Residues flanking the ARKme3T/S motif allow binding of diverse targets to the HP1 chromodomain: Insights from molecular dynamics simulations
Introduction
HP1 (Heterochromatin Protein 1) proteins have multiple functions, including participation in the formation and maintenance of heterochromatin, transcription regulation and RNA splicing, DNA repair, telomere maintenance and chromosome separation [[1], [2], [3], [4], [5]]. Their versatility can be attributed to their ability to bind numerous partner proteins as well as nucleic acids, and typically they have multiple isoforms. There are three isoforms in mammals – HP1α, β, and γ (also called CBX5, 1 and 3, respectively) – with differing chromosomal localizations and post-translational modification (PTM) patterns [[6], [7], [8]], which further tune their interactions.
HP1 proteins have two structured domains, which are connected and flanked by flexible, intrinsically disordered regions (IDRs, Fig. 1A). The chromodomain (CD, from CHROmatin MOdifier) is a well-known reader of histone protein H3 N-terminal tail lysine 9 methylation, a key heterochromatin-associated PTM marker [[9], [10], [11]]. CD has also been shown to bind other IDRs with methyllysine, namely N-terminal regions of histone H1 [12] and histone methyltransferase EHMT2 (also called G9a protein) [13], see Fig. 1C. There is also evidence of its binding to H3 methylated at K23 [14,15]. The chromo-shadow domain (CSD) is responsible for HP1 homodimerization. CSD homodimers mediate most HP1 interactions with partner proteins [5,[16], [17], [18], [19], [20]]. An IDR separating the two domains, the hinge region, mediates HP1s' interactions with nucleic acids. In this manner, HP1s not only associate with chromatin but also affect transcription and co-transcriptional splicing processes [2,[21], [22], [23]]. Flexible N- and C-terminal HP1 regions modulate binding affinities and specificities of the CD and CSD domains, respectively [18,24,25]. Broad ranges of interactions mediated by different regions of the HP1s and their ability to dynamically exchange binding partners allow them to serve as chromatin-associated hub proteins [26,27].
Structures of HP1(CD) with H3K9, H1K26 and EHMT2K185 lysine-methylated peptides have been experimentally determined [28,29]. All three of these target peptides share an ARKme3S/T motif, which is replaced by ATKme3A in the H3K23 target (Fig. 1C). CD recognizes the Kme3 (or Kme2), referred to as the 0 position of the bound peptide, via a hydrophobic cage (Fig. 1B). The backbone of the preceding segment is bound in a β-sheet style in the peptide-binding groove formed between the CD N-terminal tail and L3 loop. The −2 residue of the target, alanine, is anchored in a small hydrophobic pocket in the binding groove. –1 (arginine) and − 3 residue sidechains are oriented in the opposite direction, towards an anionic clamp formed by CD residues Glu29 (N-tail) and Asp68 (loop L3). H3 targets do not have clamp-neutralizing residues at the −3 position but the other two targets do (Fig. 1C). In the +1 position, serine or threonine form a sidechain H-bond to CD Glu62. Available experimental structures do not show any specific contacts of other residues of the bound peptides, and there is no structural information on CD binding to H3K23 peptide.
Experimental techniques are required to determine structures of proteins and their complexes, but they provide limited information on dynamics of the studied systems due to the ensemble-averaged nature of the collected data. Specific limitations of each technique inevitably restrict both the accuracy and detail of the resulting structures. Among others, potential roles of molecular motions and rare conformations involved in binding may be obscured. These limitations may be addressed by combining experimental structural data with computational approaches, particularly atomistic molecular dynamics (MD) simulations, which can portray detailed, atomic-level structural dynamics on pico- to millisecond time scales [[30], [31], [32]].
Some details of HP1 CD-H3K9 binding have already been characterized using MD simulations. Simulations of the CD-H3K9 complex with different analogues of Kme3 led to proposals that hydrophobic effects play more important roles in Kme3 readout than cation-π interactions [25]. It has also been suggested that the H3K9 peptide has higher observed affinity for the HP1β isoform than the HP1γ isoform mainly due to enhanced electrostatic interactions of R8 (R−1) and K14 (K+5) with the CD N-terminal stretch, which is more acidic in the β isoform [33]. In the same study, attempts were made to explain experimentally observed [34,35] weakening of the CD-H3K9 binding by the H3S10phos PTM. An increase in flexibility following this PTM was observed in 100 ns simulations, but the trajectories did not provide clear indications of a specific mechanism. Later microsecond-scale simulations showed that H3S10phos PTM impedes S10 binding to Glu62 and further weakens electrostatic interactions between R8 and CD by forming an intramolecular R8-S10phos salt bridge [36]. In addition, a study of effects of HP1 N-terminal tail phosphorylation on CD-H3K9 binding via 50 ns replica-exchange enhanced-sampling MD simulations in conjunction with NMR and SAXS have shown that the phosphorylation extends the tail and enhances its interactions with H3K9 peptide [25]. A recent study of full-length HP1 binding to H3K9 methylated dinucleosome using coarse-grained MD simulations [37] suggests that HP1 preference for dinucleosomes over mononucleosomes is due to interaction between the hinge region and DNA joining the nucleosomes, and that the γ isoform binds to dinucleosomes less efficiently than the α isoform because its hinge region is shorter and less basic.
In the study presented here we applied MD techniques to study HP1γ CD's binding to its four known targets. Molecular modelling allowed us to explore the behavior of flexible tails, which were omitted in experimental structures. Focusing on the role of the ARKme3S/T motif and its sequence context in binding, the simulations revealed that cationic residues as far as seven amino acids from Kme3 affect the binding and can compensate for the lack of clamp charge neutralization by the −3 residue in H3 targets. CD dynamics near the clamp also depend on the −4 residue of the bound target, since glutamine stabilizes CD-target interaction. We modelled the CD-H3K23 complex and the results suggest that the H3K23 peptide's interactions are analogous to those of the other targets but involve different residues. Results also neatly explain why H3K27 does not efficiently bind CD, in contrast to H3K23. We also studied effects of two experimentally detected HP1(CD) PTMs on the structure and dynamics of free CD and discuss here their potential effects on peptide binding. In addition, we critically evaluate the performance of AMBER ff19SB/OPC and ff14SB/TIP3P force fields for modelling these quite dynamic protein-protein complexes. Both provide reasonable agreement with available experimental data, but ff19SB provides somewhat better results for the focal systems.
Section snippets
Residue nomenclature and numbering
Here, we use three- and one-letter codes for amino acids of HP1 and its binding targets, respectively. Positions of the target peptide residues are specified relative to Kme3 by a subscript, e.g. R−1. HP1 residue numbering is based on human HP1γ. Kme3 numbering in Fig. 1C is based on human H1.4, H3.1, and EHMT2.1 sequences. The Kme3 number is also included in names of H3 targets to distinguish H3K9, H3K23 and H3K27 (in these abbreviations the sign for K methylation is omitted). If a mutation
Results
We aimed to elucidate effects of the targets' sequence variability on their recognition by the CD by modelling the HP1γ CD domain's interactions with several targets. Effects of threonine PTM modifications in the CD domain and the performance of protein simulation force fields are also addressed. Unless otherwise specified, ff19SB/OPC results are described.
Discussion
We have used MD simulations to explore structural dynamics of complexes formed by the HP1γ CD domain and its diverse IDR targets, which cannot be directly inferred from available ensemble-averaged experimental data. The simulations confirm that the bound state is associated with substantial local dynamics, which allow different residues along the target sequences to form analogous interactions that enhance the sequence flexibility of the CD targets. Besides investigating the role of structural
Conclusions
HP1 is a chromatin hub protein with a CD domain that binds specific Kme3 marks found mainly on histone tails (Fig. 1C). We report here microsecond-scale MD simulations intended to complement available experimental data on CD-target complexes through exploration of their dynamic behavior. In addition to simulations based on experimental structures, we have modelled the CD-H3K23 complex, for which there was no structural information. The simulations show that the bound state is associated with
CRediT authorship contribution statement
Pavlína Pokorná: Conceptualization, Methodology, Investigation, Writing - original draft, Writing - review & editing. Miroslav Krepl: Methodology, Validation. Jiří Šponer: Conceptualization, Resources, Funding acquisition, Writing - review & editing.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
We thank Klaudia Mráziková for help with simulation analyses. This work was supported by the Czech Science Foundation (grant number: 18-07384S) and Brno PhD Talent Financial Aid to P.P.
References (95)
- et al.
HP1a: a structural chromosomal protein regulating transcription
Trends Genet.
(2014) - et al.
Mechanisms of functional promiscuity by HP1 proteins
Trends Cell Biol.
(2014) - et al.
The HP1 protein family: getting a grip on chromatin
Curr. Opin. Genet. Dev.
(2000) - et al.
Heterochromatin protein 1 is extensively decorated with histone code-like post-translational modifications
Mol. Cell. Proteomics
(2009) - et al.
Understanding the relationship between DNA methylation and histone lysine methylation
Biochim. Biophys. Acta
(2014) - et al.
HP1 binds specifically to Lys(26)-methylated histone H1.4, whereas simultaneous Ser(27) phosphorylation blocks HP1 binding
J. Biol. Chem.
(2005) - et al.
A method for systematic mapping of protein lysine methylation identifies functions for HP1 beta in DNA damage response
Mol. Cell
(2013) - et al.
DNA microarray profiling of genes differentially regulated by three heterochromatin protein 1 (HP1) homologs in Drosophila
Biochem. Bioph. Res. Co.
(2013) - et al.
Characterization of chromoshadow domain-mediated binding of heterochromatin protein 1 alpha (HP1 alpha) to histone H3
J. Biol. Chem.
(2012) - et al.
Peptide recognition by heterochromatin protein 1 (HP1) chromoshadow domains revisited: plasticity in the pseudosymmetric histone binding site of human HP1
J. Biol. Chem.
(2017)
Recognition and specificity determinants of the human cbx chromodomains
J. Biol. Chem.
Using simulation to interpret experimental data in terms of protein conformational ensembles
Curr. Opin. Struc. Biol.
Principles of protein structural ensemble determination
Curr. Opin. Struc. Biol.
Structural ro of RKS Motifs in cromatin iteractions: a molecular dynamics study of HP1 bound to a variably modified histone tail
Biophys. J.
Interactions of HP1 bound to H3K9me3 dinucleosome by molecular simulations and biochemical assays
Biophys. J.
SPFP: speed without compromise - a mixed precision model for GPU accelerated molecular dynamics simulations
Comput. Phys. Commun.
Molecular dynamics simulations of water and biomolecules with a Monte Carlo constant pressure algorithm
Chem. Phys. Lett.
Numerical-integration of cartesian equations of motion of a system with constraints - molecular-dynamics of N-alkanes
J. Comput. Phys.
VMD: visual molecular dynamics
J. Mol. Graphics Modell.
The isolated C-terminal nuclear localization sequence of the breast cancer metastasis suppressor 1 is disordered
Arch. Biochem. Biophys.
Tudor: a versatile family of histone methylation 'readers'
Trends Biochem. Sci.
The PHD finger: a versatile epigenome reader
Trends Biochem. Sci.
Structural insight into the zinc finger CW domain as a histone modification reader
Structure
Karyopherin-beta2 recognition of a PY-NLS variant that lacks the proline-tyrosine motif
Structure
New developments in force fields for biomolecular simulations
Curr. Opin. Struc. Biol.
Computational and theoretical advances in studies of intrinsically disordered proteins
Curr. Opin. Struct. Biol.
Residue-specific alpha-helix propensities from molecular simulation
Biophys. J.
Secondary structure propensities in peptide folding simulations: a systematic comparison of molecular mechanics interaction schemes
Biophys. J.
Heterochromatin protein 1 (HP1): interactions with itself and chromatin components
Biophys. Rev.
Does heterochromatin protein 1 always follow code?
Proc. Natl. Acad. Sci. U. S. A.
10-year anniversary of a resource for post-translational modification of proteins
Nucleic Acids Res.
Localization and phosphorylation of HP1 proteins during the cell cycle in mammalian cells
Chromosoma
Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain
Nature
Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins
Nature
Automethylation of G9a and its implication in wider substrate specificity and HP1 binding
Nucleic Acids Res.
Histone peptide microarray screen of chromo and Tudor domains defines new histone lysine methylation interactions
Epigenet. Chromatin.
The HP1a disordered C terminus and chromo shadow domain cooperate to select target peptide partners
Chembiochem
SUV39h-independent association of HP1 beta with fibrillarin-positive nucleolar regions
Chromosoma
HP1 and the dynamics of heterochromatin maintenance
Nat. Rev. Mol. Cell Biol.
Histone H3 lysine 9 trimethylation and HP1 gamma favor inclusion of alternative exons
Nat. Struct. Mol. Biol.
HP1 binding to native chromatin in vitro is determined by the hinge region and not by the chromodomain
EMBO J.
Coordinated methyl and RNA binding is required for heterochromatin localization of mammalian HP1 alpha
EMBO Rep.
Extended string-like binding of the phosphorylated HP1 alpha N-terminal tail to the lysine 9-methylated histone H3 tail
Sci. Rep.
In vivo dynamics of Swi6 in yeast: evidence for a stochastic model of heterochromatin
Mol. Cell. Biol.
Maintenance of stable heterochromatin domains by dynamic HP1 binding
Science
Structural basis of the chromodomain of Cbx3 bound to methylated peptides from histone h1 and G9a
PLoS One
RNA structural dynamics as captured by molecular simulations: a comprehensive overview
Chem. Rev.
Cited by (2)
Phosphorylated HP1α-Nucleosome Interactions in Phase Separated Environments
2023, Journal of the American Chemical Society