Peng Y et al. (2021), Binding of regulatory proteins to nucleosomes i...

XB-ART-58517

Nat Commun 2021 Sep 06;121:5280. doi: 10.1038/s41467-021-25568-6.

Show Gene links Show Anatomy links

Binding of regulatory proteins to nucleosomes is modulated by dynamic histone tails.

Peng Y , Li S , Onufriev A , Landsman D , Panchenko AR .

???displayArticle.abstract???
Little is known about the roles of histone tails in modulating nucleosomal DNA accessibility and its recognition by other macromolecules. Here we generate extensive atomic level conformational ensembles of histone tails in the context of the full nucleosome, totaling 65 microseconds of molecular dynamics simulations. We observe rapid conformational transitions between tail bound and unbound states, and characterize kinetic and thermodynamic properties of histone tail-DNA interactions. Different histone types exhibit distinct binding modes to specific DNA regions. Using a comprehensive set of experimental nucleosome complexes, we find that the majority of them target mutually exclusive regions with histone tails on nucleosomal/linker DNA around the super-helical locations ± 1, ± 2, and ± 7, and histone tails H3 and H4 contribute most to this process. These findings are explained within competitive binding and tail displacement models. Finally, we demonstrate the crosstalk between different histone tail post-translational modifications and mutations; those which change charge, suppress tail-DNA interactions and enhance histone tail dynamics and DNA accessibility.

???displayArticle.pubmedLink??? 34489435
???displayArticle.link??? Nat Commun
???displayArticle.grants??? [+]

Species referenced: Xenopus laevis
Genes referenced: h2bc21 ino80 smarca5
GO keywords: chromatin [+]

???attribute.lit??? ???displayArticles.show???

	Fig. 1: Binding of histone tails to nucleosomal and linker DNA in the context of the full nucleosome. a Interconversions of DNA-bound and unbound tail conformations. The conformational snapshots are taken from the last frame of each simulation run and superimposed onto the initial models by minimizing RMSD values of Cα atoms in histone cores. While we observe multiple binding/unbinding events in the simulations, only a few snapshots are shown for clarity. b A total number of full histone tail binding/unbinding events observed in all simulations for both copies of histones. c Full histone tail residence time. Each point represents a binding/unbinding event observed in simulations for each histone copy (n(H2A_N) = 174, n(H2A_C) = 359, n(H2B) = 173, n(H3) = 31, n(H4) = 160). Data points with residence time shorter than 10 ns are excluded as this time is required for establishing stable interactions with DNA. An unbound state for the full tail is defined if no more than 10% of the tail residues maintain contacts with DNA (other cut-off values are given in Supplementary Fig. 2). Box-plot elements are defined as: center line, median; box limits, upper and lower quartiles; whiskers are drawn at values equal to 1.5× interquartile range; d The histone tail–DNA standard binding free energy derived from counting the number of bound/unbound events in histone tail conformational ensemble in all simulations for both copies of histones (number of binding/unbinding events ≥ 5, Supplementary Table 4). e A representative run shows a fraction of DNA bound residues and tail binding/unbinding events during simulation for H2B tails (see Supplementary Figs. 6–10 for other tails). Source data are provided as a Source Data file.
	Fig. 2: Nucleosomal and linker DNA solvent accessibility modulated by histone tail binding. a A mean number of contacts between histone tails and nucleosomal/linker DNA averaged over all independent simulation runs for two copies (n = 44) plotted in the DNA coordinate frame, zero corresponds to the dyad position and superhelical locations (SHL) are shown as integers; a combined conformational ensemble from both copies of histone tails is shown. The error bars represent standard errors of the mean calculated from independent simulation runs. b Mean number of contacts between histone tails and DNA mapped onto the molecular surface of the nucleosomal and linker DNA. c Mean values of changes of DNA solvent accessibility imposed by tail binding. The error bars represent standard errors of the mean calculated from independent simulation runs for two copies (n = 44). d Percentage of frames with more than 20% SASA decrease upon tail binding per DNA base pair. Source data are provided as a Source Data file
	Fig. 3: Histone tail-modulated recognition modes of nucleosomes by binding partners. a A summary of 131 available nucleosome complex structures classified based on their binding entity and function. b An example of chromatin remodeler ISWI which binds to both histone H4 tail and DNA (PDB: 6IRO). Here coordinates of histone tails are taken from the PDB structure. Nucleosome-binding proteins are colored in orange and histone tails are colored using their canonical colors. c Analysis of the contact numbers between the nucleosomal/linker DNA and partners (averaged over all structures of nucleosome complexes) plotted in DNA coordinate frame. The error bars represent standard errors calculated from the number of contacts of different nucleosome complex structures (n = 86). The top track shows the presence or absence of five or more contacts between histone tails and DNA regions (indicating the histone tail preferred binding regions). d The fraction of interface overlap between tail–DNA and partner–DNA binding interfaces. It is calculated for each nucleosome complex structure and the distribution is smoothed using the gaussian kernel function. e, f Examples of INO80 chromatin remodeler (PDB: 6HTS) and UV-damaged DNA-binding protein (PDB: 6R8Z) targeting overlapping regions on DNA. Histone tail representative conformations are taken from simulations and superimposed onto the PDB structures. The intensity of the color of the DNA surface is scaled with the mean number of contacts between histone tails and DNA as in Fig. 2b. Source data are provided as a Source Data file.
	Fig. 4: Recognition of nucleosomal DNA and histone tails by binding partners depicted via electrostatic potential analysis. Nucleosome complex structures where proteins interact with DNA are classified based on their histone tail binding modes. Positively charged DNA-binding interfaces are highlighted for chromatin remodeler INO80 (PDB: 6HTS) and UV-damaged DNA-binding protein (PDB: 6R8Z) where partners do not interact with histone tails in structures. Another two representative examples, chromatin remodeler ISWI (PDB: 6IRO) and polycomb repressive complex 2 (PDB: 6WKR), show the DNA-binding interfaces and the partner acidic patches recognized by H3 and H4 tails. Electrostatic potentials are mapped onto the molecular surfaces of nucleosome-binding proteins. Blue and red colors indicate the positive and negative electrostatic potentials, and the intensity of the color is scaled with the surface electrostatic potential values. H3 and H4 tails are colored light blue and green. The molecular surface of nucleosomal and linker DNA is highlighted with cyan color. Core histone regions are not shown for clarity.
	Fig. 5: Histone tail post-translational modifications and mutations modulate histone tail–DNA interaction modes and DNA accessibility. a Mean number of histone tail–DNA contacts for different types of modifications per residue. The result of simulations of model D with and without corresponding modifications are shown in different colors and modifications are shown by a symbol next to the residue. b Mean numbers of nucleosomal and linker DNA contacts with histone tails for different types of modifications per DNA base pair. For each type of modification, the reported values are averaged, and the error bars represent the standard errors of the mean calculated from independent simulation runs for two copies (n = 10). The locations and types of modifications and mutations are listed in Supplementary Table 2. The mean number of contacts of unmodified tails was calculated using the first 1600 ns frames from the trajectory of the simulation of Model D with the AMBER package (Supplementary Table 1). Source data are provided as a Source Data file.
	Fig. 6: A generalized model explaining how tails and their modifications can modulate nucleosomes’ interactions with nucleosome-binding proteins. a Histone tails’ interactions with DNA may in some cases modulate the accessibility of DNA to binding partners; nucleosome-binding proteins compete with histone tails for binding to DNA. b Post-translational modifications and mutations in histone tails can suppress tail–DNA interactions, enhance histone tail dynamics, and regulate the binding of proteins to nucleosomes. c Some nucleosome-binding proteins may exhibit multivalent binding modes and recognize both histones and DNA. d DNA-binding domains (DBDs) of nucleosome-binding proteins compete with histone tails if they recognize DNA and reader domains compete with DNA if they recognize tails; interactions of DBD with nucleosome can displace histone tails from their DNA preferred binding modes and increase their accessibility for recognition by reader domains.

References [+] :

Allahverdi, The effects of histone H4 tail acetylations on cation-induced chromatin folding and self-association. 2011, Pubmed