Mechanism of action of non-cisplatin type DNA-targeted platinum anticancer agents: DNA interactions of novel acridinylthioureas and their platinum conjugates

Abstract

The DNA binding of two novel acridinylthioureas, ACR-NH-(CH 2) 2-C(S)-NHCH 3 ( 1) and ACR-N(CH 3)-C(S)-NHCH 3 ( 3), and their platinum conjugates 4 and 5—derived from [PtCl 2(en)]—was studied in cell-free model systems using various physico-chemical and biophysical methods. These included: spectrophotometric drug–DNA titrations, ethidium−DNA fluorescence quenching, competitive drug displacement, high-resolution NMR spectroscopy, and unwinding of plasmid DNA monitored by agarose gel electrophoresis. The acridinium cation of 1 showed strong binding to native DNA with K i =1.5×10 6 M −1 and an excluded site size ( n) of 2 bp (McGhee–von Hippel fits of absorbance data). Compound 3 showed no measurable association with DNA. Binding of 1 was an order of magnitude stronger than that of simple 9-methylaminoacridine ( 2). In alternating copolymers, 1 exhibited slight AT preference. In poly(dA-dT) 2, enhanced association was accompanied by an increased binding site (approximately 3 bp), while parameters in poly(dG-dC) 2 were consistent with classical intercalation. Displacement of 1 by distamycin from calf thymus DNA was suggestive of non-intercalating thiourea in 1 being located in the minor groove of the duplex. 1 H NMR data of d(GGAGCTCC) 2 modified with 1 indicated intercalative binding of planar acridine, based on upfield shifts of aromatic proton signals relative to those in unbound 1 (Δδ≈−0.5 to −1 ppm). Finally, 4 and 5 were found to unwind negatively supercoiled pUC19 plasmid by 21° and 7° per adduct, respectively (electrophoretic gel mobility assays). The difference in DNA binding modes of 4 and 5 is discussed as the ultimate source of the distinctly different biological activities of the conjugates.