Improving Electrostatic Descriptions of Ions Around Double-Stranded DNA

Abstract

Ions play an essential role in governing the structure and function of nucleic acids, due to the large negative charge associated with the nucleic acid backbone. The addition of even small numbers of multivalent, positively charged ions induces intra-strand attraction in DNA and thus efficiently packages the extended polymer into compact toroids. Anomalous small angle X-ray scattering (ASAXS) has emerged as a powerful technique to report the spatial distribution of ions associated to nucleic acids with unprecedented levels of detail and resolution. To determine more detailed information about these highly mobile ions relative to the underlying nucleic acid surface requires tight coordination with theoretical or computational tools. Presently, very few robust theoretical or computational tools exist for understanding ion-nucleic interactions. Current atomically-detailed computational approaches that represent the solvent environment explicitly, as discrete water molecules and ions, are prohibitively expensive for systematic studies of biologically relevant structures on time scales needed to fully understand these interactions. The alternative approach of “implicit solvent” models, that represent solvent implicitly as a continuum, could potentially overcome this difficulty. However, most “implicit solvent” models reduce computational effort at the expense of simplifying assumptions that preclude their application to highly charged systems in the presence of concentrated salt solutions or multivalent ions. We have been examining the ability of traditional and “size-modified” Poisson-Boltzmann models to predict the distribution of multivalent ions at low concentration around B-form DNA. These predictions are tested through direct comparison with ASAXS data on similar systems. The goal of this work is to examine the role of size-exclusion in the well-known failure of traditional Poisson-Boltzmann approaches for describing multivalent ions in highly charged systems. This work lays the foundation for a systematic approach to improve implicit solvent models for nucleic acid systems.