Rhenium ( I ) phenanthrolines bearing electron withdrawing CF 3 substituents : synthesis , characterization and biological evaluation †

Department of Chemistry, University of H E-mail: c.redshaw@hull.ac.uk CSIRO Materials Science and Engineerin Avenue, Clayton South, 3169, Victoria, Aust Department of Chemistry, The University of 24th St. A5300, Austin, TX 78712-1224, USA School of Chemistry, University of East Ang School of Pharmacy, University of East Ang mueller@uea.ac.uk † Electronic supplementary informatio characterisation of 1 and 2; alternative v cell assays for 10 and 20. CCDC 755744. F or other electronic format see DOI: 10.103 Cite this: RSC Adv., 2013, 3, 23963

One of the main goals in the eld of metal-based anti-tumor drugs is to overcome the major limitations associated with cisplatin, namely drug resistance, poor selectivity and adverse side effects. 1 Amongst the metals under consideration, rhenium has shown promising apoptosis and anti-tumor activity in a variety of oxidation states. 2 Furthermore, rhenium complexes have become the logical choice for therapeutic applications, given the close similarities with the chemistry of technetium, 3 and a number of complexes are now under clinical consideration. 4Whilst the most extensive chemistry reported for these group VII congeners involves the [MO] 3+ core (metal in oxidation state +V), other studies have highlighted the potential of the organometallic fragment [fac-Re(CO) 3 ] + . 5This fragment is readily accessible as the air stable [Re(CO) 3 (H 2 O) 3 ] + species, for which the aqua ligands are substitutionally labile.Furthermore, the small size of the core allows for tagging with bioactive moieties, whilst retaining bioactivity and specicity.Radio-imaging techniques do not permit visualization at the cellular level, however by replacing the radioisotope with a uorophore, it is possible to use uorescent microscopy to identify cellular targets and thereby correlate results with in vivo radio-imaging experiments.A number of conjugated chelating ligands have recently been bound to the [fac-Re(CO) 3 ] core, including benzoimidazole, diimine and indole. 6To-date, results for the diimine systems have been restricted due to limited visible adsorption (generally emit only in the yellow/green region), 7 this despite reports on rhenium(I) tricarbonyl diimines of the type [Re(CO) 3 (N-N)X] (N-N ¼ diimine, X ¼ halide) dating back to the 1970s. 8Given the relatively low sensitivity of the HOMO energy to changes to the diimine ligand, the emission energy of the rhenium(I) complexes is, broadly speaking, proportional to the reduction potential of the ligand.The colour of the emission can therefore be shied by lowering the LUMO energy, i.e. by making the ligand reduction potential more positivethe socalled energy-gap law. 9One way in which this can be accomplished is to introduce electron withdrawing substituents at the ligand, thereby shiing the emission toward the red.Transitions to a ligand localized LUMO has recently been utilized in azadipyrromethane rhenium tricarbonyl complexes to bring about red emission. 10In the case of phenanthroline ligand systems, this can be achieved by the addition of phenyl groups, for example anionic complexes of the type [Re(CO) 3 (bathophenanthroline sulphate)(py-3-R)] (R ¼ H, CH 2 OH; bathophenanthroline ¼ 4,7-diphenyl-1,10-phenanthroline) exhibit red shis of about 20 nm for excitation and 10 nm for emission. 11It has been noted in related biquinoline chemistry that the use of electron withdrawing substituents can also lead to red shis. 12It was therefore expected that for bathophenanthroline-derived systems, phenyl groups possessing uorine containing substituents such as CF 3 groups would produce a red shi.The effect is more pronounced if the aryl group is not positioned meta to the Re-N bond.This red emission is of particular interest in biological systems, as human tissue is more transparent to red and near infra-red light potentially allowing such materials to be used in therapeutic settings, 13 as well as in OLEDs (organic light emitting diodes) which was originally our entry into such ligand sets. 14s a result, we report herein two such ligands, 2,9-bis(3-tri-uoromethylphenyl)-4,7-diphenyl-1,10-phenanthroline and 2,9bis(4-triuoromethylphenyl)-4,7-diphenyl-1,10-phenanthroline, and the rhenium(I) carbonyl chloro complexes thereof (illustrated in Scheme 1).We were attracted to the use of the chloro complexes given their desirable photophysical properties as well as the tendency to form air stable complexes. 15he cellular uptake of the two complexes in several cell lines has been investigated by uorescence microscopy, and their cytotoxicity towards a variety of cell lines has been examined by the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) assay.
Complexes 1 and 2 are readily synthesized by heating the ligand and rhenium pentacarbonyl chloride in equimolar quantities in toluene for several hours.Both complexes 1 and 2 can be puried by sublimation under high vacuum between 260 and 280 C, though complex 2 exhibits decomposition and shows the presence of the free ligand in the products.
Crystals of 1 suitable for analysis by X-ray diffraction were grown at ambient temperature from a saturated solution of 1 in toluene; a thermal ellipsoid projection is shown in Fig. 1 (alternative views are given in the ESI, see Fig. S1-S4 †).The structure adopts the expected fac-CO arrangement with a distorted octahedral rhenium centre, where the main distortion occurs as a result of the N,N-chelate bite angle (75.48(1) ).The Re-C and C-O bond lengths are typical of those observed in such fac-tricarbonyl rhenium(I) diimine complexes. 8he development of specic cancer drugs depends on a detailed understanding of how the drugs interact with different cell types.We therefore undertook a detailed analysis of the effects of these complexes bearing electron withdrawing groups on different cell types (Table 1).For toxicity screening compounds were dissolved in DMSO (1%), following which the emission shied to the yellow/green.Given this, we denote the post-DMSO complexes as 1 0 and 2 0 .
For complexes 1 and 2, the PL spectra are consistent with them acting as red emitters (solution emission for 1 and 2 were 635 nm and 596 nm, respectively, see for example ESI, Fig. S5    Leads to proliferation Leads to proliferation a Cells were tested for proliferation aer 72 hours using an MTS kit.DMSO (1%) was used as a control in all studies, data represent mean AE SEM of at least three independent experiments done in duplicate.
The IC 50 was calculated using GraphPad Prism 5 (GraphPad Soware), when data was tted to a nonlinear regression curve.
cells and is less toxic towards HeLa cells than 1 0 .Perhaps the most interesting observation of these experiments is that the rhenium complexes 1 0 and particularly 2 0 encourage the proliferation of HL-60 cells over 72 hours.Preliminary data in 16HBE cell lines also showed that 1 0 and 2 0 encourage cell growth (ESI, Fig. S13 †).We tested both compounds on the monolayer cell line CHO and the suspension cells HL60, and monitored whether or not each can enter cells.The spontaneous uorescence of the compounds exhibited aer activation under a uorescence microscope was not very bright, but it allowed us to predict cellular up-take of the compounds is indeed occurring in the HL60 cells (ESI, Fig. S14 †) (Fig. 2).We investigated how 1 0 and 2 0 can induce cell death in the CHO cells and incubated the cells for 24 hours with the compounds and stained with a nucleus and an actin bre marker (ESI, Fig. S15 †).
Neither compounds affected the actin polymerization or the shape and consistence of the nuclei in the cells, i.e. no fragmentation of the nuclei, which can be a sign of early apoptosis, was visible.
These results therefore suggest that neither 1 0 nor 2 0 induce programmed cell death (apoptosis) in CHO cells, but more work needs to be done to elucidate the actual ways how these compounds kill the cells.
The cytotoxicity of a number of other complexes containing the [fac-Re(CO) 3 ] core has been examined, including phosphine of the form [Re(CO) 3 (diphosphine)Br]. 17For pyridyl or phenanthroline containing complexes, screening against the HeLa cell lines afforded IC 50 values (mM) of about 30 for [Re(CO) 3 (2-appt)Cl] (2-appt ¼ 2-amino-4-phenylamino-6-(2pyridyl)-1,3,5-triazine) 18 and ranging from 17.5-28.5for a series of thiourea complexes, 19a 2.8 to >150 for a series of polypyridyl glucose complexes, 19b 3.6 to 40 for a series of polypyridyl orous complexes, 19c and 3.6 to >1151.7 for a series of polypyridyl poly(ethylene glycol) complexes.19d We have attempted to isolate the DMSO complexes 1 0 and 2 0 using the method of Mayer. 20Our initial attempts however have resulted in oily materials, though in the case of 1, trituration with hexane afforded a yellow/brown glass.The IR spectrum contained a stretch at 950 cm À1 which we tentatively assign to the nS-O stretch of an O-bound DMSO in the complex 1 0 . 21We note that DMSO complexes bound to Re(CO) 3 (I) have been previously structurally characterized. 22n summary, we have shown that rhenium(I) tricarbonyl complexes bearing a 2,4,7,9-tetraphenylphenanthroline ligand, where the 2,9-phenyl groups have either meta or para CF 3 groups are red emitters.Following treatment with DMSO (1%), the emission is shied to the yellow/green.In the case of the meta derivative, evaluation against several cell lines revealed anti-cancer activity against HeLa and A549 cell lines; the para derivative caused proliferation of HL-60 cells, and toxicity towards A549 cells.Further work is in progress to determine the exact nature of 1 0 and 2 0 , and also the cytotoxicity proles of the meta-and para-CF 3 ligands described herein in combination with other metals.
The Big C Cancer Appeal and the Whilelaw Frater Cancer Trust are thanked for nancial support.Dr Kevin Welham (University of Hull) is thanked for Mass Spectroscopy on 1 0 .

Fig. 2
Fig. 2 Cell viability assay in (a) A549 cells, (b) HL-60 cells and (c) CHO cells, tested for proliferation after 72 hours using an MTS kit.DMSO used as a control in all studies, data represent mean AE SEM of at least three independent experiments done in duplicate.Absorbance shown is directly proportional to the number of living cells in culture.