Structural and thermodynamical analysis of drug binding to single-stranded DNA oligomers Self-association of non-self-complementary deoxytetranucleotides of different base sequence and their complexation with ethidium bromide in aqueous solution

Abstract

Methods developed for analysing the concentration and temperature dependences of NMR experimental parameters of drug–nucleic acid complexation in solution have been used to study the binding of the drug ethidium bromide (EB) with single-stranded (ss) DNA oligomers. Non-self-complementary (nsc) deoxytetranucleotide triphosphates of different base sequence, 5′-d(CpGpApA), 5′-d(ApApGpC), 5′-d(CpTpGpA) and 5′-d(GpApApG) have been used as model systems of ss nucleotide sequences. 1D and 2D ¹ H NMR spectroscopy (500 and 600 MHz) have been used to investigate self-association of the deoxytetranucleotides, and their complexation with the drug in aqueous solution. 2D homonuclear ¹ H NMR spectroscopy (2D-TOCSY and 2D-NOESY) was used for complete assignment of the proton signals of the deoxytetranucleotides and to determine qualitatively the binding sites of the dye with tetramers. Experimental results for self-association of the tetranucleotides have been analysed using the dimer model. It has been shown that there is a relatively low probability of dimer formation for nsc compared with self-complementary (sc) tetranucleotides, so that complexation of the drug with ss tetranucleotides is expected to dominate the complex equilibrium in solution. The results show that dimerisation constants for nsc deoxytetranucleotides depend on the base sequences, being higher when there is the possibility of base-pairing in the tetranucleotide sequence. Thermodynamic parameters ΔG, ΔH and ΔS for the dimerisation reactions of nsc tetranucleotides have also been determined and confirm the role of base sequences in dimer formation. NMR data for EB complexation with nsc deoxytetranucleotides of different base sequence have been interpreted in terms of equilibrium reaction constants and limiting proton chemical shifts of different complexes (1:1, 1:2 and 2:1) in aqueous solution. Analysis of the relative content of the different complexes has been made and specific features of the dynamic equilibrium have been revealed as a function of the ratio of the drug and tetranucleotide concentrations. The results show that there is a sequence-specific binding of EB with ss DNA and that the pyrimidine–purine sequence is preferred, especially the d(CG)-site. However, it is found that the differences in binding affinities of EB to different sites containing alternating base sequence in the chain are not as great as for drug intercalation to the duplex. A much lower probability of binding is observed for formation of EB complexes with sites of ss sequence containing identical types of bases in the chain. The experimentally determined induced chemical shifts have been analysed in terms of the structures of the complexes. The most favourable structures of the 1:1 drug–tetranucleotide complexes have been calculated taking into account that two different orientations of the drug chromophore with respect to its longitudinal axis occur with equal probability in the 1:1 EB–tetranucleotide complexes. The results confirm that complexes of the dye with ss sequences have substantially higher conformational freedom compared to complexes with sc tetranucleotide duplexes. The enthalpies and entropies of complex formation between EB and nsc deoxytetranucleotides have been determined from the temperature dependence of the 500 MHz ¹ H NMR chemical shifts. The contributions have been determined for formation of the different types of complexes (1:1, 2:1 and 1:2) in solution. The nature of the intermolecular interactions involved in EB complexation with single strands and dimers of nsc tetramers of different base sequence is discussed.