose and the appropriate heterocyclic precursor 3-ethynylpiridine or 5-ethynyl-1-methyl-1H-imidazole. The peracetylated platinum(II) complexes were then prepared, starting from the chloro-precursor [PtClMe(phen)], by exchanging halogen in the presence of silver triflate, as exemplified in Scheme 1 for 4Pt-im (path i): Pt-based anticancer agents sharing an oxidation state and the nature of the ligands in two different molecular geometries [39].
In this paper, we describe the synthesis and spectroscopic characterization of the new compounds, their in-solution stability and reactivity with DNA, and their cytotoxic activity in comparison with 3Pt analogs.

Synthesis and Characterization of Complexes of Type 4Pt
The sugar ligands py, im, and imdep were obtained through a click chemistry reaction between peracetylated 1-azido-β-D-glucose and the appropriate heterocyclic precursor 3ethynylpiridine or 5-ethynyl-1-methyl-1H-imidazole. The peracetylated platinum(II) complexes were then prepared, starting from the chloro-precursor [PtClMe(phen)], by exchanging halogen in the presence of silver triflate, as exemplified in Scheme 1 for 4Ptim (path i): Scheme 1. Synthesis of the four-coordinate platinum(II) complex 4Pt-im.
After filtration of the precipitated silver chloride, the complexes were isolated as yellow powder by reducing the volume of the filtrate and adding diethyl ether. The corresponding deprotected species, 4Pt-imdep, was prepared by treatment of 4Pt-im in methanol containing a catalytic amount of potassium hydroxide. Attempts to isolate the square-planar dmphen analogs were performed by reacting the chloro-dmphen precursor in the same conditions reported in Scheme 1. Unfortunately, the reaction mixture showed no clear presence of the desired product. The unsuccessful outcome of the synthesis is probably due to the steric hindrance brought into the plane by the methyl groups, which prevents the formation of the square-planar complex. Finally, complex 4Pt-py was prepared uneventfully with a procedure analogous to the synthesis of 4Pt-im. The characterization of the products was carried out through mono-and bi-dimensional NMR spectroscopy (Figures S1-S9), which allowed the unequivocal assignment to the entire panel of protons. Figure 3 shows the proton spectrum of 4Pt-im. After filtration of the precipitated silver chloride, the complexes were isolated as yellow powder by reducing the volume of the filtrate and adding diethyl ether. The corresponding deprotected species, 4Pt-im dep , was prepared by treatment of 4Pt-im in methanol containing a catalytic amount of potassium hydroxide. Attempts to isolate the square-planar dmphen analogs were performed by reacting the chloro-dmphen precursor in the same conditions reported in Scheme 1. Unfortunately, the reaction mixture showed no clear presence of the desired product. The unsuccessful outcome of the synthesis is probably due to the steric hindrance brought into the plane by the methyl groups, which prevents the formation of the square-planar complex. Finally, complex 4Pt-py was prepared uneventfully with a procedure analogous to the synthesis of 4Pt-im. The characterization of the products was carried out through mono-and bi-dimensional NMR spectroscopy (Figures S1-S9), which allowed the unequivocal assignment to the entire panel of protons. Figure 3 shows the proton spectrum of 4Pt-im. Int  As expected, the two halves of the phen ligand are not equivalent. The signal of the methyl on platinum shifts to higher frequencies (δ 1.2-1.3) than the five-coordinate species (δ 0-0.5) [29,40], and is affected by the coupling with the 195 Pt nuclei of ca. 70 Hz, typical As expected, the two halves of the phen ligand are not equivalent. The signal of the methyl on platinum shifts to higher frequencies (δ 1.2-1.3) than the five-coordinate species (δ 0-0.5) [29,40], and is affected by the coupling with the 195 Pt nuclei of ca. 70 Hz, typical of square-planar Pt-complexes containing N,N -chelating ligands [41]. The sugar proton signals exhibit the characteristic pattern of the 4 C 1 glucose chair, with the glucose in the β-configuration. The compounds were unequivocally identified by HRMS (ESI/QTOF), carried out in methanol. In the spectra, it was possible to observe peaks relative to the cationic platinum complex at m/z 869.2225, 866.2106, and 701.1802, attributable to 4Pt-im, 4Pt-py, and 4Pt-im dep , respectively (Figures S10-S12).

In-Solution Behaviour and Reactivity with Model Nucleophiles
As mentioned before, 3Pt complexes are highly cytotoxic but poorly selective. Furthermore, similar IC 50 values were recorded for the entire panel of compounds, regardless of the nature of the sugar ligand [29]. This behavior suggests that the neutral nitrogen ligands in 3Pt are vulnerable to substitution in a physiological environment, according to the equation (L= py or im): 3Pt-L + solvent = 3Pt-solvent + L Some experiments have demonstrated the ease of substitution in coordinating solvents that is plausibly facilitated by the trans-effect of the Me ligand [29]. Therefore, the active species likely do not contain the sugar label anymore, and their activities are very similar.
To compare the solution stability of the complexes, 1 H-NMR and UV-vis spectra over time were recorded for 4Pt type species dissolved in diverse aqueous (phosphate buffer, PB; pH 7.4): organic solvent mixtures, i.e., 1:1 v/v D 2 O:acetone-d 6 ; 9:1 v/v PB:DMSO-d 6 ; 1:1 v/v PB:DMSO-d 6 and DMSO-d 6 . These solvents were selected to verify the stability in pseudo-physiological conditions in the presence of solvents with different coordinating properties and concentrations.
No appreciable structural variations were observed in the 1:1 v/v D 2 O:acetone-d 6 mixture over 3 days. As an example, Figure 4 reports the 1 H-NMR spectra recorded over time for 4Pt-im.  Analogous results were obtained in 9:1 v/v PB:DMSO-d6, where only traces of complex 4Pt-dmso-d6 were observed after 48 h ( Figure 5). Analogous results were obtained in 9:1 v/v PB:DMSO-d 6 , where only traces of complex 4Pt-dmso-d 6 were observed after 48 h ( Figure 5). Analogous results were obtained in 9:1 v/v PB:DMSO-d6, where only traces of complex 4Pt-dmso-d6 were observed after 48 h ( Figure 5). These results demonstrate that coordinating molecules such as water or acetone do not change the nature of the compounds, different from the corresponding five-coordinate 3Pt complexes that are not stable in these solvents [29]. These results demonstrate that coordinating molecules such as water or acetone do not change the nature of the compounds, different from the corresponding five-coordinate 3Pt complexes that are not stable in these solvents [29].
Upon increasing the amount of DMSO-d 6 , type 4Pt complexes slowly undergo the following exchange: 4Pt-L + DMSO-d 6 = 4Pt-DMSO-d 6 + L From the spectra recorded in 1:1 v/v PB:DMSO-d 6 ( Figure 6), it is possible to appreciate the appearance of the signals (δ 6.15, 7.23, 7.68, and 8.55) relating to the free sugar ligand im, whose intensity increases over time. After 24 h, an equilibrium is reached. Analog results were collected for 4Pt-py (Figures S15-S19). The exchange rate was, however, much higher for the analogous five-coordinate species 3Pt-im, to the point that equilibrium was established in the solution within an hour.
These results were confirmed by UV-vis absorption spectra collected in pure DMSO and in 9:1 v/v PBS:DMSO-d 6 ( Figure 7) as a function of time. Indeed, data collected in pure DMSO show significant variations of the spectral profiles, which can be explained by the exchange of a metal ligand with a solvent molecule. On the contrary, spectra collected in the mixed solvent indicate that the compounds remain rather stable in PBS, with a minimal variation of absorbance that can be explained by slight precipitation.
From the spectra recorded in 1:1 v/v PB:DMSO-d6 ( Figure 6), it is possible to appreciate the appearance of the signals (δ 6.15, 7.23, 7.68, and 8.55) relating to the free sugar ligand im, whose intensity increases over time. After 24 h, an equilibrium is reached. Analog results were collected for 4Pt-py (Figures S15-S19). The exchange rate was, however, much higher for the analogous five-coordinate species 3Pt-im, to the point that equilibrium was established in the solution within an hour. These results were confirmed by UV-vis absorption spectra collected in pure DMSO and in 9:1 v/v PBS:DMSO-d6 ( Figure 7) as a function of time. Indeed, data collected in pure DMSO show significant variations of the spectral profiles, which can be explained by the exchange of a metal ligand with a solvent molecule. On the contrary, spectra collected in the mixed solvent indicate that the compounds remain rather stable in PBS, with a minimal variation of absorbance that can be explained by slight precipitation.  These results demonstrate that four-coordinate complexes preserve their structure more efficiently than the corresponding five-coordinate species in mixed solvents. This behavior can be interpreted by considering the different trans-effects experienced by the sugar ligand in the two types of compounds. In five-coordinate complexes, the presence of a methyl destabilizes the Pt-N bond and facilitates the substitution. In the case of square-planar complexes, one phenanthroline nitrogen occupies the trans coordination site, and it is therefore reasonable to observe a lower reactivity of the complexes. This result, therefore, encourages the valuation of the biological activity of 4Pt complexes, with the expectation of a possible recognition action exerted by the sugar portion on the cellular target and a consequent beneficial effect on selectivity. These results demonstrate that four-coordinate complexes preserve their structure more efficiently than the corresponding five-coordinate species in mixed solvents. This behavior can be interpreted by considering the different trans-effects experienced by the sugar ligand in the two types of compounds. In five-coordinate complexes, the presence of a methyl destabilizes the Pt-N bond and facilitates the substitution. In the case of square-planar complexes, one phenanthroline nitrogen occupies the trans coordination site, and it is therefore reasonable to observe a lower reactivity of the complexes. This result, therefore, encourages the valuation of the biological activity of 4Pt complexes, with the expectation of a possible recognition action exerted by the sugar portion on the cellular target and a consequent beneficial effect on selectivity.

Cytotoxicity Studies
The biological effect of the Pt-compounds was tested on two human tumor cell lines, MCF-7 (breast cancer cells) and A431 (epidermoid carcinoma cells), and two nontumorigenic cell lines, H9c2 (rat cardiomyoblast cells), and HaCaT (human keratinocyte cells), using the MTT assay. After 48 h of incubation, a cytotoxic effect was observed in all the analyzed cell lines, as indicated by the IC 50 values (the amount of drug able to induce 50% cell death) reported in Table 1A. In line with the different stability in solution, the square-planar species of type 4Pt exhibit different biological activity compared to that of the corresponding five-coordinate compounds of 3Pt. Indeed, due to their inertness towards the substitution of the sugar ligand, their biological activity can be influenced by the nature of the sugar ligand, whose presence in the complex can affect its internalization or mechanism of action once inside the cell. On the other hand, in the case of the five-coordinate complexes, the loss of the sugar tag plausibly determines their substantial leveling of activities [29]. Indeed, as demonstrated by the selectivity index (Table 1B), the square-planar complexes exhibit a certain degree of selectivity, much higher than that obtained in the case of cisplatin and the corresponding five-coordinate 3Pt compounds. Although the general cytotoxicity of square-planar species of type 4Pt is reduced when compared to the coordinatively saturated 3Pt species, 4Pt-py showed much higher IC 50 values on immortalized cells than cisplatin, and a strong decrease in the IC 50 value, with respect to cisplatin, was observed for A431 cells. The behavior of 4Pt-im was very similar to that of 4Pt-py. No toxic activity was observed with 4Pt-im dep , thus confirming the importance of sugar protection, as already observed in other studies [31]. The cell death mechanism induced by 4Pt-py was analyzed on A431 cells. First, the internalization process of 4Pt-py was evaluated by studying the involvement of GLUT receptors. In particular, 4Pt-py internalization and cytotoxicity were tested using quercetin as a glucose transporter inhibitor [24,30,42]. Cells were incubated in the presence or absence of 5 µM quercetin (a non-lethal concentration, data not shown), and a dose-response assay was performed with 4Pt-py. As shown in Figure 8A, no inhibition of cytotoxicity was observed in the presence of quercetin, thus excluding the involvement of GLUT receptors in the internalization pathway of 4Pt-py. Necrosis was also excluded as no LDH release was observed after 4Pt-py cell incubation with an increasing amount of the drug ( Figure 8B). The 10 µM concentration used was very similar to the IC 50 value, and then, a double value (20 µM) was also used to definitely exclude necrosis at very high concentrations. The activation of the apoptotic pathway was analyzed by Western blotting and an analysis of the mitochondrial potential as it is well known that mitochondrial outer membrane permeabilization is essential to initiate mitochondrial apoptosis. Cells were incubated for 48 h in the presence of 12 μM 4Pt-py (the concentration of 4Pt-py at which the IC50 value was obtained), and the activation of caspase-3 and caspase-9 was analyzed. As shown in Figure 9A,B, a small but significant decrease in the signal associated with both pro-caspases was observed. Accordingly, when the mitochondrial potential (Δψm) was The activation of the apoptotic pathway was analyzed by Western blotting and an analysis of the mitochondrial potential as it is well known that mitochondrial outer membrane permeabilization is essential to initiate mitochondrial apoptosis. Cells were incubated for  48 h in the presence of 12 µM 4Pt-py (the concentration of 4Pt-py at which the IC 50 value was obtained), and the activation of caspase-3 and caspase-9 was analyzed. As shown in Figure 9A,B, a small but significant decrease in the signal associated with both pro-caspases was observed. Accordingly, when the mitochondrial potential (∆ψm) was measured, a significant decrease in A431 cells was observed at both the IC 50 value and the doubled amount (20 µM) ( Figure 9C).

DNA Binding Properties
Since the main target of Pt-based drugs is DNA, we evaluated the ability of the compounds to interact with nucleobases and double helixes. The fluorescence intensity of the probe, related to ∆ψm, is reported as a percentage of the control (%). Data shown are the means ± SD of three independent experiments; * indicates p < 0.05, *** indicates p < 0.005, and **** indicates p < 0.001 with respect to untreated cells.

DNA Binding Properties
Since the main target of Pt-based drugs is DNA, we evaluated the ability of the compounds to interact with nucleobases and double helixes.
To evaluate the ability of the compounds to interact with double-helix DNA, ethidium bromide (EtBr) displacement fluorescence assays using calf-thymus DNA (ctDNA) were carried out. Attempts to collect circular dichroism spectra of ctDNA in the presence of the compounds failed due to the formation of a yellow precipitate. Results of the fluorescence assay clearly demonstrate that the compounds belonging to the 4Pt series displace the intercalating agent EtBr from ctDNA ( Figure 10). These results indicate that the compounds bind DNA [43]. were carried out. Attempts to collect circular dichroism spectra of ctDNA in the presence of the compounds failed due to the formation of a yellow precipitate. Results of the fluorescence assay clearly demonstrate that the compounds belonging to the 4Pt series displace the intercalating agent EtBr from ctDNA ( Figure 10). These results indicate that the compounds bind DNA [43]. To shed light on the interaction between the complexes and DNA, the interaction with the model nucleobase 2′-deoxyguanosine-5′-monophosphate (dGMP) was studied by 1 H-NMR spectroscopy and ESI mass spectrometry. Indeed, N7 in guanine residues are known to be the main binding sites for platinum complexes. Complexes 4Pt-im and 4Ptpy were incubated at 37 °C in 9:1 v/v PB:DMSO-d6 in the presence of 2.5 eq of dGMP, and NMR spectra were recorded over time ( Figure 11). To shed light on the interaction between the complexes and DNA, the interaction with the model nucleobase 2 -deoxyguanosine-5 -monophosphate (dGMP) was studied by 1 H-NMR spectroscopy and ESI mass spectrometry. Indeed, N7 in guanine residues are known to be the main binding sites for platinum complexes. Complexes 4Pt-im and 4Pt-py were incubated at 37 • C in 9:1 v/v PB:DMSO-d 6 in the presence of 2.5 eq of dGMP, and NMR spectra were recorded over time ( Figure 11). After 48 h, no coordination of dGMP to the platinum center was observed, while traces to the complex 4Pt-DMSO were found. Such results were also confirmed by ESI mass spectrometry experiments performed in 9:1 v/v H2O:DMSO. The analysis of the reaction mixture at t = 0 ( Figure 12A) and after 48 h of incubation at 37 °C ( Figure 12B) confirmed the lack of coordination by dGMP and the formation of the 4Pt-DMSO complex. Analog results were obtained for the 4Pt-py complex ( Figure S19).  After 48 h, no coordination of dGMP to the platinum center was observed, while traces to the complex 4Pt-DMSO were found. Such results were also confirmed by ESI mass spectrometry experiments performed in 9:1 v/v H2O:DMSO. The analysis of the reaction mixture at t = 0 ( Figure 12A) and after 48 h of incubation at 37 °C ( Figure 12B) confirmed the lack of coordination by dGMP and the formation of the 4Pt-DMSO complex. Analog results were obtained for the 4Pt-py complex ( Figure S19).  Such results, combined with the suppressed ligand exchange in aqueous media (see Section 2.2), may indicate that the family of 4Pt agents interacts in vitro with DNA through an intercalation mechanism [44]. Indeed, classical Pt(II)-based drugs (e.g., cisplatin and its derivatives), upon the hydrolysis of leaving ligands, can be coordinated by nucleobases, resulting in covalent adducts. The inertness of 4Pt complexes toward ligand exchange, along with their planarity and the presence of aromatic phenanthroline and triazo-imidazole/pyridine ligands, can favor non-covalent interactions with the double helix [45][46][47][48].

Reactivity with S-Donor Molecules
Due to its softness, platinum forms stable bonds with sulfur-donor nucleophiles, which are ubiquitous in biological environments and play an important role in the fate of platinum-based drugs. Reduced glutathione (GSH) is a tripeptide involved in essential processes in the cell, such as redox balance and detoxification [49]. Platinum complexes promptly react with the thiol group of GSH, forming adducts that are often no longer active. Similarly, the thioether moiety of L-methionine is often involved in the sequestration of platinum complexes by proteins [50,51]. Hence, the study of the reactivity of the synthesized complexes with GSH and methionine can help to rationalize and clarify their biological activity. The interaction between compounds 4Pt-im and 4Pt-py, incubated at 37 • C in the presence of 2.5 eq. of methionine in 9:1 v/v D 2 O (PBS, pH 7.4):DMSO-d 6 , was monitored by 1 H-NMR spectroscopy. Spectra collected over time show no sign of coordination of methionine within 48 h (Figures S20 and S21).
Conversely, the compounds react with GSH in the same conditions as indicated by 1 H-NMR and ESI-MS. In the NMR spectra ( Figure 13) recorded at different times for 4Pt-im, it is possible to observe from the point of t = 0 the appearance of new signals in the aromatic and glucose regions. Such results, combined with the suppressed ligand exchange in aqueous media (see Section 2.2), may indicate that the family of 4Pt agents interacts in vitro with DNA through an intercalation mechanism [44]. Indeed, classical Pt(II)-based drugs (e.g., cisplatin and its derivatives), upon the hydrolysis of leaving ligands, can be coordinated by nucleobases, resulting in covalent adducts. The inertness of 4Pt complexes toward ligand exchange, along with their planarity and the presence of aromatic phenanthroline and triazo-imidazole/pyridine ligands, can favor non-covalent interactions with the double helix [45][46][47][48].

Reactivity with S-Donor Molecules
Due to its softness, platinum forms stable bonds with sulfur-donor nucleophiles, which are ubiquitous in biological environments and play an important role in the fate of platinum-based drugs. Reduced glutathione (GSH) is a tripeptide involved in essential processes in the cell, such as redox balance and detoxification [49]. Platinum complexes promptly react with the thiol group of GSH, forming adducts that are often no longer active. Similarly, the thioether moiety of L-methionine is often involved in the sequestration of platinum complexes by proteins [50,51]. Hence, the study of the reactivity of the synthesized complexes with GSH and methionine can help to rationalize and clarify their biological activity. The interaction between compounds 4Pt-im and 4Pt-py, incubated at 37 °C in the presence of 2.5 eq. of methionine in 9:1 v/v D2O (PBS, pH 7.4):DMSO-d6, was monitored by 1 H-NMR spectroscopy. Spectra collected over time show no sign of coordination of methionine within 48 h (Figures S20 and S21).
Conversely, the compounds react with GSH in the same conditions as indicated by 1 H-NMR and ESI-MS. In the NMR spectra ( Figure 13) recorded at different times for 4Ptim, it is possible to observe from the point of t = 0 the appearance of new signals in the aromatic and glucose regions. Such resonances can be attributed to the release of the imidazole ligand and the consequent formation of unidentified species. Such peaks decrease in intensity over time until they almost disappear (in the case of 4Pt-py, completely; Figure S22), probably because of Such resonances can be attributed to the release of the imidazole ligand and the consequent formation of unidentified species. Such peaks decrease in intensity over time until they almost disappear (in the case of 4Pt-py, completely; Figure S22), probably because of the precipitation of decomposition products. The reaction mixture analyzed by ESI-MS immediately after the incubation of 4Pt-im with GSH at 37 • C in 9:1 v/v H 2 O:DMSO showed the presence of the intact complex at m/z 869.22. The analysis, repeated after 48 h, revealed the presence of multiple peaks ( Figure 14 and Table S1) attributable to several species, including mono-and di-nuclear platinum complexes with one or two deprotonated GSH molecules, probably with bridging sulfur atoms in the di-nuclear one, as previously reported for other phenantroline Pt(II) complexes [51].  Figure 14 and Table S1) attributable to several species, including mono-and di-nuclear platinum complexes with one or two deprotonated GSH molecules, probably with bridging sulfur atoms in the di-nuclear one, as previously reported for other phenantroline Pt(II) complexes [51].

Synthesis of 4Pt-py and 4Pt-im
An equimolar solution (5.0 mL) of the appropriate ligand in dichloromethane and a suspension of [PtClMe (1,10-phen)] (0.10 g, 0.22 mmol) and CF3SO3Ag (0.058 g, 0.22 mmol) in dichloromethane (3.0 mL) were stirred at RT for 24 h. After that, the reaction mixture was filtered on Celite ® , and the filtrate was concentrated under vacuum; diethyl ether was added to afford a yellow microcrystalline solid. The solid was then isolated, washed with diethyl ether, and then vacuum dried (yield: 70-75%).

Synthesis of 4Pt-py and 4Pt-im
An equimolar solution (5.0 mL) of the appropriate ligand in dichloromethane and a suspension of [PtClMe (1,10-phen)] (0.10 g, 0.22 mmol) and CF 3 SO 3 Ag (0.058 g, 0.22 mmol) in dichloromethane (3.0 mL) were stirred at RT for 24 h. After that, the reaction mixture was filtered on Celite ® , and the filtrate was concentrated under vacuum; diethyl ether was added to afford a yellow microcrystalline solid. The solid was then isolated, washed with diethyl ether, and then vacuum dried (yield: 70-75%).  absence of 5 µM quercetin. Quercetin alone was used as the control. After 48 h incubation, cell viability was assessed by MTT assay, as described in a previous study [30]. Control experiments were performed by either growing cells in the absence of the compound or by supplementing the cell cultures with identical volumes of buffer (water:acetone, 1:1 v/v). Each value is the mean of three independent experiments, each with three determinations. Significance was determined by Student's t-test.

LDH Release
The occurrence of necrosis was determined by measuring the release of lactate dehydrogenase (LDH) in the culture medium, as described by Sucha et al. [53]. The LDH content of the medium from untreated cells was referred to as a spontaneous release, whereas the LDH total cellular content was determined upon cell lysis. The percentage of LDH release was calculated as: Each value is the mean of three independent experiments, each with three determinations. Significance was determined by Student's t-test.

Western Blot Analyses
A431 cells were plated at a density of 3 × 10 5 cells cm −2 in a complete medium for 24 h and then treated for 48 h with 12 µM 4Pt-py. At the end of the incubation, both untreated and treated cells were analyzed, as described previously for Western blot analyses [30]. Upon the determination of total protein concentration in the supernatant by the Bradford assay, samples were analyzed by SDS-PAGE and Western blot using specific antibodies directed towards procaspase-3 or -9 (Cell Signal Technology, Danvers, MA, USA). β-actin (Sigma-Aldrich) or GAPDH (Thermo Fisher, Rockford, IL, USA) was used as the loading control. Each value is the mean of three independent experiments, each with three determinations. Significance was determined by Student's t-test.

Analysis of Mitochondrial Membrane Potential
Mitochondrial membrane potential (∆ψm) was measured as we described [31]. Cells were plated at a density of 2 × 10 4 cells per well, and, after 24 h, cells were incubated for 48 h with 12-20 µM 4Pt-py. At the end of the treatment, the cells were incubated with 200 nM of the cationic lipophilic dye tetramethylrhodamine ethyl ester (TMRE) for 20 min at 37 • C. Then, the cells were gently washed with 0.2% BSA in PBS three times, and the fluorescence was measured in a microplate reader with peak λ(ex)/λ(em) = 549/575 nm. Each value is the mean of three independent experiments, each with three determinations. Significance was determined by Student's t-test.

Ethidium Bromide Displacement Fluorescence Assay
Fluorescence spectra were collected on a HORIBA Fluoromax-4 spectrofluorometer at 25 • C using 1 cm path length cuvettes. ctDNA was incubated with ethidium bromide (EtBr) in a 1:50 molar ratio for 30 min at room temperature. Then, the complex was diluted in 10 mM ammonium acetate buffer at pH 7.5 up to a ctDNA final concentration of 200 µM. The ctDNA-EtBr complex was then titrated with a 4Pt solution (final concentration from 10 to 100 µM, stock concentration 15 mM). Fluorescence emission spectra were recorded at excitation at 545 nm after an equilibration time of 5 min following each addition of the Pt complex.

1 H-NMR and ESI-MS In-Solution Studies
The appropriate 4Pt complex (5 mg) was dissolved in 0.5 mL of DMSO-d 6 or acetoned 6 . Calculated volumes of solution were diluted to 600 µL with the appropriate volumes of PB in D 2 O (25 mM, pH 7.4) and/or DMSO-d 6 to provide a 1 mM solution of the complex with the appropriate v/v ratio of solvents. Samples were incubated at 37 • C, and NMR spectra were recorded at different times.
To study the reactivity with dGMP, L-methionine, and GSH, stock solutions (10 mM) in PB of each nucleophile were prepared and the correct volume was added to a freshly prepared solution of the complexes in 9:1 v/v PB:DMSO-d 6 to afford a final concentration of 1 mM of the Pt complex and 2.5 eq. of the nucleophile.
Samples for ESI-MS analysis were prepared as described for NMR samples, using DMSO and H 2 O instead of deuterated solvents. Prior to analysis, an aliquot of the reaction mixture was withdrawn and diluted with water to get 0.1 mM of Pt-complex solutions.

Conclusions
This work is part of broad research aimed at studying the biological activity of platinum(II) complexes containing glycoconjugated neutral ligands. The expertise of this team in favoring four-or five-coordination complexes through small structural modifications of the ligands allowed a homogeneous comparison between the activity of analogous cationic complexes in the two different coordination environments. In line with expectations, the two classes of compounds revealed profound differences, starting with the higher stability of the square-planar species with respect to the substitution of the sugar ligand. This aspect determines that their biological activity depends on the nature of the latter, whereas, in the case of five-coordinate complexes, the loss of the sugar tag plausibly provokes the substantial leveling of activities. This also implies that only the square-planar complexes exhibit a certain degree of selectivity, although with reduced cytotoxicity. The ensemble of results, also accompanied by the demonstration that the complexes interact with DNA, adds a piece to the knowledge of the properties of antitumor agents based on platinum(II), highlighting how the optimization of their performance requires the careful choice of ligands in the appropriate coordination geometry. Funding: This research was funded by Università di Napoli Federico II, grant number 000023_AL-TRI_CDA_75_2021_FRA_RUFFO.

Conflicts of Interest:
The authors declare no conflict of interest.