The structure-function relationships in the nucleosome, the fundamental unit of chromatin, remain only partially understood due to limitations of experimental approaches and the high computational cost of molecular dynamics (MD) simulations. These limitations are particularly severe when it comes to capturing the full conformational ensemble of the histone tails. Here, we present an efficient simulation strategy that generates a converged conformational ensemble of histone tails by combining the accuracy of the explicit solvent with the efficiency of a novel implicit solvent model, which includes the explicit treatment of K + ions abundant in the nucleus, GBION. A close agreement between calculated and experimental tail-DNA binding thermodynamics allowed us to develop a quantitative model of the histone tail-DNA binding and its influence on DNA accessibility for DNA-binding proteins, including transcription factors (TFs). Several testable predictions are made, including tail-DNA binding strength scales linearly with the tail charge; each unit charge reduction (e.g., by lysine acetylation) weakens binding by ∼0.7 k B T nearly independent of the specific site. By contrast, the effect of acetylation of core residues is specific, varies by order of magnitude and can even change sign. Histone tails can alter the preferred TF binding motifs along nucleosomal DNA or even block the target sequences from TF binding, as tail-DNA contacts penalize extended motifs at specific sites. Histone tails hyperacetylation decreases the cumulative tail-DNA binding energy by ∼35 k B T, which corresponds to a substantial fraction of the total DNA-histone binding energy. Hyperacetylation can restore the most preferred TF target sequence.
Kolesnikov et al. (Sun,) studied this question.