Change of the Binding Mode of the DNA/Proflavine System Induced by Ethanol Begon ˜a Garcı ´a,* ,† Jose ´ M. Leal, † Rebeca Ruiz, † Tarita Biver, ‡ Fernando Secco,* ,‡ and M. Venturini ‡ Chemistry Department, UniVersity of Burgos, 09001 Burgos, Spain, and Chemistry and Industrial Chemistry Department, UniVersity of Pisa, 56126 Pisa, Italy ReceiVed: March 29, 2010; ReVised Manuscript ReceiVed: May 14, 2010 The equilibria and kinetics of the binding of proflavine to poly(dG-dC) · poly(dG-dC) and poly(dA-dT) · poly(dA- dT) were investigated in ethanol/water mixtures using spectrophotometric, circular dichroism, viscometric, and T-jump methods. All methods concur in showing that two modes of interaction are operative: intercalation and surface binding. The latter mode is favored by increasing ethanol and/or the proflavine content. Both static and kinetic experiments show that, concerning the poly(dG-dC) · poly(dG-dC)/proflavine system, intercalation largely prevails up to 20% EtOH. For higher EtOH levels surface binding becomes dominant. Concerning the poly(dA-dT) · poly(dA-dT)/proflavine system, melting experiments show that addition of proflavine stabilizes the double stranded structure, but the effect is reduced in the presence of EtOH. The ΔH° and ΔS° values of the melting process, measured at different concentrations of added proflavine, are linearly correlated, revealing the presence of the enthalpy-entropy compensation phenomenon (EEC). The nonmonotonicity of the “entropic term” of the EEC reveals the transition between the two binding modes. T-jump experiments show two relaxation effects, but at the highest levels of EtOH (>25%) the kinetic curves become monophasic, confirming the prevalence of the surface complex. A branched mechanism is proposed where diffusion controlled formation of a precursor complex occurs in the early stage of the binding process. This evolves toward the surface and/or the intercalated complex according to two rate-determining parallel steps. CD spectra suggest that, in the surface complex, proflavine is bound to DNA in the form of an aggregate. Introduction The process whereby drugs and mutagens possessing planar aromatic rings interact with double/triple-stranded nucleic acids has been the subject of extensive studies, following the original hypothesis of the intercalation model proposed by Lerman 1 as a result of the interpretation of roentgenograms and viscometric experiments on the DNA/proflavine system. The Lerman hypothesis has become so popular that the majority of the results about the binding of planar drugs to nucleic acids are interpreted according to this mode of interaction, although information provided by accurate pioneer studies showed that intercalation is not a straightforward process. A diligent analysis of spec- trophotometric data concerning the binding of proflavine to DNA demonstrated the formation of two bound forms, originally denoted as “weak” and “strong” complexes. 2 This initial observation was corroborated by the results of early kinetic experiments on the same system, which showed the occurrence of two 3 or even three 4 relaxation effects. In subsequent investigations a single relaxation effect was recorded, 5,6 but all of these studies concurred in demonstrating the biphasic nature of the process. However, some variations in the reaction parameter values, such for example the site size dependence on the concentration of added salt 7 or rate dependence on base pair nature (AT vs GC) 8 make any conclusion about the mechanism of dye binding to DNA rather uncertain. The complex- ity of the system makes it difficult to separate the contributions of the forces that are at work during the DNA-drug interaction. These forces are of electrostatic and nonelectrostatic (stacking) nature and both can play a fundamental role in the nucleic acid chemistry and in the solution chemistry of the dyes. In spite of the many efforts made to enlighten this issue, the various contributions of these forces to the overall free energy change of the binding process remain difficult to analyze. 9 The importance of changing the properties of the solvent, to investigate the origin of the electrostatic and stacking interac- tions, is well recognized. In this context, alcohols are known to remarkably affect the strength of electrostatic forces and to weaken hydrophobic effects. 10,11 The perturbation of the aqueous solvent by means of alcohols does constitute, therefore, a useful tool to elucidate the role of electrostatic and hydrophobic forces on the DNA-drug interaction. With the present work we have tried to shed light into the nature of the interaction that accompanies the binding of proflavine to poly(dA-dT) · poly(dA-dT) and poly(dG-dC) · poly(dG-dC), looking in particular to the influence exerted on the binding modes by the base-pair nature and the solvent, using water-ethanol mixtures. The static studies have been performed by employing a variety of techniques such as spectrophotometry, circular dichroism, and viscosity, while the kinetic investigation has been carried out using the T-jump relaxation technique. Experimental Section Materials. Calf thymus DNA (CT-DNA) and the synthetic polynucleotides poly(dA-dT) · poly(dA-dT) and poly(dG- dC) · poly(dG-dC) were purchased from Sigma-Aldrich as the lyophilized sodium salts and used without further purification. The CT-DNA was sonicated as described below. Stock solutions of the polynucleotides were standardized spectrophotometrically * Corresponding authors. E-mail: B.G., begar@ubu.es; F.S., ferdi@dcci.unipi.it. † University of Burgos. ‡ University of Pisa. J. Phys. Chem. B 2010, 114, 8555–8564 8555 10.1021/jp102801z  2010 American Chemical Society Published on Web 06/04/2010