← Back

A chemical-biological evaluation of rhodium(I) N-heterocyclic carbene complexes as prospective anticancer drugs.

PMID: 24243420
FULL PAPER DOI:10.1002/chem.201302819 A Chemical–Biological Evaluation of Rhodium(I) N-Heterocyclic Carbene Complexes as Prospective Anticancer Drugs Luciano Oehninger,[a] Laura Nadine K(cid:2)ster,[b] Claudia Schmidt,[a] Alvaro MuÇoz-Castro,[c] Aram Prokop,[b] and Ingo Ott*[a] Abstract: Rhodium(I) complexes bear- plexes, along with moderate inhibitory and confirmed the transport of rhodi- ing N-heterocyclic carbene (NHC) li- activity of thioredoxin reductase um into the nuclei. Changes in the mi- gandshavebeenwidelyusedincatalyt- (TrxR) andefficient bindingtobiomol- tochondrial membrane potential ic chemistry, but there are very few re- ecules (DNA, albumin). Biodistribu- (MMP) were observed as well as DNA ports of biological properties of these tion studies showed that the presence fragmentation inwild-typeand daunor- organometallics. A series of RhI-NHC of albumin lowered the cellular uptake ubicin- or vincristine-resistant Nalm-6 derivatives with 1,5-cyclooctadiene and leukemia cells. Overall, these studies CO as secondary ligands were synthe- indicated that RhI-NHC fragments Keywords: antitumor agents · car- sized, characterized, and biologically could be used as partial structures of benes · cytotoxicity · drug resist- investigated as prospective antitumor new antitumor agents, in particular in ance · rhodium · thioredoxin reduc- drug candidates. Pronounced antiproli- those drugs designed to address resist- tase ferative effects were noted for all com- antmalignanttissues. Introduction agents.[12–15] This type of complex showed among other ef- fects a strong and selective inhibition of the enzyme thiore- Over the last years organometallic complexes have experi- doxin reductase (TrxR), induction of apoptosis, or depolari- enced an increasing interest not only because of their cata- zation of the mitochondrial membrane.[14,16–19] Besides gold lytic properties but also due to their fascinating biological and silver, other metals including platinum,[20–24] palladi- effects.[1–4] Among those, organometallics bearing N-hetero- um,[25,26] ruthenium,[27–29] or rhodium[30,31] have also been cyclic carbenes (NHCs) as ligands have been increasingly in usedascentersofbioactiveNHCcomplexes. the focus of inorganic medicinal chemists.[5–7] The advantag- Rhodium complexes have been widely studied as an anti- es of NHC ligands include the fact that sufficiently stable cancer agents mainly containing RhIII coordinated to differ- coordinative bonds can be formed to many different metals ent ligands such as polypyridyls or cyclopentadienyls (see and that the versatile synthetic chemistry related to the use Figure1 forsomeselectedexamples).[32–40]Biologicalstudies of NHC ligands offers a broad variety of possible structural on rhodium(I) complexes as anticancer agents are less fre- modifications. So far, metal-NHC complexes have been quent.[31,41–44] One recent report by McAlpine and co-work- mostly studied as new antibacterial and anticancer agents. ers described the alteration of cell migration, DNA replica- Although research on anti-infectives has been largely fo- tion, and DNA condensation by a RhI-NHC derivative.[31] cused on AgI-NHC complexes,[8–11] AuI-NHC derivatives are The choice of RhI as metal center appears promising be- the most studied examples concerning new anticancer cause it is isoelectronic with platinum(II) and exhibits a square-planar geometry like cisplatin-type tumor thera- peutics. [a] L.Oehninger,C.Schmidt,Prof.Dr.I.Ott InrecentstudiesweandothersreportedonthepromisACHTUNGTRENNUNGing InstituteofMedicinalandPharmaceuticalChemistry anticancer effects of gold(I)- and ruthenium(II)-NHC TechnischeUniversit(cid:2)tBraunschweig,Beethovenstrasse55 ACHTUNGTRENNUNGcomplexes containing benzimidazol-2-ylidene li- 38106Braunschweig(Germany) gands.[14,16,28,30,45–48]StartingfromthistypeofNHCligandwe Fax:(+49)5313918456 E-mail:ingo.ott@tu-bs.de designed in this study a series of related NHC ligands with [b] L.N.K(cid:3)ster,Dr.med.habil.Dr.rer.nat.A.Prokop different surface volumes, lipophilicities, and coordinative DepartmentofPaedriatricOncology,ChildrensHospitalCologne donor strengths (see Scheme1). Perimidines have been de- AmsterdamerStrasse59,50735Cologne(Germany) scribed as DNA intercalating agents and therefore NHC li- [c] Prof.Dr.A.MuÇoz-Castro gands derived of this N-heterocyclic core were included in DepartmentofChemicalSciences the study.[49] The target RhI complexes contain the respec- UniversidadAndresBello tive NHC moiety as well as chlorido, 1,5-cyclooctadiene Republica275,Santiago(Chile) (COD), or carbon monoxide (CO) ligands, which them- SupportinginformationforthisarticleisavailableontheWWW underhttp://dx.doi.org/10.1002/chem.201302819. selves exhibit different coordinative and physicochemical Chem.Eur.J.2013,19,17871–17880 (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim 17871 a silver intermediate, followed by a transmetalation reaction by addition of bisACHTUNGTRENNUNG[chlorido(1,5- cyclooctadiene)rhodium(I)] (see Scheme1). The resulting benzimidazolylidene-rhodi- um(1,5-cyclooctadiene) deriva- tives (3a–d) and perimidinyli- dene-rhodium(1,5-cycloocta- diene) complexes (4a and 4b) were purified by filtration over Celite and recrystallized from a mixture of CHCl/n-hexane 2 2 (1:4) at 08C. In the case of 3a– d, the reaction was complete within4h,whereasanextended Figure1.Examplesofantitumoractiverhodiumcomplexes. reaction period of 24h was re- quiredinthecaseof4aand4b. The dicarbonylrhodium(I) de- rivatives 5a–d and 6 were ob- tained by stirring the respective [RhACHTUNGTRENNUNG(cod)] complex in dichloro- methane under carbon monox- ide atmosphere for 20min. The volume of the solution was re- duced, n-hexane was added, and the solution was stored at (cid:2)208C. After approximately 10h the corresponding com- plexes had precipitated as pale- yellow solids. All target com- plexes were clearly character- ized by MS and NMR spectros- Scheme1.Synthesis procedures:1)alkylhalide,KCO,CHCN,heatatreflux,6h; 2)AgOinCHCl,4h; copy (see below) and their high 2 3 3 2 2 2 3)[RhClACHTUNGTRENNUNG(cod)],2–12h;4)COatmosphereinCHCl,20min. purity was confirmed by ele- 2 2 2 mentalanalysis. properties and could hence be used to modulate the charac- In the case of CO-containing complexes, IR spectroscopic teristicsofthetargetcomplexes. analyses were also performed. In particular, the CO com- The synthesis and characterization of these RhI-NHC de- plexes were included in this study to compare the donor rivatives as wellastheir biologicalevaluation asprospective propertiesoftheNHC ligands among eachotherbyobserv- anticancer drugs are reported here. The antiproliferative ef- ing the IR band of the CO ligand (v (A)).[54,55] Ligands CO 1 fectsofthecomplexeswereevaluated,alongwithstudieson that are positioned trans to a CO ligand can have strong possible molecular mechanisms of drug action (TrxR inhibi- effect on the ability of the CO ligand to effectively p-back- tion, DNA binding, effects on mitochondrial membrane po- bond to the metal and this influences the IR-stretching fre- tential)andeffectsindrug-resistantleukemiacells. quencies of the CO ligand. Free CO shows a band at 2155cm(cid:2)1; a strong donating ligand will shift the CO band to lower frequencies.[56] This shift was observed for all CO Results and Discussion complexes but between the single complexes no strong var- iation could be noted in the v (A) bands, which were in CO 1 Synthesis and characterization: The complexes were pre- therangeof2074–2086cm(cid:2)1.The1,3-diisopropylbenzymida- pared according to procedures depicted in Scheme1. Com- zolylidene ligand of 5c had the weakest donor properties of plexes3a–cand5b,chavebeendescribedpreviously.[50–53] the used NHC ligands with v (A)=2086cm(cid:2)1, whereas CO 1 The benzimidazolium and perimidinium halides 1a–d and the methyl derivative 5a had strongest one (v (A)= CO 1 2a,b were synthesized by heating benzimidazole or perimi- 2074cm(cid:2)1). The v (A) bands of the N-ethyl complexes5b CO 1 dine at reflux with an excess of an alkyl halide in the pres- (2077cm(cid:2)1) and 6 (2079cm(cid:2)1) indicated that the s-donation enceofKCO.Next,compounds1a–dand2a,bweretreat- of the benzimidazolylidene- and perimidinylidene-NHC li- 2 3 ed with a half equivalent of AgO in CHCl to generate gandswassimilar. 2 2 2 17872 www.chemeurj.org (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim Chem.Eur.J.2013,19,17871–17880 Rhodium(I)N-HeterocyclicCarbeneComplexes FULL PAPER NMR spectroscopy: In the 1HNMR spectra of the benzi- tive labilization of the coordinative bond between Rh and midazolium and perimidinium halides, 1a,b and 2a,b, re- the respective trans-positioned alkene fragment is reflected spectively (see FigureS1, the Supporting Information) the in smaller coupling constants (trans to NHC: 1J =6.6– RhC protonsoftheC2positionin1a,b(d=9.65and9.80ppm,re- 6.8Hz,cistoNHC:14.2–14.6Hz).[57] spectively[16]) are shifted to higher ppm values compared ExchangeoftheCODwithtwoCOligandscausedanup- with those of 2a,b (d=9.01 and 8.95ppm, respectively). field shift of the C2 signals of around d=11–12ppm in the This difference is reflected in a higher acidity and hence re- 13CNMR spectra. This indicates a weaker influence of the activity of 1ab with silver oxide and explains the longer re- rhodium atom over the C2 carbon of the NHC ligand in the action time necessary to obtain the [Rh- CO derivatives compared with the COD complexes. An in- ACHTUNGTRENNUNG(cod)(perimidinylidene)]complexes(seeabove). creaseintheNHC(cid:2)Rhdistancecanbeassumed bycompar- In the 1HNMR spectra of the COD derivatives 3b and ingthecouplingconstantsbetweentheNHC(cid:2)C2carbonand 3d the signal of the N(cid:2)CH(cid:2)R protons is split into two sig- Rh(e.g.,J=51Hzfor3band43Hzfor5b,respectively).In 2 nals (Figure2, (a) signals), which is the consequence of an the 13CNMR spectra of the perimidinylidene derivative 6, the NHC(cid:2)C2 is shifted to lower ppm values compared with the benzimidazolylidene complexes 5a,b. Taken together, the signals for the cis- (1J (cid:3)74Hz) and trans- (1J RhC RhC (cid:3)55Hz) CO ligands can also be distinguished in the 13CNMR spectra according to their coupling constants (see Figure3). Figure2.1HNMR spectra of 3b (top) and 3d (bottom) in CDCl. N(cid:2) 3 CH(cid:2)R (a);trans-CH of COD (b);cis-CH ofCOD (c); equatorial-CH 2 2 ofCOD(d);axial-CH COD(e). 2 Figure3.Rh(cid:2)Ccouplingsin13CNMRspectra(CDCl)of5b(top)and6 3 (bottom). asymmetrical disposition of the N-residues regarding the [RhACHTUNGTRENNUNG(cod)] moiety. The same effect was also observed for complex 4b. However, in this case the N(cid:2)CH(cid:2)R signals Theoreticalcalculations:Thegeometry,bonding,andmolec- 2 were not well-defined and appeared as broad multiplets. Of ular properties of the studied compounds have been ana- note, similar phenomena can also be observed with RuII- lyzedthroughrelativisticdensityfunctionalcalculations.The NHCderivatives.[28] COD moiety exhibits a nearly h2,h2 coordination mode, in The [RhClACHTUNGTRENNUNG(cod)] moiety shows a characteristic pattern in which the Rh(cid:2)COD bond lengths trans to the NHC ligand the 1HNMR spectrum, in which two signals are observed areslightlyelongated((cid:3)0.1(cid:5))inrelationtotheotherRh(cid:2) that correspond to the cis- and trans-olefin protons regard- COD distances (TableS1 and FigureS3, the Supporting In- ing the NHC ligand (signals (c) and (b) in Figure2, respec- formation).Thisisingoodagreementwiththeresultsofthe tively), as well as two signals from the equatorial and axial 1HNMR measurements (see above). Similarly, the Rh(cid:2)CO CH protons (signals (d) and (e) in Figure2, respectively). distancesinthedicarbonylcounterpartsaredifferent,butto 2 This signal distribution was corroborated by an exemplary a small extent of about 0.64(cid:5). The Rh(cid:2)NHC and Rh(cid:2)Cl 2D NMR spectrum of 3a (see FigureS2, the Supporting In- distances are similar within the studied series, displaying an formation). averaged value of (2.07(cid:4)0.05) and (2.40(cid:4)0.02)(cid:5), respec- In complex 3d the equatorial and axial proton signals tively, which are in the range of other similar compounds.[58] become unfolded due to the steric influence of the benzyl Moreover, calculated 13CNMR spectroscopy and v are in CO residues of the NHC moiety over the COD ligand (see good agreement with the available experimental data (see Figure2,bottom,signals(d)and(e)). theSupportingInformation). The olefinic CH groups cis and trans to the NHC ligands The calculated bond-dissociation energies (BDE) of the show Rh(cid:2)C coupling constants in the 13CNMR spectra of NHC(cid:2)Rh bond display quite similar values for the studied 3a–d and 4a,b. The NHC moiety has a stronger trans effect series (TableS2, the Supporting Information), denoting in comparison to the chlorido ligand and the resulting rela- a more stabilizing situation in the dicarbonyl counterparts Chem.Eur.J.2013,19,17871–17880 (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim www.chemeurj.org 17873 I.Ottetal. (av: 60.18kcalmol(cid:2)1) in relation to the COD derivatives wellasthedirhodiumRhIsynthesisreagentbisACHTUNGTRENNUNG[chlorido(1,5- (av: 51.38kcalmol(cid:2)1). A deeper understanding into the cyclooctadiene)rhodium(I)] ([RhClACHTUNGTRENNUNG(cod)]) were inactive 2 terms that contribute to the BDE in the framework of Mo- (IC values>100mm,seeTable1).Incontrast,all[RhACHTUNGTRENNUNG(cod)- 50 rokuma–Ziegler energy decomposition analysis[59] reveals that the main difference is attributable to the term that ac- Table1. IC valuesofantiproliferativeeffectsinMCF-7andHT-29cells 50 counts for Pauli (steric) repulsion, which is more destabiliz- and inhibition of TrxR; the results are expressed as means ((cid:4)error) of ing in the COD derivatives (av: 17kcalmol(cid:2)1) due to the repeatedexperiments. shorter NHC(cid:2)Rh distance (TableS1, the Supporting Infor- Complex IC 50 [mm] MCF-7 HT-29 TrxR mation). In contrast, the term that accounts for the stabiliz- ing covalent character of the NHC(cid:2)Rh interaction due to 1a >100 >100 >100 ACHTUNGTRENNUNG[RhClACHTUNGTRENNUNG(cod)] >100 >100 n.d. the formation of bonds (sharing of electrons) is more favor- 2 2a >100 >100 >100 able in this series (about of av. (cid:2)4kcalmol(cid:2)1) supporting 3a 0.6(cid:4)0.2 0.9(cid:4)0.0 1.8(cid:4)0.2 the observations made by 13CNMR spectroscopy (see 3b 2.9(cid:4)0.8 4.2(cid:4)0.5 1.5(cid:4)0.1 above). 3c 2.2(cid:4)1.3 2.9(cid:4)0.7 1.5(cid:4)0.4 3d 2.3(cid:4)0.3 2.7(cid:4)0.1 2.0(cid:4)0.2 The COD ligand exhibits a BDE of about 88.03kcal 4a 2.4(cid:4)0.6 6.6(cid:4)1.3 1.7(cid:4)0.3 mol(cid:2)1, and the CO ligand exhibits values of 38.68kcalmol(cid:2)1 4b 3.0(cid:4)0.8 5.6(cid:4)0.7 1.9(cid:4)0.1 (transtotheNHCmoiety)and57.03kcalmol(cid:2)1(cis).Incon- 5a[a] 7.6(cid:4)0.1 10.2(cid:4)2.3 n.d. trast, the chlorido ligand exhibits a higher BDE ((cid:3)124kcal 5b[a] 5.0(cid:4)2.3 10.8(cid:4)3.5 n.d. mol(cid:2)1 for COD derivatives, and (cid:3)145kcalmol(cid:2)1 for the di- 5c >100 97.2(cid:4)0.6 2.3(cid:4)0.4 5d 47.5(cid:4)4.0 43.6(cid:4)1.4 1.2(cid:4)0.3 carbonyl counterparts, however, denoting a more polarized situationthatpromptsitsabstractioninpolarsolvents.Thus, [a]Obvious decomposition during the assay procedure. n.d.: not deter- mined. the differences observed for the chemical behavior of sys- tems involving the methyl- and benzyl side chains, namely, compounds 3a, 5a and 3d, 5d, concerning their behavior in ACHTUNGTRENNUNG(NHC)] derivatives (3a–d and 4ab) were active at low mi- the biochemical assays (see below) and in solution (DMF cromolar concentrations against both cancer cell lines with and HO), are ascribed to steric-hindrance effects, which IC valuesinthe0.6–3.0mmrangeinMCF-7cellsandinthe 2 50 hinder the access towards the rhodium center for external 0.9–6.6mm range in HT-29 cells. The most active [RhIACHTUNGTRENNUNG(cod)] groups or molecules. Lastly, we evaluate the rotation of the complexwas3a,whichtriggeredIC valuesbelow1.0mmin 50 benzylgroup(Figure4),which seems toprotectthemetallic bothcelllines.Compound6experiencedsolubilityproblems center from solvent molecules, in relation to the methylat- under the assay conditions and was not studied further. ed-NHCligandof3a. Whereasthecarbonylcomplexes5aand5bshowedantipro- liferative effects in the low micromolar range (IC =5– 50 Biological screening as antiproliferative agents and TrxR in- 11mm), complexes 5d and 5c were low-active or inactive hibitors: Motivated by previous results on metal-NHC com- (IC values>40mm). However, for solutions of complexes 50 plexes the initial biological screening as prospective anti- 5a,b a color change and the formation of a precipitate from cancerdrugsconsistedofmeasuringthecellgrowthinhibito- theaqueoussolutionwasnotedduringthefirsthourofincu- ry activities in cultured cancer cells as well as the inhibitory bation. This clearly indicated the formation of decomposi- effectsagainsttheenzymeTrxR.[12,14–18,28,60] tionproductsthatmighthaveinfluencedtheoutcomeofthe The triggering of antiproliferative effects by the rhodiu- biologicaltests.Thecolorchangewasnotobservedforcom- m(I) derivatives was investigated in two tumor cell lines plexes5c,d and their aqueous solutions remained visually (namely, MCF-7 human breast adenocarcinoma and HT-29 unchanged for 12h. Based on these observations complexes colon carcinoma). The rhodium-free cations 1a and 2a, as 5a,bwereexcludedfromfurtherbiologicaltests. Next, the ability to inhibit TrxR was investigated. The rhodium-freeligands1aand2aexpectedlyshowednoactiv- ityupto100mm.FortheinvestigatedNHCcomplexes3a–d, 4a,b and 5c,d, IC values were observed in a narrow range 50 around 1.5mm. Hence the activity against TrxR was largely independent on the NHC ligand structure and can be con- sidered as moderate compared with other metal-NHC com- plexesinvestigatedinexactlythesameassay.[14,16,17,28] Binding to biomolecules, cellular distribution, and effects in resistant cell lines of 3a: For further extended studies, com- plex 3awasselected asthestrongestantiproliferativeagent. Metal complexes, including NHC derivatives, are known to interact with and bind to several biomolecules such as albu- Figure4.Qualitativeenergybarrierfortherotationofthebenzylgroup in3d(halfoftheNHCligandhasbeenremovedforclarity). min or DNA. Binding of 3a to albumin and DNAwas pre- 17874 www.chemeurj.org (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim Chem.Eur.J.2013,19,17871–17880 Rhodium(I)N-HeterocyclicCarbeneComplexes FULL PAPER liminarily evaluated by precipitation assays and quantifica- tion of rhodium by atomic absorption spectroscopy (see the ExperimentalSectionfordetails). After 2h of exposure to albumin, (67(cid:4)10)% of the total rhodium was bound to albumin and the values remained rather stable over an extended exposure up to 24h. Com- plexes3b and 3c were used for comparison and showed a very similar behavior (see the Supporting Information for more details). Next, DNAwas exposed to 3a as well as the platinum anticancer drug cisplatin (as a reference) at a nu- cleotide/metal ratio of 200:1 for 4h and the percentage of thetotalamountofmetalattachedtoDNAwasdetermined. Under these conditions (72(cid:4)3)% of rhodium of 3a was boundtotheDNA,whereasDNAplatinationcausedbycis- platin was (26%(cid:4)3)%. The efficient binding to DNA is in good agreement with the recent results by McAlpine etal.[31] Overall, these experiments clearly indicatethat RhI- NHC complexes can undergo strong molecular interactions with relevant biomolecules. To understand the relevance of these interactions in a more complex system, the uptake of 3a in whole HT-29 cells and into their nuclei was deter- mined. The cellular uptake was measured in a comparative manner from the serum-free cell culture medium and from the same medium with added albumin. This experimental setup enables the evaluation of the influence of albumin on rhodium bioavailability. The experiments were performed with a concentration of 3a of 1.0mm, which is related to the IC value of the complex in this cell line. As observed in 50 Figure5a,thelevelofcellularrhodiumwashigherintheex- periments with albumin-free medium than in those contain- ingalbumin.Thisindicatedthatthepresenceofserumalbu- minhasanegativeinfluenceontheuptakeprocessof3a. Figure5.a)Cellular uptake of 3a (1.0mm) from cell culture media with and without albumin into HT-29 cells; b)Rhodium levels in nuclei ex- For the evaluation of the rhodium uptake into the nuclei posedtoof3a(5.0mm). of HT-29 cells, the experimental setup was changed to re- flect the standard cell culture conditions under which 3a would trigger strong cytotoxic effects (experiments with 3a dencedbyatleast40%apoptoticcells.Accordingly,thecal- (5.0mm) in serum containing cell culture medium, see Fig- culatedIC valuesfromtheseexperimentswerecomparable 50 ure5b). In these experiments, well-detectable rhodium (Nalm-6: (0.6(cid:4)0.1)mm, Nalm-6-DNR: (0.6(cid:4)0.1)mm, Nalm- levels were reached that were higher after short exposure 6-VCR:(1.0(cid:4)0.1)mm). (1h) than after longer incubation (6h). However, these amounts corresponded to only 1–2% of the total cellular rhodium. Conclusion The cytotoxic action and apoptosis induction of many an- ticancer drugs is related to changes in the mitochondrial RhI-NHC complexes with COD or CO secondary ligands membrane potential (MMP). For 3a, moderate MMP were prepared and characterized. Both spectroscopic and changes in Nalm-6 leukemia cells were observed at concen- theoretical evaluations indicated a higher kinetic stability of trations between 0.25 and 1.5mm (<40%), whereas drasti- the COD complexes compared with the dicarbonyl ana- cally increased effects (>80%) were noted at a concentra- logues. This was further confirmed when the complexes tionof2.0mm(seeFigure6a). were subjected to biological screening as some of the CO TheDNA fragmentationby 3a asa parameter of apopto- complexes experienced major stability problems in biologi- sisinductionwasassessedinwild-type-aswellasindaunor- cal media. Further limitations for biological applications ubicin- or vincristine-resistant P-glycoprotein (P-gp) overex- were related to insufficient solubility in aqueous environ- pressing Nalm-6 cells.[16] A strong DNA fragmentation was ment, especially in the case of the larger perimidine-derived observed in all Nalm-6 cell lines in concentrations higher NHCligands. than 0.5mm. In concentrations of 0.75mm (in daunorubicin- However, complexes of the general structure resistant cells) or 1.0mm (in vincristine-resistant cells) and [RhICl(benzimidazolylidene)ACHTUNGTRENNUNG(cod)] did not experience the higher, drug resistance could be clearly overcome as evi- mentioned problems, showed pronounced antiproliferative Chem.Eur.J.2013,19,17871–17880 (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim www.chemeurj.org 17875 I.Ottetal. Total cellular-uptake studies showed that the presence of albumin lowered the rhodium uptake. Rhodium was well- detectable in isolated nuclei, however, at a low percentage of the total. Therelevance ofthe nuclear rhodium uptake is not clear at this stage. Onthe one hand, theabsolutevalues arelowcomparedwiththatofothermetallodrugsinvestigat- ed under similar experimental conditions, on the other hand, DNA targeting by the drug cisplatin also leads to a small extent of DNA platination and the low relative uptakeofmetalsintonucleiiscommon.[69–72] Complex 3a also triggered considerable changes in the mitochondrial membrane potential and led to DNA frag- mentation,whichunderlinethepotentialofthismetallodrug as an cytotoxic and apoptosis-inducing agent. Moreover, complex 3a effectively overcame P-gp-related drug resist- ance to daunorubicin and vincristine in leukemia cells, apropertyhighlydesirableinthedesignofprospectiveanti- cancerdrugs. In consequence, [RhICl(benzimidazolylidene)ACHTUNGTRENNUNG(cod)] com- plexes showed promising properties for the design of new cancer chemotherapeutics. Further exploration of this type of bioorganometallics is highly warranted and is the subject ofongoingstudies. Experimental Section General: All reagents and the solvents were used as received from Figure6.a)ChangesinthemitochondrialmembranepotentialofNalm-6 Sigma, Aldrich, or Fluka. 1HNMR spectra were recorded on a Bruker cellsafter48hincubationwith3a;b)Apoptosisinductiondeterminedas DRX-400ASNMRspectrometerand13CNMRspectraonaBrukerAV nuclearDNAfragmentationafter72hdrugexposureinwild-typeNalm- II-600ASNMRspectrometer,MSspectrawererecordedonaFinnigan 6 cells as well as daunorubicin- (Nalm-6-DNR) or vincristine-resistant MAT4515.Thepurityofthetargetcompoundswasconfirmedbyelemen- (Nalm-6-VCR)subtypes.Valuesaregivenaspercentagesofcellswithhy- talanalysis(FlashEA112,ThermoQuestItalia).Forallcompoundsun- podiploid DNA ((cid:4)SD (n=3)). Resistance to daunorubicin (DNR) and dergoingbiologicalevaluation,theexperimentalvaluesdifferedlessthan vincristine(VCR)wasconfirmedbythemissingorverylowapoptosisin- ductionafterexposuretoDNR(52.5nm)orVCR(20nm). 0.5% from the calculated ones. Perimidine and the benzimidazolium salts 1a–d were synthesized as described in the literature.[16,49] Cell cul- ture: MCF-7 breast adenocarcinoma and HT-29 colon carcinoma cells effectsinculturedtumorcells,andweremoderateinhibitors were maintained in DMEM high glucose (PAA) supplemented with 50mgL(cid:2)1gentamycinand10%(v/v)fetalcalfserum(FCS)priortouse. of TrxR. TrxR is an enzyme that is involved in several dis- Nalm-6leukemiacellsweremaintainedinRPMImediumsupplemented ease-relevant pathways and has to be considered as a possi- with1%penicillin/streptomycinand10%(v/v)FCS. blemoleculartargetforvariousmetallodrugs.[61–68]Takento- Synthesis of perimidine (2):[49] 1,8-Diaminonaphthalene (1.140g, gether, this also suggests TrxR inhibition as a relevant con- 7.0mmol) was dissolved inethanol (20mL), followed byaddition ofan tributing factor in the chemical biology of [RhIClACHTUNGTRENNUNG(cod)- excess of formic acid (1.60mL, 40mmol). The mixture was heated at ACHTUNGTRENNUNG(NHC)] metallodrugs, although it is unlikely their major refluxfor4hundernitrogenatmosphere.Theethanolicsolutionwasdi- lutedwithwaterandneutralizedbyadditionofNHOHat08C.Thepre- modeofaction. 4 cipitatewascollectedandwashed4timeswithwateranddriedat508C Because the rhodium-free cationic ligand precursors and for 12h. Yield: 1.119g (6.7mmol, 95%) light-brown powder. 1HNMR rhodium(I) in the form of [RhClACHTUNGTRENNUNG(cod)] were devoid of any (CDCl):d=3.24(brs,1H,NH),6.47(d,2H,3J =6.8Hz,ArH ),7.11 2 3 HH 4/9 activityinthebiologicalassays,itcanbespeculatedthatthe (m,4H,ArH 5(cid:2)8 ),7.23ppm(s,1H,ArH2);elementalanalysiscalcd(%) [RhI(benzimidazolylidene)] fragment represents a useful ve- forC 11 H 8 N 2 :C78.55,H4.79,N16.66;found:C78.58,H4.69,N16.30. hicletodeliverRhItobiologicaltissues. Generalprocedureforsynthesisofthedialkylperimidiniumhalides:Peri- midine(2)(1.009g,6mmol),KCO (1.037g,7.5mmol)andanexcessof More detailed studies on the most active derivative 3a 2 3 therespectivealkylhalide(18mmol)wereheatedatrefluxinacetonitrile showed effective binding to albumin and DNA. Albumin for8–14h.Thesolventwasremovedunderreducedpressureandthere- contains an accessible cysteine residue (Cys-34) as most sultant solid was resuspended in dichloromethane and filtered off to probable binding site for metal species, whereas binding to remove the formed potassium halide along with the remaining K 2 CO 3 . Thesolventofthefiltratewasremovedunderreducedpressureandre- DNAwasnotunexpectedbasedonthesimilarchemicalfea- suspendedintetrahydrofuranetoremovetheexcessofalkylhalideand tures of RhI and platinum(II) (isoelectronic, square-planar remainingperimidine.Theobtainedwhitesolidcorrespondstothepure geometry). desiredproduct. 17876 www.chemeurj.org (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim Chem.Eur.J.2013,19,17871–17880 Rhodium(I)N-HeterocyclicCarbeneComplexes FULL PAPER 1,3-Dimethylperimidinium iodide (2a): Yield: 0.622g (1.9mmol, 32%) (1,3-Dibenzylbenzimidazol-2-ylidene)chlorido(h2,h2-cycloocta-1,5-diene)- light-brownpowder.1HNMR([D]DMSO):d=3.52(s,6H-NCH),7.04 ACHTUNGTRENNUNGrhodium(I) (3d): Yield: 0.070g (0.12mmol, 35%) yellow powder. 6 3 (d, 2H, ArH, 3J =7.6Hz), 7.54 (dd, 2H, ArH, 3J =7.6Hz, 3J = 1HNMR (CDCl): d=1.76–1.87 (m, 2H, CH COD), 1.87–1.98 (m,2H, HH HH HH 3 2 8.4Hz), 7.65 (d, 2H, ArH, 3J =8.4Hz), 9.01ppm (s, 1H, ArH2); CH COD), 2.06–2.24 (m, 2H, CH COD), 2.26–2.41 (m, 2H, CH HH 2 2 2 13CNMR([D]DMSO):d=38.67(-NCH),107.64(ArC),120.35(quater- COD),3.25–3.43(m,2H,CHCOD),5.10–5.26(m,2H,CHCOD),6.20 6 3 nary ArC), 123.53 and 128.36 (ArC), 132.84 and 133.84 (quaternary (d, 2H, -NCHAr, 2J =15.8Hz), 6.31 (d, 2H, -NCHAr, 2J = 2 HH 2 HH ArC),153.32ppm(ArC2);elementalanalysiscalcd(%)forC H NI:C 15.8Hz),6.98(m,2H,ArH ,4J =3.0Hz,3J =6.3Hz),7.02(dd,2H, 13 13 2 4/7 HH HH 48.17,H4.04,N8.64;found:C47.98,H4.31,N8.50. ArH , 4J =3.3Hz, 3J =5.9Hz), 7.29–7.33 (m, 2H, -NCHArH), 5/6 HH HH 2 1,3-Diethylperimidinium iodide (2b): Yield: 0.622g (1.9mmol, 32%) 7.34–7.38 (m, 4H, -NCH 2 ArH), 7.38–7.42ppm (m, 4H, -NCH 2 ArH); light-brownpowder.1HNMR([D 6 ]DMSO):d=1.41(t,6H,-NCH 2 CH 3 , 13CNMR (CDCl 3 ): d=28.65 (CH 2 COD), 32.74 (CH 2 COD), 52.96 3 3 2 J J H H H H H , = = 7 7 A . . 1 7 rH H H , z z ) ) , , 3 4 7 J . H . 0 5 H 3 2 = ( ( q 8 d , . d 3 4 , H H 2 z , H ) - , , N A C 8 r H . H 9 2 5 C , p 3 H J p H 3 m , H 3 = J H 7 ( H . s 7 , = H 7. 1 z 1 , H H 3 , J z H ) H , A = 7 r .1 8 H 2 .3 2) ( H ; d, z) 2 1 , 3 H C 7 , .6 N A 2 M r ( H d R , , ( C ( 1 - O N 8C C D , H , A 1 2 J A r R C h r C ) ) , , = 1 6 6 9 9 . 7 . 6 1 .8 9 H 3 ( z p d ), p , 1 m c 1 is 0 ( - . d C 9 , 3 H A ,1 r C 2 C O 2 2 . , 4 D 1 8 J , , R 1 1 h J C 2 R 7 = hC .0 5 = 8 1 , . 1 0 1 4 2 H .3 7 z . H 8 ) 2 ; z , M ) 1 , 2 S 1 8 ( 0 E . 0 9 . I 0 5 ) , 3 : 1 5 ( 3 4 d 5 4 , .0 [ t 1 r M a , n 1 + s 3 ] - 6 ; C . e 1 H l 5 - ([D 6 ]DMSO): d=11.82 (-NCH 2 CH 3 ), 46.72 (-NCH 2 CH 3 ), 107.65 (ArC), emental analysis calcd (%) for C 29 H 30 N 2 ClRh: C 63.92, H 5.55, N 5.14; 121.13 (quaternary ArC), 123.39 and 128.31 (ArC), 131.43 and 134.43 found:C64.14,H5.78,N5.26. (quaternary ArC), 152.36ppm (s, ArC2); elemental analysis calcd (%) Chlorido(h2,h2-cycloocta-1,5-diene)(1,3-dimethylperimidin-2-ylidene)- for C H NI: C 51.15, H 4.86, N 7.95; found: C 51.08, H 4.49, N 7.55. ACHTUNGTRENNUNGrhodium(I) (4a): Yield: 0.084g (0.19mmol, 61%) yellow powder. 15 17 2 Generalprocedureforsynthesisofthe[RhACHTUNGTRENNUNG(cod)ACHTUNGTRENNUNG(NHC)]derivatives:The 1HNMR (CDCl 3 ): d=1.94–2.04 (m, 4H, CH 2 COD), 2.44–2.48 (m,4H, respective benzimidazolium- or perimidinium halide (0.32mmol) and CH 2 COD),3.40(m,2H,CHCOD),4.54(s,6H,-NCH 3 ),5.08(m,2H, AgO (0.0371g, 0.16mmol) were added to a dried Schlenk tube. The CHCOD),6.67(m,2H,ArH),7.30–7.40ppm(m,4H,ArH);13CNMR 2 mixture was backflashed three times with N 2 and then dry CH 2 Cl 2 was (CDCl 3 ):d=28.86(CH 2 COD),32.52(CH 2 COD),43.16(-NCH 3 ),70.06 added(15mL).Theflaskwasclosedandstirredfor4hinthedark.Aso- (d, cis-CH COD, 1J RhC =14.6Hz), 98.22 (d, trans-CH COD, 1J RhC = lution of bisACHTUNGTRENNUNG[chlorido(h2,h2-cycloocta-1,5-diene)rhodium(I)] (0.078g, 6.8Hz), 104.35 (ArC), 119.71 (quaternary ArC), 121.04 and 127.76 2 0 – .1 1 6 2 m h m in ol) th in e C d H ar 2 k C . l 2 T w h a e s o a b d t d a e i d ne ( d 10 s m us L p ) en a s n i d on th w e a s s olu fi t l i t o e n re w d as ov s e ti r rre C d el f i o te r ( 4 A 8. r 6 C H ), z) 1 ; 34 M .17 S( a E n I d ): 13 m 4 / . z 96 : ( 4 q 4 u 2 at [ e M rn + a ] r ; y e A le rC m ) e , n 2 t 1 a 1 l .6 a 3 n p al p y m sis (d c , a A lc r d C2 ( , % 1J ) RhC fo = r (281nm)andconcentratedinavacuum.Theyellowresiduewasrecrys- C 21 H 24 N 2 ClRh:C51.99,H5.65,N7.13;found:C51.68,H5.59,N6.97. tallizedfromCHCl/n-hexane(10/40mL)at48C. Chlorido(h2,h2-cycloocta-1,5-diene)(1,3-diethylperimidin-2-ylidene)- 2 2 Chlorido(h2,h2-cycloocta-1,5-diene)(1,3-dimethylbenzimidazol-2-ylidene)- ACHTUNGTRENNUNGrhodium(I) (4b): Yield: 0.072g (0.15mmol, 48%) brown powder. ACHTUNGTRENNUNGrhodium(I) (3a): The preparation of this compound has been reported 1HNMR (CDCl 3 ): d=1.59 (t, 6H, -NCH 2 CH 3 , 3J HH =7.0Hz), 1.86–2.10 previously.[50] Yield: 0.084g (0.21mmol, 69%) yellow powder. 1HNMR (m, 4H, CH 2 COD), 2.33–2.57 (m, 4H, CH 2 COD), 3.32 (m, 2H, CH (CDCl 3 ): d=2.01–2.12 (m, 4H, CH 2 COD), 2.44–2.57 (m, 4H, CH 2 COD),4.68(m,2H,-NCH 2 CH 3 ),5.06(m,2H,CHCOD),6.29(m,2H, COD),3.40(m, 2H, CHCOD), 4.35(s, 6H, -NCH 3 ), 5.20(m,2H,CH -NCH 2 CH 3 ), 6.67 (dd, 2H, ArH, 4J HH =1.2Hz, 3J HH =7.3Hz), 7.27– COD), 7.23 (m, 2H, ArH 4/7 ), 7.28ppm (m, 2H, ArH 5/6 ); 13CNMR 7.35ppm(m,4H,ArH);13CNMR(CDCl 3 ):d=11.59(-NCH 2 CH 3 ),28.85 ( ( d C , D c C is l 3 - ) C : H d= C 2 O 8. D 85 , ( 1 C J R H hC 2 = C 1 O 4 D .9 ) H ,3 z 2 ), .97 10 ( 0 C .3 H 5 2 C (d O , D tr ) a , n 3 s 4 -C .6 H 4(- C N O C D H , 3 ) 1 , J 6 R 8 hC .3 = 5 C (C O H D 2 , C 1J O Rh D C ) = , 1 3 4 2 .6 .5 H 1 z ( ) C , H 97 2 .7 C 3 O ( D d, ), tr 4 a 9 n . s 9 - 6 CH (-N C C O H D 2 C , H 1J 3 R ) h , C = 70 6 .4 .8 7 H ( z d ) , , c 1 is 0 - 5 C .1 H 1 6.8Hz), 109.25 and 122.38ppm (ArC), 135.32 (quaternary ArC), (ArC), 120.75 (quaternary ArC), 120.94 and 127.63 (ArC), 133.25 and 1 a 9 n 6 a . l 1 y 4 sis pp c m alc ( d d, (% Ar ) C f 2 o , r 1J C Rh 1 C 7 H = 22 5 N 0. 2 6 C H l 2 R z) h ; : M C S 5 ( 1 E .9 I 9 ): , m H /z 5 : .6 3 5 9 , 2 N [M 7. + 1 ] 3 ; ; e f l o e u m n e d n : ta C l m 13 / 4 z . : 85 47 ( 0 qu [M ate + r ] n ; a e r l y em A e r n C t ) a , l 2 a 1 n 1 a .6 ly 5 s p is p c m alc (d d , ( A % r ) C2 f , or 1J C Rh 2 C 3 H = 2 4 8 8 N .5 2 C H lR z) h ; : M C S 5 ( 8 E . I 6 ) 7 : 51.68,H5.59,N6.97. H5.99,N5.95;found:C58.41,H5.69,N5.70. Chlorido(h2,h2-cycloocta-1,5-diene)(1,3-diethylbenzimidazol-2-ylidene)- General procedure for the synthesis of rhodium dicarbonyl derivatives: ACHTUNGTRENNUNGrhodium(I) (3b): The preparation of this compound has been reported Therespective[RhACHTUNGTRENNUNG(cod)] derivativewas added toadried Schlenktube. previously.[16] Yield: 0.107g (0.22mmol, 72%) yellow powder. 1HNMR The complex was back flashed three times with carbon monoxide and (CDCl 3 ): d=1.63 (t, 6H, -NCH 2 CH 3 , 3J HH =7.3Hz), 1.96–2.06 (m, 4H, then dry CH 2 Cl 2 was added (15mL). The solution was stirred under CH COD),3.41–3.51(m,4H,CH COD),3.37(m,2H,CHCOD),4.81 a constant flux of CO for 30min. The solution was concentrated in 2 2 (dq, 2H, -NCHCH, 3J =7.3Hz, 2J =14.4Hz), 5.05 (dq, 2H vacuum((cid:3)5mL)andprecipitatedbyadditionofn-hexane((cid:3)30mL)at 2 3 HH HH -NCHCH, 3J =7.3Hz, 2J =14.4Hz), 5.14 (m, 2H, CH COD), 7.21 48C. 2 3 HH HH (dd,2H,ArH ,4J =3.1Hz,3J =6.0Hz),7.32ppm(dd, 2H,ArH , Dicarbonylchlorido(1,3-dimethylbenzimidazol-2-ylidene)ACHTUNGTRENNUNGrhodium(I) (5a): 4/7 HH HH 5/6 4J =3.1Hz, 3J =6.0Hz); 13CNMR (CDCl): d=14.96 (-NCHCH), Yield: 0.021g (0.06mmol, 48%) yellow powder. 1HNMR (CDCl): d= HH HH 3 2 3 3 28.79 (CH COD), 32.90 (CH COD), 43.60 (-NCHCH), 68.58 (d, cis- 4.14 (s, 6H, -NCH), 7.40 (m, 4H, ArH), 13CNMR (CDCl): d=35.22 2 2 2 3 3 3 CH COD, 1J =14.2Hz), 99.93 (d, trans-CH COD, 1J =6.7Hz), (-NCH), 110.40 and 123.69 (ArC), 134.85 (quaternary ArC), 182.40 (d, RhC RhC 3 109.88and122.11(ArC),134.44(quaternaryArC),195.23ppm(d,ArC2, cis-CO, 1J =74.2Hz), 185.21, 185.30 (d, trans-CO, 1J =53.6Hz), RhC RhC 1J =50.6Hz); MS (EI): m/z: 420 [M+]; elemental analysis calcd (%) 185.36ppm (d, NHC, 1J =43.1Hz); FTIR: n˜=2074 (n˜(CO) ), RhC RhC sym for C H NClRh: C 52.50, H 5.87, N 5.83; found: C 52.64, H 5.72, N 1989cm(cid:2)1(v˜(CO) );MS(EI):m/z:318[M+(cid:2)CO];elementalanalysis 19 26 2 asym 5.65. calcd(%)forC H ClNORh:C38.79,H2.96,N8.23;found:C39.20, 11 10 2 2 Chlorido(h2,h2-cycloocta-1,5-diene)(1,3-diisopropylbenzimidazol-2- H3.10,N7.89. ylidene)ACHTUNGTRENNUNGrhodium(I)(3c):Thepreparationofthiscompoundhasbeenre- Dicarbonylchlorido(1,3-diethylbenzimidazol-2-ylidene)ACHTUNGTRENNUNGrhodium(I) (5b): ported previously.[53] Yield: 0.087g (0.17mmol, 55%) yellow powder. The preparation of this compound has previously been reported.[52] 1HNMR (CDCl): d=1.70 (d, 6H, -NCHACHTUNGTRENNUNG(CH) 3J =7.1Hz), 1.79 (d, Yield: 0.016g (0.04mmol, 33%) yellow powder. 1HNMR (CDCl): d= 3 32, HH 3 6H,-NCHACHTUNGTRENNUNG(CH) 3J =7.1Hz),1.94–2.04(m,4H,CH COD),2.39–2.49 1.58 (t, 6H, 3J =7.3Hz, -NCHCH), 4.62 (dq, 2H, -NCHCH, 3J = 32, HH 2 K 2 3 2 3 HH (m,4H,CH COD),3.46 (m,2H,CHCOD),5.10 (m,2H,CHCOD), 7.3Hz, 2J =14.4Hz), 4.75 (dq, 2H, -NCHCH, 3J =7.3Hz, 2J = 2 HH 2 3 HH HH 6.49(hept,2H,-NCHACHTUNGTRENNUNG(CH),3J =7.1Hz),7.14(dd,2H,ArH ,4J = 14.4Hz),7.36(dd,2H,ArH ,4J =3.1Hz,3J =6.1Hz),7.45(dd,2H, 32 HH 4/7 HH 4/7 HH HH 3.1Hz, 3J =6.1Hz), 7.50ppm (dd, 2H, ArH , 4J =3.1Hz, 3J = ArH , 4J =3.1Hz, 3J =6.1Hz); 13CNMR (CDCl): d=14.79 HH 5/6 HH HH 5/6 HH HH 3 6.1Hz); 13CNMR (CDCl): d=20.88 and 21.88 (-NCHACHTUNGTRENNUNG(CH)), 28.76 (-NCHCH),43.89(-NCHCH),110.77and123.47 (ArC),134.02(qua- 3 32 2 3 2 3 (CH COD), 32.88 (CH COD), 53.87 (-NCHACHTUNGTRENNUNG(CH)), 67.81 (d, cis-CH ternaryArC),182.61(d,cis-CO,1J =72.2Hz),184.26(d,NHC,1J = 2 2 32 RhC RhC COD, 1J =14.5Hz), 99.17 (d, trans-CH COD, 1J =6.7Hz), 112.08 42.5Hz), 185.30ppm (d, trans-CO, 1J =53.8Hz); FTIR: 2077 RhC RhC RhC and121.51(ArC).133.64(quaternaryArC),194.00ppm(d,ArC2,1J = (v(CO) ), 1991cm(cid:2)1 (v(CO) ); MS (EI): m/z: 340 [M+(cid:2)CO]; ele- RhC sym asym 50.3Hz); MS (EI): m/z: 448 [M+]; elemental analysis calcd (%) for mentalanalysiscalcd(%)forC H ClNORh:C42.36,H3.83,N7.60; 13 14 2 2 C H NClRh:C54.33,H6.34,N5.51;found:C54.14,H6.33,N5.23. found:C41.98,H3.55,N7.53. 21 30 2 Chem.Eur.J.2013,19,17871–17880 (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim www.chemeurj.org 17877 I.Ottetal. Dicarbonylchlorido(1,3-diisopropylbenzimidazol-2-ylidene)ACHTUNGTRENNUNGrhodium(I) TrxR inhibition assay: To determine the inhibition of TrxR, an estab- (5c):Thepreparationofthiscompoundhaspreviouslybeenreported.[53] lishedmicroplatereader-basedassaywasperformedwithminormodifica- Yield: 0.016g (0.04mmol, 37%) yellow powder. 1HNMR (CDCl): d= tions.[16]Forthispurpose,commerciallyavailableratrecombinatedTrxR1 3 1.72(dd,12H,-NCHACHTUNGTRENNUNG(CH),4J =1.4Hz,3J =7.0Hz),5.77(hept,2H, (from IMCO Corporation) was used and diluted with distilled water to 32 HH HH -NCHACHTUNGTRENNUNG(CH), 3J =7.0Hz), 7.29 (dd, 2H, ArH , 4J =3.2Hz, 3J = achieveaconcentrationof0.01UmL(cid:2)1.Thecompoundswerefreshlydis- 32 HH 4/7 HH HH 6.2Hz), 7.62 (dd, 2H, ArH , 4J =3.2Hz, 3J =6.2Hz)ppm; solved as stock solutions in DMF. Potassium phosphate buffer pH7.0 5/6 HH HH 13CNMR (CDCl): d=20.91 and 21.19 (-NCHACHTUNGTRENNUNG(CH)), 54.58 (-NCH- (25mL) containing the compounds in graded concentrations or vehicle 3 32 ACHTUNGTRENNUNG(CH)), 112.97 and 122.84 (ArC), 133.41 (quaternary ArC), 182.58 (d, (DMF) without compounds (control probe) were added to each aliquot 32 cis-CO,1J =74.4Hz),182.89(d,ArC2,1J =42.4Hz),185.52ppm(d, oftheenzymesolution(25mL)andtheresultingsolutions(finalconcen- RhC RhC trans-CO, 1J =53.7Hz); FTIR: n˜=2086 (v(CO) ), 1990cm(cid:2)1 tration of DMF: 0.5% v/v) were incubated with moderate shaking for RhC sym (v(CO) asym ); MS (EI): 368 [M+(cid:2)CO]; elemental analysis calcd (%) for 75minat378Cina96-wellplate.Thereactionmixture(225of1000mL; C 15 H 18 ClN 2 O 2 Rh:C45.42,H4.57,N7.06;found:C45.37,H4.49,N6.88. the reaction mixture consisted of 500mL potassium phosphate buffer Dicarbonylchlorido(1,3-dibenzylbenzimidazol-2-ylidene)ACHTUNGTRENNUNGrhodium(I) (5d): pH7.0, 80mL 100mm EDTA solution pH7.5, 20mL BSA solution Yield: 0.013g (0.03mmol, 28%) yellow powder. 1HNMR (CDCl): d= 0.05%,100mLof20mmNADPHsolution,and300mLofdistilledwater) 3 5.83 (d, 2H, -NCHAr, 2J =15.7Hz), 6.05 (d, 2H, -NCHAr, 2J = wasaddedtoeachwell,andthereactionstartedbyadditionofan20mm 2 HH 2 HH 15.7Hz),7.25–7.17(m,4H,ArH),7.41–7.29ppm(m,10H,-NCHArH); ethanolic 5,5’-dithiobis(2-nitrobenzoic acid) solution (25mL). After 2 13CNMR (CDCl): d=53.02 (-NCHAr), 111.67, 123.78, 128.29, 128.98, proper mixing,theformationof 5-thio-2-nitrobenzoicacid(5-TNB)was 3 2 132.55, 135.05(18C,ArC),182.04(d,cis-CO,1J =74.2Hz),185.06 (d, monitoredwithamicroplatereader(PerkinElmerVictorX4)at405nm RhC trans-CO, 1J =53.8Hz), 186.93ppm (d, ArC2, 1J =43.5Hz). FTIR: in 10s intervals for 6min. The increase in 5-TNB concentration over RhC RhC n˜=2076(v(CO) ),1994cm(cid:2)1(v(CO) );MS(EI):464[M+(cid:2)CO];ele- timefollowedalineartrend(r2(cid:5)0.99),andtheenzymaticactivitieswere sym asym mentalanalysiscalcd(%)forC H ClNORh:C56.06H3.68N5.69;C calculatedastheslopes(increaseinabsorbancepersecond)thereof.For 23 18 2 2 56.34,H3.75,N5.77. each tested compound, the noninterference with the assay components Dicarbonylchlorido(1,3-diethylperimidin-2-ylidene)ACHTUNGTRENNUNGrhodium(I) (6): Yield: wasconfirmedbyanegativecontrolexperimentbyusinganenzyme-free 0.012g(0.03mmol, 27%)yellow powder.1HNMR(CDCl 3 ):d=1.51(t, solution. The IC 50 values were calculated as the concentration of com- 6H, -NCHCH, 3J =7.1Hz), 4.76 (m, 2H, -NCHCH), 5.03 (m, 2H, pounddecreasingtheenzymaticactivityoftheuntreatedcontrolby50% 2 3 HH 2 3 -NCHCH), 6.73 (dd, 2H, ArH, 4J =2.3Hz, 3J =6.3Hz), 7.40– andaregivenasthemeansanderrorofrepeatedexperiments. 2 3 HH HH 7.34ppm(m,4H,ArH);13CNMR(CDCl):d=11.06(-NCHCH),50.53 Bindingtoalbumin:Proteinbindingwasstudiedaccordingtoarecently 3 2 3 (-NCHCH),105.80(ArC),121.41(quaternaryArC),121.86and127.79 described method with minor modifications (see refs.[82] and [83]). 2 3 (ArC), 133.20 and 135.00 (quaternary ArC), 182.43 (d, cis-CO, 1J = Bovine serum albumin (BSA, 400mg, Sigma Aldrich) was dissolved in RhC 75.4Hz), 185.56 (d, trans-CO, 1J =54.4Hz), 199.29ppm (d, NHC, Dulbecco’smodifiedEagle(cid:6)smedium(DMEM)highglucose(PAA)cell RhC 1J =41.3Hz); FTIR: n˜=2079 (v(CO) ), 2000cm(cid:2)1 (v(CO) ); MS culturemedium(10.0mL)supplementedwithgentamycin(50mgL(cid:2)1).A RhC sym asym (EI): 390 [M+(cid:2)CO]; elemental analysis for calcd (%) for stocksolution(10mLof5.0mm)ofthecorrespondentcomplexinDMF C H ClNORh:C48.77H3.85N6.69;found:C49.04,H3.99,N6.97. was added to the BSA-treated medium and incubated at 378C in the 17 16 2 2 Computational details: Relativistic density functional theory calcula- dark under gentle shaking. After 0, 1, 2, 4, 6, 12, and 24h, an aliquot tions[73] were carried out by using the ADF 2012 code[74] incorporating (250mL) thereof was taken, treated with of cold ((cid:2)208C) ethanol scalar corrections through the ZORA Hamiltonian.[75,76] Triple-x Slater (500mL)andstoredat(cid:2)208Cfor2htoallowanoptimalprecipitationof basis set plus two polarization functions (STO-TZ2P) for valence elec- theproteinfraction.Afterwardsthesolutionwascentrifugedat400gfor trons were employed within the generalized gradient approximation 5minat48C,analiquotofthesupernatant(400mL)wastaken andthe (GGA) according to the Perdew–Burke–Ernzerhof (PBE) nonlocal ex- rhodium content was measured by atomic absorption spectroscopy change-correlation functional.[77,78] The frozen core approximation was (AAS;seebelow).Theresultswereexpressedaspercentageofrhodium appliedtothe[1s2–3d10]coreforRh,and[1s2]forC,N,andCl,leaving boundtoalbuminandaregivenasthemeansanderrorofrepeatedex- the remaining electrons to be treated variationally. Geometry optimiza- periments. tions were done without any symmetry restrain through the analytical BindingtoDNA:Experimentswereperformedaccordingtoapreviously energy gradient method implemented by Verluis and Ziegler.[79] Bond- describedmethod.[69]Salmontestes DNA (Sigma)was dissolvedinPBS dissociation energies are corrected to the basis set superposition error pH7.4 and the drugs were added as stock solutions in DMF. The final (BSSE). solutionscontained3a(4.0mm),salmontestesDNA(247.5mgmL(cid:2)1)and NMR properties were evaluated by employing the gauge-including 0.1%(v/v)DMF.Aftervortexing,thesolutionswereincubatedat378C atomic orbitals (GIAO) methodology[80] as implemented in the NMR for2h.Aliquots(200mL)weremixedwith0.9msodiumacetate(100mL) propertymoduleoftheADFcodeinconjunctiontotheGGAexchange andice-coldethanol(900mL).Sampleswerestoredat(cid:2)208Cfor30min. expressionproposedbyHandyandCohen[81]andthecorrelationexpres- Thepelletswereisolatedbycentrifugation(2390g,10min,48C)andre- sionproposedbyPerdew,Burke,andErnzerhof(OPBE).[77,78]Therela- suspended in 0.3m sodium acetate (300mL). Ice-cold ethanol (900mL) tive13CNMRparameters(d=s TMS(cid:2)s calcd)werecalculatedbyusingthe was then added, and the precipitate was collected after centrifugation C C GIAOOPBE/ZORAleveloftheoryandreferencedtotetramethylsilane (2390g,10min,48C).Sampleswere washedtwice with ice-cold ethanol (TMS,s TMS=191.19ppm). andstoredat(cid:2)208C.Thepelletsweredissolvedinwater(500mL,twice C AntiproliferativeeffectsinMCF-7andHT-29cells:Theantiproliferative distilled),andtheDNAcontentwasdeterminedbyabsorptionreadingat effectsofthecompoundsweredeterminedfollowinganestablishedpro- 260nm in a microplate reader (PerkinElmer Victor X4). Salmon testes cedure. In short, cells were suspended in cell culture medium (HT-29: DNAdissolvedinwaterwereusedforcalibrationpurposes.Therhodium 3000cellsmL(cid:2)1, MCF-7: 10000cellsmL(cid:2)1), and 100mL aliquots thereof from the samples was determined by AAS. The amount of rhodium were plated in 96-well plates and incubated at 378C: 5% CO for 48h boundtoDNAwasexpressedaspercentageoftotalrhodiuminthesam- 2 (HT-29)or72h(MCF-7).Stocksolutionsoftherutheniumcomplexesin ples.Resultswerecalculatedasmeansofthreeindependentexperiments, DMFwerefreshlypreparedanddilutedwithcellculturemediumtothe whichwereperformedwithtworeplicates. desired concentrations (final DMF concentration: 0.1% v/v). The Uptakeintothecells:Forcellular-uptakestudiesHT-29cellsweregrown medium in the plates was replaced with medium containing the com- untilatleast70%confluencyin75cm2cellcultureflasks.Stocksolutions pounds in graded concentrations (six replicates, 200mL per well). After ofcomplex3ainDMFwerefreshlypreparedanddilutedwithcellcul- furtherincubationfor72h(HT-29)or96h(MCF-7)thecellbiomasswas ture medium to the desired concentrations (final DMF concentration: determinedbycrystalvioletstainingandtheIC valuesweredetermined 0.1%v/v,finalcomplexconcentration:1.0mm).Thecellculturemedium 50 asthoseconcentrationscausing50%inhibitionofcellproliferation.Re- of the cell culture flasks was replaced with 10mL of the cell culture sultsarerepresentedasmeansofindependentexperiments. medium (with or without bovine serum albumin (40mg per mL)) solu- 17878 www.chemeurj.org (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim Chem.Eur.J.2013,19,17871–17880 Rhodium(I)N-HeterocyclicCarbeneComplexes FULL PAPER tions containing 3a and the flasks were incubated for 1, 4, and 6h at PBS2%(v/v)formaldehydeonicefor30min.Afterfixation,cellswere 378C/5%CO.Afterwardsthemediumwasremovedandthecellswere incubated with ethanol/PBS (2:1, v/v) for 15min, pelleted, and resus- 2 washed with phosphate-buffered saline pH7.4. After trypsinization, cell pended in PBS containing RNase A (40mgmL(cid:2)1). After incubation for pelletswereisolatedbycentrifugation,resuspendedin1–5mLtwice-dis- 30minat378C,cellswerepelletedagainandfinallyresuspendedinPBS tilled water, lysed by ultrasonication and appropriately diluted using containing propidium iodide (50mgmL(cid:2)1). Nuclear DNA fragmentation twice-distilled water. The rhodium content of the samples was deter- was then quantified by flow cytometric determination of hypodiploid minedbyAAS(seebelow)andtheproteincontentofseparatealiquots DNA. Data were collected and analyzed by using a FACScan (Becton wasdeterminedbytheBradfordmethod. Dickinson, Heidelberg, Germany) equipped with the CELLQuest soft- Uptakeintothenuclei:ThenucleiofHT-29cellswereisolatedaccording ware. Data are given in % hypoploidy (subG1), which reflects the to previously described procedure[69,71] with some modifications: Cells numberofapoptoticcells. were grownin 175cm2cell culture flasks untilat least 70% confluency. The cell culture medium was removed and replaced with cell culture medium (20mL) containing 5mm of 3a. The flasks were incubated at 378C/5% CO for 1 and 6h and the cell pellets were isolated as de- 2 scribed above. After centrifugation (923g, 5min) the pellet was resus- pendedin0.9%NaClsolution(1.0mL)andanaliquot(200mL)wasre- Acknowledgements moved before nuclei isolation to determine the total cellular rhodium uptake (see above). For nuclei isolation, the cell suspension was centri- ThefinancialsupportbyDFG(granttosupporttheinitiationofinterna- fuged at 923g for 5min and resuspendedin 300mL RSB-1 (0.01m Tris- tional collaboration), FONDECYT 11100027, FCI (Fonds der Chemi- HCl, 0.01m NaCl, 1.5mm MgCl, pH7.4) and left for 10min in an ice schen Industrie), Dr. Kleist-Foundation, Berlin, Dr. Koch-Foundation, 2 bath. Swollen cells were centrifuged (1231g, 5min), resuspended in Berlin, as well as the excellent technical support by Davida Fernandez 300mL of RSB-2 (RSB-1 containing each 0.3% v/v Nonidet-P40 and Fernandezaregratefullyacknowledged. sodiumdesoxycholate)andhomogenizedby10–15up/down-pushesusing a dounce homogenizer. The resultant homogenate was centrifuged at 3583g for 5min and the resulting crude nuclei were taken up in 0.25m [1] G. Gasser,I. Ott, N. Metzler-Nolte, J. Med. Chem. 2011, 54, 3–25. sucrose (150mL) containing CaCl 2 (3mm). The suspension was under- [2] M.Patra,G.Gasser,ChemBioChem2012,13,1232–1252. layedwith 0.88msucrose(150mL)andcentrifuged for10minat3583g. [3] C.G. Hartinger, N. Metzler-Nolte, P.J. Dyson, Organometallics The nuclei pellets were stored at (cid:2)208C or immediately dissolved in 2012,31,5677–5685. EDTA solution (200mL of 10gL(cid:2)1) and disrupted by ultrasonification. [4] A.L. Noffke, A. Habtemariam, A.M. Pizarro, P.J. Sadler, Chem. Therhodiumcontent ofthe samples was determined by high-resolution Commun.2012,48,5219–5246. continuum source atomic absorption spectrometry (HR-CS AAS, see [5] L. Oehninger, R. Rubbiani, I. Ott, Dalton Trans. 2013, 42, 3269– below)andthenuclearproteincontentwasdeterminedbytheBradford 3284. method.Resultsareexpressedasmeansoftwoindependentexperiments [6] W.Liu,R.Gust,Chem.Soc.Rev.2013,42,755–773. aspmoldrugpermgnuclearprotein.Theratioofdrugtakenupintothe [7] A.Gautier,F.Cisnetti,Metallomics2012,4,23–32. nucleusiscalculatedaspercentageofthetotalrhodiumcontentofeach [8] A.Kascatan-Nebioglu,M.J.Panzner,C.A.Tessier,C.L.Cannonb, pellet. W.J.Youngs,Coord.Chem.Rev.2007,251,884–895. AAS measurements: AAS measurements were performed using a con- [9] F.Hackenberg,G.Lally,H.M(cid:3)ller-Bunz,F.Paradisi,D.Quaglia,W. trAA700high-resolutioncontinuumsourceatomicabsorptionspectrom- Streciwilk,M.Tacke,Inorg.Chim.Acta2013,395,135–144. eter(HRCS-AAS)forthequantificationofrhodiumandplatinum.Rho- [10] S.Patil,K.Dietrich,A.Deally,B.Gleeson,H.M(cid:3)ller-Bunz,F.Para- dium was detected at a wavelength of 343.4893nm, whereas platinum disi,M.Tacke,Helv.Chim.Acta2010,93,2347–2364. was measured at 265.9450nm. Standards for calibration purposes were [11] J.G.Leid,A.J.Ditto,A.Knapp,P.N.Shah,B.D.Wright,R.Blust, prepared in a matrix-matched manner (in ethanol in case of albumin L. Christensen, C.B. Clemons, J.P. Wilber, G.W. Young, A.G. binding,in100mgmL(cid:2)1aqueousDNAsolutionincaseofDNAbinding, Kang, M.J. Panzner,C.L. Cannon, Y.H. Yun, W.J. Youngs, N.M. andcellularmatrixincaseofcellularandnuclearuptakeexperiments,re- Seckinger,E.K.Cope,J.Antimicrob.Chemother.2012,67,138–148. [12] J.L. Hickey, R.A. Ruhayel, P.J. Barnard, M.V. Baker, S.J. Bern- spectively) using a standard stock solution of 3a or cisplatin. Triton X- ers-Price,A.Filipovska,J.Am.Chem.Soc.2008,130,12570–12571. 100 (20mL, 1%) and HNO (20mL, 13%) were added to each 200mL 3 [13] S.J.Berners-Price,A.Filipovska,Metallomics2011,3,863–873. sampleorstandardsolution.Avolumeof25mLthereofwasinjectedinto [14] R.Rubbiani,S.Can,I.Kitanovic,H.Alborzinia,M.Stefanopoulou, thegraphitetubes.Drying,pyrolysis,andatomizationinthegraphitefur- M.Kokoschka,S.Mçnchgesang,W.S.Sheldrick,S.Wçlfl,I.Ott,J. nacewereperformedaccordingtotheconditionspreviouslyreportedfor Rh[40]and Pt[84]with minor modifications. Themean absorbances ofdu- Med.Chem.2011,54,8646–8657. [15] E.Schuh,C.Pfl(cid:3)ger,A.Citta,A.Folda,M.P.Rigobello,A.Bindoli, plicateinjectionswereusedthroughoutthestudy. A.Casini,F.Mohr,J.Med.Chem.2012,55,5518–5528. Measurement of the mitochondrial membrane potential: Nalm-6 cells [16] R.Rubbiani,I.Kitanovic,H.Alborzinia,S.Can,A.Kitanovic,L.A. were incubated with different concentrations of complex3a for 48h at Onambele, M. Stefanopoulou, Y. Geldmacher, W.S. Sheldrick, G. 378Cand5%CO.Afterwards,1mLofthecellsuspensionwas centri- 2 Wolber,A.Prokop,S.Wçlfl,I.Ott,J.Med.Chem.2010,53,8608– fuged at 3000rpm at 48C for 5min and the supernatant was removed 8618. carefully. JC-1 (6.25mL, 0.2mgmL(cid:2)1 in DMSO) was added to every [17] R. Rubbiani, E. Schuh, A. Meyer, J. Lemke, J. Wimberg, N. Met- sample and vortexed briefly. All samples were incubated for 30min at zler-Nolte,F.Meyer,F.Mohr,I.Ott,MedChemComm2013,4,942– 378C, 300rpm and vortexed briefly every 10min. After incubation the 948. samples were centrifuged at 4000rpm, 48C for 5min. The supernatant [18] O.Rackham,A.-M.J.Shearwood,R.Thyer,E.McNamara,S.M.K. was removed carefully and the pellet resuspended in PBS (200mL) and Davies, B.A. Callus, A. Miranda-Vizuete, S.J. Berners-Price, Q. analyzed using a FACScan (Becton Dickinson, Heidelberg, Germany) Cheng, E.S.J. Arn(cid:8)r, A. Filipovska, Free Radical Biol. Med. 2011, equippedwiththeCELLQuestsoftware. 50,689–699. Measurement of DNA fragmentation: Apoptotic cell death was deter- [19] M. Baker, P.J. Barnard, S.J. Berners-Price, S.K. Brayshaw, J.L. mined by a modifiedcell cycleanalysis, which detects DNA fragmenta- Hickey,B.W.Skelton,A.H.White,DaltonTrans.2006,3708–3715. tion at the single-cell level. For measurement of DNA fragmentation, [20] M.Skander,P.Retailleau,B.Bourrie,L.Schio,P.Mailliet,A.Ma- cellswereseededatadensityof1(cid:7)105cellsmL(cid:2)1andtreatedwithdiffer- rinetti,J.Med.Chem.2010,53,2146–2154. entconcentrationsof3a.After72hofincubation,cellswerecollectedby [21] S.D.Adhikary,D.Bose,P.Mitra,K.D.Saha,V.Bertolasi,J.Dinda, centrifugation at 30g for 5min, washed with PBS at 48C, and fixed in NewJ.Chem.2012,36,759–767. Chem.Eur.J.2013,19,17871–17880 (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim www.chemeurj.org 17879 I.Ottetal. [22] R. Wai-YinSun, A. Lok-FungChow, X.-H. Li, J.J. Yan, S. Sin-Yin [53] B. Hildebrandt, S. Raub, W. Frank, C. Ganter,Chem. Eur. J. 2012, Chui,C.-M.Che,Chem.Sci.2011,2,728–736. 18,6670–6678. [23] G. Alves, L. Morel, M. El-Ghozzi, D. Avignant, B. Legeret, L. [54] C.A.Tolman,Chem.Rev.1977,77,313–348. Nauton, F. Cisnetti, A. Gautier, Chem. Commun. 2011, 47, 7830– [55] W.A.Herrmann,J.Sch(cid:3)tz,G.D.Frey,E.Herdtweck,Organometal- 7832. lics2006,25,2437–2448. [24] T.Zou,C.-N.Lok,Y.M.E.Fung,C.-M.Che,Chem.Commun.2013, [56] K.Nakamoto,InfraredandRamanSpectraofInorganicandCoordi- 49,5423–5425. nation Compounds, Applications in Coordination, Organometallic, [25] C.-H.Wang,W.-C.Shih,H.C.Chang,Y.-Y.Kuo,W.-C.Hung,T.-G. andBioinorganicChemistry,Wiley-VCH,Weinheim,2009. Ong,W.-S.Li,J.Med.Chem.2011,54,5245–5249. [57] P.S. Pregosin, NMR in Organometallic Chemistry, Wiley-VCH, [26] S.Ray,R.Mohan,J.K.Singh,M.K.Samantaray,M.M.Shaikh,D. Weinheim,2012. Panda,P.Ghosh,J.Am.Chem.Soc.2007,129,15042–15053. [58] A.Bittermann,E.Herdtweck,P.H(cid:2)rter,W.A.Herrmann,Organo- [27] O. Ciftci, A. Beytur, N. Vardi, I. Ozdemir, Drug Dev. Ind. Pharm. metallics2009,28,6963–6968. 2012,38,40–46. [59] F.M.Bickelhaupt,E.J.Baerends,K.B.Lipkowitz, D.B.Boyd,Re- [28] L. Oehninger, M. Stefanopoulou, H. Alborzinia, J. Schur, S. Lude- viewsinComputationalChemistry,Wiley-VCH,Weinheim,2005. wig,K.Namikawa,A.MuÇoz-Castro,R.W.Kçster,K.Baumann,S. [60] M. Pellei, V. Gandin, M. Marinelli, C. Marzano, M. Yousufuddin, Wçlfl,W.S.Sheldrick,I.Ott,DaltonTrans.2013,42,1657–1666. H.V.R.Dias,C.Santini,Inorg.Chem.2012,51,9873–9882. [29] L.Oehninger,H.Alborzinia,S.Ludewig,K.Baumann,S.Wçlfl,I. [61] P. Mura, M. Camalli, A. Bindoli, F. Sorrentino, A. Casini, C. Gab- Ott,ChemMedChem2011,6,2142–2145. biani, M. Corsini, P. Zanello, M.P. Rigobello, L. Messori, J. Med. [30] B. Cetinkaya, E. Cetinkaya, H. Kucubay, R. Durmaz, Arzneim.- Chem.2007,50,5871–5874. Forsch./DrugRes.1996,46,821–823. [62] A. Casini, C. Gabbiani, F. Sorrentino, M.P. Rigobello, A. Bindoli, [31] J.R. McConnell, D.P. Rananaware, D.M. Ramsey, K.N. Buys, T.J. Geldbach, A. Marrone, N. Re, C.G. Hartinger,P.J. Dyson, L. M.L. Cole, S.R. McAlpine, Bioorg. Med. Chem. Lett. 2013, 23, Messori,J.Med.Chem.2008,51,6773–6781. 2527–2531. [63] A.-B. Witte, K. Anestal, E. Jerremalm, H. Ehrsson, E.S.J. Arner, [32] M.A.Scharwitz,I.Ott,Y.Geldmacher,R.Gust,W.S.Sheldrick,J. FreeRadicalBiol.Med.2005,39,696–703. Organomet.Chem.2008,693,2299–2309. [64] W. Cai, L. Zhang, Y. Song, B. Wang, B. Zhang, X. Cui, G. Hu, Y. [33] M.Harlos,I.Ott,R.Gust,H.Alborzinia,S.Wçlfl,A.Kromm,W.S. Liu,J.Wu,J.Fang,FreeRadicalBiol.Med.2012,52,257–265. Sheldrick,J.Med.Chem.2008,51,3924–3933. [65] A. Bindoli, M.P. Rigobello, G. Scutari, C. Gabbiani, A. Casini, L. [34] Y.Geldmacher,I.Kitanovic,H.Alborzinia,K.Bergerhoff,R.Rub- Messori,Coord.Chem.Rev.2009,253,1692–1707. biani,P.Wefelmeier,A.Prokop,R.Gust,I.Ott,S.Wçlfl,W.S.Shel- [66] C.Gabbiani,G.Mastrobuoni,F.Sorrentino,B.Dani,M.P.Rigobel- drick,ChemMedChem2011,6,429–439. lo,A.Bindoli,M.A.Cinellu,G.Pieraccini,L.Messoria,A.Casini, [35] M. Dobroschke, Y. Geldmacher, I. Ott, M. Harlos, L. Kater, L. MedChemComm2011,2,50–54. Wagner,R.Gust,W.S.Sheldrick,A.Prokop,ChemMedChem2009, [67] J.Lu,E.-H. Chew,A.Holmgren,Proc.Natl.Acad. Sci.USA2007, 4,177–187. 104,12288–12293. [36] Y. Geldmacher, M. Oleszak, W.S. Sheldrick, Inorg. Chim. Acta [68] K.N. deOliveira, V. Andermark, S. vonGrafenstein, L.A. Onam- 2012,393,84–102. bele,G.Dahl,R.Rubbiani,G.Wolber,C.Gabbiani,L.Messori,A. [37] M.B. Schwarz, A. Kurzwernhart, A. Roller, W. Kandioller, B.K. Prokop,I.Ott,ChemMedChem2013,8,256–264. Keppler, C.G. Hartinger, Z. Anorg. Allg. Chem. 2013, 1648–1654. [69] J. Schur, C.M. Manna, A. Deally, R.W. Kçster, M. Tacke, E.Y. [38] R.Payne,P.Govender,B.Therrien,C.M.Clavel,P.J.Dyson,G.S. Tshuva,I.Ott,Chem.Commun.2013,49,4785–4787. Smith,J.Organomet.Chem.2013,729,20–27. [70] M.Groessl,O.Zava,P.J.Dyson,Metallomics2011,3,591–599. [39] R.J. Ernst, A.C. Komor, J.K. Barton, Biochemistry 2011, 50, [71] I.Ott,K.Schmidt,B.Kircher,P.Schumacher,T.Wiglenda,R.Gust, 10919–10928. J.Med.Chem.2005,48,622–629. [40] F.Hackenberg,L.Oehninger,H.Alborzinia,S.Can,I.Kitanovic,Y. [72] A.G. Weidmann, A.C. Komor, J.K. Barton, Philos. Trans. R. Soc. Geldmacher, M. Kokoschka, S. Wçlfl, I. Ott, W.S. Sheldrick, J. A1995,371,20120117. Inorg.Biochem.2011,105,991–999. [73] G. L(cid:3)mmen, H. Sperling, H. Luboldt, T. Otto, H. R(cid:3)bben, Cancer [41] G.Sava,T.Giraldi,G.Mestroni,G.Zassinovich,Chem.-Biol.Inter- Chemother.Pharmacol.1998,42,415–417. act.1983,45,1–6. [74] AmsterdamDensityFunctional (ADF),VrijeUniversiteit,Amster- [42] Y.Li,C.d.Kock,P.J.Smith,H.Guzgay,D.T.Hendricks,K.Naran, dam(TheNetherlands),2012. V. Mizrahi, D.F. Warner, K. Chibale, G.S. Smith, Organometallics [75] E. vanLenthe, E.J. Baerends, J.G. Snijders, J. Chem. Phys. 1994, 2013,32,141–150. 101,9783–9792. [43] U.Sliwinska-Hill, F.P.Pruchnik,M. Latocha, D.Nawrocka-Musial, [76] K.G.Dyall,K.Fægri,IntroductiontoRelativisticQuantumChemis- S.Ułaszewski,Inorg.Chim.Acta2013,400,26–31. try,OxfordUniversityPress,Oxford,NewYork,2007. [44] G.Rubner,K.Bensdorf,A.Wellner,S.Bergemann,I.Ott,R.Gust, [77] J.P. Perdew, K. Burke, Y. Wang, Phys. Rev. B 1996, 54, 16533– Eur.J.Med.Chem.2010,45,5157–5163. 16539. [45] S. Patil, A. Deally, B. Gleeson, H. M(cid:3)ller-Bunz, F. Paradisi, M. [78] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, Tacke,Metallomics2011,3,74–88. 3865–3868. [46] I. (cid:9)zdemir, N. Temelli, S. G(cid:3)nal, S. Demir, Molecules 2010, 15, [79] L.Versluis,T.Ziegler,J.Chem.Phys.1988,88,322–328. 2203–2210. [80] K. Wolinski, J.F. Hinton, P. Pulay, J. Am. Chem. Soc. 1990, 112, [47] S. Patil, A. Deally, F. Hackenberg, L. Kaps, H. M(cid:3)ller-Bunz, R. 8251–8260. Schobert,M.Tacke,Helv.Chim.Acta2011,94,1551–1562. [81] N.C.Handy,A.J.Cohen,Mol.Phys.2001,99,403–412. [48] S. Patil, A. Deally, B. Gleeson, F. Hackenberg, H. M(cid:3)ller-Bunz, F. [82] I.Ott,R.Gust,Biometals2005,18,171–177. Paradisi,M.Tacke,Z.Anorg.Allg.Chem.2011,637,386–396. [83] J. Ma, G. Stoter, J. Verweij, J.H.M. Schellens, Cancer Chemother. [49] J.M. Herbert, P.D. Woodgate, W.A. Denny, J. Med. Chem. 1987, Pharmacol.1996,38,391–394. 30,2081–2086. [84] R.Gust,B.Schnurr,R.Krauser,G.Bernhardt,M.Koch,B.Schmid, [50] E. C¸etinkaya, P.B. Hitchcock, H. K(cid:3)Å(cid:3)kbay, M.F. Lappert, S. Al- E. Hummel, H. Schçnenberger, J. Cancer Res. Clin. Oncol. 1998, Juaid,J.Organomet.Chem.1994,481,89–95. 124,585–597. [51] F.E. Hahn, T. vonFehren, L. Wittenbecher, R. Frçhlich, Z. Natur- forsch.2004,59b,541–543. Received:July18,2013 [52] J.A.V.Er,A.G.Tennyson,J.W.Kamplain,V.M.Lynch,C.W.Bie- Revised:September17,2013 lawski,Eur.J.Inorg.Chem.2009,1729–1738. Publishedonline:November15,2013 17880 www.chemeurj.org (cid:4)2013Wiley-VCHVerlagGmbH&Co.KGaA,Weinheim Chem.Eur.J.2013,19,17871–17880