The application of electroanalytical methods to the measurement of metal complex-nucleic acid interactions

Abstract

This thesis reports voltammetric and quartz crystal microgravimetric studies of the binding of metal complexes to nucleic acids in solution and immobilised on metal surfaces. Cyclic voltam iet y and steady-state microelectrode voltammetsy were applied to the solution phase interactions between metal complexes and nucleic acids. The binding constants were obtained by the analysis of bound and free metal complex concentrations. The binding of N, N, N-1-propylthyminedimethylaminomethylferrocene (Fc-Th), N, N, Ntrimethylaminomethylferrocene (Fc-NMe3), bis(hexamethybenzene)iron(II), hexamuiiineruthenium( III), tris(1,10-phenantlu-oline)iron(II) and tris-(bipyridyl)iron(II) to DNA and RNA was observed. The application of microelectrode voltammetry for metal complexnucleic acid binding studies has not been reported before and this thesis demonstrates the advantages of the method due to increased signal-to noise ratio and better discrimination between free and nucleic acid-bound metal complex. These voltammetric results showed the binding of Fc-Th to DNA is stronger than the binding of Fc-NMe3 to DNA, indicating that even a single nucleobase can influence the binding. The binding of singly charged feirocenyl derivatives to DNA or RNA is mainly electrostatic plus some non-electrostatic contribution from interaction of the thymine with DNA. Fe(bz)22 binds to DNA electrostatically and the binding is senstive to ionic strength. Ru(NH3)63+ binds more strongly to DNA due to its higher charge. The binding of Fe(phen)32+, a known intercalator, is stronger than the binding of Fe(bipy)32+ to DNA and the measured binding constants were in agreement with previous reports, however more precise data could be obtained using the microelectrode technique devised in this thesis. This thesis also describes a novel modification of gold and platinum surfaces by the adsorption of 4-mercaptopyridine and subsequent methylation with methyliodide to produce a positively charged suface at which DNA adsorbs strongly. Cyclic voltammetry was applied to quantify the binding of Fc-Th, Fc-NMe3 , hexammineruthenium(II), and tris(1,10-phenanthroline) iron(II) to DNA or RNA immobilised on a gold electrode. A detectable binding of Fc-Th, Ru(NH3)63+ to DNA was observed, while no bound. FcNMe3 and Fe(phen)32 were detected using cyclic voltammetry. The difference in binding to immobilised DNA compared to dissolved DNA could be rationalised by the effect of the electrostatic interactions of the metal complexes with the charged pyridinium monolayer. Quartz crystal microfravimetry was used to estimate the surface coverage of DNA, Fc-Th, Ru(NH3)63+ and Fe(phen); '+on gold and platinum crystals modified as above. Crystal admittance measurements showed no significant change on DNA adsorption indicating approximate rigid-layer behaviour. In agreement with the CV studies no binding of Fc- NMe3 was detected. Some binding of Fe(phen)32 was observed and the negative result of the CV experiment may be due to instability of the monolayer at the high potentials required to oxidise Fe(phen)32+. In general, the QCM results showed higher surface coverages than detected by CV. Two factors may be important, the absence of solution phase metal complex in the CV experiment leads to some desorption and the QCM measurements are complicated by the unknown extent of solvation of the metal complexes.