As part of an overall program to characterize the impact of mutagenic lesions on the physiochemical properties of DNA, we report here the results of a comparative spectroscopic and calorimetric study on a family of DNA duplexes both with and without the oxidative lesion 2'-deoxy-7-hydro-8-oxoguanosine (8-oxodG). Specifically, we have studied a family of eight 13-mer duplexes of the form [5'-GCGTAC[G* or G]CATGCG-3'].[3'-CGCATG[C, A, T, or G]GTACGC-5'] in which G* is the 8-oxodG lesion. These eight duplexes, which we designate by the identity of the variable central base pair (e.g., G*C), reflect two subsets: four duplexes in which the modified guanine base is positioned opposite each of the four possible canonical residues (G*C, G*A, G*G, G*T) and the corresponding four "control" duplexes in which the guanine is not modified (GC, GA, GG, GT). The data derived from our spectroscopic and calorimetric measurements on these eight duplexes allow us to evaluate the influence of the 8-oxodG lesion, as well as the base opposite the lesion, on the conformation, the thermal and thermodynamic stability, and the melting thermodynamics of the host DNA duplex. We find that modification of dG to 8-oxodG (G*) does not change the global DNA duplex conformation as judged by circular dichroism spectra. Despite this structural similarity, our data reveal that the dG to dG* modification does influence duplex thermal and thermodynamic properties, some of which depend on the base opposite the lesion. Thus, apparent structural identity does not mean that two duplexes necessarily will exhibit equivalent thermal and/or thermodynamic properties. In general, we find that the thermodynamic effects induced by the lesion (e.g., GC vs G*C) or by mismatched base pairs (e.g., GC vs GG) can result in relatively large changes in enthalpy which are partially or wholly compensated entropically to produce relatively modest changes in free energy. Our data also suggest that the biologically observed differential recognition of 8-oxodG duplexes and the preferential nucleotide insertion opposite 8-oxodG residues cannot be rationalized simply in terms of large thermodynamic differences.

No abstract

Using high-precision densitometric and ultrasonic measurements, we have determined, at 25 degrees C, the apparent molar volumes, phi V, and the apparent molar compressibilities, phi KS, of five natural and three synthetic B-form DNA duplexes with varying base compositions and base sequences. We find that phi V ranges from 152.0 to 186.6 cm3 mol-1, while phi KS ranges from -73.0 x 10(-4) to -32.6 x 10(-4) cm3 mol-1 bar-1. We interpret these data in terms of DNA hydration which, by the definition employed in this work, refers to those water molecules whose density and compressibility differ from those of bulk water due to interactions with the DNA solute. This definition implies that hydration depends not just on the quantity but also on the quality of the solvent molecules perturbed by the solute. In fact, we find that the number of water molecules perturbed by the DNA duplexes (the quantity of water in their hydration shells) is approximately the same for all of the B-form double helixes studied, while the quality of this water differs as measured by its density and compressibility, thereby yielding differences in the overall hydration properties. Specifically, we find a linear relationship between the density and the coefficient of adiabatic compressibility, beta Sh, of water in the hydration shell of the DNA duplexes, with the range of values for beta Sh being only 65-80% of the value of bulk water.(ABSTRACT TRUNCATED AT 250 WORDS)

No abstract

We use high-precision acoustic and densimetric techniques to determine, at 25 degrees C, the changes in volume, delta V, and adiabatic compressibility, delta Ks, that accompany the binding of netropsin to the poly(dAdT).poly(dAdT) and poly(dA).poly(dT) duplexes, as well as to the poly(dT).poly(dA).poly(dT) triplex. We find that netropsin binding to the heteropolymeric poly(dAdT).poly(dAdT) duplex is accompanied by negative changes in volume, delta V, and small positive changes in compressibility, delta Ks. By contrast, netropsin binding to the homopolymeric poly(dA).poly(dT) duplex is accompanied by large positive changes in both volume, delta V, and compressibility, delta Ks. Furthermore, netropsin binding to the poly(dT).poly(dA).poly(dT) triplex causes changes in both volume and compressibility that are nearly twice as large as those observed when netropsin binds to the poly(dA).poly(dT) duplex. We interpret these macroscopic data in terms of binding-induced microscopic changes in the hydration of the DNA structures and the drug. Specifically, we find that netropsin binding induces the release of approximately 22 waters from the hydration shell of the poly(dAdT).poly(dAdT) heteropolymeric duplex, approximately 40 waters from the hydration shell of the poly(dA).poly(dT) homopolymeric duplex, and about 53 waters from the hydration shell of the poly(dA).poly(dT), induces the release of 18 more water molecules than netropsin binding to the heteropolymeric duplex, poly(dAdT).poly(dAdT). On the basis of apparent molar volume, phi V, and apparent molar adiabatic compressibility, phi Ks, values for the initial drug-free and final drug-bound states of the two all-AT duplexes, we propose that the larger dehydration of the poly(dA).poly(dT) duplex reflects, in part, the formation of a less hydrated poly(dA).poly(dT)-netropsin complex compared with the corresponding poly(dAdT).poly(dAdT)-netropsin complex. In conjunction with our previously published entropy data [Marky, L. A., & Breslauer, K. J. (1987) Proc. Natl. Acad. Sci. U.S.A. 84, 4359-4363], we calculate that each water of hydration released to the bulk solvent by ligand binding contributes 1.6 cal K-1 mol-1 to the entropy of binding. This value corresponds to the average difference between the partial molar entropy of water in the bulk state and water in the hydration shells of the two all-AT duplexes. When netropsin binds to the poly(dT).poly(dA).poly(dT) triplex, the changes in both volume and compressibility suggest that the binding event induces more dehydration of the triplex than of the duplex state. Specifically, we calculate that netropsin binding to the poly(dT).poly(dA).poly(dT) triplex causes the release of 13 more waters than netropsin binding to the poly(dA).poly(dT) duplex.(ABSTRACT TRUNCATED AT 400 WORDS)