The anti-tumour properties of novel gold compounds

Abstract

Since the introduction of Auranofin in 1985 there has been no new clinically approved gold containing drugs introduced. Although promising results were achieved with a gold(I) phosphine complex [Au(dppe) 2]Cl (Hoke et al., 1991; Mckeage et al., 2002), this compound was never entered into clinical trials due to its toxicity to normal tissue such as the liver and heart (Smith et al.,1989). Six novel derivatives of [Au(dppe) 2]Cl were developed and synthesized to identify possible new candidates with improved tumour specificity compared to [Au(dppe) 2]Cl and cisplatin. Human cervical carcinoma cells (HeLa) were used for an initial toxicity screening. IC50’s obtained for [Au(dppe) 2]Cl and cisplatin were 0.661 and 0.710 µM respectively. Three mixed novel derivatives (MM4, MM5 and MM6) displayed IC50’s ranging between 0.026 and 0.103 µM. These compounds were then selected to be tested further for selectivity and cytotoxicity on various malignant and normal cell lines. MM4 showed selectivity for ovarian, prostate, cervical and breast cancer cells, while MM5 was the most effective against ovarian, colon, prostate, cervical and breast cancer cells. MM6 was most active against ovarian, colon, prostate, cervical and breast cancer cells. The experimental compounds had much higher IC50’s when tested on the normal cells, which indicates selectivity for cancer cells. The octanol/water partition coefficient (lipophilicity) of all the experimental compounds was measured to determine the lipophilicity of the compounds. [Au(dppe) 2]Cl was found to be strongly lipophilic; while the novel compounds had varying degrees of lipo- and hydrophilicity. The octanol/water partition coefficient (lipophilicity) was also used to establish whether there is a correlation between the lipophilicity, IC50 and tumour specificity. In this study no correlation was found between these parameters. [Au(dppe) 2]Cl is known to have an effect on the mitochondrial membrane potential of cells. MM4, MM5, MM6 and cisplatin were compared to [Au(dppe) 2]Cl for effects on mitochondrial membrane potential. PHA stimulated human lymphocytes and a human undifferentiated leukemia T-cell line (Jurkat cells) were used in these experiments. [Au(dppe) 2]Cl, MM4, MM5 and MM6 depolarized the mitochondrial membranes of PHA stimulated lymphocytes significantly, while only [Au(dppe) 2]Cl depolarized the mitochondrial membranes of the Jurkat cells significantly, indicating that a different mechanism of action might be operational. MM4, MM5, MM6 and cisplatin were compared to [Au(dppe) 2]Cl for effects on plasma membrane potential. PHA stimulated human lymphocytes and Jurkat cells were used in these experiments. [Au(dppe) 2]Cl and MM6 depolarized the plasma membranes of both PHA stimulated lymphocytes and the Jurkat cells significantly. In order to determine whether the depolarization of mitochondrial and plasma membranes was a precursor for apoptosis, experiments were done to determine whether MM4, MM5 and MM6 induce apoptosis in Jurkat cells. [Au(dppe) 2]Cl and cisplatin were added for comparison. [Au(dppe) 2]Cl, cisplatin, MM4 and MM6 did induce apoptosis in Jurkat cells, but MM5 did not. The effect of [Au(dppe) 2]Cl, cisplatin, MM4, MM5 and MM6 on the cell cycle of Jurkat cells was determined to establish whether the experimental compounds altered this process. [Au(dppe) 2]Cl, MM4, MM5 and MM6 arrested the cell cycle in the G1 phase and cisplatin did so in the S phase. In order to determine whether the inhibition of cell growth and partition coefficient of the experimental compounds is related to the uptake of the drug, radio labeled drug uptake experiments were carried out with 198Au labeled [Au(dppe) 2]Cl, MM5 and MM6. Two different types of ovarian cancer cells were used for these studies. One cell line was sensitive to cisplatin (A2780) and the other was resistant to cisplatin (A2780 cis). Results obtained from these experiments showed that the uptake of these experimental compounds was dependent on their octanol/water partition coefficient. However, the inhibition of cell growth did not correlate with the uptake of these compounds by the cells that were tested. To confirm the octanol/water partition coefficient and drug uptake results, 198Au labelled [Au(dppe) 2]Cl, MM5 and MM6 were testedin vivo for bio distribution in rats. [Au(dppe) 2]Cl (lipophilic) had higher bio distribution compared to MM5 and MM6 (hydrophilic). Conclusion The experimental compounds show low IC50’s combined with increased tumour specificity. This indicates that these compounds have great potential to target tumour cells selectively and should be investigated further as anti-cancer agents.