Interactions between chromodomains and trimethyllysine marks on histone H3 peptides

Recent findings suggest that a variety of post-translational modifications (PTMs) found on N-terminal tails of histones are intricately involved in DNA packaging and directly control levels of gene expression. These modifications include methyllysine, methylarginine, phosposerine, phosphothreonine, and acetyllysine. Methyllysine marks on the N-terminal tail histone H3 are known to recruit effector proteins that can modify chromatin structure and regulate gene expression. Trimethyllysine 4 of histone H3 recruits CHD1 (chromo-ATPase/helicase-DNA binding domain 1), which is part of a chromatin-remodeling complex associated with active transcription. Additionally, trimethyllysine 9 of histone H3 recruits heterochromatin protein 1alpha(HP1alpha), which stabilizes heterochromatin, which is typically associated with repressed gene expression. I investigated the protein-protein interactions in each of these complexes to explore the driving force and the selectivity of recognition. The tandem chromodomain of CHD1 binds H3 K4Me3 with an aromatic cage consisting of two tryptophan residues (Trp64 and Trp67) forming cation-gamma interactions with the trimethyllysine. Arginine 2 of histone H3 is involved in an H-bond with the backbone of Gly66 of the tandem chromodomain and a cation-gamma interaction with Trp67. The effect of incorporating methylarginine and citrulline at position 2 in H3 K4Me3 on CHD1 binding affinity was explored. The results show that symmetric dimethylarginine and citrulline weakened binding affinity while asymmetric dimethylarginine enhanced binding affinity. This study demonstrates the significance of these three modifications and how they may play a role in regulating gene expression by affecting protein-protein interactions. The HP1alpha chromodomain binds to H3 K9Me2 with a KD of 20 muM H3 K9Me3 with a KD of 17 muM. The chromodomain contains a three-membered aromatic cage and a glutamate (Glu52) around di- and trimethyllysine. The aromatic residues are involved in a cation-gamma interaction with methyllysine and Glu52 forms a water-mediated H-bond to dimethyllysine. The histone tail is also inserted between two beta-strands of the chromodomain to form a 3-stranded beta-sheet. First, Glu52 was modified to enhance selectivity for H3 K9Me3 over H3 K9Me2. The E52Q mutant had a 2.5 fold weaker binding affinity to H3 K9Me2 (KD=52 muM) and maintained the same binding affinity to H3 K9Me3 (KD=15 muM) most likely because glutamine is a weak H-bond acceptor compared to glutamate. Second, beta-sheet interactions between HP1alpha chromodomain and H3 K9Me3 were investigated. Residue Thr6 of the histone tail forms cross-strand interactions with Ala25 and Asp62 of the chromodomain. Each of these three residues was systematically substituted for amino acids known to have high beta-sheet propensity and form favorable sidechain-sidechain interactions. These studies demonstrated the applicability of information gleaned from model systems and statistical studies to protein-protein recognition. Lastly, two PTM-recognition domains, derived from naturally occurring effector proteins were coupled together to create a coupled-receptor construct to visualize dual modifications on a single histone tail. Two HP1alpha chromodomains were coupled to determine if they were functional for detecting a synthetic peptide with two H3 K9Me2 sequences as proof that coupled receptors can have cooperative binding for a peptide with two dimethyllysine marks. Taken together, these studies provide a new mechanistic insight into the proteinprotein interactions between chromodomains and methyllysine marks on N-terminal histone H3 tails, which are important for sequence selectivity and binding affinity