Synthesis and Investigations of the Antitumor Effects of First-Row Transition Metal(II) Complexes Supported by Two Fluorinated and Non-Fluorinated β-Diketonates

The 3d transition metal (Mn(II), Fe(II), Co(II), Ni(II), Cu(II) and Zn(II)) complexes, supported by anions of sterically demanding β-diketones, 1,3-dimesitylpropane-1,3-dione (HLMes) and 1,3-bis(3,5-bis(trifluoromethyl)phenyl)-3-hydroxyprop-2-en-1-one (HLCF3), were synthesized and evaluated for their antitumor activity. To assess the biological effects of substituents on phenyl moieties, we also synthesized and investigated the analogous metal(II) complexes of the anion of the less bulky 1,3-diphenylpropane-1,3-dione (HLPh) ligand. The compounds [Cu(LCF3)2], [Cu(LMes)2] and ([Zn(LMes)2]) were characterized by X-ray crystallography. The [Cu(LCF3)2] crystallizes with an apical molecule of solvent (THF) and features a rare square pyramidal geometry at the Cu(II) center. The copper(II) and zinc(II) complexes of diketonate ligands, derived from the deprotonated 1,3-dimesitylpropane-1,3-dione (HLMes), adopt a square planar or a tetrahedral geometry at the metal, respectively. We evaluated the antitumor properties of the newly synthesized (Mn(II), Fe(II), Co(II), Ni(II), Cu(II) and Zn(II)) complexes against a series of human tumor cell lines derived from different solid tumors. Except for iron derivatives, cellular studies revealed noteworthy antitumor properties, even towards cancer cells endowed with poor sensitivity to the reference drug cisplatin.

The complexes [Mn(L CF3 ) 2 (H 2 O) 2 ] (1), [Fe(L CF3 ) 2 ] (2), [Co(L CF3 ) 2 (H 2 O) 2 ] (3) and [Ni(L CF3 ) 2 (H 2 O) 2 ] (4) (Scheme 1) were synthesized by dissolving the ligand HL CF3 in an ethanol solution containing the metal acetate acceptor in a 2:1 stoichiometric ratio.A similar procedure was used for the synthesis in good yield of [Cu(L CF3 ) 2 ] (5) and [Zn(L CF3 ) 2 ] (6) by dissolving the corresponding ligand HL CF3 in a water/ethanol solution containing the copper(II) acetate monohydrate and the zinc(II) acetate, respectively.The IR spectra were carried out on solid samples of 1-6.They show all the expected absorption bands; in particular, absorptions in the range 2898-3104 cm −1 are due to the C-H bonds, medium absorptions at 1564-1629 cm −1 are attributable to the asymmetric stretching of the C=O groups and absorptions in the range 1515-1555 cm −1 are assigned to the ν(C=C) stretching modes.Bands due to C-F stretching and CF 3 deformation are at 1169-1173 and 1125-1129 cm −1 , respectively.The ligand coordination sites involved in bonding with the metal ions were determined by careful comparison of the infrared absorption bands of the complexes with those of the parent ligand HL CF3 , suggesting the coordination of the ligand as an enolate.The 3226-3374 cm −1 regions of the spectra of compounds 1, 3 and 4 show broad bands, which may be due to lattice and/or coordinated water molecules associated with the complexes.The proton nuclear magnetic resonance ( 1 H-NMR) spectrum of [Zn(L CF3 ) 2 ] (6), recorded in acetone-d 6 solution at room temperature, show a single set of resonances for the β-diketone moiety, indicating that the protons of the aromatic rings are equivalents, with a shift due to the ligand coordination to the metal center.The 19 F NMR spectrum of 6 in acetone-d 6 displayed a singlet at δ -63.35.The electrospray ionization mass spectra (ESI-MS) study was performed by dissolving complexes 1-6 in CH 3 CN or CH 3 OH/CH 3 CN and recording the spectrum in positive-and negative-ion mode.The structure of the complexes was confirmed by the presence of peaks attributable to positive fragments of the dissociation of the ligand from the complexes ([Mn(L CF3 ) + CH 3 or negative adducts of the complexes such as [Mn(L CF3 ) 2 + Cl] − , [Co(L CF3 ) 2 + Cl] − and [Ni(L CF3 ) 2 + Cl] − .In addition, in the negative-ion mode spectra, we observed peaks at 495 due to the [L CF3 ] − fragment.Compound [Cu(L CF3 ) 2 ] (5) was characterized by X-ray crystallography.It crystallizes as a THF adduct from acetone/chloroform/tetrahydrofuran mixture.The molecular structure is illustrated in Figure 1 and Figure S3.Detailed bond distances and angles are provided in the Supporting Information (Tables S7-S9).[Cu(L CF3 ) 2 (THF)] is a square pyramidal complex with an apical THF ligand.Such square pyramidal diketonate molecules are very rare in copper(II), and the copper(II) complex [Cu(L C6F5 ) 2 (OH 2 )] derived from the deprotonated 1,3-bis(pentafluorophenyl)propane-1,3-dione (HL C6F5 ) is the only other structurally characterized example in the literature, to our knowledge [89] 12) was synthesized in a water/ethanol solution, using zinc(II) chloride and NaL Mes precursors.The IR spectra were carried out on solid samples of 7-12.Absorptions in the range 2852-3081 cm −1 are due to the C-H bonds, medium absorptions in the range 1569-1613 cm −1 are attributable to the asymmetric stretching of the C=O groups and absorptions at 1500-1552 cm −1 are due to the ν(C=C) stretching modes; all these shifts confirm the coordination of the ligands as enolates.The carbonyl stretching frequencies are comparable with those reported in the literature for analogous Cu(II) complexes supported by bulky β-diketones [90,91].The 3318-3405 cm −1 regions of the spectra of compounds 7, 9 and 10 show broad bands, which may be due to lattice and/or coordinated water molecules associated with the complexes.O (18) [94] (Scheme 1), were synthesized, modifying the procedure reported in the literature by dissolving the ligand HL Ph in an ethanol solution containing the corresponding metal acetate acceptor.The IR spectra were carried out on solid samples of 13-18.They show all the expected absorption bands: absorptions in the range 2970-3061 cm −1 due to the C-H bonds, medium absorptions at 1590-1620 cm −1 due to the asymmetric stretching of the C=O groups and absorptions in the range 1519-1548 cm −1 , attributable to the ν(C=C) stretching modes; all these shifts suggest the coordination of the ligands as enolates.The carbonyl stretching frequencies fall in a similar range to compounds 1-12, suggesting little sensitivity to steric or electronic properties.The 3298-3349 cm −1 regions of the spectra of compounds 13, 15, 16 and 18 show broad bands, which may be due to lattice and/or coordinated water molecules associated with the complexes.The

Biological Studies
The stability of complexes 1-18 in 0.5% dimethyl sulfoxide (DMSO)/physiological solution was evaluated by UV-Vis spectroscopy (Figure S57).Spectra were collected every 24 h in the range of 240-640 nm over 72 h.Exemplificative spectra collected at t = 0 and t = 72 h are reported in the Supplementary Material, showing that all compounds were sufficiently stable under physiological conditions.The newly synthesized metal complexes were evaluated for their cytotoxic activity towards various human cancer cell lines representative of different solid tumors.In particular, the in-house cancer cell panel included examples of human colon (HCT-15), pancreatic (BxPC3), testis (NTERA-2), breast (MCF-7) and small cell lung cancer (SCLC) (U-1285) cancer.The cytotoxicity parameters, expressed in terms of 50% inhibitory concentration (IC 50 ) obtained after 72 h of exposure to the colorimetric tetrazolium dye (MTT) assay, are reported in Table 1.Due to its simple structure and well-known pharmacological and toxicological profiles, cisplatin is largely applied as a reference drug for the preliminary in vitro testing of new metallodrug candidates.For comparison reasons, the cytotoxicity of the reference metal-based chemotherapeutic drug cisplatin was assessed under the same experimental conditions.Cytotoxicity data for the corresponding uncoordinated ligands and their salts have been previously reported [88].The cytotoxicity results showed that, except the iron derivatives 2, 8 and 14, all tested complexes demonstrated significant antiproliferative activity towards all cancer cell lines belonging to the in-house cancer cell panel, showing IC 50 values in the micromolar range.Even if it seems rather difficult to define straightforward structure-activity relationships, as a general consideration, all the metal complexes bearing the 1,3-dimesitylpropane-1,3-dione (HL Mes ) were on average more effective than the corresponding derivatives, including the 3-bis(3,5-bis(trifluoromethyl)phenyl)-3-hydroxyprop-2-en-1-one (HL CF3 ) and 1,3-diphenylpropane-1,3-dione (HL Ph ) ligands.In particular, complexes [Mn(L Mes ) 2 (H 2 O) 2 ] (7) and [Cu(L Mes ) 2 ] (11) were the most effective derivatives of the series, being much more effective than cisplatin towards all tested cancer cell lines.It is noteworthy that, against human colon carcinoma HCT-15 cells, [Mn(L Mes ) 2 (H 2 O) 2 ] (7) was up to 15-fold more efficacious than cisplatin in decreasing cell proliferation.Similarly, [Mn(L CF3 ) 2 (H 2 O) 2 ] (1) proved to be much more effective than the reference metallodrug against colon, testis, pancreatic and breast carcinoma cells whereas compound [Co(L Mes ) 2 (H 2 O) 2 ] (9) proved to be much more effective than cisplatin against colon, testis and pancreatic human cancer cells whereas it was slightly less effective on breast carcinoma and SCLC cells.

Materials and General Methods
All the reagents were obtained from commercial sources and used as received.Melting point (MP) analysis was performed by an SMP3 Stuart Scientific Instrument (Bibby Sterilin Ltd., London, UK).Elemental analyses (C, H, N, S) (EA) were performed with a Fisons Instruments EA-1108 CHNS-O Elemental Analyzer (Thermo Fisher Scientific Inc., Waltham, MA, USA).Fourier-transform infrared (FT-IR) spectra were recorded from 4000 to 700 cm −1 on a PerkinElmer Frontier Instrument (PerkinElmer Inc., Waltham, MA, USA), equipped with an attenuated total reflection (ATR) unit using a universal diamond top-plate as a sample holder.Abbreviations used in the analyses of the FT-IR spectra are as follows: br = broad, m = medium, mbr = medium broad, s = strong, sbr = strong broad, vs = very strong, vsbr = very strong broad, sh = shoulder, w = weak, vw = very weak and wbr = weak broad.Nuclear magnetic resonance (NMR) spectra for the nuclei 1 H and 13 C were recorded with a Bruker 500 Ascend Spectrometer (Bruker BioSpin Corporation, Billerica, MA, USA; 500.13MHz for 1 H, 125.78 MHz for 13 C and 470.59 MHz for 19 F).Tetramethylsilane (SiMe 4 ) was used as an external standard for the 1 H-and 13 C-NMR spectra.The chemical shifts (δ) are reported in ppm, and the coupling constants (J) are reported in hertz (Hz).Abbreviations used in the analyses of the NMR spectra are as follows: br = broad, d = doublet, m = multiplet, s = singlet, sbr = singlet broad, t = triplet and q = quartet.Electrospray ionization mass spectrometry (ESI-MS) spectra were recorded in positive-(ESI-MS (+)) or negative-ions (ESI-MS (−)) mode on a Waters Micromass ZQ Spectrometer, equipped with a single quadrupole (Waters Corporation, Milford, MA, USA), using a methanol or acetonitrile mobile phase.The compounds were added to reagent grade methanol or acetonitrile to give approximately 0.1 mM solutions.These solutions were injected (1 µL) into the spectrometer fitted with an autosampler.The pump delivered the solutions to the mass spectrometer source at a flow rate of 200 µL/min, and nitrogen was employed both as a drying and nebulizing gas.Capillary voltage was typically 2500 V.The temperature of the source was 100 • C, while the temperature of the desolvation was 400 • C. In the analyses of the ESI-MS spectra, the confirmation of major peaks was supported by the comparison of the observed and predicted isotope distribution patterns, with the latter calculated using the IsoPro 3.1 computer software (T-Tech Inc., Norcross, GA, USA).
The ligand HL Ph was obtained from commercial sources and used as received.The ligands HL CF3 [86], HL Mes [87] and the related sodium salt NaL Mes [88] were prepared according to procedures detailed in the literature and were fully characterized.For comparison with the related metal complexes, we reported in the Supplementary Material the spectroscopic characterizations of the sodium salts NaL CF3 , NaL Mes and NaL Ph (FT-IR (Figures S4, S9 and S13), 1 H-NMR (Figures S5, S10 and S14), 13 C{ 1 H}-NMR (Figures S6, S11 and S15), 19 F{ 1 H}-NMR (Figure S7) and ESI-MS (+) (Figures S8, S12 and S16) spectra).

X-ray Data Collection and Structure Determination
A suitable crystal covered with a layer of hydrocarbon/Paratone-N oil was selected and mounted on a Cryo-loop, then immediately placed in the low temperature nitrogen stream.The X-ray intensity data of [Cu(L Mes ) 2 ] (11) and [Zn(L Mes ) 2 ] (12) were measured at 100 K while the data of [Cu(L CF3 ) 2 (THF)] were measured at 200 K (due to crystal cracking issues at 100 K) on a Bruker SMART APEX II CCD area detector system, equipped with an Oxford Cryosystems 700 series cooler, a graphite monochromator, and a Mo Kα finefocus sealed tube (λ = 0.71073 Å).Intensity data were processed using the Bruker APEX3, Version 2016.5-0 program suite.Absorption corrections were applied using SADABS 2016/2 [100].Initial atomic positions were located by SHELXT Version 2018/2 [101], and the structures of the compounds were refined by the least-squares method using SHELXL Version 2019/3 [102] within Olex2-1.5 GUI [103].All the non-hydrogen atoms were refined anisotropically.Hydrogen atoms were included at calculated positions and refined using appropriate riding models.The copper center of [Cu(L Mes ) 2 ] molecule sits on a center of inversion.[Cu(L CF3 ) 2 (THF)] crystallizes in the P-1 space group, with two additional molecules of THF in the asymmetric unit.All three THF molecules and fluorine atoms of six CF 3 groups show positional disorder, which was resolved satisfactorily.The X-ray structural figures were generated using Olex2.CCDC 2279303-2279305 files contain the supplementary crystallographic data.These data files have been deposited at Cambridge Crystallographic Data Centre (CCDC), 12 Union Road, Cambridge, CB2 1EZ, UK).Additional details are provided in the Supporting Information section.

Biology
Metal complexes were dissolved in DMSO just before the experiment, and a calculated volume of the drug solution was added to the cell growth medium to a final solvent concentration of 0.5%, which had no detectable effects on cell viability.Cisplatin was dissolved in 0.9% sodium chloride solution.MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) and cisplatin were obtained from Sigma Chemical Co, St. Louis, MO, USA.

MTT Assay
The growth inhibitory effect toward tumor cells was evaluated by means of the MTT assay, as previously described [104].Briefly, 3-8 × 10 3 cells/well, dependent upon the growth characteristics of the cell line, were seeded in 96-well microplates in a growth medium (100 µL).After 24 h, the medium was removed and replaced with a fresh one containing the compound to be studied at the appropriate concentration (0.5-50 µM concentration range).Triplicate cultures were established for each treatment.IC 50 values, the drug concentrations that reduce the mean absorbance at 570 nm to 50% of those in the untreated control wells, were calculated using the four-parameter logistic (4-PL) model.The evaluation was based on means from at least three independent experiments.
In general, apart from Fe(II) complexes, the newly synthetized metal(II) complexes showed a marked cytotoxic activity, even against human colon cancer cells with poor sensitivity to cisplatin.The preliminary screening allowed the identification of the complexes [Mn(L Mes ) 2 (H 2 O) 2 ] (7) and [Cu(L Mes ) 2 ] (11) as the most effective derivatives of the series, being much more effective than cisplatin towards all tested cancer cell lines.In particular, against HCT-15 cancer cells, [Mn(L Mes ) 2 (H 2 O) 2 ] (7) was up to 15-fold more efficacious than cisplatin in decreasing cell proliferation, thus providing a foundation for also developing β-diketonate Mn(II) complexes, other than Cu(II) ones, as anticancer agents.

Table 1 .
Cont. × 10 3 × well) were treated for 72 h with increasing concentrations (0.5-50 µM range) of tested compounds.Cytotoxicity was assessed by MTT test.The IC 50 values were calculated by the four-parameter logistic model (p < 0.05).