A Comparative Analysis of the In Vitro Anticancer Activity of Iridium(III) {η5-C5Me4R} Complexes with Variable R Groups

Piano-stool iridium complexes based on the pentamethylcyclopentadienyl ligand (Cp*) have been intensively investigated as anticancer drug candidates and hold much promise in this setting. A systematic study aimed at outlining the effect of Cp* mono-derivatization on the antiproliferative activity is presented here. Thus, the dinuclear complexes [Ir(η5-C5Me4R)Cl(μ-Cl)]2 (R = Me, 1a; R = H, 1b; R = Pr, 1c; R = 4-C6H4F, 1d; R = 4-C6H4OH, 1e), their 2-phenylpyridyl mononuclear derivatives [Ir(η5-C5Me4R)(kN,kCPhPy)Cl] (2a–d), and the dimethylsulfoxide complex [Ir{η5-C5Me4(4-C6H4OH)}Cl2(κS-Me2S=O)] (3) were synthesized, structurally characterized, and assessed for their cytotoxicity towards a panel of six human and rodent cancer cell lines (mouse melanoma, B16; rat glioma, C6; breast adenocarcinoma, MCF-7; colorectal carcinoma, SW620 and HCT116; ovarian carcinoma, A2780) and one primary, human fetal lung fibroblast cell line (MRC5). Complexes 2b (R = H) and 2d (4-C6H4F) emerged as the most active ones and were selected for further investigation. They did not affect the viability of primary mouse peritoneal cells, and their tumoricidal action arises from the combined influence on cellular proliferation, apoptosis and senescence. The latter is triggered by mitochondrial failure and production of reactive oxygen and nitrogen species.


Introduction
Few platinum compounds have been routinely administered in clinical treatments against various types of cancer [1]; however, despite their undoubtful efficacy, they present significant limitations, such as negative side effects, associated phenomena of intrinsic and acquired resistance, a limited number of treatable tumors, and the needing of hospitalization for intravenous administration [2]. These facts have stimulated the research towards new drugs based on different transition metal elements [3]. Indeed, transition metal complexes possess peculiar properties associated with the metal center, which are not available on organic compounds, and thus provide a superior medicinal potential [4]. In this regard, some categories of organometallic complexes have shown a great promise [5], and in particular piano-stool iridium(III) complexes of general formula [IrCp*(XˆY)Cl] 0/+ (Cp* = η 5 -C 5 Me 5 ; XˆY = bidentate neutral ligand or anionic 2-arylpyridyl ligand) have been intensively investigated (Figure 1, structure I) [6,7]. Compounds of this type, as well as a variety of their derivatives (see below), are typically accessible from dinuclear precursors type, as well as a variety of their derivatives (see below), are typically accessible from dinuclear precursors upon addition of the bidentate ligand (X^Y) via cleavage of chloride bridges. The presence of the strong donor pentamethylcyclopentadienyl ring provides a stabilizing effect towards the +III oxidation state of the metal, and favors the substitution of the relatively labile chloride ligand. It was found that the replacement of one methyl substituent belonging to Cp* with an aryl group (phenyl or biphenyl, see structures II and III in Figure 1) generally determines a leap in the antiproliferative activity [8]. Sadler and co-workers attributed this effect to increased lipophilicity and ability to intercalate DNA [9]. Otherwise, the opposite activity trend (i.e., I > II > III) was recently reported for 2arylpyridyl complexes on A549, HeLa, and BEAS-2B cells by Liu and co-workers, explained on the basis of electronic factors regulating chloride dissociation [10]. Neutral 2arylpyridyl complexes usually exert a relatively strong cytotoxicity by means of a multimodal action, including alteration of cellular redox balance. Further, significant enhancement of the cytotoxicity is achieved by replacing the chloride ligand with pyridine, which retards hydrolysis and results in a marked propension to trigger ROS production [11]. More recently, Pizarro and co-workers reported that a tether ring structure, accessible via modification of the Cp* with a methylene-pyridine pendant (Figure 1, structure IV), exhibits exceptional potency against MCF-7 cells, with respect to analogous benzyl species (structure V) which in turn revealed more cytotoxic than the Cp* analogues [12]. Alternative functionalization of Cp* with an alkyl-alcohol group (structure VI) has also been evaluated, resulting in modest activity against the A2780 cell line [13].
In this scenario, a comprehensive evaluation of the effects of the modification of the Cp* ring has not been fully addressed hitherto. A recent study by some of us outlined that replacement of one methyl with a range of substituents deeply influences the catalytic activity of the resulting iridium(III) tetramethylcyclopentadienyl complexes in water oxidation, through a balance of electronic and steric factors [14]. Here, we report the synthesis More recently, Pizarro and co-workers reported that a tether ring structure, accessible via modification of the Cp* with a methylene-pyridine pendant (Figure 1, structure IV), exhibits exceptional potency against MCF-7 cells, with respect to analogous benzyl species (structure V) which in turn revealed more cytotoxic than the Cp* analogues [12]. Alternative functionalization of Cp* with an alkyl-alcohol group (structure VI) has also been evaluated, resulting in modest activity against the A2780 cell line [13].
In this scenario, a comprehensive evaluation of the effects of the modification of the Cp* ring has not been fully addressed hitherto. A recent study by some of us outlined that replacement of one methyl with a range of substituents deeply influences the catalytic activity of the resulting iridium(III) tetramethylcyclopentadienyl complexes in water oxidation, through a balance of electronic and steric factors [14]. Here, we report the synthesis of a series of piano-stool 2-phenylpyridyl complexes containing a {η 5 -C 5 Me 4 R} ligand with variable R, and a study of their in vitro anticancer activity.

Synthesis and Characterization of Compounds
The neutral di-iridium complexes 1a-e were obtained from the commercial iridium(III) chloride hydrate by reaction with the appropriate substituted cyclopentadiene precursor in refluxing methanol, following the published procedures (Scheme 1). The synthesis of 1e required the preliminary protection of the hydroxyl group; thus, the reaction between IrCl 3 ·nH 2 O and C 5 HMe 4 (4-C 6 H 4 OCMe 2 OMe) directly afforded 1e after silica chromatography. All the products showed very low solubilities in water, otherwise 1a-d were well soluble in chlorinated solvents (i.e., dichloromethane and chloroform), methanol, and dimethylsulfoxide. The hydroxyl function is crucial to solubility; thus, 1e is slightly soluble in methanol and dimethylsulfoxide but not soluble in chlorinated solvents. The mononuclear 2-phenylpyridyl complexes 2a-d, including the unprecedented ones 2b-d, were synthesized by allowing the respective parent compounds 1a-d to react with a two-fold excess of 2-phenylpyridine in the presence of sodium acetate, in dichloromethane at reflux temperature. After the work-up, 2a-c were isolated in approximately 75% yields, whereas 2d was isolated in 40% yield. It was not possible to obtain the 2-phenylpyridyl derivative of 1e, essentially due to the solubility issues; on the other hand, the mono-iridium dimethylsulfoxide adduct 3 could be easily obtained from 1e and finally isolated in 71% yield (Scheme 1). The unprecedented mononuclear complexes 2b-d and 3 are insoluble in water. In contrast to 2b-d, which are well soluble in dichloromethane and chloroform, 3 manifests a very low solubility even in organic solvents, apart from DMSO. The new compounds were fully characterized by elemental analysis, IR, and NMR spectroscopy (Figures S1-S15). Due to the presence of the chiral iridium center, the NMR spectra of 2b-d (in CDCl 3 ) reveal four non-equivalent methyl groups within the {C 5 Me 4 } moiety; conversely, two resonances for such methyls are recognized in the NMR spectra of 3 (in DMSO-d 6 ), this complex lacking asymmetry. The 19 F NMR resonance due to the fluorine atom in 2d occurs at −114.5 ppm. The hydroxyl unit in 3 manifests itself with an infrared band at 3262 cm −1 and a 1 H NMR singlet at 9.68 ppm. of a series of piano-stool 2-phenylpyridyl complexes containing a {η 5 -C5Me4R} ligand with variable R, and a study of their in vitro anticancer activity.

Synthesis and Characterization of Compounds
The neutral di-iridium complexes 1a-e were obtained from the commercial iridium(III) chloride hydrate by reaction with the appropriate substituted cyclopentadiene precursor in refluxing methanol, following the published procedures (Scheme 1). The synthesis of 1e required the preliminary protection of the hydroxyl group; thus, the reaction between IrCl3·nH2O and C5HMe4 (4-C6H4OCMe2OMe) directly afforded 1e after silica chromatography. All the products showed very low solubilities in water, otherwise 1a-d were well soluble in chlorinated solvents (i.e., dichloromethane and chloroform), methanol, and dimethylsulfoxide. The hydroxyl function is crucial to solubility; thus, 1e is slightly soluble in methanol and dimethylsulfoxide but not soluble in chlorinated solvents. The mononuclear 2-phenylpyridyl complexes 2a-d, including the unprecedented ones 2b-d, were synthesized by allowing the respective parent compounds 1a-d to react with a two-fold excess of 2-phenylpyridine in the presence of sodium acetate, in dichloromethane at reflux temperature. After the work-up, 2a-c were isolated in approximately 75% yields, whereas 2d was isolated in 40% yield. It was not possible to obtain the 2-phenylpyridyl derivative of 1e, essentially due to the solubility issues; on the other hand, the mono-iridium dimethylsulfoxide adduct 3 could be easily obtained from 1e and finally isolated in 71% yield (Scheme 1). The unprecedented mononuclear complexes 2b-d and 3 are insoluble in water. In contrast to 2b-d, which are well soluble in dichloromethane and chloroform, 3 manifests a very low solubility even in organic solvents, apart from DMSO. The new compounds were fully characterized by elemental analysis, IR, and NMR spectroscopy (Figures S1-S15). Due to the presence of the chiral iridium center, the NMR spectra of 2b-d (in CDCl3) reveal four non-equivalent methyl groups within the {C5Me4} moiety; conversely, two resonances for such methyls are recognized in the NMR spectra of 3 (in DMSO-d6), this complex lacking asymmetry. The 19 F NMR resonance due to the fluorine atom in 2d occurs at −114.5 ppm. The hydroxyl unit in 3 manifests itself with an infrared band at 3262 cm −1 and a 1 H NMR singlet at 9.68 ppm.
The structures of 2b and 3 were ascertained by single-crystal X-ray diffraction studies. Views of the ORTEP molecular structures are shown in Figures 2 and 3 and details of these studies are provided in Table 1. Both complexes adopt a three-legged piano stool geometry, with bonding parameters comparable to those reported for analogous Ir III -complexes containing functionalized Cp ligands and, respectively, additional phenylpyridine [10] and DMSO [15][16][17][18][19] ligands. Scheme 1. Synthesis of di-and mono-iridium complexes bearing variable tetramethyl-cyclopentadienyl ligands. a From C5HMe4(4-C6H4OCMe2OMe).
The structures of 2b and 3 were ascertained by single-crystal X-ray diffraction studies. Views of the ORTEP molecular structures are shown in Figures 2 and 3 and details of these studies are provided in Table 1. Both complexes adopt a three-legged piano stool geometry, with bonding parameters comparable to those reported for analogous Ir III -complexes containing functionalized Cp ligands and, respectively, additional phenylpyridine [10] and DMSO [15][16][17][18][19] ligands.    The structures of 2b and 3 were ascertained by single-crystal X-ray diffraction studies. Views of the ORTEP molecular structures are shown in Figures 2 and 3 and details of these studies are provided in Table 1. Both complexes adopt a three-legged piano stool geometry, with bonding parameters comparable to those reported for analogous Ir III -complexes containing functionalized Cp ligands and, respectively, additional phenylpyridine [10] and DMSO [15][16][17][18][19] ligands.    Upon dissolution of 2a-d and 3 into dimethylsulfoxide/water solution (1:3 v/v), only one set of signals was NMR detected. The NMR patterns did not change after maintaining the solutions at 37 • C for three days. Using dimethylsulfone as an internal standard, it was established that the fraction of residual compound approached 100% in every cases. Analogous evaluation was not reliable for 1b-e, due to limited solubility of these complexes at 37 • C. The NMR evidence highlights the substantial stability of mono-iridium complexes in the selected aqueous environment, while Sadler previously recognized fast, partial chloride/water exchange from 2a in CD 3 OD/D 2 O (1:4 v/v) at room temperature [21]. Overall, these facts suggest that 2a-d are relatively inert in the presence of water, although sensitive to modifications of the composition of the aqueous medium, and a viable mechanism for their first activation in physiological environment may reasonably involve chloride dissociation. According to UV-Vis spectroscopy, the species derived from dissolution of 2a, 2c, and 2d in RPMI-1640 cell culture medium did not change after being stored at 37 • C for 24 h; conversely, 2b underwent fast decomposition under these conditions, and the same behavior was recognized for this complex in octanol/water mixture (Log P ow analysis, vide infra).

Cytotoxicity
The new mono-iridium complexes 2b-d and 3, as well as the corresponding precursors 1b-e, were assessed for their cytotoxic activity ( Table 2). The viability of cells was analyzed by two different tests, i.e., MTT (based on mitochondrial respiration) and CV (based on the quantification of adherent, vital cells), with an aim to overcome deficiencies of each method and thus increase the accuracy of the study. In general, IC 50 values obtained by MTT are significantly lower than those supplied by CV. However, microscopical evaluation of the cells exposed to the treatments showed that the complexes inhibited cell respiration prior affecting cell viability (data not shown), thus rendering CV data more representative of the real viability reduction.
The previously studied Cp* complex 2a and cisplatin were used as references. The diiridium complexes 1b and 1e, bearing respectively hydrogen and 4-hydroxyphenyl group as R substituents, resulted as substantially inactive, while 1c and 1d exhibited a moderate cytotoxicity which is comparable to that of cisplatin only in the case of 1d against SW620 cells ( Figure S16). The mono-iridium derivatives 2b-d generally showed significantly lower IC 50 values compared to the corresponding parent compounds 1b-d. Otherwise, the conversion of 1e into its dimethylsulfoxide mononuclear adduct is not beneficial to the activity: as a matter of fact, 3 is inactive towards all the investigated cell lines, including those ones on which 1e exhibits IC 50 values in the range 80-90 µM. With respect to the reference Cp* compound 2a, a considerable enhancement of the in vitro anticancer action was recognized with 2d, in alignment with previous findings highlighting the favorable effect of the introduction of an aryl-moiety on the cyclopentadienyl ring (Table 1, Figure S17). On the other hand, the replacement of one methyl (2a) with a longer alkyl chain (propyl, 2c) does not provide an appreciable effect, except for a slight decrease of activity against SW620 cells and MCF-7. The performance of the tetramethylcyclopentadienyl species 2b does not significantly differ from that of 2a in several cell lines, although it is slightly better in B16, SW620, and C6 cell lines, where it approaches 2d. It does not seem that differences in the antiproliferative activity of 2a-d are imputable to significantly different degrees of lipophilicity, according to measured octanol-water partition coefficients (Log P ow , see Experimental). As a matter of fact, the Log P ow values of 2a-d fall in the restricted range 1.3-1.6.
In general, the effect on the viability of human transformed lung fibroblasts, MRC5, is almost the same as on tumor cell lines. The high sensitivity of nonmalignant, primary MRC5 cells toward iridium complexes can be ascribed to the properties of this cell line, resembling the aggressive malignant phenotype, such as high proliferative rate and intracellular features analogous to stem phenotype [22]. Among all tested compounds, 2b and 2d were selected as the most promising ones for further studies. Interestingly, both compounds did not affect viability of primary mouse peritoneal cells, thus outlining some tendency to selectivity towards the tumor phenotype ( Figure 4).

Mechanism of Action
The analysis of the mechanism of action of 2b and 2d revealed inhibition of cell proliferation upon 72 h of incubation ( Figure 4A). Inhibited cell division was accompanied by a modest induction of apoptotic cell death for both drugs in caspase dependent manner ( Figure 4B,C). The number of cells exhibiting the higher levels of activated caspases was in accordance with the percentage of detected apoptotic cells (see Experimental for details), highlighting the more potent proapoptotic effect of 2d. Hyperpolarization of mitochondrial membrane preceded mitochondrial collapse and subsequent induction of cell death ( Figure 4D) [23]. More in detail, exposure to 2b and 2d led to mitochondrial membrane hyperpolarization in the first 24 h of treatment (32.3% and 48.6% in comparison with untreated cells, respectively), suggesting strong influence of iridium complexes on cell respiration and mitochondrial function. Dysfunction of mitochondrial membrane is one of the hallmarks of triggered intrinsic apoptotic pathway, starting as hyperpolarization and finalizing as a loss of membrane potential. Previously, some of us found that a class of ansa-titanocene complexes was able to provoke hyperpolarization prior the total mitochon-drial collapse [24]. In the present study, staining of cells with propidium iodide permitted to disclose cells with abnormal shape, compressed volume of nuclei, and condensed chromatin, which is typical of apoptotic cells [25], besides cells with giant nuclei, indicating that 2b and 2d possess the ability to induce senescence ( Figure 4E,F). Cell senescence was further confirmed by increased activity of beta galactosidase, a major senescence biomarker, in cultures exposed to 2b and 2d (increase by 38.4 and 31%, respectively, in comparison to the control), Figure 4F. However, a high percentage of autophagosomes was not recognized in treated cultures, indicating irrelevance of autophagic process to the action of iridium complexes ( Figures S18-S19). The fact that compounds inhibited cell division and induced cell senescence in parallel with cell death deserves more comments. In fact, while the induction of apoptosis has been reported for diverse iridium complexes, to the best of our knowledge we describe here the first case clearly evidencing the ability of iridium complexes to initiate senescence [26,27]. In recent years, it has been established that senescence is a special form of durable cell cycle arrest, and might represent a primary mechanism for tumor prevention and suppression. There are some examples in the literature demonstrating that the co-existence of senescence and apoptosis can be triggered by the same stressors [28]. To date, it is not clear which of these two alternatives determines the cell destiny, but it is believed that senescence is triggered in cells resistant to apoptosis [28][29][30]. In this regard, several examples are known of cytotoxic drugs able to trigger cell senescence, e.g., cisplatin, bleomycin, and ritonavir [28,29], and much effort is currently directed to develop the so called pro-senescent therapy, as part of differentiation-based therapies. The possibility of inducing cellular senescence and/or differentiation was previously observed for other metal-based drugs such as organotin compounds, either in free form or linked to nano-carriers [31]. The advantage of cell senescence over the killing-based strategies might consist in avoiding apoptosis-induced cell division, which limits the efficacy of chemotherapy [32].
For the evaluation of the production of reactive oxygen and nitrogen species (ROS/RNS), two different probes were used. More precisely, the DAF-FM probe detects the intracellular status of nitric oxide, while the DHR123 probe measures the total amount of hydrogen peroxide, hypochlorous acid, and peroxynitrite, accumulated inside the cells during the whole time of incubation. Using the mentioned redox sensitive dyes, a strong increase of intracellular levels of NO (40.9% for 2b and 38% for 2d) and of H 2 O 2 , HOCl and ONOO − (in total, 46.4% for 2b and 36.3% for 2d) was detected in A2780 cells upon treatment with the complexes, with respect to the untreated cultures ( Figure 5). Consistently, Sadler and co-workers observed stimulation of ROS production on analogous complexes. Briefly, they found that iridium(III) cyclopentadienyl complexes with iodide and azopyridine ligands induced ROS burst in human lung cancer cells [11,33]. In addition, related iridium complexes containing a pyridine ligand elicited a high level of ROS production in A2780 cancer cells.
In the present case, the observed affected mitochondrial respiration should be tightly connected to the oxidative stress promoted by the complexes. Having in mind that ROS/RNS species play an essential role in cellular physiology as well as in pathology, from control of cell signaling pathways to induction of death, a fine tuning between proliferation inhibition, apoptosis, and senescence may be of crucial relevance for the activity of a drug candidate.

Conclusions
Iridium-Cp* complexes constitute a well-established family of organometallic anticancer drug candidates, and previous findings outlined the significant impact of monosubstitution of one Cp*-methyl group on the cytotoxicity. Here, we reported a systematic study aimed at assessing the effect of different substituents. Two series of homologous diand mononuclear iridium complexes, respectively, were prepared with variable {C 5 Me 4 R} rings, characterized, and assessed for the antiproliferative activity towards a panel of cell lines. The results showed a higher activity for mono-iridium phenyl-pyridyl complexes compared to the respective dinuclear precursors and some effect associated with R. In general, differences in the antiproliferative activity might arise, at least in part, from a different stability of the complexes, leading to a different speciation in the cell environment. While the relatively weak binding of the chloro-bridges in di-iridium compounds is expected to accelerate their disruption, the properties of the R substituent may affect the inertness of the iridium coordination set. Notably, the replacement of one methyl with hydrogen has been proven sufficient to determine fast degradation of the related mononuclear complex in aqueous media; however, this phenomenon leads to enhanced cytotoxicity thus rendering the overall picture not easily rationalizable.
Mononuclear complexes with R = H (2b) and 4-C 6 H 4 F (2d) exhibited the best performance within the series, displaying IC 50 values in the low micromolar range, and were thus selected for targeted experiments aimed at elucidating the mode of action. While 2b and 2d were inactive towards nontumoral mouse peritoneal cells, they supplied a triple effect on tumor cells, consisting in proliferation inhibition, apoptotic cell death, and senescence induction. Note that induction of senescence has been rarely recognized on organoiridium complexes. The possibility of developing new drugs able to initiate a senescent program is highly desirable, in the light of a convenient mode of action affecting the growth of the malignant cells.

Materials and Methods
Iridium(III) chloride hydrate (99.9%) was purchased from Strem, while organic reactants were purchased from Alfa Aesar, Apollo Sci., and TCI Europe, and were of the highest purity available. Compounds 1a [34] and 1b-e [14] were synthesized according to the respective literature procedures. Unless otherwise specified, operations were conducted in air. Silica gel (Merck, 70-230 mesh) was dried at 150 • C overnight. NMR spectra were recorded at 298 K on a Bruker Avance II DRX400 instrument equipped with a BBFO broadband probe. Chemical shifts (expressed in parts per million) are referenced to the residual solvent peaks ( 1 H, 13 C) or to external standard ( 19 F, CFCl 3 ). 1 H and 13 C spectra were assigned with the assistance of DEPT-135 spectra and 1 H-13 C (gs-HSQC and gs-HMBC) correlation experiments [35]. IR spectra of solid samples (650-4000 cm −1 ) were recorded on a Perkin Elmer Spectrum One FT-IR spectrometer, equipped with a UATR sampling accessory. Elemental analyses were performed on a Vario MICRO cube instrument (Elementar).

Synthesis and Characterization of Compounds
Synthesis of (1,2,3,4,5-pentamethylcyclopentadienyl)(2-phenylpyridine)iridium(III)chloride, 2a (Scheme 2). thus selected for targeted experiments aimed at elucidating the mode of action. While 2b and 2d were inactive towards nontumoral mouse peritoneal cells, they supplied a triple effect on tumor cells, consisting in proliferation inhibition, apoptotic cell death, and senescence induction. Note that induction of senescence has been rarely recognized on organoiridium complexes. The possibility of developing new drugs able to initiate a senescent program is highly desirable, in the light of a convenient mode of action affecting the growth of the malignant cells.

Materials and Methods
Iridium(III) chloride hydrate (99.9%) was purchased from Strem, while organic reactants were purchased from Alfa Aesar, Apollo Sci., and TCI Europe, and were of the highest purity available. Compounds 1a [34] and 1b-e [14] were synthesized according to the respective literature procedures. Unless otherwise specified, operations were conducted in air. Silica gel (Merck, 70-230 mesh) was dried at 150 °C overnight. NMR spectra were recorded at 298 K on a Bruker Avance II DRX400 instrument equipped with a BBFO broadband probe. Chemical shifts (expressed in parts per million) are referenced to the residual solvent peaks ( 1 H, 13 C) or to external standard ( 19 F, CFCl3). 1 H and 13 C spectra were assigned with the assistance of DEPT-135 spectra and 1 H-13 C (gs-HSQC and gs-HMBC) correlation experiments [35]. IR spectra of solid samples (650-4000 cm −1 ) were recorded on a Perkin Elmer Spectrum One FT-IR spectrometer, equipped with a UATR sampling accessory. Elemental analyses were performed on a Vario MICRO cube instrument (Elementar). The title compound was prepared by means of a slightly modified literature procedure [36], which was then applied to the syntheses of 2b-d. A mixture of [Ir(η 5 -Cp * )Cl(μ-Cl)]2 (0.100 g, 0.126 mmol), 2-phenylpyridine (37 μL, 0.25 mmol), and sodium acetate (0.041 g, 0.50 mmol) in CH2Cl2 (15 mL) was refluxed with stirring for 24 h under N2 atmosphere. The resulting solution was filtered through celite and the filtrate was evaporated to dryness on a rotary evaporator and then washed with pentane. The obtained suspension was filtered, and the orange solid was dried under vacuum affording 2a. The title compound was prepared by means of a slightly modified literature procedure [36], which was then applied to the syntheses of 2b-d. A mixture of [Ir(η 5 -Cp * )Cl(µ-Cl)] 2 (0.100 g, 0.126 mmol), 2-phenylpyridine (37 µL, 0.25 mmol), and sodium acetate (0.041 g, 0.50 mmol) in CH 2 Cl 2 (15 mL) was refluxed with stirring for 24 h under N 2 atmosphere. The resulting solution was filtered through celite and the filtrate was evaporated to dryness on a rotary evaporator and then washed with pentane. The obtained suspension was filtered, and the orange solid was dried under vacuum affording 2a. Yield: 0.093 g, 72%. Analyses calculated for C 21        (3) detector using Mo-Kα radiation. Data were corrected for Lorentz polarization and absorption effects (empirical absorption correction SADABS) [37]. The structures were solved by direct methods and refined by full-matrix least-squares based on all data using F 2 [38]. Hydrogen atoms were fixed at calculated positions and refined by a riding model. All non-hydrogen atoms were refined with anisotropic displacement parameters.

Behavior in Aqueous Solutions
(a) Octanol/water partition coefficients (Log P ow ). Partition coefficients (P ow ; IU-PAC: K D partition constant [39]), defined as P ow = c org /c aq , where c org and c aq are molar concentrations of the selected compound in the organic and aqueous phase, respectively, were determined by the shake-flask method and UV-Vis measurements [40]. Deionized water and 1-octanol were vigorously stirred for 24 h, to enable saturation of both phases, then separated by centrifugation. A stock solution of the selected iridium compound (ca. 2 mg) was prepared by first adding acetone (50 µL, to help solubilization), followed by water-saturated octanol (2.5 mL). The solution was diluted with water-saturated octanol (ca. 1:3 v/v ratio, c Ir ≈ 10 −4 M, so that 1.5 ≤ A ≤ 2.0 at λ max ) and its UV-Vis spectrum was recorded (A 0 org ). An aliquot of the solution (V org = 1.2 mL) was transferred into a test tube and octanol-saturated water (V org = V aq = 1.2 mL) was added. The mixture was vigorously stirred for 30 min at 21 • C then centrifuged (5000 rpm, 5 min). The UV-Vis spectrum of the organic phase was recorded (A f org ) and the partition coefficient was calculated as P ow = A f org /(A 0 org − A f org ) where A 0 org and A f org are the absorbance in the organic phase before and after partition with the aqueous phase, respectively. The wavelength of the maximum absorption of each compound (ca. 300 nm) was used for UV-Vis quantitation. The procedure was repeated three times for each sample (from the same stock solution); results are given as mean ± standard deviation. Log P ow values were as follows: 2a: 1.47 ± 0.07; 2b: n/a (extensive degradation of the complex occurred during analysis); 2c: 1.5 ± 0.2; 2d: 1.3 ± 0.2. Literature value for 2a: 1.57 ± 0.08 [41]. , then dimethyl sulfone (0.05 mL, 0.164 mmol) was added as internal standard. The mixtures were filtered through a celite pad, and the filtrated solutions were analyzed by 1 H NMR spectroscopy. NMR data from 2a-d and 3 are reported in the following (Figures S10-S15). Line broadening observed for 2a and 3 is probably consequence of the low solubility of these complexes in the employed medium.
2a. 1  The solutions were then stored for 72 h at 37 • C, and new NMR analyses were subsequently performed. Integral ratios referred to the standard, measured respectively after and before heating, allowed us to estimate the fraction of unaltered compound: 2a, 98%; 2b, 98%; 2c, 98%; 2d, 97%; 3, 98%. In the cases of 1b, 1c, and 1d, progressive precipitation of a solid was observed from the solutions over 72 h, presumably due to limited solubility of the compounds; however, no significant variations were detected in the 1 H NMR spectra. Complex 1e could not be NMR analyzed due to insufficient solubility in aqueous media, except in the presence of DMSO where, however, 1e undergoes rapid conversion into 3.
(c) Stability in cell culture medium Complexes 2a-d (ca. 2 mg) were dissolved in the minimum volume of methanol, then the solutions were diluted with RPMI-1640 cell culture medium up to ca. 10 −6 M [Ir] concentration. UV-Vis spectra were recorded immediately after the preparation of these solutions, showing the maximum absorption at ca. 300 nm except in the case of 2b. The solutions were stored at 37 • C for 24 h, then new spectra were recorded without displaying significant changes. at concentration of 20 mM and kept at −20 • C, while 10 mM stock solution of cisplatin was prepared in DMF just before the usage. Desired final concentrations were obtained by dilution in culture medium.

Colorimetric Assays for Cellular Viability
Cells were treated with various concentrations (up to 50 or 100 µM) of the abovementioned complexes for 72 h. For the detection of mitochondrial respiration, cells were cultivated in the presence of MTT staining solution (0.5 mg/mL) for approximately 1 h at 37 • C. The dye was then discarded, and the formed formazan crystals (purple) were dissolved in DMSO. For an evaluation of viable attached cells, cells were fixed with 4% PFA for 10 min at room temperature, and subsequently stained for 15 min with 0.02% CV solution. Cells were then washed with tap water, dried in air, and the CV dye was dissolved in 33% acetic acid. The absorbance was measured with an automated microplate reader at 540 nm, with a reference wavelength of 670 nm. The IC 50 values were calculated using a four-parameter logistic function and presented as mean ± SD from three independent experiments.

Flow Cytometry
For better insight into the mechanisms of action of complexes 2b and 2d on A2780 cell line, cells were incubated with 2b and 2d (IC 50 doses) and analyzed by flow cytometry. Several staining protocols were performed: (a) AnnV/propidium iodide (PI) for the detection of apoptotic cell death; (b) ApoStat for detection of caspase activity; (c) AO for the detection of autophagosomes; (d) CFSE for monitoring the influence on cellular proliferation; (e) DAF-FM diacetate for detection of intracellular nitric oxide (NO); (f) DHR for the detection of reactive oxygen/nitrogen species (ROS/RNS); (g) FDG assay for detecting senescence; and h) DiOC6(3) to measure mitochondrial membrane potential. Results were obtained with CyFlow ® Space Partec using the PartecFloMax ® software. Experiments were carried out in three independent replicates. For AnnV/PI, AO and ApoStat staining, cells were treated with 2b and 2d during 48 h, and then detached and washed with PBS. Afterwards, cells were stained, in accordance with the manufacturer's protocols, with AnnV/PI (15 min, room temperature) in AnnV-binding buffer, or ApoStat (30 min, 37 • C) in PBS 5% FBS, or AO (100 µM; 15 min, 37 • C), in PBS. Finally, cells were washed, resuspended in PBS (or in AnnV-binding buffer for AnnV/PI), and analyzed. For CFSE staining, cells were pre-stained with a PBS solution of CFSE (1 µM; 10 min, 37 • C), washed, seeded, and then exposed to 2b and 2d for 72 h. Cells were then washed, trypsinized, dissolved in PBS, and analyzed. Similarly to CFSE, DHR staining was performed with cells being pre-stained first, with 1 µM DHR for 20 min at 37 • C, and then exposed to the experimental compounds for 48 h. At the end of cultivation, cells were washed, detached, and analyzed. For DAF-FM staining, cells were exposed the experimental compounds for 48 h, washed with PBS, trypsinized, and stained with 5 µM of DAF-FM diacetate for 1 h at 37 • C in phenol red-free RPMI 1640. Thereafter, cells were washed and additionally incubated for 15 min in fresh RPMI 1640 without phenol red and serum, to finish the reaction of de-esterification. For the FDG assay, cells were treated with 2b and 2d during 48 h, and then stained with β-galactosidase substrate FDG to a final concentration of 1 mM. After 1 min incubation at 37 • C, cells were analyzed. DiOC6(3) staining protocol considered the exposure of treated cells (2b and 2d during 24 h) to 70 nM of DiOC6(3) in PBS during 20 min at 37 • C. Channels FL1 (green emission), FL2 (orange emission), and/or FL3 (dark red emission) were used for fluorescence detection, according to the specific staining agent.

Fluorescence Microscopy
Cells were cultivated on chamber slides overnight (4 × 10 4 /well), then treated with 2b and 2d during 48 h. Cells were washed with PBS, fixed with 4% PFA during 15 min at room temperature, and stained with a solution of propidium iodide (50 µg/mL) with 0.1% Triton X-100, 0.1 mM EDTA pH 8.0, and RNase (85 µg/mL) in PBS for 2 min. Finally, cells were washed with PBS and prepared for fluorescence microscopy by layering with fluorescent mounting medium (Dako, Glostrup, Denmark). The slides were analyzed with a Zeiss AxioObserver Z1 inverted fluorescence microscope (Carl Zeiss AG, Oberkochen, Germany) at 200× magnification.

Statistical Analysis
Analysis of variance (ANOVA) followed with a Student-Newman-Keuls test was used for significance of the differences between treatments, and a p-value less than 0.05 was taken as statistically significant.

Data Availability Statement:
The data presented in this study are available on request from the corresponding authors.