- Email: [email protected]
Synthesis and structural characterization of copper(I) complexes bearing N-methyl-1,3,5-triaza-7-phosphaadamantane (mPTA)
Journal of Inorganic Biochemistry 103 (2009) 1644–1651
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
Synthesis and structural characterization of copper(I) complexes bearing N-methyl-1,3,5-triaza-7-phosphaadamantane (mPTA) Cytotoxic activity evaluation of a series of water soluble Cu(I) derivatives containing PTA, PTAH and mPTA ligands Marina Porchia a,*, Franco Benetollo a, Fiorenzo Refosco a, Francesco Tisato a, Cristina Marzano b, Valentina Gandin b a b
ICIS-CNR, Corso Stati Uniti 4, 35127 Padova, Italy Dipartimento di Scienze Farmaceutiche, Università di Padova, via Marzolo 5, 35131 Padova, Italy
a r t i c l e
i n f o
Article history: Received 17 April 2009 Received in revised form 8 July 2009 Accepted 8 September 2009 Available online 13 September 2009 Keywords: Water-soluble copper(I) complexes PTA Cytotoxic activity
a b s t r a c t New copper(I) complexes containing the water soluble N-methyl-1,3,5-triaza-7-phosphaadamantane (mPTA) phosphine have been synthesized by ligand-exchange reactions starting from [Cu(CH3CN)4][BF4] or [Cu(CH3CN)4][PF6] precursors and (mPTA)X (X = CF3SO3, I). Depending on the ligand counter ion, the hydrophilic [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a and [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c complexes or the iodine-coordinated [Cu(mPTA)3I]I3 4 species were obtained respectively and fully characterized by spectroscopic methods. Single crystal structural characterization was undertaken for [Cu(mPTA)3I]I3H2O, 4H2O, and [Cu(mPTA)4][(CF3SO3)2(BF4)3] 0.25H2O, 3b0.25H2O, the latter obtained by crystallization of [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a. The cytotoxicity of analogous tetrahedral homoleptic Cu(I) derivatives [Cu(PTA)4](BF4) 1, [Cu(PTAH)4][Cl4(BF4)] 2, [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a and [Cu(mPTA)4][ (CF3SO3)4(PF6)] 3c was evaluated against a panel of several human tumor cell lines. All the complexes showed in vitro antitumor activity comparable to that of the reference metallodrug cisplatin. Tests performed on cisplatin sensitive and resistant cell lines showed that against human ovarian 2008/C13 cell line pair, the resistance factor of copper derivatives was roughly 7-fold lower than that of cisplatin, whereas against human cervix cancer A431/A431-Pt cell line pair it was about 2.5-fold lower. These results, confirming the circumvention of cisplatin resistance, support the hypothesis that phosphine copper(I) complexes follow different cytotoxic mechanisms than do platinum drugs. Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction Since the discovery of the antitumor activity of cisplatin (cis-[PtCl2(NH3)2]) for the treatment of several human tumors [1], thousands of platinum and other metal-based compounds have been tested for their potential antitumor properties in the last 40 years. A series of compounds showing encouraging perspectives were the phosphine complexes of group 11 metal ions [2]. Their biological properties were little explored until late 1970s when a thioglucose derivative of triethylphosphine gold(I) (auranofin) was found to possess antiarthritic activity [3] and subsequently shown to have in vivo antitumor activity in murine models, although only against P388 leukemia [4]. In an attempt to identify
* Corresponding author. Fax: +39 049 8295951. E-mail address: [email protected] (M. Porchia). 0162-0134/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2009.09.005
gold complexes with a wider spectrum of activity, Sadler and coworkers demonstrated the cytotoxicity of mono-cationic bisdiphosphine gold(I) compounds toward a panel of human tumor cell lines including B16 melanoma, P388 leukemia, and M5076 reticulum cell carcinoma [5]. In particular, it was found that [Au(dppe)2][Cl] (dppe = 1,2-bis(diphenylphosphino)ethane) was 10-fold more cytotoxic than dppe alone, suggesting, at that time, that metal ions could potentiate the cytotoxic properties of dppe [5]. In general, the higher drug tolerance profiles of gold-based agents compared to the severe toxic effects on normal tissues and/or the occurrence of inherited or acquired resistance induced by cisplatin have represented a hopeful prospect for the chemotherapeutic application of gold-based drugs. Despite this promising outlook, clinical trials in humans of gold compounds remain still elusive. A rational extension of diphosphine gold(I) chemistry to the first row congener copper indicated that analogous copper(I) complexes of ‘‘CuP4” stoichiometry could be efficiently prepared and
1645
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
tested in vitro [6,7]. Among them, [Cu(dppe)2][Cl] and [Cu(dppey)2][Cl] (dppey = 1,2-bis(diphenylphosphino)ethylene) produced cytotoxic effects on cell metabolism comparable to those exhibited by Au(I) analogs, which were finally postulated to be the result of the uncoupling of mitochondrial oxidative phosphorylation [8]. However, the presence of several phenyl groups appended to the phosphorus donors in both [Cu(dppey)2][Cl] and [Au(dppe)2][Cl] species caused undesired nephrotoxicity [2] and cardiovascular toxicity [9] in animal models, respectively, thus precluding clinical trials in humans. Aiming at the formation of less-toxic copper compounds, we have recently prepared a series of hydrophilic copper(I) derivatives including the water soluble tris(hydroxymethyl)phosphine (thp) ligand, either alone [10,11] or in combination with hydrophilic scorpionates [12]. These water soluble compounds were proved to be easier to handle during the in vitro tests and, more importantly, have shown cytotoxic activity against a large panel of human tumor cell lines belonging to a variety of tumor types, including cisplatin and multidrug resistant phenotypes [11–13]. Moreover, it has been found that the cytotoxic activity of these copper(I) complexes may be correlated to their ability to induce a non-apoptotic mechanism of cell death [11,12]. Such results, together with recent studies which reported that the antiproliferative action of auranofin is connected with the inhibition of the thioredoxin reductase leading to augmented apoptosis [14], corroborate the view that mechanisms of action different from the DNA damage induced by cisplatin could underlie the cytotoxic activity of phosphine Au(I) and Cu(I) drugs. In other words, other molecular targets, in addition to DNA, could be identified in order to inhibit cancer cell proliferation, thus offering further motivation for the design of novel metal-based anticancer drugs. The water-soluble 1,3,5-triaza-7-phosphaadamantane (PTA) phosphine ligand and its derivatives are receiving growing attention in recent years [15–17]. The good water solubility of related transition metal PTA complexes makes possible their efficient application in aqueous phase catalysis [18–20]. The combination of PTA physico-chemical properties including water solubility, low steric hindrance (comparable to that of PMe3) and resistance to oxidation in water, joint to the ability to reduce various metal ions, include this phosphine among those few ligands suitable for stabilization of Cu(I) in aqueous media. While a large variety of lipophilic Cu(I) phosphine complexes stable in classic organic media has already been reported [21] very few examples of water-soluble Cu(I) compounds stable to disproportionation in water are known. These comprise the polymeric {Na5[Cu(TPPTS)2]5H2O}n complex [22] (TPPTS = tris(m-sulfonatophenyl)phosphine), [Cu(thp)4][PF6] (thp = tris(hydroxymethyl)phosphine), [Cu(bhpe)2][PF6] (bhpe = bis[bis(hydroxymethyl)phoshino]ethane) [11], the nitrate and chloride salts [Cu(PTA)4][X] (X = NO3, Cl) and [Cu(HPTA)4][NO3]5 [23,24]. Other recent water soluble Cu(I) derivatives present a mixed coordination sphere with a monodentate tertiary phosphine and a tripodal scorpionate ligand [12,25–27]. Aim of our study was to prepare a series of water-soluble copper(I) compounds whose steric and electronic properties could be tuned by slight modifications of the ligands, in order to compare and to correlate their physico-chemical properties with a possible biological activity. The PTA phosphine and its N-protonated (PTAH) and N-methylated (mPTA) derivatives (Chart 1) proved to be an ideal series of ligands for this purpose, as they give rise to different charged complexes conferring indeed a good hydrosolubility. In this study we report on the synthesis and characterization of new mPTA copper complexes, as well as the cytotoxic activity evaluation of stable water-soluble Cu(I) compounds containing PTA derivatives towards a panel of human tumor cell lines.
P
P N
N N PTA
P N
N N PTAH
+
H
+ N CH3
N N mPTA
Chart 1. 1,3,5-Triaza-7-phosphaadamantane (PTA) ligand and its N-protonated (PTAH) and N-methylated (mPTA) derivatives.
2. Experimental 2.1. General procedures All solvents and commercially available substances were of reagent grade and used without further purification. The Cu(I) starting compound [Cu(CH3CN)4][BF4] was prepared by reaction of [Cu(H2O)6][BF4]2 with metallic copper in acetonitrile whereas [Cu(CH3CN)4][PF6] was prepared from Cu2O and HPF6 according to Ref. [28]. The ligands PTA, (mPTA)I and (mPTA)(CF3SO3) were synthesized in accordance with published methods [16,17]. Elemental analyses were performed on a Carlo Erba 1106 Elemental Analyzer .1H, 31P, 13C and 63Cu NMR spectra were recorded on a Bruker AMX-300 instrument. The splitting of nuclear resonances in the reported NMR spectra is defined as s = singlet, d = doublet, q = quartet, m = multiplet, and bs = broad singlet. FT IR spectra were recorded on a Mattson 3030 Fourier transform spectrometer in the range 4000–400 cm1 in KBr pellets. The intensity of reported IR signals is defined as w = weak, m = medium, s = strong, and vs = very strong. Mass spectra have been recorded by an electrospray LCQ ThermoFinnigan mass spectrometer. 2.2. Syntheses of the complexes 2.2.1. [Cu(PTA)4][BF4] (1) To an acetonitrile solution of [Cu(CH3CN)4][BF4] (136 mg, 0.43 mmol) an excess of PTA (320 mg, 2.03 mmol) was added at room temperature. The reaction mixture was stirred overnight and then filtered. The white residue was washed three times with chloroform and diethyl ether and then dried under vacuum. Yield: 90% 1H NMR (D2O, ppm): 4.06 (s, 6H, P–CH2), 4.44–4.58 (q, 6H, N– CH2). 31P{H} NMR (D2O, ppm): 80.5 (q, 1JP–Cu = 750 Hz). 13C{H}(D2O, ppm): 71.5 (s, N–CH2), 50.8 (s, P–CH2). 63Cu NMR (D2O, ppm): 137.6 (quintet, 1JCu–P = 750 Hz). Crystal of [CuPTA4]BF46H2O, 16H2O, suitable for X-ray analysis were obtained from a acetonitrile/methanol solution. ESI-MS (electrospray ionization–mass spectrometry) (m/z assignment, % intensity): 377 ([Cu(PTA)2]+, 100), 220 ([CuPTA]+, 5), 158 (PTA+, 10). Anal. Calcd. for [CuPTA4]BF43H2O CuP4N12C24H48BF43H2O: C 34.60, H 6.53, N 20.18. Found: C 34.43, H 6.52, N 20.13. 2.2.2. [Cu(PTAH)4][Cl4(BF4)] (2) Concentrated hydrochloric acid (37%, 0.17 mL, 2.04 mmol) was added to an aqueous solution of (1) (398 mg, 0.51 mmol) at room temperature. After removal of water the white residue was identified as (2). 1H NMR (D2O, ppm): 4.18 (s, 6H, P–CH2), 4.88–4.74 (m, 6H, N–CH2). 31P{H} NMR (D2O, ppm): 80.4 (bs). Anal. Calcd. for CuP4N12C24H52Cl4BF4: C 31.17, H 5.67, N 18.17. Found: C 31.50, H 5.55, N 18.30. 2.2.3. [Cu(mPTA)4][(CF3SO3)4(BF4)] (3a) To an acetonitrile solution of (mPTA)(SO3CF3) (410 mg, 1.28 mmol) the Cu(I) precursor [Cu(CH3CN)4][BF4] (100 mg, 0.32 mmol) was added at room temperature. After 3 h stirring
1646
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
the solvent was removed and the white residue washed with diethylether. Yield: 87% 1H NMR (D2O, ppm): 2.72 (s, 3H, CH3), 4.07–3.85 (m, 4H, N–CH2–P), 4.58–4.37 (m, 6H, N–CH2–N and N+–CH2–P), 4.95–4.80 (m, 4H, N+–CH2–N). 31P{H} NMR (D2O, ppm): 78.0 (bs). 13C{H}(D2O, ppm): 81.2 (s, N+–CH2–N), 69.9 (s, N–CH2–N), 56.3 (d, N+–CH2–P), 50.6 (s, CH3), 46.5 (s, N–CH2–P). IR (KBr, cm1): m 1460(m), 1271(s), 1258(s), 1155(m), 1086(m), 1030(s), 637(s), 557(m), 518(m). Anal. Calcd. for [Cu(mPTA)4][(CF3SO3)4(BF4)] CuP4N12C28H60C4F12S4O12BF4: C 26.78, H 4.21, N 11.71. Found: C 27.11, H 4.22, N 11.95. Slow evaporation of a 50/50 water/methanol solution of (3a) instead produced X-ray quality crystal of [Cu(mPTA)4][(CF3SO3)2(BF4)3] 0.25H2O (3b0.25H2O).
2.2.4. [Cu(mPTA)4][(CF3SO3)4(PF6)] (3c) Reaction of [Cu(CH3CN)4][PF6] and (mPTA)(CF3SO3) was carried on analogously to the synthesis of (3a). Yield: 85% 1H NMR (D2O, ppm): 2.72 (s, 3H, CH3), 4.05–3.83 (m, 4H, N–CH2–P), 4.57–4.35 (m, 6H, N–CH2–N and N+–CH2–P), 4.93–4.74 (m 4H, N+–CH2–N). 31 P{H} NMR (D2O, ppm): 79.6 (bs), 144.6 (septet PF6). 13 C{H}(D2O, ppm): 80.9 (N+–CH2–N), 69.8 (N–CH2–N), 56.5 (d, N+– CH2–P), 49.5 (CH3), 46.1 (N–CH2–P). IR (KBr, cm1): m1460(m), 1271(vs), 1258(vs), 1155(m), 1031(s), 846(s), 637(s), 557(m), 517(m). Anal. Calcd. for [Cu(mPTA)4][(CF3SO3)4(PF6)]CH3CN CuP4N12C28H60C4F12S4O12PF6CH3CN C 26.61, H 4.14, N 11.86. Found: C 26.32, H 4.33, N 11.55.
2.2.5. [Cu(mPTA)3I][I]3 (4) To an acetonitrile solution of [Cu(CH3CN)4][BF4] (63 mg, 0.2 mmol) an excess of (mPTA)I (202 mg, 1.02 mmol) was added at room temperature. The reaction mixture was stirred overnight, filtered and the white residue was washed with chloroform, diethyl ether and water. Yield: 80%. 1 H NMR (D2O, ppm): 2.7 (s, 3H, CH3), 3.8–4.1 (m, 4H, N–CH2–P), 4.3–4.6 (m, 6H, N–CH2–N and N+–CH2–P), 4.7–5.1 (m, 4H, N+–CH2– N). 31P{H} NMR (D2O, ppm): 81.5 (bs). 13C{H}(DMSO, ppm): 80.8 (s, N+–CH2–N), 69.9 (s, N–CH2–N), 49.9 (s, CH3), 47.6 (N–CH2–P). IR (KBr, cm1): m 1459(m), 1310(s), 1091(s), 922(m), 810(m), 557(m). Anal. Calcd. for [Cu(mPTA)I][I]3 CuP3N9C21H45I4: C 23.19, H 4.17, N 11.59. Found: C 23.50, H 4.13, N 11.17. X-ray quality single crystals of [Cu(mPTA)3I][I]3H2O, 4H2O, were obtained from a 50/50 acetonitrile/methanol solution.
2.3. X-ray measurements and structure determination Crystal data, collected reflections and parameters of the final refinement are reported in Table 1. Intensity data were collected using a Philips PW1100 diffractometer (FEBO system) with graphite-monochromated (Mo Ka) radiation, following the standard procedures. There were no significant fluctuations of intensities other than those expected from Poisson statistics. All intensities were corrected for Lorentz polarization and absorption [29]. The positions of the heavy atoms were obtained from Patterson syntheses [30]. All non-H atoms were located in the subsequent Fourier maps. Refinement was carried out by full-matrix least-squares procedures (based on (F 2o ) using anisotropic temperature factors for all non-hydrogen atoms. The H-atoms were placed in calculated positions with fixed, isotropic thermal parameters (1.2Uequiv of the parent carbon atom). Calculations were performed with the SHELX-97 program [31] implemented in the WinGX package [32]. 2.4. Experiments with human cells Copper complexes and the corresponding uncoordinated ligands and counteranions were dissolved in purified water just before the experiment. Cisplatin was dissolved in dimethylsulfoxide just before the experiment, and a calculated amount of drug solution was added to the growth medium containing cells to a final solvent concentration of 0.5%, which had no discernible effect on cell killing. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) and cisplatin were obtained from Sigma Chemical Co., St. Louis, MO. 2.5. Cell cultures A549, MCF7, HCT-15 and HeLa are human lung, breast, colon and cervix carcinoma cell lines, respectively, which were obtained from ATCC (Rockvill, MD), along with the malignant melanoma (A375). Human ovarian cancer cell line (2008) and its cisplatinresistant variant, C13 cells, were kindly provided by Prof. G. Marverti (Department of Biomedical Science of Modena University, Italy). A431 human cervix carcinoma and the cisplatin-resistant counterparts, A431-Pt cells, were kindly provided by Prof. Zunino (Division of Experimental Oncology B, Istituto Nazionale dei Tumori, Milan, Italy). All cisplatin-resistant sublines were selected after continuous in vitro exposure to increasing concentrations of
Table 1 Crystal data and structure parameters for compounds 16H2O, 3b0.25H2O, 4H2O. Compound
16H2O [Cu(PTA)4] [BF4]6H2O
3b0.25H2O [Cu(mPTA)4][(CF3SO3)2 (BF4)3]0.25H2O
4H2O [Cu(MePTA)3I][I]3H2O
Empirical formula Formula weight Crystal system Space group
C24H48N12O6P4BF4Cu 874.97 Cubic
C30H60.5N12O6.25P4S2B3F18Cu 1315.37 Monoclinic P21/n (14)
C21H45N9OP3I4Cu 1103.71 Orthorhombic Pnma (62)
12.074(3) 23.323(3) 19.509(3) 104.38(3) 5322(2) 4 1.641 0.725 2688 9619 8309 698 0.068 0.170
14.012(2) 17.791(3) 15.058(3)
a(Å) b(Å) c(Å) ß(°) Volume (A3) Z qcalcd (g cm3) l(Mo Ka) (mm1) F(0 0 0) No. reflections collected No. reflections used [I P 2r(I)] Number of parameters P P R¼ jFoj jmj= jFoj nP o1=2 P Rw ¼ ½wðFo2 Fc2 Þ2 = ½wðFo2 Þ2
Fd 3 (203) 20.319(3) – – – 8389(2) 8 1.386 0.741 3632 5786 654 41 0.064 0.148
3754(1) 4 1.953 4.029 2112 4125 3033 193 0.068 0.162
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
cisplatin. Cell lines were maintained in the logarithmic phase at 37 °C in a 5% carbon dioxide atmosphere, using the following culture media: (i) RPMI-1640 medium (Euroclone, Celbio, Milan, Italy) containing 10% fetal calf serum (Biochrom-Seromed GmbH&Co., Berlin, Germany) and supplemented with 25 mM HEPES buffer (HEPES = 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), Lglutamine, and the antibiotics penicillin (50 units mL1) and streptomycin (50 g mL1) for MCF7, HCT-15, 2008, C13, A431, and A431-Pt cells; (ii) F-12 HAM’S (Sigma Chemical Co.) containing 10% fetal calf serum, L-glutamine, penicillin (50 units mL1), and streptomycin (50 g mL1) for HeLa cells; (iii) D-MEM (Dulbecco’s modified eagle’s medium; Euroclone) supplemented with 10% fetal calf serum (Euroclone), L-glutamine, penicillin (50 units mL1), and streptomycin (50 g mL1) for A549 and A375 cells. 2.6. Cytotoxicity assay The growth inhibitory effect toward tumor cell lines was evaluated by means of MTT (MTT = 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide) assay (tetrazolium salt reduction). Briefly, 3–8 103 cells/well, dependent upon the growth characteristics of the cell line, were seeded in 96-well microplates in growth medium (100 lL) and then incubated at 37° C in a 5% carbon dioxide atmosphere. After 24 h, the medium was removed and replaced with a fresh one containing the compounds to be studied at the appropriate concentration. Triplicate cultures were established for each treatment. After 48 h, each well was treated with 10 lL of a 5 mg mL1 MTT saline solution, and after 5 h of incubation, 100 lL of a sodium dodecylsulfate (SDS) solution in 0.01 M HCl was added. Following overnight incubation, the inhibition of cell growth induced by the tested complexes was detected by measuring the absorbance of each well at 570 nm using a Bio-Rad 680 microplate reader. Mean absorbance for each drug dose was expressed as a percentage of the control untreated well absorbance and plotted vs. drug concentration. IC50 values represent the drug concentrations that reduced the mean absorbance at 570 nm to 50% of those in the untreated control wells. 3. Results and discussion 3.1. Synthesis and characterization of Cu(I) complexes All the copper complexes were obtained starting from Cu(I) precursors according to Scheme 1. The syntheses were carried out in acetonitrile at room temperature and consist of a ligand-exchange reaction between the labile acetonitrile molecules of [Cu(CH3CN)4]+ and the appropriate phosphine ligands. This procedure differs from the reported synthesis of [Cu(PTA)4][NO3] and [Cu(PTAH)4][NO3]5, analogous to 1 and 2 except for the counterions, which were obtained from Cu(NO3)2 exploiting the reducing ability of the PTA phosphine [23]. The use of pre-reduced Cu(I) starting products allowed to
PTA
(ii) [Cu(PTAH) 4Cl4]+ [Cu(PTA) 4]+ 1 HCl 2
(i) (iii) [Cu(CH3CN)4]+
(mPTA)(CF 3SO3)
[Cu(mPTA) 4(CF3SO3)4]+ 3
(iv) (mPTA)I [Cu(mPTA) 3I][I]3 4 Scheme 1. Synthesis of compounds 1, 2, 3 and 4 starting from a Cu(I) precursor.
1647
perform reactions at room temperature with high yields (85– 90%), and contemporary to use a reduced amount of phosphine. Protonation of 1 was performed with a stoichiometric amount of hydrochloric acid and, as already observed elsewhere [23], the interconversion between 1 and 2 is a reversible, pH dependant process. In particular, from 31P NMR it was found that deprotonation of the PTAH derivative was complete at pH P 8. Calculation of the pKa value of 1 would be an important clue to determine the pH range where 1 and 2 are in equilibrium. Recently, determination of the pKa of some ruthenium-arene PTA complexes by means of 31P NMR has been reported and interestingly the obtained pKa values were lower with respect to that of the free ligand (3 vs. 5.6) [33]. Unfortunately, in our case the peculiarity of 31P spectrum of 1 (presence of a 1:1:1:1 quartet, vide infra), the slight difference in chemical shift between 1 and 2 and the presence of four PTA ligands to be protonated in each molecule made more difficult to carry out a pH titration and to determinate a proper pKa value. However, the NMR unequivocal determination of pure 1 and 2 was possible as the value of their autogeneous pH in unbuffered D2O solution is consistent with the protonated (2) or unprotonated species (1). Notwithstanding the straightforwardness of PTA N-alkylation, only Ru, Pd, Au, Rh and Ni derivatives containing mPTA have been reported so far [15]. Reaction of [Cu(CH3CN)4]+ salts with [mPTA]X (X = I, CF3SO3) were carried on in acetonitrile at room temperature, analogously to the synthesis of 1, using a 1:4 Cu:ligand ratio. As reported in Scheme 1, the nature of mPTA counter ion greatly affected the products of the reaction. In fact, by using (mPTA)I a water insoluble product was invariably obtained. Its IR spectrum did not show any bands attributable to the BF4 or PF6 group, nor the 31P{1H} NMR spectrum showed the septet typical of PF6. By means of elemental analysis the compound was identified as the mixed-ligand complex [Cu(mPTA)3I][I]3 4. Coordination of three phosphines to copper ion was confirmed by X-ray diffraction studies of crystals grown from a 50/50 methanol/acetonitrile solution and consistent of [Cu(mPTA)3I][I]3H2O, 4H2O, molecules. The low solubility of 4 both in water and in common organic solvents prevented its use in biological studies. Attempts to obtain a fully phosphine substituted derivative were made forcing the reaction conditions, i.e. by increasing the reaction temperature and the mPTA:Cu ratio, but the iodine-coordinated complex 4 was invariably obtained. The coordination ability of the iodine favoured its direct bonding to the metal thus forming the neutral, water insoluble compound 4. The choice of the counter ion both of the metal precursor and of the mPTA salt is crucial in determining the composition and the hydrophilicity of the products. Looking for a poorly coordinating anion able to ensure good hydro-solubility, we used the triflate salt mPTA(OSO2CF3) [34] instead of the iodine derivative. Using a synthetic procedure similar to that employed for 1, the ‘‘CuP4” species [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a and [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c were straightforwardly obtained in good yields (P85%). As expected both compounds possess a good water solubility (ca. 5 mg/ml) and are stable to Cu(I) disproportionation in the solid and in the solution states, so allowing in vitro citotoxicity studies. The use of BF4 or PF6 anions did not greatly affect the reactivity and the solubility of the final copper complexes; the choice of PF6 can be more convenient as it displays a septet in the 31P NMR which can be a diagnostic tool for characterization. The infrared spectra of 3a, 3c and 4 show typical bands of coordinated mPTA and are generally poorly diagnostic. IR spectra have been useful only in the characterization of 4 as no bands typical of BF4 or PF6 anions were detectable. Conversely, spectra of 3a and 3c showed bands due to the anion groups, in particular vibrations at 1271 and 1259 cm1 due to triflate, 846 cm1 due to PF6 and 1086 cm1 due to BF4 group.
1648
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
NMR studies of all compounds were carried out in D2O and, in the case of 4, also in DMSO. In 1H NMR spectra the effect of the counterion (BF4 or PF6) on the chemical shift of the pertinent complex is negligible. Moreover, no significant differences affect the 1H NMR spectrum of 4 in comparison with those of 3a and 3c. 1H NMR spectra of mPTA derivatives show a singlet attributable to the methyl group (d 2.72 ± 0.01) and signals due to four types of methylene protons. In all the spectra the PCH2NCH3 protons, which give rise to a doublet in the spectrum of the free ligand, show a broad singlet around 4.35 ppm. The 31P–1H coupling is instead maintained for the PCH2N signal which appears as a multiplet centred at ca. 3.9 ppm. The patterns of NCH2NCH3 and NCH2N are similar to those observed in free mPTA, each consisting of an ABMX and an ABMQX spin system centred at 4.4 and at 4.8 ppm, respectively. It is worth noting that protonation of PTA does not remove the magnetic equivalence of the three PCH2N and of the three NCH2N methylene groups both in the free ligand and in the copper complexes, whereas N-methylation determines a relevant change in 1 H NMR spectra. A slight downfield shift due to PF6, with respect to BF4, is evident in the 13C spectra of 3a and 3c, together with a remarkable increase of 1JC–P coupling costant value from 9 to 16 Hz. No 31P–13C coupling is evident in 13C spectrum of 4. The 31P{1H} spectrum of 1 displays a characteristic 1:1:1:1 quartet at room temperature [23] while 3a, 3c and 4 display a broad singlet respectively at 78, 79,6 and 81,5 ppm with no evidence of 31P–63Cu coupling notwithstanding the high symmetry of 3a and 3c. Lacking of such coupling has already been evidenced in the case of PTAH derivatives and has been ascribed to a fluxional behaviour in solution rather than to marked geometry distortions [23]. A fast ligand-exchange can similarly be supposed for 3; in fact, addition of an excess of free mPTA to 3a or 3c solutions determines a slight upfield shift of the 31P singlets signals without showing a distinct resonance of free mPTA.
such as [Cu(thp)3(P(CH2OH)2other tetrahedral CuP4 complexes 0 A)[10] and [Cu(PMe)30 ]X, whose mean (CH2O))] (average 2.281(2) Å Cu–P bond lengths are 2.270, 2.271 and 2.278 Å A for X = Cl, Br, I respectively [35]. A three-dimensional H-bond network involving the six water molecules for every [Cu(PTA)4]+ cation, and the nitrogen atoms of the PTA ligands characterizes the whole structure of 1 as also found in [(PTA)4Cu]Cl6H2O [24] and in [(PTA)4Au]Cl6H2O [36]. The asymmetric unit of 3b0.25H2O consists of the [Cu(mPTA)4]5+ cation (Fig. 2) which is balanced by two triflate groups and three BF4 moieties. The copper ion presents tetrahedral coordination geometry due to the coordination of four independent phosphorous atoms of the ligands with P–Cu–P angles ranging from 105.8(1) to 112.7(1) (Table 2). The Cu–P bond distances are equal 0 in the limit of their e.s.d.s (2.281(3) Å A average value) and comparable to those found in the [Cu(PTAH)4][NO3]56H2O complex 0 (2.298(1) Å A average value) [23]. The four methylated phosphane
3.2. X-ray crystallography The structure of 16H2O is highly symmetric, with the copper atom in special position that imposes a perfect tetrahedral geometry to the [Cu(PTA)4]+ cation (Fig. 1). 0 The Cu–P bond distances of 2.258(2) Å A in this BF4 derivative are significantly shorter with respect to those observed in the analogous complex with the nitrate as counter anion (Cu–Pav 0 2.284(2) Å A) that presents also a less symmetric complex cation [23] and is comparable to the0 highly symmetric analogous chloride derivative (Cu–P 2.2596(13) Å A) [24]. It is worth noting that the Cu– P distances of 1 are smaller with respect to the values found for
Fig. 2. Perspective view of the cation in compound 3b0.25H2O. Thermal ellipsoids are shown at 40% probability levels.
Table 2 Relevant bond distances and angles in the coordination spheres of [Cu(PTA)4][BF4]6H2O 16H2O, [Cu(mPTA)4][(CF3SO3)2(BF4)3]0.25H2O 3b0.25H2O and [Cu(mPTA)3I][I]3H2O 4H2O complexes.
0
Fig. 1. Perspective view of the cation in compound 16H2O. Thermal ellipsoids are shown at 40% probability levels.
Compound
16H2O
3b0.25H2O
4H2O
Cu–P(1) Cu–P(2) Cu–P(3) Cu–P(4) Cu–I(1) N–Ca (Me)N–Cb N–CMec P(1)–Cu–P(1)0 P(1)–Cu–P(2) P(1)–Cu–P(3) P(1)–Cu–P(4) P(2)–Cu–P(3) P(2)–Cu–P(4) P(3)–Cu–P(4) P(1)–Cu–I(1) P(2)–Cu–I(1)
2.258(2) – – –
2.284(1) 2.284(1) 2.280(1) 2.277(1)
1.468(6)
1.46(1) 1.53(1) 1.50(1) – 109.4(1) 110.6(1) 109.8(1) 112.7(1) 105.8(1) 108.3(1)
2.254(2) 2.247(3) – – 2.631(2) 1.48(1) 1.53(1) 1.46(1) 118.4(1) 112.7(1) – – – – – 105.0(1) 100.8(1)
109.5(2) – – –
at x + 1/4, y + 1/4, z for compound 1; x, y + 1/2, z for compound 4. a Average distance calculated for tertiary nitrogens. b Average distance calculated for quaternary nitrogens. c Average.
1649
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
3.3. Cytotoxicity assays
Fig. 3. Perspective view of the cation in compound 4H2O. Thermal ellipsoids are shown at 40% probability levels.
ligands are characterized by a lengthening of the C–N distances relative to the quaternarized nitrogens with respect to the tertiary ones (av. 1.53(1) vs. 1.46(1)). Such a lengthening is stressed compared to those complexes bearing protonated PTA such as [Cu(PTAH)4][23] and [M(PTAH)4]Cl4 (M = Ni, Pd, Pt) derivatives [37]. The average N–CMe distance is 1.50(1) compatible with a N–C single bond. Compound 4H2O is a three iodine salt with three mPTA ligands and one additional iodine linked to the metal in about a tetrahedral fashion (Fig. 3). A crystallographic mirror plane passing through iodine and P(2) atom characterizes the 0complex cation. The Cu–P bond distances of shorter with respect to those found in 2.254(2) and 2.247(3) Å A are 0 compound 3b (2.281(3) Å A average) due to the substitution of the sterically hindered mPTA ligand with iodine, so allowing a closer approach of the remaining three mPTA ligands towards a trigonal geometry as shown by the values of the P–Cu–P angles of 118.4(1)° and 112.7(1)°. An analogous behaviour has already been described in [(mPTA)3AuI][I]32H2O and [(EtPTA)3AuI][I]33H2O complexes [38,39]. The three iodide anions of compound 4H2O are hydrogen bonded to the water molecules present in the crystal lattice. We were unable to locate the H atoms of the water molecule, but the values of the distances of O I(2)0 3.56(1), O I(3)00 3.39(1) (0 at A (00 at x + 1/ x + 1/2 + 1, y + 1, +z + 1/2); and O I(3)00 3.39(1) Å 2 + 1, +y 1/2, + z + 1/2) are consistent with the presence of O– H I hydrogen bonding.
Compounds 1, 2, 3a and 3c and the corresponding uncoordinated ligands and counterions were examined for their cytotoxic properties against a panel of human tumor cell lines containing examples of lung (A549), cervix (HeLa), colon (HCT-15) and breast (MCF-7) cancer and melanoma (A375). For comparison purposes, the cytotoxicity of cisplatin was evaluated under the same experimental conditions. IC50 values, calculated from the dose-survival curves obtained after 48 h of drug treatment by MTT test, are shown in Table 3. Uncoordinated ligands and the various counterions proved to be ineffective in all tumor cell lines. On the contrary, all the tested complexes showed a growth inhibitory potency in the micromolar range towards the different types of tumor cells. They displayed an in vitro antitumor activity similar to that shown by cisplatin, the mean IC50 values (lM) being 15.0, 17.6, 20.0, 18.8 and 21.8 for 1, 2, 3a, 3c and cisplatin, respectively. In particular, on A549 lung adenocarcinoma cells, that are poor chemosensitive to cisplatin [40], the cytotoxicities of copper complexes exceeded that of the reference drug by a factor ranging from about 2 to 3. The cytotoxic effect elicited by these copper compounds seems to be uninfluenced by changes in the steric demand and charge of the ligand (PTA vs PTAH and mPTA). In the case of 2, the similarity of cytotoxic efficiency with 1 could be explained by deprotonation of 2 to give 1 at physiological pH (pH ffi 7.2) determined by the use of a buffered milieu such as the growth medium. Moreover, as many cancer cells have a lower pH compared to normal cells, an equilibrium between 1 and 2 may take place into the cell as well, determining a similar cytotoxic effect for 1 or 2. Also the insertion of a methyl group in the PTA framework does not induce significant changes in the cytotoxic profiles of the corresponding 3a and 3c, which were found roughly as effective as the parent compound 1. Interestingly in the case of several ruthenium compounds a different activity and specificity was observed by using PTA or mPTA. [33,41]. Also the specificity against a particular cancer type was unaffected by the ligand sphere as the higher activity was invariably shown by all derivatives against HeLa cervix carcinoma and A549 lung adenocarcinoma cells, whereas the lower against MCF-7 breast carcinoma cells. Interestingly, on HeLa and A549 cell lines the activity shown by the original compound 1 was comparable to that of [Cu(thp)4][PF6] [11]. It is worth mentioning that experiments performed with [64Cu(thp)4]+ and [64Cu(PTA)4]+ and EMT-6 mouse mammary tumor cells have assessed a slightly higher serum stability and a significative higher cellular uptake of the PTA derivative probably related to their different lipophilicity (log P 0.01 ± 0.01 for [64Cu(PTA)4]+ and 2.26 ± 0.04 for [64Cu(thp)4]+) [42].
Table 3 In vitro antitumor activity. Compound
IC50 (lM) ± S.D. MCF-7
HeLa
A549
HCT-15
A375
[Cu(PTA)4][BF4] 1 [Cu(PTAH)4][Cl4(BF4)] 2 [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c Cisplatin (mPTA)(CF3SO3) PTA PTAHCl NaBF4 KPF6
23.2 ± 1.2 23.6 ± 0.6 25.8 ± 1.2 23.8 ± 2.5 23.2 ± 1.3 nd nd nd nd nd
7.58 ± 0.19 12.2 ± 2.0 15.8 ± 1.1 13.6 ± 1.3 11.2 ± 1.5 nd nd nd nd nd
8.47 ± 0.10 11.4 ± 2.6 14.6 ± 1.9 13.9 ± 1.7 29.3 ± 1.7 nd nd nd nd nd
19.2 ± 2.1 21.9 ± 1.3 23.4 ± 1.4 23.2 ± 0.8 25.3 ± 1.2 nd nd nd nd nd
16.4 ± 1.2 18.8 ± 2.0 19.8 ± 1.8 19.4 ± 1.6 20.3 ± 2.2 nd nd nd nd nd
S.D. = standard deviation. nd = Not detectable. IC50 values were calculated by probit analysis (P < 0.05, v2 test). (5–8) 104 mL1 cells were treated for 48 h with increasing concentrations of tested compounds. Cytotoxicity was assessed by MTT test.
1650
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
Table 4 Cross-resistance profiles. Compound
Human ovarian adenocarcinoma cells [Cu(PTA)4][BF4] 1 [Cu(PTAH)4][Cl4(BF4)] 2 [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c Cisplatin
IC50 (lM) ± SD 2008
C13
R.F.
14.24 ± 2.18 14.89 ± 1.67 9.43 ± 1.01 10.24 ± 0.87 12.69 ± 1.23
11.45 ± 1.07 13.28 ± 2.31 10.08 ± 1.32 9.41 ± 0.65 89.21 ± 2.24
1.24 1.12 0.93 1.09 7.02
Human cervix squamous carcinoma cells A431
A431-Pt
R.F.
[Cu(PTA)4][BF4] 1 [Cu(PTAH)4][Cl4(BF4)] 2 [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c Cisplatin
10.52 ± 0.55 17.01 ± 2.13 9.23 ± 0.98 10.75 ± 0.87 57.76 ± 1.45
1.03 1.00 1.01 0.96 2.61
10.14 ± 0.94 16.91 ± 1.98 9.10 ± 0.21 11.15 ± 0.70 22.06 ± 1.27
SD = standard deviation. IC50 values were calculated by probit analysis (P < 0.05, v2 test). (3–5) 104 mL1 cells were treated for 48 h with increasing concentrations of tested compounds. Cytotoxicity was assessed by MTT test. Resistant factor (RF) is defined as IC50 resistant/parent line.
The encouraging results obtained against the in-house panel of cell lines prompted us to test the cytotoxic activity of our Cu(I) complexes onto two additional cell line pairs, which were selected for their resistance to cisplatin: 2008/C13 ovarian cancer cells and A431/A431-Pt cervix carcinoma cells. Although cisplatin resistance is multifactorial, the main molecular mechanisms involved in resistance in these sublines have almost been defined. In C13 cells, resistance is correlated to reduced cellular drug uptake, high cellular glutathione levels, and enhanced repair of DNA damage [43]. In human squamous cervix carcinoma A431-Pt cells, resistance is due to defect in drug uptake and to decreased levels of proteins involved in DNA mismatch repair (MSH2), causing an increased tolerance to cisplatin-induced DNA damage [44]. Cytotoxicity of tested compounds in sensitive and resistant cells was assessed after a 48 h drug exposure by MTT test. Cross-resistance profiles were evaluated by means of the resistance factor (RF), which is defined as the ratio between IC50 values calculated for the resistant cells and those arising from the sensitive ones (Table 4). All the tested copper(I) complexes induced a similar pattern of response across the parental and the resistant sublines indicative of different cross-resistance profiles than that of cisplatin. Against 2008/C13 cell pair, the resistance factor of copper derivatives was roughly 7-fold lower than that of cisplatin, whereas against A431/A431-Pt cell line pair it was about 2.5-fold lower. The circumvention of cisplatin resistance, recently observed for other phosphine-containing Cu(I) derivatives, supports the hypothesis of a different mechanism of action of copper(I) complexes than that of cisplatin.
4. Conclusions A series of water-soluble Cu(I) compounds has been synthesized and their cytotoxic activity checked against an in-house panel of cell lines. The data reported in the present paper show that all the studied complexes containing PTA derivatives are effective against the panel of human tumor cell lines tested with IC50 values ranging from 8 to 25 lM, mostly comparable with those elicited by cisplatin. Cytotoxicity was found not to be significantly affected by both the nature of the PTA ligand and the anionic counterion. Interestingly, the original compound [Cu(PTA)4](BF4) appears roughly 1.5-fold most powerful than cisplatin and, towards human cervix and non-small lung carcinoma cells, [Cu(PTA)4][BF4]
shows an antiproliferative activity very similar to that already described for [Cu(thp)4][PF6]. Moreover, such class of ‘‘CuP4” compounds can be equally cytotoxic in both the cisplatin-sensitive and -resistant sublines of human ovarian and cervix squamous carcinoma cells ruling out the occurrence of cross resistance phenomena and suggesting a mechanism of action different from the covalent binding to DNA shown by cisplatin, as already observed for other phosphine–copper(I) agents. The reported results confirm the high potentiality of water-soluble phosphine copper(I) complexes as a new class of non-platinum anticancer drugs. Actually, expanding the molecular targets for metallo-drugs will open new opportunities to the strategic development of more effective and less-toxic anticancer drugs. Acknowledgements This work was financially supported by University of Padova (Progetto di Ateneo CPDA065113/06), and by Ministero dell’Istruzione dell’Università e della Ricerca (PRIN 2007). We are grateful to CIRCMSB (Consorzio Interuniversitario di Ricerca in Chimica dei Metalli nei Sistemi Biologici). The authors thank A. Moresco for elemental analyses. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jinorgbio.2009.09.005. References [1] B. Rosemberg, L. Van Camp, T. Krigas, Nature 205 (1965) 698–699. [2] S.J. Berners-Price, P.J. Sadler, in: Structure and Bonding, Bioinorganic Chemistry, vol. 70, Springer, Berlin/Heidelberg, 1988, pp. 27–102. [3] C.F. Shaw III, Chem. Rev. 99 (1999) 2589–2600. [4] T.M. Simon, D.H. Kunishima, G.J. Vibert, A. Lorber, Cancer Res. 41 (1981) 94–97. [5] S.J. Berners-Price, C.K. Mirabelli, R.K. Johnson, M.R. Mattern, F.L. McCabe, L.F. Faucette, C.M. Sung, S.M. Mong, P.J. Sadler, S.T. Crooke, Cancer Res. 46 (1986) 5486–5493. [6] T. Wang, Z. Guo, Curr. Med. Chem. 13 (2006) 525–537. [7] C. Marzano, M. Pellei, F. Tisato, C. Santini, Anti-Cancer Agents Med. Chem. 9 (2009) 185–211. [8] S.J. Berners-Price, M.E. Sant, R.I. Christopherson, P.W. Kuchel, Magn. Reson. Med. 18 (1991) 142–158. [9] G.D. Hoke, R.A. Macia, P.C. Meunier, P.J. Bugelski, C.K. Mirabelli, G.F. Rush, W.D. Matthews, Toxicol. Appl. Pharmacol. 100 (1989) 293–306. [10] D. Saravana Bharathi, M.A. Sridhar, J. Shashidhara Prasad, A.G. Samuelson, Inorg. Chem. Commun. 4 (2001) 490–492. [11] C. Marzano, V. Gandin, M. Pellei, D. Colavito, G. Papini, G. Gioia Lobbia, E. Del Giudice, M. Porchia, F. Tisato, C. Santini, J. Med. Chem. 51 (2008) 798–808. [12] C. Marzano, M. Pellei, D. Colavito, S. Alidori, G. Gioia Lobbia, V. Gandin, F. Tisato, C. Santini, J. Med. Chem. 49 (2006) 7317–7324. [13] C. Marzano, M. Pellei, S. Alidori, A. Brossa, G. Gioia Lobbia, F. Tisato, C. Santini, J. Inorg. Biochem. 100 (2006) 299–304. [14] C. Marzano, V. Gandin, A. Folda, G. Scutari, A. Bindoli, M.P. Rigobello, Free Radic. Biol. Med. 42 (2007) 872–881. [15] A.D. Phillips, L. Gonsalvi, A. Romerosa, F. Vizza, M. Peruzzini, Coord. Chem. Rev. 248 (2004) 955–993. [16] D.J. Daigle, Inorg. Synth. 32 (1998) 40–45. [17] A. Mena-Cruz, P. Lorenzo-Luis, A. Romerosa, M. Saoud, M. Serrano-Ruiz, Inorg. Chem. 46 (2007) 6120–6126. [18] M. Erlandsson, V.R. Landaeta, L. Gonsalvi, M. Peruzzini, A.D. Phillips, P.J. Dyson, G. Laurenczy, Eur. J. Inorg. Chem. (2008) 620–627. [19] C.A. Mebi, B.J. Frost, Organometallics 24 (2005) 2339–2346. [20] D.A. Krogstad, S.B. Owens, J.A. Halfen, V.G. Young Jr., Inorg. Chem. Commun. 8 (2005) 65–69. [21] W. Lewasin, in: F.R. Hartley (Ed.), The Chemistry of Organophosphorous Compounds, vol. 1, Wiley-Interscience, New York, 1990 (Chapter 15). [22] F. Tisato, F. Refosco, G. Bandoli, G. Pilloni, B. Corain, Inorg. Chem. 40 (2001) 1394–1396. [23] A.M. Kirillov, P. Smolenski, M.F.C. Guedes da Silva, A.J.L. Pombeiro, Eur. J. Inorg. Chem. (2007) 2686–2692. [24] A.M. Kirillov, P. Smolenski, M.F.C. Guedes da Silva, M.N. Kopylovich, A.J.L. Pombeiro, Acta Crystallogr. Section E – Struct Rep Online 64 (2008) M603– U32. [25] R. Wanke, P. Smolenski, M.F.C. Guedes da Silva, L.M.D.R.S. Martins, A.J.L. Pombeiro, Inorg. Chem. 47 (2008) 10158–10168.
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651 [26] M. Pellei, S. Alidori, M. Camalli, G. Campi, G. Gioia Lobbia, M. Mancini, G. Papini, R. Spagna, C. Santini, Inorg. Chim. Acta 361 (2008) 1456–1462. [27] M. Pellei, G. Gioia Lobbia, C. Santini, R. Spagna, M. Camalli, D. Fedeli, G. Falcioni, Dalton Trans. (2004) 2822–2828. [28] G.J. Kubas, Inorg. Synth. 26 (1990) 68–70. [29] A.T.C. North, D.C. Philips, F.S. Mathews, Acta Crystallogr. A24 (1968) 351–359. [30] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo, A. Guagliardi, A.G.G. Moliterni, G. Polidori, R. Spagna, SIR-97, J. Appl. Crystallogr. 32 (1999) 115–119. [31] G.M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal Structures, University of Göttingen, Germany, 1997. [32] L.J.J. Farrugia, Appl. Crystallogr. 32 (1999) 837–838. [33] C. Scolaro, A. Bergamo, L. Brescacin, R. Delfino, M. Cocchietto, G. Laurenczy, T.J. Geldbach, G. Sava, P.J. Dyson, J. Med. Chem. 48 (2005) 4161–4171. [34] A. Romerosa, T. Campos-Malpartida, C. Lidrissi, M. Saoud, M. Serrano-Ruiz, M. Peruzzini, J.A. Garrido-Cárdenas, F. García-Maroto, Inorg. Chem. 45 (2006) 1289–1298.
1651
[35] G.A. Bowmaker, P.C. Healy, L.M. Engelhardt, J.D. Kildea, B.W. Skelton, A.H. White, Aust. J. Chem. 43 (1990) 1697–1705. [36] J.M. Forwar, Z. Assefa, R.J. Staples, J.P. Fackler Jr., Inorg. Chem. 35 (1996) 16–22. [37] D.J. Darensbourg, T.J. Decuir, N.W. Stafford, J.B. Robertson, J.D. Draper, J.H. Reibenspies, A. Kathó, F. Joó, Inorg. Chem. 36 (1997) 4218–4226. [38] J.M. Forwar, J.P. Fackler Jr., R.J. Staples, Organometallics 14 (1995) 4194–4198. [39] E. Suresh, M.M. Bhadbhade, Acta Crystallogr. C53 (1997) 195–197. [40] P. Zhang, W.Y. Gao, S. Turner, B.S. Ducatman, Mol. Cancer 2 (2003) 1–9. [41] A. Garcia-Fernandez, J. Diez, A. Manteca, J. Sanchez, M.P. Gamasa, E. Lastra, Polyhedron 27 (2008) 1214–1228. [42] S. Alidori, G. Gioia Lobbia, G. Papini, M. Pellei, M. Porchia, F. Refosco, F. Tisato, J.S. Lewis, C. Santini, J. Biol. Inorg. Chem. 13 (2008) 307–315. [43] G. Marverti, P.A. Andrews, G. Piccinini, S. Ghiaroni, D. Barbieri, M.S. Moruzzi, Eur. J. Cancer 33 (1997) 669–675. [44] C. Lanzi, P. Perego, R. Supino, S. Romanelli, T. Pensa, N. Carenini, I. Viano, D. Colangelo, R. Leone, P. Apostoli, G. Cassinelli, R.A. Gambetta, F. Zunino, Biochem. Pharmacol. 55 (1998) 1247–1254.
Contents lists available at ScienceDirect
Journal of Inorganic Biochemistry journal homepage: www.elsevier.com/locate/jinorgbio
Synthesis and structural characterization of copper(I) complexes bearing N-methyl-1,3,5-triaza-7-phosphaadamantane (mPTA) Cytotoxic activity evaluation of a series of water soluble Cu(I) derivatives containing PTA, PTAH and mPTA ligands Marina Porchia a,*, Franco Benetollo a, Fiorenzo Refosco a, Francesco Tisato a, Cristina Marzano b, Valentina Gandin b a b
ICIS-CNR, Corso Stati Uniti 4, 35127 Padova, Italy Dipartimento di Scienze Farmaceutiche, Università di Padova, via Marzolo 5, 35131 Padova, Italy
a r t i c l e
i n f o
Article history: Received 17 April 2009 Received in revised form 8 July 2009 Accepted 8 September 2009 Available online 13 September 2009 Keywords: Water-soluble copper(I) complexes PTA Cytotoxic activity
a b s t r a c t New copper(I) complexes containing the water soluble N-methyl-1,3,5-triaza-7-phosphaadamantane (mPTA) phosphine have been synthesized by ligand-exchange reactions starting from [Cu(CH3CN)4][BF4] or [Cu(CH3CN)4][PF6] precursors and (mPTA)X (X = CF3SO3, I). Depending on the ligand counter ion, the hydrophilic [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a and [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c complexes or the iodine-coordinated [Cu(mPTA)3I]I3 4 species were obtained respectively and fully characterized by spectroscopic methods. Single crystal structural characterization was undertaken for [Cu(mPTA)3I]I3H2O, 4H2O, and [Cu(mPTA)4][(CF3SO3)2(BF4)3] 0.25H2O, 3b0.25H2O, the latter obtained by crystallization of [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a. The cytotoxicity of analogous tetrahedral homoleptic Cu(I) derivatives [Cu(PTA)4](BF4) 1, [Cu(PTAH)4][Cl4(BF4)] 2, [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a and [Cu(mPTA)4][ (CF3SO3)4(PF6)] 3c was evaluated against a panel of several human tumor cell lines. All the complexes showed in vitro antitumor activity comparable to that of the reference metallodrug cisplatin. Tests performed on cisplatin sensitive and resistant cell lines showed that against human ovarian 2008/C13 cell line pair, the resistance factor of copper derivatives was roughly 7-fold lower than that of cisplatin, whereas against human cervix cancer A431/A431-Pt cell line pair it was about 2.5-fold lower. These results, confirming the circumvention of cisplatin resistance, support the hypothesis that phosphine copper(I) complexes follow different cytotoxic mechanisms than do platinum drugs. Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction Since the discovery of the antitumor activity of cisplatin (cis-[PtCl2(NH3)2]) for the treatment of several human tumors [1], thousands of platinum and other metal-based compounds have been tested for their potential antitumor properties in the last 40 years. A series of compounds showing encouraging perspectives were the phosphine complexes of group 11 metal ions [2]. Their biological properties were little explored until late 1970s when a thioglucose derivative of triethylphosphine gold(I) (auranofin) was found to possess antiarthritic activity [3] and subsequently shown to have in vivo antitumor activity in murine models, although only against P388 leukemia [4]. In an attempt to identify
* Corresponding author. Fax: +39 049 8295951. E-mail address: [email protected] (M. Porchia). 0162-0134/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.jinorgbio.2009.09.005
gold complexes with a wider spectrum of activity, Sadler and coworkers demonstrated the cytotoxicity of mono-cationic bisdiphosphine gold(I) compounds toward a panel of human tumor cell lines including B16 melanoma, P388 leukemia, and M5076 reticulum cell carcinoma [5]. In particular, it was found that [Au(dppe)2][Cl] (dppe = 1,2-bis(diphenylphosphino)ethane) was 10-fold more cytotoxic than dppe alone, suggesting, at that time, that metal ions could potentiate the cytotoxic properties of dppe [5]. In general, the higher drug tolerance profiles of gold-based agents compared to the severe toxic effects on normal tissues and/or the occurrence of inherited or acquired resistance induced by cisplatin have represented a hopeful prospect for the chemotherapeutic application of gold-based drugs. Despite this promising outlook, clinical trials in humans of gold compounds remain still elusive. A rational extension of diphosphine gold(I) chemistry to the first row congener copper indicated that analogous copper(I) complexes of ‘‘CuP4” stoichiometry could be efficiently prepared and
1645
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
tested in vitro [6,7]. Among them, [Cu(dppe)2][Cl] and [Cu(dppey)2][Cl] (dppey = 1,2-bis(diphenylphosphino)ethylene) produced cytotoxic effects on cell metabolism comparable to those exhibited by Au(I) analogs, which were finally postulated to be the result of the uncoupling of mitochondrial oxidative phosphorylation [8]. However, the presence of several phenyl groups appended to the phosphorus donors in both [Cu(dppey)2][Cl] and [Au(dppe)2][Cl] species caused undesired nephrotoxicity [2] and cardiovascular toxicity [9] in animal models, respectively, thus precluding clinical trials in humans. Aiming at the formation of less-toxic copper compounds, we have recently prepared a series of hydrophilic copper(I) derivatives including the water soluble tris(hydroxymethyl)phosphine (thp) ligand, either alone [10,11] or in combination with hydrophilic scorpionates [12]. These water soluble compounds were proved to be easier to handle during the in vitro tests and, more importantly, have shown cytotoxic activity against a large panel of human tumor cell lines belonging to a variety of tumor types, including cisplatin and multidrug resistant phenotypes [11–13]. Moreover, it has been found that the cytotoxic activity of these copper(I) complexes may be correlated to their ability to induce a non-apoptotic mechanism of cell death [11,12]. Such results, together with recent studies which reported that the antiproliferative action of auranofin is connected with the inhibition of the thioredoxin reductase leading to augmented apoptosis [14], corroborate the view that mechanisms of action different from the DNA damage induced by cisplatin could underlie the cytotoxic activity of phosphine Au(I) and Cu(I) drugs. In other words, other molecular targets, in addition to DNA, could be identified in order to inhibit cancer cell proliferation, thus offering further motivation for the design of novel metal-based anticancer drugs. The water-soluble 1,3,5-triaza-7-phosphaadamantane (PTA) phosphine ligand and its derivatives are receiving growing attention in recent years [15–17]. The good water solubility of related transition metal PTA complexes makes possible their efficient application in aqueous phase catalysis [18–20]. The combination of PTA physico-chemical properties including water solubility, low steric hindrance (comparable to that of PMe3) and resistance to oxidation in water, joint to the ability to reduce various metal ions, include this phosphine among those few ligands suitable for stabilization of Cu(I) in aqueous media. While a large variety of lipophilic Cu(I) phosphine complexes stable in classic organic media has already been reported [21] very few examples of water-soluble Cu(I) compounds stable to disproportionation in water are known. These comprise the polymeric {Na5[Cu(TPPTS)2]5H2O}n complex [22] (TPPTS = tris(m-sulfonatophenyl)phosphine), [Cu(thp)4][PF6] (thp = tris(hydroxymethyl)phosphine), [Cu(bhpe)2][PF6] (bhpe = bis[bis(hydroxymethyl)phoshino]ethane) [11], the nitrate and chloride salts [Cu(PTA)4][X] (X = NO3, Cl) and [Cu(HPTA)4][NO3]5 [23,24]. Other recent water soluble Cu(I) derivatives present a mixed coordination sphere with a monodentate tertiary phosphine and a tripodal scorpionate ligand [12,25–27]. Aim of our study was to prepare a series of water-soluble copper(I) compounds whose steric and electronic properties could be tuned by slight modifications of the ligands, in order to compare and to correlate their physico-chemical properties with a possible biological activity. The PTA phosphine and its N-protonated (PTAH) and N-methylated (mPTA) derivatives (Chart 1) proved to be an ideal series of ligands for this purpose, as they give rise to different charged complexes conferring indeed a good hydrosolubility. In this study we report on the synthesis and characterization of new mPTA copper complexes, as well as the cytotoxic activity evaluation of stable water-soluble Cu(I) compounds containing PTA derivatives towards a panel of human tumor cell lines.
P
P N
N N PTA
P N
N N PTAH
+
H
+ N CH3
N N mPTA
Chart 1. 1,3,5-Triaza-7-phosphaadamantane (PTA) ligand and its N-protonated (PTAH) and N-methylated (mPTA) derivatives.
2. Experimental 2.1. General procedures All solvents and commercially available substances were of reagent grade and used without further purification. The Cu(I) starting compound [Cu(CH3CN)4][BF4] was prepared by reaction of [Cu(H2O)6][BF4]2 with metallic copper in acetonitrile whereas [Cu(CH3CN)4][PF6] was prepared from Cu2O and HPF6 according to Ref. [28]. The ligands PTA, (mPTA)I and (mPTA)(CF3SO3) were synthesized in accordance with published methods [16,17]. Elemental analyses were performed on a Carlo Erba 1106 Elemental Analyzer .1H, 31P, 13C and 63Cu NMR spectra were recorded on a Bruker AMX-300 instrument. The splitting of nuclear resonances in the reported NMR spectra is defined as s = singlet, d = doublet, q = quartet, m = multiplet, and bs = broad singlet. FT IR spectra were recorded on a Mattson 3030 Fourier transform spectrometer in the range 4000–400 cm1 in KBr pellets. The intensity of reported IR signals is defined as w = weak, m = medium, s = strong, and vs = very strong. Mass spectra have been recorded by an electrospray LCQ ThermoFinnigan mass spectrometer. 2.2. Syntheses of the complexes 2.2.1. [Cu(PTA)4][BF4] (1) To an acetonitrile solution of [Cu(CH3CN)4][BF4] (136 mg, 0.43 mmol) an excess of PTA (320 mg, 2.03 mmol) was added at room temperature. The reaction mixture was stirred overnight and then filtered. The white residue was washed three times with chloroform and diethyl ether and then dried under vacuum. Yield: 90% 1H NMR (D2O, ppm): 4.06 (s, 6H, P–CH2), 4.44–4.58 (q, 6H, N– CH2). 31P{H} NMR (D2O, ppm): 80.5 (q, 1JP–Cu = 750 Hz). 13C{H}(D2O, ppm): 71.5 (s, N–CH2), 50.8 (s, P–CH2). 63Cu NMR (D2O, ppm): 137.6 (quintet, 1JCu–P = 750 Hz). Crystal of [CuPTA4]BF46H2O, 16H2O, suitable for X-ray analysis were obtained from a acetonitrile/methanol solution. ESI-MS (electrospray ionization–mass spectrometry) (m/z assignment, % intensity): 377 ([Cu(PTA)2]+, 100), 220 ([CuPTA]+, 5), 158 (PTA+, 10). Anal. Calcd. for [CuPTA4]BF43H2O CuP4N12C24H48BF43H2O: C 34.60, H 6.53, N 20.18. Found: C 34.43, H 6.52, N 20.13. 2.2.2. [Cu(PTAH)4][Cl4(BF4)] (2) Concentrated hydrochloric acid (37%, 0.17 mL, 2.04 mmol) was added to an aqueous solution of (1) (398 mg, 0.51 mmol) at room temperature. After removal of water the white residue was identified as (2). 1H NMR (D2O, ppm): 4.18 (s, 6H, P–CH2), 4.88–4.74 (m, 6H, N–CH2). 31P{H} NMR (D2O, ppm): 80.4 (bs). Anal. Calcd. for CuP4N12C24H52Cl4BF4: C 31.17, H 5.67, N 18.17. Found: C 31.50, H 5.55, N 18.30. 2.2.3. [Cu(mPTA)4][(CF3SO3)4(BF4)] (3a) To an acetonitrile solution of (mPTA)(SO3CF3) (410 mg, 1.28 mmol) the Cu(I) precursor [Cu(CH3CN)4][BF4] (100 mg, 0.32 mmol) was added at room temperature. After 3 h stirring
1646
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
the solvent was removed and the white residue washed with diethylether. Yield: 87% 1H NMR (D2O, ppm): 2.72 (s, 3H, CH3), 4.07–3.85 (m, 4H, N–CH2–P), 4.58–4.37 (m, 6H, N–CH2–N and N+–CH2–P), 4.95–4.80 (m, 4H, N+–CH2–N). 31P{H} NMR (D2O, ppm): 78.0 (bs). 13C{H}(D2O, ppm): 81.2 (s, N+–CH2–N), 69.9 (s, N–CH2–N), 56.3 (d, N+–CH2–P), 50.6 (s, CH3), 46.5 (s, N–CH2–P). IR (KBr, cm1): m 1460(m), 1271(s), 1258(s), 1155(m), 1086(m), 1030(s), 637(s), 557(m), 518(m). Anal. Calcd. for [Cu(mPTA)4][(CF3SO3)4(BF4)] CuP4N12C28H60C4F12S4O12BF4: C 26.78, H 4.21, N 11.71. Found: C 27.11, H 4.22, N 11.95. Slow evaporation of a 50/50 water/methanol solution of (3a) instead produced X-ray quality crystal of [Cu(mPTA)4][(CF3SO3)2(BF4)3] 0.25H2O (3b0.25H2O).
2.2.4. [Cu(mPTA)4][(CF3SO3)4(PF6)] (3c) Reaction of [Cu(CH3CN)4][PF6] and (mPTA)(CF3SO3) was carried on analogously to the synthesis of (3a). Yield: 85% 1H NMR (D2O, ppm): 2.72 (s, 3H, CH3), 4.05–3.83 (m, 4H, N–CH2–P), 4.57–4.35 (m, 6H, N–CH2–N and N+–CH2–P), 4.93–4.74 (m 4H, N+–CH2–N). 31 P{H} NMR (D2O, ppm): 79.6 (bs), 144.6 (septet PF6). 13 C{H}(D2O, ppm): 80.9 (N+–CH2–N), 69.8 (N–CH2–N), 56.5 (d, N+– CH2–P), 49.5 (CH3), 46.1 (N–CH2–P). IR (KBr, cm1): m1460(m), 1271(vs), 1258(vs), 1155(m), 1031(s), 846(s), 637(s), 557(m), 517(m). Anal. Calcd. for [Cu(mPTA)4][(CF3SO3)4(PF6)]CH3CN CuP4N12C28H60C4F12S4O12PF6CH3CN C 26.61, H 4.14, N 11.86. Found: C 26.32, H 4.33, N 11.55.
2.2.5. [Cu(mPTA)3I][I]3 (4) To an acetonitrile solution of [Cu(CH3CN)4][BF4] (63 mg, 0.2 mmol) an excess of (mPTA)I (202 mg, 1.02 mmol) was added at room temperature. The reaction mixture was stirred overnight, filtered and the white residue was washed with chloroform, diethyl ether and water. Yield: 80%. 1 H NMR (D2O, ppm): 2.7 (s, 3H, CH3), 3.8–4.1 (m, 4H, N–CH2–P), 4.3–4.6 (m, 6H, N–CH2–N and N+–CH2–P), 4.7–5.1 (m, 4H, N+–CH2– N). 31P{H} NMR (D2O, ppm): 81.5 (bs). 13C{H}(DMSO, ppm): 80.8 (s, N+–CH2–N), 69.9 (s, N–CH2–N), 49.9 (s, CH3), 47.6 (N–CH2–P). IR (KBr, cm1): m 1459(m), 1310(s), 1091(s), 922(m), 810(m), 557(m). Anal. Calcd. for [Cu(mPTA)I][I]3 CuP3N9C21H45I4: C 23.19, H 4.17, N 11.59. Found: C 23.50, H 4.13, N 11.17. X-ray quality single crystals of [Cu(mPTA)3I][I]3H2O, 4H2O, were obtained from a 50/50 acetonitrile/methanol solution.
2.3. X-ray measurements and structure determination Crystal data, collected reflections and parameters of the final refinement are reported in Table 1. Intensity data were collected using a Philips PW1100 diffractometer (FEBO system) with graphite-monochromated (Mo Ka) radiation, following the standard procedures. There were no significant fluctuations of intensities other than those expected from Poisson statistics. All intensities were corrected for Lorentz polarization and absorption [29]. The positions of the heavy atoms were obtained from Patterson syntheses [30]. All non-H atoms were located in the subsequent Fourier maps. Refinement was carried out by full-matrix least-squares procedures (based on (F 2o ) using anisotropic temperature factors for all non-hydrogen atoms. The H-atoms were placed in calculated positions with fixed, isotropic thermal parameters (1.2Uequiv of the parent carbon atom). Calculations were performed with the SHELX-97 program [31] implemented in the WinGX package [32]. 2.4. Experiments with human cells Copper complexes and the corresponding uncoordinated ligands and counteranions were dissolved in purified water just before the experiment. Cisplatin was dissolved in dimethylsulfoxide just before the experiment, and a calculated amount of drug solution was added to the growth medium containing cells to a final solvent concentration of 0.5%, which had no discernible effect on cell killing. MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) and cisplatin were obtained from Sigma Chemical Co., St. Louis, MO. 2.5. Cell cultures A549, MCF7, HCT-15 and HeLa are human lung, breast, colon and cervix carcinoma cell lines, respectively, which were obtained from ATCC (Rockvill, MD), along with the malignant melanoma (A375). Human ovarian cancer cell line (2008) and its cisplatinresistant variant, C13 cells, were kindly provided by Prof. G. Marverti (Department of Biomedical Science of Modena University, Italy). A431 human cervix carcinoma and the cisplatin-resistant counterparts, A431-Pt cells, were kindly provided by Prof. Zunino (Division of Experimental Oncology B, Istituto Nazionale dei Tumori, Milan, Italy). All cisplatin-resistant sublines were selected after continuous in vitro exposure to increasing concentrations of
Table 1 Crystal data and structure parameters for compounds 16H2O, 3b0.25H2O, 4H2O. Compound
16H2O [Cu(PTA)4] [BF4]6H2O
3b0.25H2O [Cu(mPTA)4][(CF3SO3)2 (BF4)3]0.25H2O
4H2O [Cu(MePTA)3I][I]3H2O
Empirical formula Formula weight Crystal system Space group
C24H48N12O6P4BF4Cu 874.97 Cubic
C30H60.5N12O6.25P4S2B3F18Cu 1315.37 Monoclinic P21/n (14)
C21H45N9OP3I4Cu 1103.71 Orthorhombic Pnma (62)
12.074(3) 23.323(3) 19.509(3) 104.38(3) 5322(2) 4 1.641 0.725 2688 9619 8309 698 0.068 0.170
14.012(2) 17.791(3) 15.058(3)
a(Å) b(Å) c(Å) ß(°) Volume (A3) Z qcalcd (g cm3) l(Mo Ka) (mm1) F(0 0 0) No. reflections collected No. reflections used [I P 2r(I)] Number of parameters P P R¼ jFoj jmj= jFoj nP o1=2 P Rw ¼ ½wðFo2 Fc2 Þ2 = ½wðFo2 Þ2
Fd 3 (203) 20.319(3) – – – 8389(2) 8 1.386 0.741 3632 5786 654 41 0.064 0.148
3754(1) 4 1.953 4.029 2112 4125 3033 193 0.068 0.162
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
cisplatin. Cell lines were maintained in the logarithmic phase at 37 °C in a 5% carbon dioxide atmosphere, using the following culture media: (i) RPMI-1640 medium (Euroclone, Celbio, Milan, Italy) containing 10% fetal calf serum (Biochrom-Seromed GmbH&Co., Berlin, Germany) and supplemented with 25 mM HEPES buffer (HEPES = 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid), Lglutamine, and the antibiotics penicillin (50 units mL1) and streptomycin (50 g mL1) for MCF7, HCT-15, 2008, C13, A431, and A431-Pt cells; (ii) F-12 HAM’S (Sigma Chemical Co.) containing 10% fetal calf serum, L-glutamine, penicillin (50 units mL1), and streptomycin (50 g mL1) for HeLa cells; (iii) D-MEM (Dulbecco’s modified eagle’s medium; Euroclone) supplemented with 10% fetal calf serum (Euroclone), L-glutamine, penicillin (50 units mL1), and streptomycin (50 g mL1) for A549 and A375 cells. 2.6. Cytotoxicity assay The growth inhibitory effect toward tumor cell lines was evaluated by means of MTT (MTT = 3-(4,5-dimethylthiazol-2-yl)-2,5diphenyltetrazolium bromide) assay (tetrazolium salt reduction). Briefly, 3–8 103 cells/well, dependent upon the growth characteristics of the cell line, were seeded in 96-well microplates in growth medium (100 lL) and then incubated at 37° C in a 5% carbon dioxide atmosphere. After 24 h, the medium was removed and replaced with a fresh one containing the compounds to be studied at the appropriate concentration. Triplicate cultures were established for each treatment. After 48 h, each well was treated with 10 lL of a 5 mg mL1 MTT saline solution, and after 5 h of incubation, 100 lL of a sodium dodecylsulfate (SDS) solution in 0.01 M HCl was added. Following overnight incubation, the inhibition of cell growth induced by the tested complexes was detected by measuring the absorbance of each well at 570 nm using a Bio-Rad 680 microplate reader. Mean absorbance for each drug dose was expressed as a percentage of the control untreated well absorbance and plotted vs. drug concentration. IC50 values represent the drug concentrations that reduced the mean absorbance at 570 nm to 50% of those in the untreated control wells. 3. Results and discussion 3.1. Synthesis and characterization of Cu(I) complexes All the copper complexes were obtained starting from Cu(I) precursors according to Scheme 1. The syntheses were carried out in acetonitrile at room temperature and consist of a ligand-exchange reaction between the labile acetonitrile molecules of [Cu(CH3CN)4]+ and the appropriate phosphine ligands. This procedure differs from the reported synthesis of [Cu(PTA)4][NO3] and [Cu(PTAH)4][NO3]5, analogous to 1 and 2 except for the counterions, which were obtained from Cu(NO3)2 exploiting the reducing ability of the PTA phosphine [23]. The use of pre-reduced Cu(I) starting products allowed to
PTA
(ii) [Cu(PTAH) 4Cl4]+ [Cu(PTA) 4]+ 1 HCl 2
(i) (iii) [Cu(CH3CN)4]+
(mPTA)(CF 3SO3)
[Cu(mPTA) 4(CF3SO3)4]+ 3
(iv) (mPTA)I [Cu(mPTA) 3I][I]3 4 Scheme 1. Synthesis of compounds 1, 2, 3 and 4 starting from a Cu(I) precursor.
1647
perform reactions at room temperature with high yields (85– 90%), and contemporary to use a reduced amount of phosphine. Protonation of 1 was performed with a stoichiometric amount of hydrochloric acid and, as already observed elsewhere [23], the interconversion between 1 and 2 is a reversible, pH dependant process. In particular, from 31P NMR it was found that deprotonation of the PTAH derivative was complete at pH P 8. Calculation of the pKa value of 1 would be an important clue to determine the pH range where 1 and 2 are in equilibrium. Recently, determination of the pKa of some ruthenium-arene PTA complexes by means of 31P NMR has been reported and interestingly the obtained pKa values were lower with respect to that of the free ligand (3 vs. 5.6) [33]. Unfortunately, in our case the peculiarity of 31P spectrum of 1 (presence of a 1:1:1:1 quartet, vide infra), the slight difference in chemical shift between 1 and 2 and the presence of four PTA ligands to be protonated in each molecule made more difficult to carry out a pH titration and to determinate a proper pKa value. However, the NMR unequivocal determination of pure 1 and 2 was possible as the value of their autogeneous pH in unbuffered D2O solution is consistent with the protonated (2) or unprotonated species (1). Notwithstanding the straightforwardness of PTA N-alkylation, only Ru, Pd, Au, Rh and Ni derivatives containing mPTA have been reported so far [15]. Reaction of [Cu(CH3CN)4]+ salts with [mPTA]X (X = I, CF3SO3) were carried on in acetonitrile at room temperature, analogously to the synthesis of 1, using a 1:4 Cu:ligand ratio. As reported in Scheme 1, the nature of mPTA counter ion greatly affected the products of the reaction. In fact, by using (mPTA)I a water insoluble product was invariably obtained. Its IR spectrum did not show any bands attributable to the BF4 or PF6 group, nor the 31P{1H} NMR spectrum showed the septet typical of PF6. By means of elemental analysis the compound was identified as the mixed-ligand complex [Cu(mPTA)3I][I]3 4. Coordination of three phosphines to copper ion was confirmed by X-ray diffraction studies of crystals grown from a 50/50 methanol/acetonitrile solution and consistent of [Cu(mPTA)3I][I]3H2O, 4H2O, molecules. The low solubility of 4 both in water and in common organic solvents prevented its use in biological studies. Attempts to obtain a fully phosphine substituted derivative were made forcing the reaction conditions, i.e. by increasing the reaction temperature and the mPTA:Cu ratio, but the iodine-coordinated complex 4 was invariably obtained. The coordination ability of the iodine favoured its direct bonding to the metal thus forming the neutral, water insoluble compound 4. The choice of the counter ion both of the metal precursor and of the mPTA salt is crucial in determining the composition and the hydrophilicity of the products. Looking for a poorly coordinating anion able to ensure good hydro-solubility, we used the triflate salt mPTA(OSO2CF3) [34] instead of the iodine derivative. Using a synthetic procedure similar to that employed for 1, the ‘‘CuP4” species [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a and [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c were straightforwardly obtained in good yields (P85%). As expected both compounds possess a good water solubility (ca. 5 mg/ml) and are stable to Cu(I) disproportionation in the solid and in the solution states, so allowing in vitro citotoxicity studies. The use of BF4 or PF6 anions did not greatly affect the reactivity and the solubility of the final copper complexes; the choice of PF6 can be more convenient as it displays a septet in the 31P NMR which can be a diagnostic tool for characterization. The infrared spectra of 3a, 3c and 4 show typical bands of coordinated mPTA and are generally poorly diagnostic. IR spectra have been useful only in the characterization of 4 as no bands typical of BF4 or PF6 anions were detectable. Conversely, spectra of 3a and 3c showed bands due to the anion groups, in particular vibrations at 1271 and 1259 cm1 due to triflate, 846 cm1 due to PF6 and 1086 cm1 due to BF4 group.
1648
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
NMR studies of all compounds were carried out in D2O and, in the case of 4, also in DMSO. In 1H NMR spectra the effect of the counterion (BF4 or PF6) on the chemical shift of the pertinent complex is negligible. Moreover, no significant differences affect the 1H NMR spectrum of 4 in comparison with those of 3a and 3c. 1H NMR spectra of mPTA derivatives show a singlet attributable to the methyl group (d 2.72 ± 0.01) and signals due to four types of methylene protons. In all the spectra the PCH2NCH3 protons, which give rise to a doublet in the spectrum of the free ligand, show a broad singlet around 4.35 ppm. The 31P–1H coupling is instead maintained for the PCH2N signal which appears as a multiplet centred at ca. 3.9 ppm. The patterns of NCH2NCH3 and NCH2N are similar to those observed in free mPTA, each consisting of an ABMX and an ABMQX spin system centred at 4.4 and at 4.8 ppm, respectively. It is worth noting that protonation of PTA does not remove the magnetic equivalence of the three PCH2N and of the three NCH2N methylene groups both in the free ligand and in the copper complexes, whereas N-methylation determines a relevant change in 1 H NMR spectra. A slight downfield shift due to PF6, with respect to BF4, is evident in the 13C spectra of 3a and 3c, together with a remarkable increase of 1JC–P coupling costant value from 9 to 16 Hz. No 31P–13C coupling is evident in 13C spectrum of 4. The 31P{1H} spectrum of 1 displays a characteristic 1:1:1:1 quartet at room temperature [23] while 3a, 3c and 4 display a broad singlet respectively at 78, 79,6 and 81,5 ppm with no evidence of 31P–63Cu coupling notwithstanding the high symmetry of 3a and 3c. Lacking of such coupling has already been evidenced in the case of PTAH derivatives and has been ascribed to a fluxional behaviour in solution rather than to marked geometry distortions [23]. A fast ligand-exchange can similarly be supposed for 3; in fact, addition of an excess of free mPTA to 3a or 3c solutions determines a slight upfield shift of the 31P singlets signals without showing a distinct resonance of free mPTA.
such as [Cu(thp)3(P(CH2OH)2other tetrahedral CuP4 complexes 0 A)[10] and [Cu(PMe)30 ]X, whose mean (CH2O))] (average 2.281(2) Å Cu–P bond lengths are 2.270, 2.271 and 2.278 Å A for X = Cl, Br, I respectively [35]. A three-dimensional H-bond network involving the six water molecules for every [Cu(PTA)4]+ cation, and the nitrogen atoms of the PTA ligands characterizes the whole structure of 1 as also found in [(PTA)4Cu]Cl6H2O [24] and in [(PTA)4Au]Cl6H2O [36]. The asymmetric unit of 3b0.25H2O consists of the [Cu(mPTA)4]5+ cation (Fig. 2) which is balanced by two triflate groups and three BF4 moieties. The copper ion presents tetrahedral coordination geometry due to the coordination of four independent phosphorous atoms of the ligands with P–Cu–P angles ranging from 105.8(1) to 112.7(1) (Table 2). The Cu–P bond distances are equal 0 in the limit of their e.s.d.s (2.281(3) Å A average value) and comparable to those found in the [Cu(PTAH)4][NO3]56H2O complex 0 (2.298(1) Å A average value) [23]. The four methylated phosphane
3.2. X-ray crystallography The structure of 16H2O is highly symmetric, with the copper atom in special position that imposes a perfect tetrahedral geometry to the [Cu(PTA)4]+ cation (Fig. 1). 0 The Cu–P bond distances of 2.258(2) Å A in this BF4 derivative are significantly shorter with respect to those observed in the analogous complex with the nitrate as counter anion (Cu–Pav 0 2.284(2) Å A) that presents also a less symmetric complex cation [23] and is comparable to the0 highly symmetric analogous chloride derivative (Cu–P 2.2596(13) Å A) [24]. It is worth noting that the Cu– P distances of 1 are smaller with respect to the values found for
Fig. 2. Perspective view of the cation in compound 3b0.25H2O. Thermal ellipsoids are shown at 40% probability levels.
Table 2 Relevant bond distances and angles in the coordination spheres of [Cu(PTA)4][BF4]6H2O 16H2O, [Cu(mPTA)4][(CF3SO3)2(BF4)3]0.25H2O 3b0.25H2O and [Cu(mPTA)3I][I]3H2O 4H2O complexes.
0
Fig. 1. Perspective view of the cation in compound 16H2O. Thermal ellipsoids are shown at 40% probability levels.
Compound
16H2O
3b0.25H2O
4H2O
Cu–P(1) Cu–P(2) Cu–P(3) Cu–P(4) Cu–I(1) N–Ca (Me)N–Cb N–CMec P(1)–Cu–P(1)0 P(1)–Cu–P(2) P(1)–Cu–P(3) P(1)–Cu–P(4) P(2)–Cu–P(3) P(2)–Cu–P(4) P(3)–Cu–P(4) P(1)–Cu–I(1) P(2)–Cu–I(1)
2.258(2) – – –
2.284(1) 2.284(1) 2.280(1) 2.277(1)
1.468(6)
1.46(1) 1.53(1) 1.50(1) – 109.4(1) 110.6(1) 109.8(1) 112.7(1) 105.8(1) 108.3(1)
2.254(2) 2.247(3) – – 2.631(2) 1.48(1) 1.53(1) 1.46(1) 118.4(1) 112.7(1) – – – – – 105.0(1) 100.8(1)
109.5(2) – – –
at x + 1/4, y + 1/4, z for compound 1; x, y + 1/2, z for compound 4. a Average distance calculated for tertiary nitrogens. b Average distance calculated for quaternary nitrogens. c Average.
1649
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
3.3. Cytotoxicity assays
Fig. 3. Perspective view of the cation in compound 4H2O. Thermal ellipsoids are shown at 40% probability levels.
ligands are characterized by a lengthening of the C–N distances relative to the quaternarized nitrogens with respect to the tertiary ones (av. 1.53(1) vs. 1.46(1)). Such a lengthening is stressed compared to those complexes bearing protonated PTA such as [Cu(PTAH)4][23] and [M(PTAH)4]Cl4 (M = Ni, Pd, Pt) derivatives [37]. The average N–CMe distance is 1.50(1) compatible with a N–C single bond. Compound 4H2O is a three iodine salt with three mPTA ligands and one additional iodine linked to the metal in about a tetrahedral fashion (Fig. 3). A crystallographic mirror plane passing through iodine and P(2) atom characterizes the 0complex cation. The Cu–P bond distances of shorter with respect to those found in 2.254(2) and 2.247(3) Å A are 0 compound 3b (2.281(3) Å A average) due to the substitution of the sterically hindered mPTA ligand with iodine, so allowing a closer approach of the remaining three mPTA ligands towards a trigonal geometry as shown by the values of the P–Cu–P angles of 118.4(1)° and 112.7(1)°. An analogous behaviour has already been described in [(mPTA)3AuI][I]32H2O and [(EtPTA)3AuI][I]33H2O complexes [38,39]. The three iodide anions of compound 4H2O are hydrogen bonded to the water molecules present in the crystal lattice. We were unable to locate the H atoms of the water molecule, but the values of the distances of O I(2)0 3.56(1), O I(3)00 3.39(1) (0 at A (00 at x + 1/ x + 1/2 + 1, y + 1, +z + 1/2); and O I(3)00 3.39(1) Å 2 + 1, +y 1/2, + z + 1/2) are consistent with the presence of O– H I hydrogen bonding.
Compounds 1, 2, 3a and 3c and the corresponding uncoordinated ligands and counterions were examined for their cytotoxic properties against a panel of human tumor cell lines containing examples of lung (A549), cervix (HeLa), colon (HCT-15) and breast (MCF-7) cancer and melanoma (A375). For comparison purposes, the cytotoxicity of cisplatin was evaluated under the same experimental conditions. IC50 values, calculated from the dose-survival curves obtained after 48 h of drug treatment by MTT test, are shown in Table 3. Uncoordinated ligands and the various counterions proved to be ineffective in all tumor cell lines. On the contrary, all the tested complexes showed a growth inhibitory potency in the micromolar range towards the different types of tumor cells. They displayed an in vitro antitumor activity similar to that shown by cisplatin, the mean IC50 values (lM) being 15.0, 17.6, 20.0, 18.8 and 21.8 for 1, 2, 3a, 3c and cisplatin, respectively. In particular, on A549 lung adenocarcinoma cells, that are poor chemosensitive to cisplatin [40], the cytotoxicities of copper complexes exceeded that of the reference drug by a factor ranging from about 2 to 3. The cytotoxic effect elicited by these copper compounds seems to be uninfluenced by changes in the steric demand and charge of the ligand (PTA vs PTAH and mPTA). In the case of 2, the similarity of cytotoxic efficiency with 1 could be explained by deprotonation of 2 to give 1 at physiological pH (pH ffi 7.2) determined by the use of a buffered milieu such as the growth medium. Moreover, as many cancer cells have a lower pH compared to normal cells, an equilibrium between 1 and 2 may take place into the cell as well, determining a similar cytotoxic effect for 1 or 2. Also the insertion of a methyl group in the PTA framework does not induce significant changes in the cytotoxic profiles of the corresponding 3a and 3c, which were found roughly as effective as the parent compound 1. Interestingly in the case of several ruthenium compounds a different activity and specificity was observed by using PTA or mPTA. [33,41]. Also the specificity against a particular cancer type was unaffected by the ligand sphere as the higher activity was invariably shown by all derivatives against HeLa cervix carcinoma and A549 lung adenocarcinoma cells, whereas the lower against MCF-7 breast carcinoma cells. Interestingly, on HeLa and A549 cell lines the activity shown by the original compound 1 was comparable to that of [Cu(thp)4][PF6] [11]. It is worth mentioning that experiments performed with [64Cu(thp)4]+ and [64Cu(PTA)4]+ and EMT-6 mouse mammary tumor cells have assessed a slightly higher serum stability and a significative higher cellular uptake of the PTA derivative probably related to their different lipophilicity (log P 0.01 ± 0.01 for [64Cu(PTA)4]+ and 2.26 ± 0.04 for [64Cu(thp)4]+) [42].
Table 3 In vitro antitumor activity. Compound
IC50 (lM) ± S.D. MCF-7
HeLa
A549
HCT-15
A375
[Cu(PTA)4][BF4] 1 [Cu(PTAH)4][Cl4(BF4)] 2 [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c Cisplatin (mPTA)(CF3SO3) PTA PTAHCl NaBF4 KPF6
23.2 ± 1.2 23.6 ± 0.6 25.8 ± 1.2 23.8 ± 2.5 23.2 ± 1.3 nd nd nd nd nd
7.58 ± 0.19 12.2 ± 2.0 15.8 ± 1.1 13.6 ± 1.3 11.2 ± 1.5 nd nd nd nd nd
8.47 ± 0.10 11.4 ± 2.6 14.6 ± 1.9 13.9 ± 1.7 29.3 ± 1.7 nd nd nd nd nd
19.2 ± 2.1 21.9 ± 1.3 23.4 ± 1.4 23.2 ± 0.8 25.3 ± 1.2 nd nd nd nd nd
16.4 ± 1.2 18.8 ± 2.0 19.8 ± 1.8 19.4 ± 1.6 20.3 ± 2.2 nd nd nd nd nd
S.D. = standard deviation. nd = Not detectable. IC50 values were calculated by probit analysis (P < 0.05, v2 test). (5–8) 104 mL1 cells were treated for 48 h with increasing concentrations of tested compounds. Cytotoxicity was assessed by MTT test.
1650
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651
Table 4 Cross-resistance profiles. Compound
Human ovarian adenocarcinoma cells [Cu(PTA)4][BF4] 1 [Cu(PTAH)4][Cl4(BF4)] 2 [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c Cisplatin
IC50 (lM) ± SD 2008
C13
R.F.
14.24 ± 2.18 14.89 ± 1.67 9.43 ± 1.01 10.24 ± 0.87 12.69 ± 1.23
11.45 ± 1.07 13.28 ± 2.31 10.08 ± 1.32 9.41 ± 0.65 89.21 ± 2.24
1.24 1.12 0.93 1.09 7.02
Human cervix squamous carcinoma cells A431
A431-Pt
R.F.
[Cu(PTA)4][BF4] 1 [Cu(PTAH)4][Cl4(BF4)] 2 [Cu(mPTA)4][(CF3SO3)4(BF4)] 3a [Cu(mPTA)4][(CF3SO3)4(PF6)] 3c Cisplatin
10.52 ± 0.55 17.01 ± 2.13 9.23 ± 0.98 10.75 ± 0.87 57.76 ± 1.45
1.03 1.00 1.01 0.96 2.61
10.14 ± 0.94 16.91 ± 1.98 9.10 ± 0.21 11.15 ± 0.70 22.06 ± 1.27
SD = standard deviation. IC50 values were calculated by probit analysis (P < 0.05, v2 test). (3–5) 104 mL1 cells were treated for 48 h with increasing concentrations of tested compounds. Cytotoxicity was assessed by MTT test. Resistant factor (RF) is defined as IC50 resistant/parent line.
The encouraging results obtained against the in-house panel of cell lines prompted us to test the cytotoxic activity of our Cu(I) complexes onto two additional cell line pairs, which were selected for their resistance to cisplatin: 2008/C13 ovarian cancer cells and A431/A431-Pt cervix carcinoma cells. Although cisplatin resistance is multifactorial, the main molecular mechanisms involved in resistance in these sublines have almost been defined. In C13 cells, resistance is correlated to reduced cellular drug uptake, high cellular glutathione levels, and enhanced repair of DNA damage [43]. In human squamous cervix carcinoma A431-Pt cells, resistance is due to defect in drug uptake and to decreased levels of proteins involved in DNA mismatch repair (MSH2), causing an increased tolerance to cisplatin-induced DNA damage [44]. Cytotoxicity of tested compounds in sensitive and resistant cells was assessed after a 48 h drug exposure by MTT test. Cross-resistance profiles were evaluated by means of the resistance factor (RF), which is defined as the ratio between IC50 values calculated for the resistant cells and those arising from the sensitive ones (Table 4). All the tested copper(I) complexes induced a similar pattern of response across the parental and the resistant sublines indicative of different cross-resistance profiles than that of cisplatin. Against 2008/C13 cell pair, the resistance factor of copper derivatives was roughly 7-fold lower than that of cisplatin, whereas against A431/A431-Pt cell line pair it was about 2.5-fold lower. The circumvention of cisplatin resistance, recently observed for other phosphine-containing Cu(I) derivatives, supports the hypothesis of a different mechanism of action of copper(I) complexes than that of cisplatin.
4. Conclusions A series of water-soluble Cu(I) compounds has been synthesized and their cytotoxic activity checked against an in-house panel of cell lines. The data reported in the present paper show that all the studied complexes containing PTA derivatives are effective against the panel of human tumor cell lines tested with IC50 values ranging from 8 to 25 lM, mostly comparable with those elicited by cisplatin. Cytotoxicity was found not to be significantly affected by both the nature of the PTA ligand and the anionic counterion. Interestingly, the original compound [Cu(PTA)4](BF4) appears roughly 1.5-fold most powerful than cisplatin and, towards human cervix and non-small lung carcinoma cells, [Cu(PTA)4][BF4]
shows an antiproliferative activity very similar to that already described for [Cu(thp)4][PF6]. Moreover, such class of ‘‘CuP4” compounds can be equally cytotoxic in both the cisplatin-sensitive and -resistant sublines of human ovarian and cervix squamous carcinoma cells ruling out the occurrence of cross resistance phenomena and suggesting a mechanism of action different from the covalent binding to DNA shown by cisplatin, as already observed for other phosphine–copper(I) agents. The reported results confirm the high potentiality of water-soluble phosphine copper(I) complexes as a new class of non-platinum anticancer drugs. Actually, expanding the molecular targets for metallo-drugs will open new opportunities to the strategic development of more effective and less-toxic anticancer drugs. Acknowledgements This work was financially supported by University of Padova (Progetto di Ateneo CPDA065113/06), and by Ministero dell’Istruzione dell’Università e della Ricerca (PRIN 2007). We are grateful to CIRCMSB (Consorzio Interuniversitario di Ricerca in Chimica dei Metalli nei Sistemi Biologici). The authors thank A. Moresco for elemental analyses. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jinorgbio.2009.09.005. References [1] B. Rosemberg, L. Van Camp, T. Krigas, Nature 205 (1965) 698–699. [2] S.J. Berners-Price, P.J. Sadler, in: Structure and Bonding, Bioinorganic Chemistry, vol. 70, Springer, Berlin/Heidelberg, 1988, pp. 27–102. [3] C.F. Shaw III, Chem. Rev. 99 (1999) 2589–2600. [4] T.M. Simon, D.H. Kunishima, G.J. Vibert, A. Lorber, Cancer Res. 41 (1981) 94–97. [5] S.J. Berners-Price, C.K. Mirabelli, R.K. Johnson, M.R. Mattern, F.L. McCabe, L.F. Faucette, C.M. Sung, S.M. Mong, P.J. Sadler, S.T. Crooke, Cancer Res. 46 (1986) 5486–5493. [6] T. Wang, Z. Guo, Curr. Med. Chem. 13 (2006) 525–537. [7] C. Marzano, M. Pellei, F. Tisato, C. Santini, Anti-Cancer Agents Med. Chem. 9 (2009) 185–211. [8] S.J. Berners-Price, M.E. Sant, R.I. Christopherson, P.W. Kuchel, Magn. Reson. Med. 18 (1991) 142–158. [9] G.D. Hoke, R.A. Macia, P.C. Meunier, P.J. Bugelski, C.K. Mirabelli, G.F. Rush, W.D. Matthews, Toxicol. Appl. Pharmacol. 100 (1989) 293–306. [10] D. Saravana Bharathi, M.A. Sridhar, J. Shashidhara Prasad, A.G. Samuelson, Inorg. Chem. Commun. 4 (2001) 490–492. [11] C. Marzano, V. Gandin, M. Pellei, D. Colavito, G. Papini, G. Gioia Lobbia, E. Del Giudice, M. Porchia, F. Tisato, C. Santini, J. Med. Chem. 51 (2008) 798–808. [12] C. Marzano, M. Pellei, D. Colavito, S. Alidori, G. Gioia Lobbia, V. Gandin, F. Tisato, C. Santini, J. Med. Chem. 49 (2006) 7317–7324. [13] C. Marzano, M. Pellei, S. Alidori, A. Brossa, G. Gioia Lobbia, F. Tisato, C. Santini, J. Inorg. Biochem. 100 (2006) 299–304. [14] C. Marzano, V. Gandin, A. Folda, G. Scutari, A. Bindoli, M.P. Rigobello, Free Radic. Biol. Med. 42 (2007) 872–881. [15] A.D. Phillips, L. Gonsalvi, A. Romerosa, F. Vizza, M. Peruzzini, Coord. Chem. Rev. 248 (2004) 955–993. [16] D.J. Daigle, Inorg. Synth. 32 (1998) 40–45. [17] A. Mena-Cruz, P. Lorenzo-Luis, A. Romerosa, M. Saoud, M. Serrano-Ruiz, Inorg. Chem. 46 (2007) 6120–6126. [18] M. Erlandsson, V.R. Landaeta, L. Gonsalvi, M. Peruzzini, A.D. Phillips, P.J. Dyson, G. Laurenczy, Eur. J. Inorg. Chem. (2008) 620–627. [19] C.A. Mebi, B.J. Frost, Organometallics 24 (2005) 2339–2346. [20] D.A. Krogstad, S.B. Owens, J.A. Halfen, V.G. Young Jr., Inorg. Chem. Commun. 8 (2005) 65–69. [21] W. Lewasin, in: F.R. Hartley (Ed.), The Chemistry of Organophosphorous Compounds, vol. 1, Wiley-Interscience, New York, 1990 (Chapter 15). [22] F. Tisato, F. Refosco, G. Bandoli, G. Pilloni, B. Corain, Inorg. Chem. 40 (2001) 1394–1396. [23] A.M. Kirillov, P. Smolenski, M.F.C. Guedes da Silva, A.J.L. Pombeiro, Eur. J. Inorg. Chem. (2007) 2686–2692. [24] A.M. Kirillov, P. Smolenski, M.F.C. Guedes da Silva, M.N. Kopylovich, A.J.L. Pombeiro, Acta Crystallogr. Section E – Struct Rep Online 64 (2008) M603– U32. [25] R. Wanke, P. Smolenski, M.F.C. Guedes da Silva, L.M.D.R.S. Martins, A.J.L. Pombeiro, Inorg. Chem. 47 (2008) 10158–10168.
M. Porchia et al. / Journal of Inorganic Biochemistry 103 (2009) 1644–1651 [26] M. Pellei, S. Alidori, M. Camalli, G. Campi, G. Gioia Lobbia, M. Mancini, G. Papini, R. Spagna, C. Santini, Inorg. Chim. Acta 361 (2008) 1456–1462. [27] M. Pellei, G. Gioia Lobbia, C. Santini, R. Spagna, M. Camalli, D. Fedeli, G. Falcioni, Dalton Trans. (2004) 2822–2828. [28] G.J. Kubas, Inorg. Synth. 26 (1990) 68–70. [29] A.T.C. North, D.C. Philips, F.S. Mathews, Acta Crystallogr. A24 (1968) 351–359. [30] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo, A. Guagliardi, A.G.G. Moliterni, G. Polidori, R. Spagna, SIR-97, J. Appl. Crystallogr. 32 (1999) 115–119. [31] G.M. Sheldrick, SHELXL-97, Program for the Refinement of Crystal Structures, University of Göttingen, Germany, 1997. [32] L.J.J. Farrugia, Appl. Crystallogr. 32 (1999) 837–838. [33] C. Scolaro, A. Bergamo, L. Brescacin, R. Delfino, M. Cocchietto, G. Laurenczy, T.J. Geldbach, G. Sava, P.J. Dyson, J. Med. Chem. 48 (2005) 4161–4171. [34] A. Romerosa, T. Campos-Malpartida, C. Lidrissi, M. Saoud, M. Serrano-Ruiz, M. Peruzzini, J.A. Garrido-Cárdenas, F. García-Maroto, Inorg. Chem. 45 (2006) 1289–1298.
1651
[35] G.A. Bowmaker, P.C. Healy, L.M. Engelhardt, J.D. Kildea, B.W. Skelton, A.H. White, Aust. J. Chem. 43 (1990) 1697–1705. [36] J.M. Forwar, Z. Assefa, R.J. Staples, J.P. Fackler Jr., Inorg. Chem. 35 (1996) 16–22. [37] D.J. Darensbourg, T.J. Decuir, N.W. Stafford, J.B. Robertson, J.D. Draper, J.H. Reibenspies, A. Kathó, F. Joó, Inorg. Chem. 36 (1997) 4218–4226. [38] J.M. Forwar, J.P. Fackler Jr., R.J. Staples, Organometallics 14 (1995) 4194–4198. [39] E. Suresh, M.M. Bhadbhade, Acta Crystallogr. C53 (1997) 195–197. [40] P. Zhang, W.Y. Gao, S. Turner, B.S. Ducatman, Mol. Cancer 2 (2003) 1–9. [41] A. Garcia-Fernandez, J. Diez, A. Manteca, J. Sanchez, M.P. Gamasa, E. Lastra, Polyhedron 27 (2008) 1214–1228. [42] S. Alidori, G. Gioia Lobbia, G. Papini, M. Pellei, M. Porchia, F. Refosco, F. Tisato, J.S. Lewis, C. Santini, J. Biol. Inorg. Chem. 13 (2008) 307–315. [43] G. Marverti, P.A. Andrews, G. Piccinini, S. Ghiaroni, D. Barbieri, M.S. Moruzzi, Eur. J. Cancer 33 (1997) 669–675. [44] C. Lanzi, P. Perego, R. Supino, S. Romanelli, T. Pensa, N. Carenini, I. Viano, D. Colangelo, R. Leone, P. Apostoli, G. Cassinelli, R.A. Gambetta, F. Zunino, Biochem. Pharmacol. 55 (1998) 1247–1254.