Post-translational modifications (PTMs) on histone proteins are known as epigenetic marks that demarcate the status of chromatin. These modifications are ‘read' by specific reader proteins, which in turn recruit additional factors to modulate chromatin accessibility and the activity of the underlying DNA. Accumulating evidence suggests that these modifications are not restricted solely to histones, many non-histone proteins may function in a similar way through mimicking the histones. In this commentary, we briefly discuss a systematic study of the discovery of histone H3 N-terminal mimicry proteins (H3TMs), and their implications in chromatin regulation and drug discoveries.
Post-translational modifications (PTMs) on histone proteins play important roles in regulating the dynamics and accessibility of chromatin. These modifications, including methylation, acetylation, and phosphorylation that mainly occur in the N-terminal unstructured tails of histones, have been proposed as the ‘Histone Code' that can demarcate the status of the underlying DNA . A critical function of the ‘Histone Code' is to recruit reader proteins that can recognize specific modifications in a sequence- and state-dependent manner and recruit additional factors, which in turn modulate the status of chromatin or the activity of the underlying DNA . Accumulating evidence suggests that these modifications are not restricted solely to histones; hundreds, or even thousands of non-histone proteins have been identified as substrates of the histone-modifying enzymes . This raises a question that whether the modifications on non-histone proteins can mimic the ‘Histone Code' in binding to histone readers.
The study led by Chen et al. was set out to determine whether in the human proteome there are proteins that have an H3-like N-terminal motif and whether they interact with known H3K4me3-reader proteins in a methylation-dependent manner . A systematic analysis of ∼20 000 annotated human proteins identified 257 candidate proteins that were named as the H3 N-terminal mimicry proteins (H3TMs). All these H3TMs contain the histone H3-like ‘M-Z-R-X-K' motif, in which Z represents one of seven amino acids that are permissive for initiator methionine cleavage and X represents any amino acid. Among all the H3TMs, 48 proteins possess the motif at their N-termini, which are more likely to be functional mimics of the unstructured H3 N-terminal tail. To identify potential interactions between these proteins and histone readers, Chen et al. synthesized methylated and unmethylated peptides of seven representative H3TMs to probe an in-house protein domain microarray that contains ∼200 methyl-binding domains. All the tested methyl peptides showed bindings to known histone H3K4me3-binding PHD fingers and Tudor domains, albeit each with a unique interaction profile. Importantly, these interactions were validated by individual peptide pulldowns using both recombinant protein domains and full-length proteins from whole-cell extracts. In addition, these H3TMs can be modified in vitro by the histone H3K4 methyltransferases, including SET7/9, MLL4, and PRDM9, suggesting potential crosstalk in cells .
An interesting finding Chen et al. made is that some known histone readers showed even higher binding affinities to non-histone proteins than histones. For example, compared with the H3K4me3 peptide, methyl-VRK1 showed a three folds higher binding affinity to the PHF2 PHD finger. The crystal structure of PHF2 PHD finger in complex with the methyl-VRK1 peptide revealed extensive interactions between PHF2 and the first seven residues of VRK1 involving electrostatic, hydrogen bonds, aromatic cage, and van der Waals contacts, accounting for the high binding affinity . PHF2 belongs to the KDM7 histone demethylase family and contains a PHD finger and a Jumonji-C (JMJC) domain. PHF2 demethylates H3K9me2 through the catalytic JMJC domain and this process relies on the interaction between its PHD finger and histone H3K4me3 . The actual role of VRK1 in modulating PHF2-mediated H3K9 demethylation remains to be determined. It is possible that the methylated VRK1 may positively or negatively impact PHF2's recruitment to chromatin through either cooperation or competition with H3K4me3 in binding to PHF2 PHD finger (Figure 1). One example for the histone competition model of H3TMs is the non-structure protein 1 (NS1) of influenza A virus H3N2 subtype, which contains an H3K4-like motif (A-R-S-K). NS1 interacts with the human PAF1 transcription elongation complex (hPAF1C) through its histone mimicry motif, and this interaction is critical for suppression of the host antiviral immune response . Another example is the protein E of SARS-CoV-2, which caused the global COVID-19 pandemic. The transmembrane segment of protein E shares local sequence similarity to the N terminus of histone H2A. This histone mimic, likely in its acetylated form, interacts with the bromodomain-containing proteins BRD2 and BRD4, thus disrupting their histone binding activity and consequently the host immune defense . Examples of cooperation between histone and histone mimics in reader binding have also been reported. For instance, the histone H3K9 methyltransferase G9a contains an H3K9-like sequence, methylation of which facilitates the recruitment of heterochromatin protein 1 (HP1) to chromatin . Moreover, the transcription factor Snail can recruit lysine specific demethylase 1 (LSD1) to target gene promoters through its H3 mimicry sequence residing in its SNAG domain . In all these cases, both histones and histone mimics are recognized by the same reader proteins, how the dynamics between readers and histones or histone mimics are controlled remains to be revealed.
Working models for the H3 N-terminal mimicry proteins (H3TMs).
In this study, Chen et al. focused on the study of the bindings between H3TMs and the known H3K4me3 readers, including PHD fingers and Tudor domains. The strategy, however, can be applied to the studies of other types of PTMs on H3, such as acetylation and phosphorylation, as well as mimicry proteins to other histone tails. Finally, the concept of histone mimicry also has a great potential in drug development. Indeed, several histone-mimic inhibitors have been developed, such as the synthetic antagonist of the BET bromodomains, I-BET, and the ENL YEATS domain inhibitor XL-13m [10,11]. This novel strategy opens a new window for the development of drugs for the treatment of viral infections and other human diseases.
The authors declare that there are no competing interests associated with the manuscript.
This work was supported in part by grants from NCI (CA255506) to H.W. and (CA204020) to X.S.