| Abstract: | This article is a first step in an attempt to reevaluate the relative role of different contributions to the energetics of DNA in salt solutions. To identify individual terms yielding such contributions a new derivation is given of the generalized Poisson–Boltzmann equation, which includes correlation effects, and explicitly shows terms ignored in the regular Poisson–Boltzmann approach. A general method based on the Boundary Element Technique is discussed, which can be used to evaluate these terms in the next steps of the reevaluation. An implementation of this method for the solution of the nonlinear Poisson–Boltzmann equation is described in detail, and is used to compute the ionic atmosphere around DNAs modeled as cylinders with helical distributions of charges. In the B-type DNA models, it is found that the ion densities in the minor and major grooves near the DNA surface differ by up to threefold. This difference is ca. 10-fold for Z-type DNA models. There are 20–25% differences in the magnitude of the maximum ionic charge density between DNA models of the same type. The addition of excess salt (up to 0.15 M) changes this maximum by only 10–15%. This change is not proportional to the concentration of excess salt. The contributions of different factors to the stabilization of alternative forms of DNA are evaluated. These factors are: (1) interactions between the phosphates, (2) interactions of phosphates with water, (3) interactions of phosphates with the ionic cloud, (4) interactions within the ionic cloud, (5) entropy of the ionic cloud. It is found that regardless of large variations in the counterion distributions around different DNAs, energetic contributions from these distributions are similar (?12.65 ± 0.6 kcal/mol · cell). The calculated change in stabilization per unit cell of models of B and Z-type DNAs due to 0.15 M excess NaCl is only ?0.56 ± 0.02 kcal/mol, indicating no tendency toward B-Z transition in this concentration range. Significantly larger variations of the order of 10 kcal/mol per unit cell can result from factors 1–2. Possible effects of the realistic DNA-solvent boundaries on the energetics of DNA solutions are discussed. |
