The use of metallo-pharmaceuticals, such as the platinum drugs, for cancer treatment illustrates the utility of metal complexes as therapeutic agents. Platinum group metal complexes therefore offer potential as anti-tumour agents to fight cancer. This study was aimed at synthesizing and evaluating the effects of palladium(II) and platinum(II) complexes as anticancer agents.

Since the discovery of cisplatin as an anti-cancer agent, there has been a broad and multidisciplinary interest over four decades in the development of metal complexes as metallotherapeutic drugs. The principal objective of this thesis was to develop and characterize a novel library of gold(III) complexes of aromatic and non-aromatic quinoline- and pyridine-amido ligands and to test their efficacy as cytotoxic agents against multiple human cancer cell lines. To this end, fifteen novel (16 in total) gold(III) complexes have been prepared and studied by multiple methods including FTIR, NMR, MS, and UV-visible spectroscopy, and in numerous cases, single crystal X-ray diffraction. Ligands H2L1–HL14 were prepared via the reaction of the relevant pyridine or quinoline carboxylic acid in the presence of triphenylphosphite and either picolylamine or 8-aminoquinoline in pyridine, in moderate to good yields. Ligands HL15 and HL16 were prepared via the reaction between benzoyl or 1-naphthoyl chloride and 8-aminoquinoline in good yields. The synthesis of complexes [Au(HL1)Cl2]–[Au(L4)Cl2] were prepared by the reaction between the respective ligand, K[AuCl4] and NaOAc in 1:1 MeOH:DCM. Metal complexes [Au(L5)Cl](PF6)–[Au(L14)Cl](PF6) were synthesised by encouraging the formation of a AuCl4- counter ion in acetic acid and 3-fold excess of NaHCO3. Subsequent metathesis afforded the desired PF6- anion. Complexes [Au(L15)Cl] and [Au(L16)Cl] were synthesised by the reaction of H[AuCl4] and respective ligand in acetic acid and a 3-fold excess of NaHCO3. The solubility of all complexes was assessed, with complexes [Au(L8)Cl](PF6)–[Au(L14)Cl](PF6) proving to be the most stable in biologically relevant media (TBS 50 mM, NaCl 10mM, pH 7.34, 37°C). Complex [Au(L12)Cl](PF6) was further evaluated for its stability in the presence of glutathione and imidazole and found to be sensitive to reduction by thiols, but substitution-inert to N-donor heterocycles such as imidazole. The DNA binding constants of [Au(L8)Cl](PF6)–[Au(L11)Cl](PF6) were subsequently evaluated by UV-vis spectroscopy and found to be in the range of 2.7(5) x 105 to 4.7(6) x 105 M-1. Complexes [Au(L12)Cl](PF6)–[Au(L14)Cl](PF6) were similarly assessed using ethidium bromide displacement fluorescence assays, however their ability to bind DNA could not be conclusively proven. The log Po/w values of complexes [Au(L12)Cl](PF6)–[Au(L14)Cl](PF6) were measured and spanned the range -0.8 to -2.16, consistent with significant hydrophilic character. The solid state structures of all complexes, with the exception of [Au(L10)Cl](PF6), [Au(L14)Cl](PF6) and [Au(L16)Cl](PF6), were determined by X-ray crystallography with the gold(III) ion co-ordinated to the ligand in a square planar geometry. The co-ordination mode in complexes Au(HL1)Cl2]–[Au(HL3)Cl2] was unexpected with the metal centre only co-ordinating to half the tetradentate ligand with a pair of cis-dichloro ions completing the square planar geometry. The average Au–Npy/qu distance is 2.02(2) Å while the average Au–Namide distance is 1.97(4) Å. In all complexes the trans labilising effect of the anionic amide nitrogen was observed through a structural elongation of the respective Au–Cl bond length. Almost all complexes studied exhibited π-stacking interactions, with compound [Au(L12)Cl](PF6) exhibiting a mean plane separation between rings of 3.307 Å. This is a result of the extended aromatic rings present in all compounds DFT geometry optimizations, frequency, NMR, and energy calculations were carried out on all the gold(III) complexes at the HSEH1PBE/6-311G(d,p)/LanL2DZ level of theory. The 6-311G(d,p) basis set was used for all atoms with the exception of the gold atom for which the LanL2DZ basis set was used. In general, the chosen level of theory satisfactorily correlates with the experimental data for all complexes and was instrumental in deconvoluting the UV-vis spectra of all complexes. The lowest energy transitions (300–500 nm) were assigned to a LMCT while the higher energy transitions were assigned to π-π* transitions. The cytotoxicity profiles of all compounds, with the exception of [Au(HL1)Cl2] and [Au(L16)Cl], were evaluated through one-dose screens against the 60 human cancer cell lines at the NCI, where [Au(HL3)Cl2], [Au(L6)Cl](PF6)–[Au(L8)Cl](PF6), [Au(L10)Cl](PF6)–[Au(L13)Cl](PF6) and [Au(L15)Cl](PF6) were deemed sufficiently cytotoxic to proceed further to five-dose screening. The cytotoxicity results for compound [Au(L12)Cl](PF6) were most encouraging with GI50, TGI and LC50 values of 0.11(0.1), 0.70(0.7) and 26.5(1.5) μM, respectively, against the breast cancer cell line MDA-MB-468. Statistical analysis of the GI50 values for complexes [Au(HL3)Cl2] and [Au(L12)Cl](PF6) revealed they may exert their cytotoxicity through the inhibition/poisoning of topoisomerase II and I enzymes, respectively. Both compounds were assessed for this through a topoisomerase IB DNA unwinding assay and a topoisomerase IIα decatenation assay. [Au(HL3)Cl2] was found to be a dual catalytic inhibitor and poison of topoisomerase IIα between concentrations of 500 nM and 50 μM while [Au(L12)Cl](PF6) was found to be a dual catalytic inhibitor and poison of topoisomerase IB between concentrations of 1 and 100 μM. Electrophoretic mobility shift assays were performed on both complexes, with [Au(HL3)Cl2] indicating DNA binding at a concentration of 50 μM, while [Au(L12)Cl](PF6) displayed no evidence for DNA binding despite an unexpected increase in mobility shift of the substrate DNA. This is indicative of an alternative mechanism of DNA interaction such as electrostatic binding. In summary, we present in this thesis, the discovery, synthesis and application of a novel series of gold(III) amide-based metal complexes as anti-cancer agents with the mechanism of action by which the complexes exert their cytotoxic activity being elucidated. The compounds show immense potential in the metallo-drug discovery field of research, and with further development, a leading class of metallotherapeutic drugs may be developed from this research. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2012.

The use of metallo-pharmaceuticals, such as the platinum drugs, for cancer treatment illustrates the utility of metal complexes as therapeutic agents. Platinum group metal complexes therefore offer potential as anti-tumour agents to fight cancer. This study was aimed at synthesizing and evaluating the effects of palladium(II) and platinum(II) complexes as anticancer agents.

This work focuses on the design and synthesis of Au(I) and Ag(I) complexes from ligand systems that aim to combine both N-heterocyclic carbene (NHC) and phosphine ligand types. A number of synthetic approaches towards both the ligands and the prepared metal complexes have been developed, with a concerted effort on achieving the desired Au(I) or Ag(I) complexes with minimal reaction steps and synthetic style. The thesis body is divided into two main sections. The first section addresses the preparation of suitable ligand precursors of potential Au(I) and Ag(I) complexes in the form of halo- and phosphino-functionalised imidazolium salts. Several series of haloalkylimidazolium salts were prepared that encompass a range of halogens (Cl, Br, I), alkyl substituents (Me, i-Pr, t-Bu, n-Bu), differing alkyl linker length (n = 0-3), and a variety of organic spacers employed to bridge multi-imidazolium moieties. Novel bidentate and multidentate phosphinoalkylimidazolium salts were synthesised from the various haloalkylimidazolium salts, via the substitution of a halide with nucleophilic diphenylphosphide. A new approach towards rare methylene bridged phosphinomethylimidazolium salts was achieved from the reactions of halomethylimidazolium salts with diphenylphosphine. The second section investigates the preparation of Au(I) and Ag(I) complexes from the halo- and phosphino-functionalised imidazolium salts. A series of dicationic 10, 12, and 14-membered metallacyclic Ag(I) complexes were prepared from the bidentate phosphinoalkylimidazolium salts. The dinuclear Ag(I) metallacycles combine two phosphino-functionalised NHC ligands that are bridged by two coordinated Ag(I) ions in an exclusively head-to-head arrangement. A dinuclear Ag(I) metallacycle was investigated for transmetallation potential to a Au(I) complex and found to selectively transmetallate at the Ag(I) coordinated to the NHC ligands to form a bimetallic metallacycle. Unexpected phosphine oxidation of a 10-membered dinuclear Ag(I) metallacycle resulted in complex disproportionation to an isolable and rare silver(I) trimer. Metal-NHC complexes from haloalkylimidazolium salts have not been reported previously, a novel approach to the synthesis of a series of Au(I) complexes from haloalkylimidazolium salts and a respective gold source was developed and is reported herein. Different synthetic approaches towards Au(I) complexes with the phosphinoalkylimidazolium salts explored a variety of ways to generate the NHC from an imidazolium in the presence of the phosphine. A one-pot, high yielding synthesis of a dinuclear Au(I) complex from PPh3 was also devised, with controlled assembly of the complex resulting in a similar head-to-head ligand arrangement to the dinuclear Ag(I) metallacycles. As an aside, a family of mononuclear [Au(R2NHC)2]+ complexes (R = Me, i-Pr, t- Bu, n-Bu, Cy) prepared previously in our research group, was expanded because of the promising antimitochondrial activity shown by [Au(i-Pr2NHC)2]+. Two new [Au(R2NHC)2]+ complexes with simple alkyl chain functionality were prepared with fine-tuned lipophilicity in close proximity to that of [Au(i-Pr2NHC)2]+.