A series of cyclometalated gold(III) compounds [Au(m)(C(wedge)N(wedge)C)mL]n+ (m = 1-3; n = 0-3; HC(wedge)N(wedge)CH = 2,6-diphenylpyridine) was prepared by ligand substitution reaction of L with N-donor or phosphine ligands. The [Au(m)(C(wedge)N(wedge)C)mL]n+ compounds are stable in solution in the presence of glutathione. Crystal structures of the gold(III) compounds containing bridging bi- and tridentate phosphino ligands reveal the presence of weak intramolecular pi pi stacking between the [Au(C(wedge)N(wedge)C)]+ units. Results of MTT assays demonstrated that the [Au(m)(C(wedge)N(wedge)C)mL]n+ compounds containing nontoxic N-donor auxiliary ligands (2) exert anticancer potency comparable to that of cisplatin, with IC50 values ranging from 1.5 to 84 microM. The use of [Au(C(wedge)N(wedge)C)(1-methylimidazole)]+ (2 a) as a model compound revealed that the gold(III)-induced cytotoxicity occurs through an apoptotic cell-death pathway. The cell-free interaction of 2 a with double-stranded DNA was also examined. Absorption titration showed that 2 a binds to calf-thymus DNA (ctDNA) with a binding constant of 4.5 x 10(5) dm3 mol(-1) at 298 K. Evidence from gel-mobility-shift assays and viscosity measurements supports an intercalating binding mode for the 2 a-DNA interaction. Cell-cycle analysis revealed that 2 a causes S-phase cell arrest after incubation for 24 and 48 hours. The cytotoxicity of 3 b-g toward cancer cells (IC50 = 0.04-4.3 microM) correlates to that of the metal-free phosphine ligands (IC50 = 0.1-38.0 microM), with [Au2(C(wedge)N(wedge)C)2(mu-dppp)]2+ (3 d) and dppp (dppp = 1,2-bis(diphenylphosphino)propane) being the most cytotoxic gold(III) and metal-free compounds, respectively. Compound 3 d shows a cytotoxicity at least ten-fold higher than the other gold(III) analogues; in vitro cellular-uptake experiments reveal similar absorptions for all the gold(III) compounds into nasopharyngeal carcinoma cells (SUNE1) (1.18-3.81 ng/cell; c.f., 3 d = 2.04 ng/cell), suggesting the presence of non-gold-mediated cytotoxicity. Unlike 2 a, both gold(III) compounds [Au(C(wedge)N(wedge)C)(PPh3)]+ (3 a) (PPh3 = triphenylphosphine) and [Au2(C(wedge)N(wedge)C)2(mu-dppp)]2+ (3 d) interact only weakly with ctDNA and do not arrest the cell cycle.

A stable gold(III)-phosphine complex [(C^N^C) 2 Au 2 (m-dppp)](CF 3 SO 3 ) 2 [Au3, HC^N^CH ¼ 2,6diphenylpyridine; dppp ¼ bis(diphenylphosphino)propane] displays potent in vitro cytotoxicity towards various cancers with sub-micromolar range cytotoxic IC 50 values, and is significantly more potent than its structural and iso-electronic platinum(II) analog [(C^N^N) 2 Pt 2 (m-dppp)](CF 3 SO 3 ) 2 (HC^N^N ¼ 6-phenyl-2,2 0 -bipyridine) and gold(III)-carbene complexes. Complex Au3 displays promising inhibition on tumor growth in animal models, and its acute and sub-chronic toxicities have been examined in mice and beagle dogs. Transcriptomic and connectivity map analyses have revealed that the transcriptional profile of Au3 is similar to those of inhibitors of thioredoxin reductase (TrxR) and inducers of endoplasmic reticulum (ER) stress. As we found that Au3 is also a nanomolar inhibitor of TrxR, a model of ER stress-induced cell death mediated by inhibition of TrxR is proposed. The transcriptomic analysis also leads to the identification of TRAIL, a ligand for death receptor 5 (DR5), as a synergistic agent of the anti-tumor activity of Au3. Collectively, our results demonstrate that the gold(III) complex Au3 effectively inhibits tumor growth in vivo, and displays promising cytotoxicity towards cancer cells in association with the inhibition of TrxR, induction of ER stress and also a death-receptor-dependent apoptotic pathway.

In the design of physiologically stable anticancer gold(III) complexes, we have employed strongly chelating porphyrinato ligands to stabilize a gold(III) ion [Chem. Commun. 2003, 1718; Coord. Chem. Rev. 2009, 253, 1682]. In this work, a family of gold(III) tetraarylporphyrins with porphyrinato ligands containing different peripheral substituents on the meso-aryl rings were prepared, and these complexes were used to study the structure-bioactivity relationship. The cytotoxic IC(50) values of [Au(Por)](+) (Por=porphyrinato ligand), which range from 0.033 to >100 microM, correlate with their lipophilicity and cellular uptake. Some of them induce apoptosis and display preferential cytotoxicity toward cancer cells than to normal noncancerous cells. A new gold(III)-porphyrin with saccharide conjugation [Au(4-glucosyl-TPP)]Cl (2a; H(2)(4-glucosyl-TPP)=meso-tetrakis(4-beta-D-glucosylphenyl)porphyrin) exhibits significant cytostatic activity to cancer cells (IC(50)=1.2-9.0 microM) without causing cell death and is much less toxic to lung fibroblast cells (IC(50)>100 microM). The gold(III)-porphyrin complexes induce S-phase cell-cycle arrest of cancer cells as indicated by flow cytometric analysis, suggesting that the anticancer activity may be, in part, due to termination of DNA replication. The gold(III)-porphyrin complexes can bind to DNA in vitro with binding constants in the range of 4.9 x 10(5) to 4.1 x 10(6) dm(3) mol(-1) as determined by absorption titration. Complexes 2a and [Au(TMPyP)]Cl(5) (4a; [H(2)TMPyP](4+)=meso-tetrakis(N-methylpyridinium-4-yl)porphyrin) interact with DNA in a manner similar to the DNA intercalator ethidium bromide as revealed by gel mobility shift assays and viscosity measurements. Both of them also inhibited the topoisomerase I induced relaxation of supercoiled DNA. Complex 4a, a gold(III) derivative of the known G-quadruplex-interactive porphyrin [H(2)TMPyP](4+), can similarly inhibit the amplification of a DNA substrate containing G-quadruplex structures in a polymerase chain reaction stop assay. In contrast to these reported complexes, complex 2a and the parental gold(III)-porphyrin 1a do not display a significant inhibitory effect (<10%) on telomerase. Based on the results of protein expression analysis and computational docking experiments, the anti-apoptotic bcl-2 protein is a potential target for those gold(III)-porphyrin complexes with apoptosis-inducing properties. Complex 2a also displays prominent anti-angiogenic properties in vitro. Taken together, the enhanced stabilization of the gold(III) ion and the ease of structural modification render porphyrins an attractive ligand system in the development of physiologically stable gold(III) complexes with anticancer and anti-angiogenic activities.