Effect of Tetrameric Base-Pair Context on the Sequence-Dependent Configurations of DNA Minicircles

Abstract

The biophysical characteristics of various DNA configurations, ranging from changes in localized base-pair twisting to global writhe, are known to be influenced by nucleotide sequence and composition. Decades worth of experimental findings have shown how regions of high adenine content exhibit high degrees of bending or how pyrimidine-purine dimers exhibit a high twist/low twist polymorphism, examples critical in protein interactions and genetic expression. Calculations and subsequent modeling in the past have been limited to dimeric base pair steps but the recent influx of high-resolution structural data has shown these steps to be influenced by both their 3′ and 5′ nearest neighbors, such as conformational differences of the internal AA dimers in the context of A-AA-A and A-AA-C tetramers. The growing database of high-resolution nucleic acid-containing structures can be harnessed to generate intrinsic rigid-body parameters and elastic constants of successive base pairs in all tetrameric contexts. Optimizing the elastic energy of DNA minicircles at the dinucleotide level has been used to determine the effects of sequence on global shape and elasticity. These optimizations make use of an array of intrinsic dimer models ranging from a simple linear model, a curved three-state model, or one of higher complexity with rigid-body parameters and elastic constants unique to each dimer, as well as impose spatial constraints to mimic behavior observed in protein-DNA interactions. These models have been further refined to include nearest neighbor interactions. Elastic energy minimization calculations were used on multiple circular configurations with sequences of high A⋅T content. Comparative analysis between optimization results are useful in exploring how purine and pyrimidine nucleotides before and after a dinucleotide sequence may affect polymer shape and elasticity.