In silico evidence for sequence-dependent nucleosome sliding

Abstract

Nucleosomes represent the basic building block of chromatin and provide an important mechanism by which cellular processes are controlled. The locations of nucleosomes across the genome are not random but instead depend on both the underlying DNA sequence and the dynamic action of other proteins within the nucleus. These processes are central to cellular function, and the molecular details of the interplay between DNA sequence and nucleosome dynamics remain poorly understood. In this work, we investigate this interplay in detail by relying on a molecular model, which permits development of a comprehensive picture of the underlying free energy surfaces and the corresponding dynamics of nucleosome repositioning. The mechanism of nucleosome repositioning is shown to be strongly linked to DNA sequence and directly related to the binding energy of a given DNA sequence to the histone core. It is also demonstrated that chromatin remodelers can override DNA-sequence preferences by exerting torque, and the histone H4 tail is then identified as a key component by which DNA-sequence, histone modifications, and chromatin remodelers could in fact be coupled.nucleosome repositioning | chromatin dynamics | molecular simulation | advanced sampling techniques T he basic building block of eukaryotic chromatin is the nucleosome, a DNA-protein complex containing 147 bp of DNA wrapped around a disk-like protein complex known as the histone octamer (1). Since nucleosomal DNA is inaccessible to other DNA-binding proteins, such as transcription factors and polymerases (2-4), the locations of nucleosomes represent an important mechanism by which cellular processes are controlled. Notably, nucleosome positions are dynamic, with changes in transcription levels, cellular state, and environmental factors resulting in different packagings of chromatin (5, 6). Proper packaging of genomic DNA is critical to cellular function, and a wide range of human diseases have been associated with defects in chromatin structure (7,8). Understanding the molecular factors that control the locations of nucleosomes, and how they are dynamically modulated, therefore represents a central goal of molecular biology and biophysics.It is now appreciated that the DNA sequence itself represents a key factor that governs the locations of nucleosomes. Different DNA sequences exhibit different affinities for the histone proteins, and as such, they form nucleosomes with probabilities that can differ by several orders of magnitude (9, 10). The dependence of nucleosome locations on DNA sequence originates from subtle differences in the intrinsic shape and flexibility of a specific DNA sequence, which lead to favorable electrostatic interactions between the DNA backbone and residues on the histone surface (11). In fact, this pronounced dependence on DNA sequence has led several authors to propose that a genetic code exists (12, 13) where the positions of 50% of nucleosomes in vivo are dictated by DNA sequence alone. Such a view, however, is not without controversy (14, 15),...

Critical Role of Histone Tail Entropy in Nucleosome Unwinding

“…In this study we model the nucleosome through the combination of two coarse-grained models: the three sites per nucleotide (3SPN2.C) DNA model and the associative memory, watermediated, structure and energy model (AWSEM) for proteins (23)(24)(25). The 3SPN2.C model is a three-site per nucleotide coarse-grained model for DNA which was parameterized to replicate key mechanical properties such as persistence length, melting temperature, and geometrical properties at varying ionic concentrations.…”

Static Kinks or Flexible Hinges: Conformational Distributions of Bent DNA Bound to Integration Host Factor Mapped by Fluorescence Lifetime Measurements

“…To examine how the presence or absence of the A-tract affects the population distribution of differently bent DNA conformations in the IHF-H1 complex, we analyzed the fluorescence 21 decay traces measured on the end-labeled IHF-H1 complex with the MEM, as before. The lifetime distributions measured for IHF-H1 also show two peaks ( Figure 5 and SI Figure S3); however, the DNA in the IHF-H1 complex prefers the low-FRET conformation, with only 36 ± 2% in the high FRET state (E ≈ 0.49 ± 0.03) and a larger fraction, 64 ± 2%, in the low FRET conformation (E ≈ 0.11 ± 0.01).…”

“…12 Precise measurements of the accessible protein-DNA conformations in solution are needed to elucidate the functional roles of these dynamic fluctuations as well as to provide experimental checkpoints in the development and improvement of computational models of sequence-dependent DNA deformability and stabilization of bent DNA by proteins. [13][14][15][16][17][18][19][20][21][22] Several lines of evidence suggest that DNA in nonspecific complexes indeed adopts a range of bent conformations. Evidence for conformational heterogeneity is well documented in the case of the nonspecific architectural histone-like nucleoprotein HU that is known to bend DNA into a U-shape.…”

Chromatin remodelers couple inchworm motion with twist-defect formation to slide nucleosomal DNA

“…Modes of nucleosome repositioning can be classified in two types depending on whether sliding is accompanied by the rotation of DNA around its axis (Niina, Brandani, Tan, & Takada, 2017). In the rotation-uncoupled mode, sliding proceeds via large steps of about a DNA turn (~10 bp) (Lequieu, Schwartz, & de Pablo, 2017;Niina et al, 2017;Schiessel, Widom, Bruinsma, & Gelbart, 2001), possibly facilitated by the formation of loops (Lequieu et al, 2017;Schiessel et al, 2001). On the other hand, in the rotation-coupled mode, DNA sliding proceeds at small steps of 1 bp via a screw-like motion (Niina et al, 2017), facilitated by the formation of twist defects (Brandani, Niina, Tan, & Takada, 2018;Edayathumangalam, Weyermann, Dervan, Gottesfeld, & Luger, 2005;Kulić & Schiessel, 2003;Richmond & Davey, 2003;Shaytan et al, 2016;van Holde & Yager, 1985).…”

“…Notably, coarse-graining approaches have been successfully applied to the study of nucleosome dynamics (Chang & Takada, 2016;B. Zhang, Zheng, Papoian, & Wolynes, 2016), including spontaneous repositioning (Brandani et al, 2018;Lequieu et al, 2017;Niina et al, 2017), and ATP-dependent molecular motors (Flechsig & Mikhailov, 2010;Koga & Takada, 2006).…”