← Back

The contrasting chemical reactivity of potent isoelectronic iminopyridine and azopyridine osmium(ii) arene anticancer complexes

Original citation: Fu, Y. (Ying), Romero, Maria, Habtemariam, Abraha, Snowden, Michael E., Song, Lijiang, Clarkson, Guy J., Qamar, Bushra, Pizarro, Ana M., Unwin, Patrick R. and Sadler, Peter J.. (2012) The contrasting chemical reactivity of potent isoelectronic iminopyridine and azopyridine osmium(ii) arene anticancer complexes. Chemical Science, Vol.3 (No.8). pp. 2485-2494. Permanent WRAP url: http://wrap.warwick.ac.uk/53174 Copyright and reuse: The Warwick Research Archive Portal (WRAP) makes the work of researchers of the University of Warwick available open access under the following conditions. Copyright © and all moral rights to the version of the paper presented here belong to the individual author(s) and/or other copyright owners. To the extent reasonable and practicable the material made available in WRAP has been checked for eligibility before being made available. Copies of full items can be used for personal research or study, educational, or not-forprofit purposes without prior permission or charge. Provided that the authors, title and full bibliographic details are credited, a hyperlink and/or URL is given for the original metadata page and the content is not changed in any way. Publisher’s statement: http://dx.doi.org/10.1039/C2SC20220D A note on versions: The version presented here may differ from the published version or, version of record, if you wish to cite this item you are advised to consult the publisher’s version. Please see the ‘permanent WRAP url’ above for details on accessing the published version and note that access may require a subscription. For more information, please contact the WRAP Team at: wrap@warwick.ac.uk http://go.warwick.ac.uk/lib-publications CREATED USING THE RSC ARTICLE TEMPLATE (VER. 3.1) - SEE WWW.RSC.ORG/ELECTRONICFILES FOR DETAILS ARTICLE TYPE www.rsc.org/xxxxxx | XXXXXXXX The contrasting chemical reactivity of potent isoelectronic iminopyridine and azopyridine osmium(II) arene anticancer complexes Ying Fu, María J. Romero, Abraha Habtemariam, Michael E. Snowden, Lijiang Song, Guy J. Clarkson, Bushra Qamar, Ana M. Pizarro, Patrick R. Unwin and Peter J. Sadler* 5 Received (in XXX, XXX) Xth XXXXXXXXX 200X, Accepted Xth XXXXXXXXX 200X First published on the web Xth XXXXXXXXX 200X DOI: 10.1039/b000000x A wide variety of steric and electronic features can be incorporated into transition metal coordination complexes, offering the prospect of rationally-designed therapeutic agents with novel 10 mechanisms of action. Here we compare the chemical reactivity and anticancer activity of organometallic OsII complexes [Os(η6-arene)(XY)Z]PF6 where arene = p-cymene or biphenyl, XY = N,N΄-chelated phenyliminopyridine or phenylazopyridine derivatives, and Z = Cl or I. The X-ray crystal structure of [Os(η6-p-cym)(Impy-OH)I]PF6⋅0.5CH2Cl2⋅H2O (Impy-OH = 4-[(2pyridinylmethylene)amino]-phenol) is reported. Like the azopyridine complexes we reported 15 recently (Dalton Trans. 2011, 40, 10553-10562), some iminopyridine complexes are also potently active towards cancer cells (nanomolar IC50 values). However we show that, unlike the azopyridine complexes, the iminopyridine complexes can undergo aquation, bind to the nucleobase guanine, and oxidize coenzyme nicotine adenine dinucleotide (NADH). We report the first detection of an Oshydride adduct in aqueous solution by 1H NMR (-4.2 ppm). Active iminopyridine complexes 20 induced a dramatic increase in the levels of reactive oxygen species (ROS) in A549 lung cancer cells. The anticancer activity may therefore involve interference in the redox signalling pathways in cancer cells by a novel mechanism. Introduction 25 Transition metal coordination complexes offer a variety of electronic and structural features for the design of therapeutic agents, including the choice of the metal and its oxidation state, the number and types of coordinated ligands, and their geometries.1-9 The heavier, third-row (5d) 30 coordination transition elements in particular tend to exhibit slow ligand exchange kinetics and so might reach biological target sites with at least some of the initial ligands still bound. Organometallic arene and cyclopentadienyl complexes are 10-13 35 attractive since their potential amphiphilicity can be beneficial for drug transport and target recognition. Subtle changes in structure can lead to dramatic changes in chemical reactivity and biological activity. For example, certain inert N,N΄-chelated half-sandwich organometallic iodido Os II arene 40 azopyridine complexes exhibit potent in vitro and in vivo anticancer activity.14, 15 Their mechanism of action is different from that of N,O-chelated Os II arene picolinate complexes which can hydrolyze and then bind to DNA bases, 16 with DNA being a likely target, as it is for cisplatin.17,18 In II 45 contrast, with the strong π-acceptor azopyridine present, Ru arene complexes with iodide as the monodentate ligand are relatively inert towards substitution reactions and appear to be activated by redox reactions, for example with glutathione.19 Other than glutathione, the couple NADH and oxidized 50 nicotinamide adenine dinucleotide (NAD+) is also an important part of the cellular redox balance system.20 Under This journal is © The Royal Society of Chemistry [year] physiological conditions, the ratio of [NADH]/[NAD+] in cells is low.21 The cellular concentration of free NADH is in the nanomolar range,21 which makes it a potential and sensitive 55 target for anticancer drugs. Early work by Steckhan22 and Fish23 demonstrated an efficient process for the regeneration of NADH from NAD + using formate as an hydride source and rhodium catalysts. Similarly we have shown that ruthenium arene anticancer + 60 complexes can reduce NAD regioselectively to NADH through transfer hydrogenation.24 Recent studies have shown that both organometallic Ru II arene complexes and Ir III Cp* complexes can oxidize NADH to NAD+.25-27 However, there are no reports of osmium complexes as transfer hydrogenation 65 catalysts in aqueous solution. There are some intriguing differences between the chemical properties of organometallic Ru II and OsII arene complexes even though their three-dimensional structures can be almost identical.28 For example, complexes of OsII (a low-spin 5d6 16, 29 70 metal ion) are often more inert, and aqua adducts of OsII 16 arenes are more acidic. Several organometallic osmium anticancer complexes have been designed based on their ruthenium analogues.30-32 Studies of both RuII and OsII azopyridine anticancer complexes have been reported. 75 However, no anticancer studies of ruthenium arene complexes with iminopyridine ligands have been reported. Osmium(II) arene phenylazopyridine complexes exhibit anticancer activity in the nanomolar range in vitro.14 Moreover the complex ([Os(η6-p-cym)(Azpy-NMe2)I]PF6 (FY026) is active against Journal Name, [year], [vol], 00–00 | 1 coloreactal cancer in vivo and has low toxicity.15 The exploration of bioisosterism provides a useful approach to the study of the anticancer activity33 of osmium arene complexes. Here we compare the chemical reactivity and biological II 5 activity of Os arene azopyridine (N=N) complexes with isoelectronic iminopyridine (HC=N) complexes. Anticancer activity has been studied in A2780 ovarian cancer cells as well as in the NCI 60-cell line panel. To investigate possible redox mechanisms, the production of ROS in lung cancer cells studied as well as the influence of combined 10 was administration with the glutathione synthase inhibitor Lbuthionine-sulfoximine (L-BSO).34 Attempts are made to correlate biological activity with chemical reactivity through studies of aquation, binding to the nucleobase guanine, and 15 redox reactions with glutathione and NADH. The data suggest that these osmium arene complexes possess novel mechanisms of action that can be finely tuned through the choice of the ligands. 20 Results PF6 Arene R2 X Os N 25 R1 N CH 30 p -cym bip 70 Complex 1 2 3 4 5 Arene bip p-cym bip p-cym bip R1 H H H H H R2 H H OH OH NMe2 X I I I I I Chelating ligand Impy Impy Impy-OH Impy-OH Impy-NMe2 6 p-cym H NMe2 I Impy-NMe2 7 bip OMe NMe2 I OMe-Impy-NMe2 8 9 10 11 12 13 p-cym bip p-cym bip p-cym bip OMe H H H H H NMe2 H H OH OH NMe2 I Cl Cl Cl Cl Cl OMe-Impy-NMe2 Impy Impy Impy-OH Impy-OH Impy-NMe2 14 p-cym H NMe2 Cl Impy-NMe2 15 16 bip p-cym OMe OMe NMe2 NMe2 Cl Cl OMe-Impy-NMe2 OMe-Impy-NMe2 Chart 1 Osmium (II) arene iminopyridine complexes synthesized and studied in this work. 35 2 | Journal Name, [year], [vol], 00–00 Synthesis and characterization Sixteen osmium(II) arene iminopyridine complexes (Chart 1) of general formula [Os(η6-arene)(R1-Impy-R2)X]PF6 [arene = 40 biphenyl (bip) or p-cymene (p-cym); X = Cl or I] containing different chelating iminopyridine ligands [Impy (R1= R2= H), Impy-OH (R1= H, R2= OH), Impy-NMe2 (R1= H, R2= NMe2) and OMe-Impy-NMe2 (R1= OMe, R2= NMe2)] were synthesized. The general method involved stirring an osmium dimer and the chelating iminopyridine ligand (synthesized 45 appropriate following a literature method)35 in a methanol solution, similar to that reported previously for the synthesis of the azopyridine analogues.14 In general, the iminopyridine complexes were obtained in good yields and were well characterized by 1H-NMR 50 spectroscopy, mass spectrometry and CHN elemental analysis (for details see ESI). Half-sandwich organometallic arene complexes containing an unsymmetrical chelating ligand and a monodentate ligand, are chiral. 1H NMR spectra recorded before and after adding the 55 chiral anionic shift-reagent ∆–trisphat to a solution of complex 14 in CDCl3, 298 K, showed a splitting of the peaks (Figure S1), indicating the presence of two enantiomers in ca. 1:1 ratio, consistent with a previous report on osmium/ruthenium arene picolinamide anticancer complexes.17 No attempt was made to 60 separate the enantiomers, although the chirality at the metal centre may affect anticancer activity if the biological target is chiral. Recrystallization of [Os(η6-p-cym)(Impy-OH)I]PF6 (4) from DCM/methanol afforded purple single crystals corresponding to 65 4⋅0.5CH2Cl2⋅H2O suitable for X-ray diffraction (Figure 1). Crystallographic data are listed in Table S1 and selected bond lengths and angles in Table S2. The asymmetric unit contains two crystallographically independent cations [Os(η6-p-cym)(ImpyOH)I]+ comprising of either enantiomer. 75 80 85 Fig. 1 X-ray structure of the cation [Os(η6-p-cym)(ImpyOH)I]+ in crystals of 4⋅0.5CH2Cl2⋅H2O showing the atom numbering scheme. The thermal ellipsoids are drawn at 50% probability. The hydrogen atoms, counterion (PF6 ), and solvent molecules have been omitted for clarity. 90 In addition, the two enantiomers are partially solvated by one dichloromethane and two water molecules in the asymmetric unit. The osmium complex shows a pseudo-octahedral “piano-stool” structure with the p-cymene ligand π-bonded to the metal ion and This journal is © The Royal Society of Chemistry [year] the Impy-OH chelating ligand coordinated through the pyridine and imine nitrogen atoms. The imine bond adopts an E conformation, minimising the steric hindrance arising from the bulky phenol group. The coordination sphere of the OsII ion is 5 completed by a terminal monodentate iodide. The Os−N (2.077(4)-2.087(4) Å), Os−I (2.7091(4) Å, 2.7247(4) Å) and Os−arene centroid bond lengths (1.6845(2) Å, 1.6962(2) Å) in this complex are similar to those of the osmium analogues previously reported.16 The crystal structure of 4⋅0.5CH2Cl2⋅H2O 6 10 shows differences from the reported crystal structure of [Os(η 14 bip)(Azpy-O)I]⋅0.5H2O. In complex 4⋅0.5CH2Cl2⋅H2O there is no deprotonation of the hydroxyl group on the chelating ligand which allows both molecules to establish an hydrogen bond between the phenolic OH group of the iminopyridine ligand and a 15 water molecule [O115-H11H⋅⋅⋅O400 1.84 Å, O215-H21H⋅⋅⋅O300 1.89 Å] as shown in Figure S2. Aquation, pKa of aqua complexes and binding to 9ethylguanine 20 The possible activation of these complexes by aquation and their interaction with 9-ethylguanine (9-EtG), a model DNA nucleobase, were studied by 1H NMR spectroscopy. It was found that the osmium iminopyridine complexes [Os(η6-p-cym)(Impy6 25 NMe2)I]PF6 (6) and [Os(η -p-cym)(Impy-NMe2)Cl]PF6 (14), can undergo aquation. After incubation of the samples at 310 K for 24 h in D2O, the extent of aquation reached 50% and 99%, respectively. The formation of the aqua product (14A) was confirmed by the presence of new signals in the 1H NMR spectra 30 shifted with respect to the signals observed for 14 (Figure S3). These signals were confirmed as corresponding to the aqua adduct after repeating the experiment in the presence of 500 mM NaCl where only one set of signals for the chlorido complex was observed. 35 40 45 50 60 for the coordinated water and 2.3 (pKa 2.6) for the dimethylamine substituent; (B) Reaction of the hydroxido complex 14A (1 mM) with 9-ethylguanine (2 mM, 2 mol equiv) in 10%D2O/90%H2O phosphate buffer, pH* 7.4, followed by 1H NMR. The spectra were recorded 10 min after preparing the mixture and after incubation for 18 h and 38 h at 310 K. 65 The pKa value of the aqua adduct (14A) was determined by H NMR spectroscopy (Figure 2A). The chemical shifts of the 1H NMR signals gradually shifted upfield with increase in pH* from c.a. pH* 1.5 to 13. A plot of the chemical shift of the pyridine * 70 proton Ha against pH was fitted to the Henderson-Hasselbalch equation, giving a pKa value of 5.2.36 The low pKa value obtained for the aqua adduct suggests that the hydroxido adduct will be the predominant species under physiological conditions. The pKa value of 2.6 for the dimethylamine substituent on the 75 iminopyridine chelating ligand Impy-NMe2 is the first reported pKa of an NMe2 substituent in this type of ligand bound to metal (Figure 2A). To study the potential for DNA binding, the reaction of 14A with 9-EtG in 0.1 M phosphate buffer (10%D2O/90%H2O, pH* 1 80 7.4) was monitored over various time intervals by H NMR spectroscopy after incubation of the solution at 310 K. 9-EtG binding was confirmed by the appearance of a new set of peaks assignable to a 14-9-EtG adduct shifted downfield in comparison to the peaks of 14A, while the signals corresponding to 85 coordinated 9-EtG shifted upfield with respect to the free nucleobase. The extent of binding reached only 14% after 18 h and increased only slightly to 19% after another 20 h of incubation at 310 K (Figure 2B). 1 90 Anticancer Activity The activity of all sixteen osmium arene iminopyridine complexes towards the A2780 human ovarian cancer cell line was determined and of selected complexes towards the A549 95 human lung cancer cell line (Table 1). They exhibit a broad range of anticancer activity with IC50 values (the concentration that inhibits cell growth by 50%) towards A2780 cells ranging from 0.14 µM for [Os(η6-bip)(Impy-NMe2)I]PF6 (5) to 35.5 µM for [Os(η6-p-cym)(OMe-Impy-NMe2)I]PF6 (8) (Table 1A). When the 100 ortho -OMe substituent on the pyridine ring in 8 was replaced by H in [Os(η6-p-cym)(Impy-NMe2)I]PF6 (6), the anticancer activity increased 44-fold (IC50 decreased from 35.5 µM to 0.80 µM). This trend was also observed when biphenyl is the arene and the monodentate ligand is chloride where a 35-fold increase in 105 anticancer activity was observed. Anticancer efficacy was also observed towards A549 cells. A significant enhancement in activity was obtained by combination treatment with Lbuthionine-sulfoximine (L-BSO), a specific inhibitor of γglutamylcysteine synthetase (Table 1B). L-BSO depletes 110 intracellular glutathione, which plays an important role in the maintenance of the redox balance in cancer cells. 55 Fig. 2 (A) Dependence of the 1H NMR chemical shift of the pyridine ortho proton Ha on pH* in complex 14A. The line represents a computer best fit giving pKa* values of 5.1 (pKa 5.2) This journal is © The Royal Society of Chemistry [year] 115 Journal Name, [year], [vol], 00–00 | 3 Table 1 Anticancer activity in vitro towards (A) A2780 human ovarian cancer cells, (B) A549 human lung cancer cells and combination treatment with 50 µM L-BSO, (C) NCI 60-cell line screening (data for 58 cell lines) of complexes 6 and 14. 5 (A) A2780 Ovarian Cancer Cells Complex IC50(µM) (1) [Os(η -bip)(Impy)I]PF6 (2) [Os(η6-p-cym)(Impy)I]PF6 (3) [Os(η6-bip)(Impy-OH)I]PF6 (4) [Os(η6-p-cym)(Impy-OH)I]PF6 (5) [Os(η6-bip)(Impy-NMe2)I]PF6 (6) [Os(η6-p-cym)(Impy-NMe2)I]PF6 (7) [Os(η6-bip)(OMe-Impy-NMe2)I]PF6 (8) [Os(η6-p-cym)(Ome-Impy-NMe2)I]PF6 (9) [Os(η6-bip)(Impy)Cl]PF6 (10) [Os(η6-p-cym)(Impy)Cl]PF6 (11) [Os(η6-bip)(Impy-OH)Cl]PF6 (12) [Os(η6-p-cym)(Impy-OH)Cl]PF6 (13) [Os(η6-bip)(Impy-NMe2)Cl]PF6 (14) [Os(η6-p-cym)(Impy-NMe2)Cl]PF6 (15) [Os(η6-bip)(OMe-Impy-NMe2)Cl]PF6 (16) [Os(η6-p-cym)(OMe-Impy-NMe2)Cl]PF6 Cisplatin 18.6 (±0.9) 29.4 (±5.3) 5.4 (±1.1) 31.8 (±3.8) 0.14 (±0.01) 0.80 (±0.05) 9.14 (±0.93) 35.5 (±3.2 ) 4.6 (±0.4) 26.2 (±2.8) 2.4 (±1.1) 5.5 (±0.6) 0.44 (±0.01) 1.50 (±0.047) 9.50 (±0.22) 32.9 (±1.2) 1.8 (±0.1) 6 Cancer Institute (DTP of NCI), which includes nine tumour-type subpanels. The mean values of IC50, TGI (the concentration which inhibits cell growth by 100 %) and LC50 (the concentration that kills original cells by 50 %) are listed in Table S3. Both complexes showed anticancer activity within the same range as 25 cisplatin (Table 1C). Complexes 6 and 14 showed a broad spectrum of activity, with IC50 values ranging from 0.5 µM to more than 100 µM, and a particular selectivity for melanoma and breast cancer cells. The high activity towards MDA-MB-468 breast cancer cells is particularly notable, with some IC50 values 30 in the nanomolar-micromolar range (0.46 µM for 6 and 2.06 µM for 14). Both complex 14 and cisplatin (CDDP) showed higher activity towards A549 human lung cancer cells (8.7 µM for cisplatin and 6.7 µM for 14, Figure 3) compared to MRC-5 35 human fetal lung fibroblast-like cells (IC50 values of 16.6 µM and 51.3 µM, respectively). Hence the ratio of IC50 values in normal cells compared to this cancer cell line is ca. 2 for cisplatin but increases to ca. 8 for complex 14. 20 40 45 (B) A549 Lung Cancer Cells 50 Complex IC50(µM) IC50(µM) with L-BSO 3 4.55(±1.09) 1.59(±0.36) 6 3.7(±0.2) 0.7 (±0.1) 9 >100 25.88(±12.29) 10 0.9(±0.09) NAa 13 2.55(±1.35) NAa 14 Cisplatin 6.65(±0.06) 8.68(±2.11) 2.34(±1.18) NAa 55 60 10 (C) NCI 60-Cell Line Complexb IC50(µM) TGI(µM) LC50(µM) 6 3.72 13.2 49 8.3 33.9 75.8 14 b Cisplatin 10.3 50.7 90.5 a NA = not acquired. b Data from NCI/DTP screening. 6 = NSC755639; 14 = NSC755640. Mean-graph midpoint (MGMID) for IC50, TGI and LC50 values of NCI all cell panels. 15 The potency of complexes [Os(η6-p-cym)(Impy-NMe2)I]PF6 (6) and [Os(η6-p-cym)(Impy-NMe2)Cl]PF6 (14) towards A2780 cancer cells was similar to that of the anticancer drug cisplatin and so were further screened in the human tumour 60-cell line panel of the Developmental Therapeutics Program of the National 4 | Journal Name, [year], [vol], 00–00 Fig. 3 In vitro cytotoxicity data for cisplatin and complex 14 towards A549 human lung cancer cells and MRC-5 human fetal lung fibroblast-like cells. Induction of Reactive Oxygen Species (ROS) in Lung Cancer Cells OsII arene azopyridine complexes can increase ROS levels in cancer cells.14 Consequently, the effect of iminopyridine complexes on cellular ROS levels was also investigated. The 65 level of ROS induced in A549 human lung cancer cells by osmium complexes 6 and 14 was monitored using the probe 2,7dichlorodihydrofluorescein-diacetate (DCFH-DA). This hydrolyzes to 2,7-dichlorodihydrofluorescein (DCFH) in live cells, and in turn is oxidized to 2,7-dichlorofluorescein (DCF) in 37, 38 70 the presence of ROS, exhibiting a green fluorescence. Using this probe, the level of general oxidative stress induced in A549 cells by 6 and 14 was determined. The accumulation of ROS during combined exposure to 6 or 14 with L-BSO was also studied. 75 This journal is © The Royal Society of Chemistry [year] 60 still observable after 30 h. However, no hydride peak was detected after 45 h. This suggested that the oxidation of NADH in the presence of 14 occurs via hydride transfer to Os but the resulting Os-H adduct is unstable and decomposes. 5 65 10 70 15 Fig. 4 Time dependence of the ROS level in A549 human lung cancer cells after treatment with 6 (4 µM), 14 (4 µM), LBSO (50 µM), 6 or 14 combined with L-BSO (50 µM), and H2O2 (10 µM, positive control) during 4 h at 310 K. C= control. In A549 cells, a relative increase in DCF fluorescence was detected over a period of 4 h after treating the cells with 6 (4 µM), 14 (4 µM), 6 (4 µM) plus L-BSO (50 µM), 14 (4 µM) plus L-BSO (50 µM) or 10 µM H2O2 for comparison (Figure 4). After treatment only with 6 or 14, the ROS level increased dramatically compared to the control and even the positive control (H2O2, 10 25 µM). This level increased further in the presence of L-BSO. The further increase of ROS by the combination treatment may explain why L-BSO can enhance the cytotoxicity of complex 6 and 14 towards A549 cancer cells significantly (Table 1B). LBSO alone at a dose of 50 µM had no significant effect on the 30 increase of ROS (C+L-BSO, Fig.4) and the growth of A549 cancer cells, but greatly enhanced the cytotoxicity of osmium iminopyridine complexes towards A549 cancer cells. 20 Oxidation of NADH to NAD+ 35 After observing the increase in ROS levels in cells induced by osmium iminopyridine complexes, we investigated reactions of the complexes with potential cellular reducing agents. It was found that 6 and 14 do not catalytically oxidize GSH, in contrast 19 40 to their ruthenium azopyridine iodido analogues (Figure S4). + Since NADH/NAD is an important redox couple which maintains the redox balance in cells, reactions with NADH were also studied. Reactions of NADH with complexes 6, 8, 14, 14A and 16 were followed by 1H NMR spectroscopy. These + 45 complexes oxidized NADH to NAD to different extents: 2.0 ± 0.2 mol equiv of NADH per mol of 14A, 1.6 ±0.1 for 6 and 14, and 1.2 ±0.1 for 8 and 16, after incubation in a 10%MeOD/90%D2O phosphate buffer solution (pH* 7.4) for 24 h at 310 K. Under these conditions the azopyridine complex 6 50 [Os(η -p-cym)(Azpy-NMe2)I]PF6 (FY026) does not oxidize NADH (Figures S5 and S6). The control NADH was stable towards NAD+ formation under the same conditions at milimolar concentrations. Since more than one mol equiv of NADH was oxidized per mol equiv of osmium complex, this implies a 1 55 catalytic mechanism. The H NMR spectrum for the reaction with 14A at 310 K (10%D2O/90%H2O phosphate buffer) gave a weak signal at –4.2 ppm (Figure 5) assignable to an osmium hydride adduct. This signal was detected after 27 h of reaction and was This journal is © The Royal Society of Chemistry [year] 75 Fig. 5 Decay of the Os-H 1H NMR peak from reaction of the hydroxido complex 14A (2.5 mM) with 4 mol equiv of NADH in * 80 10%D2O/90%H2O phosphate buffer, pH 7.4 after different periods of incubation at 310 K. LC-MS was also employed to analyse the products from the reaction between NADH (0.5 mM) and 14A (0.5 mM) after 85 incubation at 310 K for 24 h. The decrease in intensity of the peak with a retention time of ca. 11.5 min (NADH; calcd for C21H30N7O14P2 m/z = 666.1, found m/z = 666.1) and the appearance of a new peak with a retention time of 4.8-5.2 min (NAD+ calcd for C21H28N7O14P2 m/z = 664.1, found m/z = 664.1) 90 were observed after the reaction with 14A (Figure S7). The reaction of the chlorido complex 14 with NADH (8 mol equiv) was monitored by UV-Vis over 18 h at 310 K in a phosphate buffer pH 7.4 (Figure 6). The kinetic experiment showed an isosbestic point at 300 nm corresponding to the + 95 oxidation of NADH to NAD and another isosbestic point at 432 nm related to the aquation of 14, suggesting that the aquation of the chlorido complex 14 accompanies the oxidation. The formation of NAD+ was confirmed by a decrease in intensity of the characteristic NADH band at 338 nm and the simultaneous 100 increase in intensity at 260 nm. 105 110 115 Fig. 6 (A) UV-Vis spectra for the reaction between the chlorido complex 14 (0.025 mM) and NADH (8 mol equiv) in 1 mM phosphate buffer (pH* 7.4) during 18 h at 310 K. The spectra Journal Name, [year], [vol], 00–00 | 5 were recorded at intervals of 20 min. Inset: (B) Dependence of the turnover number (TON) of complex 14 (0.025 mM) on the mol ratio of NADH. The reaction was followed by the decrease in intensity of the NADH absorption band at 338 nm. 5 The dependence of the oxidative activity of 14 on the concentration of NADH in solution was studied by following the conversion of 2 and 8 mol equivalents of NADH to NAD+ (Figure S8A) under the same conditions described above. The 10 rate of reaction of 14 (25 µM) was similar for reactions of 2 and 8 mol equiv NADH suggesting that the formation of an initial adduct is rate-limiting. An increase in NADH concentration resulted in higher turnover numbers for complex 14 (Figure 6B). The NAD+/NADH ratio after 18 h was determined to be 0.14 15 for the reaction between 14 and 2 equiv of NADH, 0.11 for 4 equiv, 0.12 for 6 equiv and 0.13 for 8 equiv, respectively. The rate of oxidation of NADH (8 mol equiv) was the same in the presence of either 25 µM or 10 µM 14 (Figure S8B). The control solutions showed a slight decomposition of the NADH under 20 these conditions attributable to the effect of phosphate buffer, consistent with previous reports.39, 40 Since the instability of the osmium hydride adduct formed by reaction with NADH might be due to protonation and liberation of H2,41 an attempt was made to detect H2 as a product 25 by gas chromatography. However, none was detected (for conditions see ESI). The reaction between osmium complex 14 (0.1 mM) and NADH (0.4 mM) under aerobic and anaerobic (bubbling argon for 10 min) conditions was compared as monitored by 1H NMR 30 spectroscopy (18-h incubation, 310 K in 10%MeOD/90%D2O phosphate buffer, pH* 7.4). Under argon there was almost no conversion of NADH to NAD+ (3 ± 0.5% compared to 17 ± 1% in air; Figure 7). It can therefore be concluded that O2 plays a role in the conversion of NADH to NAD+. 35 After saturating a solution containing 14 (0.1 mM) and NADH (0.5 mM) with O2 (bubbling for 10 min), the conversion + 60 of NADH to NAD increased from 10% (aerobic) to 25% (O2saturated), whereas such saturation had little effect on a similar control solution of NADH alone (no significant NAD+ formation after O2 saturation and incubation for 18 h at 310 K; Figure 8). 65 70 75 80 Fig. 8 1H NMR spectra for the reaction between the osmium complex 14 (0.1 mM) and NADH (5 mol equiv.) (A) under oxygen saturation, (B) in air, (C) NADH alone under oxygen * 85 saturation. Solvent: 10%MeOD/90%D2O phosphate buffer, pH 7.4). Incubation conditions: 18 h at 310 K. To investigate whether aquation is a necessary step for 14 to oxidize NADH to NAD+, the reaction of 14 (0.1 mM) and NADH 90 (0.4 mM) was carried out under three conditions: (1) without NaCl, (2) with 100 mM NaCl (3) with 500 mM NaCl. The resulting NMR spectra after 24 h incubation at 310 K are shown in Figure 9. 95 40 100 45 105 50 Fig. 7 1H NMR spectra for the reaction between osmium complex 14 (0.1 mM) and NADH (4 mol equiv) in 10%MeOD/90%D2O phosphate buffer, pH* 7.4, after 18 h of 55 incubation at 310 K. (A) Under aerobic conditions, and (B) under argon. 6 | Journal Name, [year], [vol], 00–00 110 Fig. 9 Reactions of complex 14 [Os(η6-p-cym)(ImpyNMe2)Cl]PF6 (0.1 mM) with NADH (0.4 mM) in the absence and presence of NaCl (to suppress aquation), followed by 1H NMR 115 spectroscopy. (A) 500 mM NaCl; (B) 100 mM NaCl; (C) no This journal is © The Royal Society of Chemistry [year] NaCl. Solvent: 10% MeOD-d4/90% D2O phosphate buffer, pH* 7.4). 1H NMR spectra were recorded 24 h after incubation at 310 K. 14A is the aqua complex. The presence of NaCl suppresses aquation but the extent of oxidation of NADH to NAD+ (40 ± 5 3%) is not affected. The extent of aquation decreased from 99% to 11% and 1% at NaCl concentrations of 0 mM, 100 mM and 500 mM, respectively. After 24 h incubation with NADH at 310 K, the + 10 same amount of NAD product (40 ± 3%) was observed for each of these three solutions (Figure 9), this result indicating that NADH can react directly with the chlorido complex. Since complex 14 appeared to oxidize NADH by hydride transfer, we studied the effect of complex 14 on the + 15 NAD /NADH ratio in human ovarian A2780 cancer cells. After treatment of the cancer cells with 1.5 µM of complex 14 for 6 h (see Supporting Information), the NAD+/NADH ratio in cell lysates increased from 2.54±0.46 to 5.44±0.26. The increase may suggest that the anticancer activity of these OsII arene 20 iminopyridine complexes can be related to their reactivity with the cellular reducing agent NADH and the consequent change in the redox status of the cancer cells. Electrochemistry 25 Cyclic voltammetry studies of iodido complexes 3 and 6, and chlorido complexes 10 and 14 together with 3 and 10 showed that these compounds underwent two irreversible electrochemical reductions in dimethylformamide with potentials ranging from – 30 0.58 to –0.76 V for the first and –0.82 to –1 V for the second. Subtle variations in the CV morphology and forward-reverse peak separations for the different complexes suggests some small differences in heterogeneous electron transfer kinetics (Figure S9). previously reported osmium picolinate DNA-targeted complexes,16 which suggests that other targets may be more important. The contrast between these iminopyridine complexes and their isoelectronic azopyridines analogues is quite striking. The azopyridine complexes do not readily hydrolyse and do not bind to guanine. In accordance with this, the azopyridines appear 65 to exert their anticancer activity partly by redox mechanisms. We also investigated whether redox mechanisms are important for iminopyridine complexes (vide infra). 60 70 Structure-Activity Relationships (SAR) for Human Ovarian Cancer Cells When the anticancer activity of osmium arene phenyliminopyridine complexes with various substituents on the phenyl ring is compared with their phenylazopyridine analogues, 75 the correlation coefficient is low (Figure 10), which suggests that the mechanism of action is different for these two families of osmium anticancer complexes. For a subset of 8 of the 11 complexes however, there is a stronger correlation, Figure 10. The iminopyridine complexes bearing a biphenyl arene exhibit 80 higher anticancer activity towards A2780 cells than their pcymene analogues (Table 1, Chart 1). It seems likely that the greater hydrophobicity of the biphenyl complexes enhances cell uptake, although intercalation into DNA may also contribute if DNA is a target.43 85 90 35 Discussion 95 Aquation and Binding to 9-Ethylguanine The chlorido complex 14 [Os(η6-p-cym)(Impy-NMe2)Cl]PF6 hydrolysed almost completely after 24 h in water at 310 K, whereas the iodido complex 6 [Os(η6-p-cym)(Impy-NMe2)I]PF6 was only ca. 50% hydrolysed. This is consistent with the expected strengthening of the Os-halide bond for the heavier 45 halides. However, both these complexes are much less stable in water than the analogous azopyridine complexes.14 The pKa of the aqua adduct 14A was determined to be 5.2, which implies that the major product of aquation at physiological pH (7.4) will be the hydroxido adduct. This may account for the 50 low extent and slowness of binding to the nucleobase 9-ethyl guanine (only 19% after 38 h of incubation at 310K, Figure 2), since Os-OH bonds are expected to be much less reactive than Os-OH2 bonds. We chose to study guanine as the nucleobase since G N7 is the most electron-dense site on DNA and is a 42 55 known target for platinum anticancer drugs. The findings for guanine binding suggest that DNA (and RNA) could be a target for osmium iminopyridine complexes, but the extent of binding of 14 to 9-EtG was much less than for the 40 This journal is © The Royal Society of Chemistry [year] 100 105 110 115 Fig. 10 Correlations of the IC50 values of osmium arene iminopyridine complexes with their azopyridine analogues. The Journal Name, [year], [vol], 00–00 | 7 numbers in brackets refer to iminopyridine complexes (see Chart 1). Complexes with biphenyl arene (red) and para-cymene (blue) arene are labelled with different colours; chlorido complexes are indicated by Cl in the brackets; different substituents on the 5 chelating ligands are designated using different shapes: R= H (square), R= OH (circle) and R= NMe2 (triangle). Overall the correlation is poor (R2= 0.0673) although if outliers 4, 9 and 13 are ignored, then there is a reasonable correlation for the remaining 8 complexes (R2= 0.9395). 10 In general, the chlorido iminopyridine complexes are as active or more active than their iodido analogues. This contrasts with the azopyridine series for which the opposite is true, except for those with a –NMe2 electron-donating substituent in the 15 phenyl ring of the phenyliminopyridine ligand. For the inert azopyridine complexes, addition of an ortho substituent such as CF3 or Cl on the pyridine ring maintained or increased the anticancer activity. However, in the present case, ortho functionalization of the pyridine ring with a methoxy group 20 decreased the anticancer activity significantly (Table 1). Unlike the azopyridine complexes, the Os-Cl/I bond is reactive in iminopyridine complexes and an ortho substituent may hinder the reactivity. Substitution reactions at the osmium centre may therefore be important in the mechanism of action of 25 iminopyridine complexes. Additionally, it is notable that 6 and 14 are more potent anticancer complexes than previously-reported ruthenium and osmium arene complexes which bind more strongly to DNA bases and appear to have DNA as a major target.16, 44 Redox 30 mediated mechanisms of activity may be more effective since they can affect multi-specific targets in cells. Accumulation of ROS in Cancer Cells ROS are highly reactive O2 metabolites that include superoxide radicals (O2-·), hydrogen peroxide (H2O2) and hydroxyl radicals (OH·).45 Because cancer cells have increased ROS levels compared to normal cells, this difference can be exploited as a biochemical basis for 46, 47 40 therapeutic selectivity. Increased ROS levels in cancer cells can reach a threshold and induce cell death but such increases can be tolerated by normal cells.30 This may explain why complex 14 is 8-fold more cytotoxic towards A549 human lung cancer cells compared to normal human fetal lung 45 fibroblast-like MRC-5 cells (Fig. 3). Although ROS are generally believed to be toxic to cells, they act as a ‘doubleedged sword’ for the proliferation of cells. When the ROS level in cells is up-regulated on a small scale, ROS may contribute to the proliferation; if the ROS level is raised 50 further, it can cause DNA damage which will induce cell cycle arrest and apoptosis.48 Therefore selectivity to cancer cells rather than normal cells can be achieved by increasing the ROS levels because the basal ROS levels in cancer cells are higher than in normal cells. These increases in ROS levels 55 provide a basis for a strategy involving combination treatment of 6 or 14 with the glutathione synthesis inhibitor buthionine sulfoximine (L-BSO), Fig. 4. 35 8 | Journal Name, [year], [vol], 00–00 Because the GSH concentration is 100-10,000 fold higher in cells compared to other reductants (e.g. NADH, NADPH and 60 thioredoxin), the level of GSH usually determines the steady-state value of intracellular redox potentials.49 However, complexes 6 and 14 do not readily oxidize GSH unlike ruthenium azopyridine analogues.14 The half-wave reduction potentials of osmium complexes 3, 6, 10 and 14 in dimethylformamide of –0.58 to – 65 0.76 V and –0.82 to –1 V are more negative than those of the ruthenium azopyridine complexes,19 indicating that the iminopyridine complexes are more difficult to reduce, perhaps explaining why no reaction was observed with GSH (Figures S4 and S9). In contrast, novel reactions with the reduced coenzyme 70 NADH, a stronger reducing agent than GSH, were observed. Oxidation of NADH to NAD+ Apart from a role in the generation of ROS, the facile oxidation of NADH may help to rationalize the surprisingly good anticancer activity of these OsII iminopyridine complexes. NMR, LC-MS and UV-Vis studies showed that osmium iminopyridine complexes can oxidize NADH to NAD+. The reaction appears to involve the transfer of hydride from NADH to 80 the Os centre with displacement of the halide ligand (I for 6 and 8, and Cl for 14 and 16) or aqua ligand. During the conversion of NADH to NAD+ in the presence of 14A, an osmium-hydride peak was detected by 1H NMR at –4.2 ppm (Figure 5). This value is consistent with previous reports of Os-H adducts.50, 51 The organometallic hydride osmium(II) compounds 85 reported containing arene52, 53 and/or imine ligands54, 55 showed Os-H signals ranging from -9 ppm to -13 ppm (terminal hydride) and from -15 ppm to -20 ppm (bridging hydride). In all the reported compounds, the hydride peaks were detected for complexes in 90 deuterated organic solvents such as CD2Cl2, acetone-d6 or C6D6. The current work appears to be the first detection of an osmium(II) hydride NMR peak for an osmium hydride in an aqueous solution.56-64 These findings suggest that osmium arene complexes can modulate the redox balance in cells by a novel 95 mechanism. 1 H NMR studies show that complexes 6 and 14 are more effective in oxidising NADH than 8 and 16 after 24 h incubation at 310K (Figure S5). This may be due to the steric hindrance exerted by the methoxy substituent in the ortho-position of the 100 iminopyridine chelating ligand for 8 and 16. This trend is also observed for the anticancer activity of these compounds: 6 and 14 are ca. 20 times more potent than 8 and 16. It is possible therefore that NADH oxidation is involved in the mechanism of anticancer activity of these osmium iminopyridine anticancer complexes. A possible mechanism for the oxidation of NADH in the 105 presence of these iminopyridine complexes involves hydride transfer to Os followed by protonation of bound hydride,65 liberation of H2, and regeneration of the aqua/hydroxido Os adduct, similar to that of IrIII cyclopentadienyl complexes studied 27 110 in our laboratory. However, no H2 was detected after the incubation of NADH with 14 or 14A for 24 h at 310 K (100 mM phosphate buffer, pH 7.2). The presence of O2 did appear to influence the course of the oxidation significantly, being inhibited under argon and enhanced by O2 saturation (Figures 7 and 8). 115 One possibility is that hydride is transferred from Os-H to O2 to 75 This journal is © The Royal Society of Chemistry [year] form H2O2,37 giving a cycle such as that shown in Scheme 1. However attempts to detect H2O2 production electrochemically were unsuccessful (for conditions see ESI). Suppression of aquation of the chlorido complex 14 still led 5 to the oxidation of NADH (Figure 9) suggesting that aquation is not an essential step in the mechanism and that hydride transfer can occur with direct chloride displacement. O H 2N N 10 H X X- Os H 2O 2 N N Ribo ADP H N NADH CH X- O2 (X- = OH -, Cl- , I-) 2 15 N N H Os H2N N N Os O N H CH N N CH H dinucleotide (NADH). The Os II arene iminopyridine 60 compounds followed opposite trends in redox reactivity compared to the azo analogues: the iminopyridine complexes were inert toward GSH but they could oxidize NADH to different extents depending on the substituents on the chelating ligand. A mechanism for the oxidation of NADH to + 65 NAD through the formation of an Os-hydride adduct in aqueous solution detected by 1H NMR (singlet at -4.2 ppm) is proposed. Furthermore, an increase in the ratio of NAD+/NADH in human ovarian A2780 cancer cells after treatment with the chlorido complex [Os(η6-p-cym)(Impy70 NMe 2 )Cl]PF 6 was found. We conclude from these studies that the anticancer activity of these osmium arene iminopyridine complexes may involve the modulation of redox pathways in cancer cells by a novel mechanism. In addition, we have shown that the rational II 75 design of the chelating ligands and structure of the Os arene complexes plays an important role in controlling the reactivity and biological behaviour of these anticancer drugs. b Ri o 20 A DP O NH 2 ADP Ribo 80 N Table 2 Comparison between the biological and chemical behaviour of osmium(II) arene iminopyridine complexes and azopyridine complexes. NAD+ 25 Scheme 1 A possible mechanism for the oxidation of NADH to NAD+ catalysed by an OsII arene iminopyridine complex, X- = OH-, Cl- or I-.23, 27 [Os(η6-arene)(XY)Z]+ Complexes XY = 30 Property Iminopyridine Azopyridine Activity to A2780 cells Nanomolar Nanomolar Increase in ROS Yes Yes Oxidation of GSH No No Oxidation of NADH Yes No Aquation Yes No Binding to 9-EtG Yes No Conclusions In this work, we have prepared new OsII arene iminopyridine 35 anticancer complexes to explore the effect of bioisosterism on the biological activity of our previously studied OsII arene azopyridine complexes. We have contrasted the chemical reactivity and anticancer activity of isoelectronic organometallic Os II complexes [Os(η6-arene)(XY)Z]PF6 arene = p-cym or bip, XY = N,N΄-chelated 40 where phenyliminopyridine or phenylazopyridine derivatives, and Z = Cl or I. The X-ray crystal structure of [Os(η6-p-cym)(ImpyOH)I]PF6·0.5CH2Cl2·H2O is reported. Some OsII arene iminopyridine complexes are potently active towards cancer 45 cells (nanomolar IC 50 values), like the azopyridine complexes we reported recently.66 Nevertheless, unlike the azopyridine analogues, the iminopyridine complexes hydrolyse in aqueous solution and bind to the model nucleobase 9-ethylguanine, suggesting that DNA could be a possible target for these compounds, although other targets are not 50 anticancer excluded. Additionally, we have found that active iminopyridine complexes induce a dramatic increase in the levels of reactive oxygen species (ROS) in A549 lung cancer cells. For this we investigated the possibility that these Os II 55 reason, complexes can be involved in biologically-relevant redox chemistry, for example with the cellular reducing agents glutathione (GSH) and reduced coenzyme nicotine adenine This journal is © The Royal Society of Chemistry [year] Acknowledgements 85 We thank the ERC (grant no. 247450 BIOINCMED), EPSRC and Science City/EU ERDF/AWM (MaXis mass spectrometer and the X-ray diffractometer) for funding. M. J. R. thanks the Ministerio de Educación-FECYT (Spain) for her postdoctoral 90 fellowship. We thank colleagues in the EC COST Action D39 for stimulating discussions, Dr. Michael Khan (Life Sciences) for provision of facilities for cell culture, Joan Soldevila for discussions on catalytic reactions, Prof. Timothy D. H. Bugg and Mr. Darren Braddick for use of the microplate reader, Dr. 95 Andrew Crombie and Prof. Colin Murrell for use of gas chromatography, Dr. Salzitsa Anastasova-Ivanov and Prof. Journal Name, [year], [vol], 00–00 | 9 23. Pankaj Vadgama (QMW University of London) for H2O2 detection by electrochemistry, and the National Cancer Institute (NCI) for 60-cancer-cell-line screening. 24. 75 5 Notes and references Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK. Tel: +44-2476523818; Fax: +44-2476523819. Email: P.J.Sadler@warwick.ac.uk † Electronic Supplementary Information (ESI) available: Experimental 10 Section (synthesis and characterisation of complexes), Methods (electrochemistry, LC-MS, GC, X-ray crystallography, Cell culture and IC50 determinations, ROS detection, NAD+/NADH ratio in cells), Instrumentation, Tables S1 and S2 of crystallographic data, Table S3 of NCI 60-cell line anticancer data, and Figures S1-S9. CCDC reference 15 number 867420. See DOI: 10.1039/b000000x/. 26. 27. 80 28. 85 29. 90 20 1. 2. 25 3. 4. 5. 30 6. 7. 35 8. 9. 40 10. 11. 45 12. 13. 50 14. 15. 55 16. 17. 60 18. 19. 65 20. 21. 22. 70 C. G. Hartinger, S. Zorbas-Seifried, M. A. Jakupec, B. Kynast, H. Zorbas and B. K. Keppler, J. Inorg. Biochem., 2006, 100, 891-904. M. Galanski, V. B. Arion, M. A. Jakupec and B. K. Keppler, Curr. Pharm. Des., 2003, 9, 2078-2089. C. G. Hartinger and P. J. Dyson, Chem. Soc. Rev., 2009, 38, 391-401. U. Schatzschneider and N. Metzler-Nolte, Angew. Chem., Int. Ed., 2006, 45, 1504-1507. G. Gasser, I. Ott and N. Metzler-Nolte, J. Med. Chem., 2010, 54, 3-25. G. I. Pascu, A. C. G. Hotze, C. Sanchez-Cano, B. M. Kariuki and M. J. Hannon, Angew. Chem., Int. Ed., 2007, 119, 44524456. E. Hillard, A. Vessières, L. Thouin, G. Jaouen and C. Amatore, Angew. Chem., Int. Ed., 2006, 45, 285-290. G. Sava, A. Bergamo, S. Zorzet, B. Gava, C. Casarsa, M. Cocchietto, A. Furlani, V. Scarcia, B. Serli, E. Iengo, E. Alessio and G. Mestroni, Eur. J. Cancer, 2002, 38, 427-435. W.-X. Ni, W.-L. Man, M. T.-W. Cheung, R. W.-Y. Sun, Y.-L. Shu, Y.-W. Lam, C.-M. Che and T.-C. Lau, Chem. Commun., 2011, 47, 2140-2142. O. Novakova, J. Kasparkova, O. Vrana, P. M. van Vliet, J. Reedijk and V. Brabec, Biochemistry, 1995, 34, 12369-12378. P. Annen, S. Schildberg and W. S. Sheldrick, Inorg. Chim. Acta, 2000, 307, 115-124. O. R. Allen, L. Croll, A. L. Gott, R. J. Knox and P. C. McGowan, Organometallics, 2003, 23, 288-292. D. R. van Staveren and N. Metzler-Nolte, Chem. Rev., 2004, 104, 5931-5986. Y. Fu, A. Habtemariam, A. M. Pizarro, S. H. van Rijt, D. J. Healey, P. A. Cooper, S. D. Shnyder, G. J. Clarkson and P. J. Sadler, J. Med. Chem., 2010, 53, 8192-8196. S. D. Shnyder, Y. Fu, A. Habtemariam, S. H. van Rijt, P. A. Cooper, P. M. Loadman and P. J. Sadler, Med. Chem. Commun., 2011, 2, 666-668. A. F. A. Peacock, S. Parsons and P. J. Sadler, J. Am. Chem. Soc., 2007, 129, 3348-3357. S. H. van Rijt, A. J. Hebden, T. Amaresekera, R. J. Deeth, G. J. Clarkson, S. Parsons, P. C. McGowan and P. J. Sadler, J. Med. Chem., 2009, 52, 7753-7764. H. Kostrhunova, J. Florian, O. Novakova, A. F. A. Peacock, P. J. Sadler and V. Brabec, J. Med. Chem., 2008, 51, 3635-3643. S. J. Dougan, A. Habtemariam, S. E. McHale, S. Parsons and P. J. Sadler, Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 1162811633. W. Ying, Antioxidants & Redox Signaling, 2008, 10, 179-206. Q. Zhang, D. W. Piston and R. H. Goodman, Science, 2002, 295, 1895-1897. E. Steckhan, S. Herrmann, R. Ruppert, E. Dietz, M. Frede and E. Spika, Organometallics, 1991, 10, 1568-1577. 10 | Journal Name, [year], [vol], 00–00 25. 30. 31. 95 32. 33. 34. 100 35. 105 36. 37. 38. 110 39. 40. 41. 115 42. 43. 120 44. 45. 46. 125 47. 48. 49. 130 50. 51. 135 52. 53. 54. 55. 140 56. H. C. Lo, C. Leiva, O. Buriez, J. B. Kerr, M. M. Olmstead and R. H. Fish, Inorg. Chem., 2001, 40, 6705-6716. Y. Yan, M. Melchart, A. Habtemariam, A. Peacock and P. Sadler, J. Biol. Inorg. Chem., 2006, 11, 483-488. Z. Liu, A. Habtemariam, J. J. Soldevila, A. M. Pizarro, P. J. Sadler, PCT Int. Appl. 2011, 88p, WO/2011/148124. Y. Maenaka, T. Suenobu and S. Fukuzumi, J. Am. Chem. Soc., 2012, 134, 367-374. S. Betanzos-Lara, Z. Liu, A. Habtemariam, A. M. Pizarro, B. Qamar and P. J. Sadler, Angew. Chem., Int. Ed., 2012, 51, 3897-3900. J. Maksimoska, D. S. Williams, G. E. Atilla-Gokcumen, K. S. M. Smalley, P. J. Carroll, R. D. Webster, P. Filippakopoulos, S. Knapp, M. Herlyn and E. Meggers, Chem.-Eur. J., 2008, 14, 4816-4822. A. Dorcier, W. H. Ang, S. Bolaño, L. Gonsalvi, L. JuilleratJeannerat, G. Laurenczy, M. Peruzzini, A. D. Phillips, F. Zanobini and P. J. Dyson, Organometallics, 2006, 25, 40904096. A. F. A. Peacock, A. Habtemariam, S. A. Moggach, A. Prescimone, S. Parsons and P. J. Sadler, Inorg. Chem., 2007, 46, 4049-4059. G. E. Büchel, I. N. Stepanenko, M. Hejl, M. A. Jakupec, B. K. Keppler and V. B. Arion, Inorg. Chem., 2011, 50, 7690-7697. B. Cebrián-Losantos, A. A. Krokhin, I. N. Stepanenko, R. Eichinger, M. A. Jakupec, V. B. Arion and B. K. Keppler, Inorg. Chem., 2007, 46, 5023-5033. Y. C. Martin, J. Med. Chem., 1981, 24, 229-237. U. Vanhoefer, S. Cao, H. Minderman, K. Toth, B. S. Skenderis, M. L. Slovak and Y. M. Rustum, Clin. Cancer Res., 1996, 2, 1961-1968. S. S. Hindo, A. M. Mancino, J. J. Braymer, Y. Liu, S. Vivekanandan, A. Ramamoorthy and M. H. Lim, J. Am. Chem. Soc., 2009, 131, 16663-16665. A. Krezel and W. Bal, J. Inorg. Biochem., 2004, 98, 161-166. H. Wang and J. A. Joseph, Free Radical Biol. Med. 1999, 27, 612-616. A. Gomes, E. Fernandes and J. L. F. C. Lima, J. Biochem. Biophys. Methods, 2005, 65, 45-80. J. Wu, L. Wu and J. Knight, Clin. Chem., 1986, 32, 314-319. R. M. Burton and N. O. Kaplan, Arch. Biochem. Biophys., 1963, 101, 139-149. C. Creutz, M. H. Chou, H. Hou and J. T. Muckerman, Inorg. Chem., 2010, 49, 9809-9822. B. Song, J. Zhao, R. Griesser, C. Meiser, H. Sigel and B. Lippert, Chem.-Eur. J., 1999, 5, 2374-2387. H.-K. Liu, S. J. Berners-Price, F. Wang, J. A. Parkinson, J. Xu, J. Bella and P. J. Sadler, Angew. Chem., Int. Ed., 2006, 118, 8333-8336. H.-K. Liu and P. J. Sadler, Acc. Chem. Res., 2011, 44, 349359. V. J. Thannickal and B. L. Fanburg, Am. J. Physiol. Lung Cell Mol. Physiol., 2000, 279, L1005-L1028. E. Hileman, J. Liu, M. Albitar, M. Keating and P. Huang, Cancer Chemother. Pharmacol., 2004, 53, 209-219. J. Wang and J. Yi, Cancer Biol. Ther., 2008, 7, 1875-1884. B. Halliwell, Biochem. J., 2007, 401, 1-11. F. Q. Schafer and G. R. Buettner, Free Radical Biol. Med., 2001, 30, 1191-1212. T. Bolaño, R. Castarlenas, M. A. Esteruelas, F. J. Modrego and E. Oñate, J. Am. Chem. Soc., 2005, 127, 11184-11195. A. V. Marchenko, H. Gerard, O. Eisenstein and K. G. Caulton, New J. Chem., 2001, 25, 1244-1255. S. Stahl and H. Werner, Organometallics, 1990, 9, 1876-1881. J. P.-K. Lau and W.-T. Wong, Dalton Trans., 2005, 25792587. J. A. Cabeza, I. del Río, F. Grepioni and V. Riera, Organometallics, 2000, 19, 4643-4646. S. Aime, E. Diana, R. Gobetto, M. Milanesio, E. Valls and D. Viterbo, Organometallics, 2001, 21, 50-57. C. L. Gross and G. S. Girolami, Organometallics, 2006, 26, 160-166. This journal is © The Royal Society of Chemistry [year] 57. 58. 5 59. 60. 61. 10 62. 63. 15 64. 65. 66. 20 R. Castro-Rodrigo, M. A. Esteruelas, A. M. López and E. Oñate, Organometallics, 2008, 27, 3547-3555. M. A. Esteruelas, F. J. Fernández-Alvarez, A. M. López, M. Mora and E. Oñate, J. Am. Chem. Soc., 2010, 132, 5600-5601. T.-Y. Cheng and R. M. Bullock, Organometallics, 2002, 21, 2325-2331. M. A. Esteruelas, A. M. López, E. Oñate and E. Royo, Organometallics, 2004, 23, 3021-3030. M. A. Esteruelas and A. M. López, Organometallics, 2005, 24, 3584-3613. J. Zhang, D. C. Grills, K.-W. Huang, E. Fujita and R. M. Bullock, J. Am. Chem. Soc., 2005, 127, 15684-15685. C. A. Bayse, M. Couty and M. B. Hall, J. Am. Chem. Soc., 1996, 118, 8916-8919. C. L. Gross and G. S. Girolami, Organometallics, 1996, 15, 5359-5367. H. F. Haarman, F. R. Bregman, P. W. N. M. van Leeuwen and K. Vrieze, Organometallics, 1997, 16, 979-985. Y. Fu, A. Habtemariam, A. M. B. H. Basri, D. Braddick, G. J. Clarkson and P. J. Sadler, Dalton Trans., 2011, 40, 1055310562. This journal is © The Royal Society of Chemistry [year] Journal Name, [year], [vol], 00–00 | 11