Abstract
Heterochromatin plays an important role in eukaryotic cellular functions, including gene silencing, higher-order chromatin structure, genome stability and so on. Heterochromatin protein 1 (HP1), a key component of heterochromatin, is conserved from fission yeast to mammals. HP1 binds to histone H3K9me, a hallmark of heterochromatin, through its N-terminal chromodomain (CD) and self-dimerizes and recruits other chromatin proteins through its C-terminal chromo shadow domain (CSD), acting as an epigenetic reader. Between the CD and CSD is an unstructured, less conserved hinge region, which has been implicated in nucleic acid binding. The molecular dissection of the fission yeast HP1 orthologue, Chp2, recently reported in this journal, elucidated the cooperative DNA binding of the hinge and N-terminus of the CSD, which contributes to the stable association with heterochromatin and gene silencing. In this commentary, we focus on the mechanisms involving the HP1 hinge region, which is more than a simple linker.
Abbreviations
- HP1
Heterochromatin protein 1
- H3K9me
histone H3K9me
- CD
chromodomain
- CSD
chromo shadow domain
- Chp2
Chromodomain-containing protein2
- Swi6
Switching 6
- SHREC
Snf2/Hdac-containing Repressor Complex
- Clr4
Cryptic loci regulator 4
- MAT
mating-type
- IDRs
intrinsically disordered regions
- CPC
chromosome passenger complex
- INCENP
inner centromere protein
Gene expression regulation is critical for virtually every functional aspect of cells and organisms. HP1 was initially identified by immunofluorescence as a protein localized at constitutive heterochromatin in the polytene chromosomes from Drosophila melanogaster larval salivary glands (1,). Subsequent studies revealed its function in gene silencing, as it suppressed the position effect variegation of genes near heterochromatin (2). Currently, it is well accepted that HP1 assembles condensed heterochromatin and silences genes within it.
HP1 is evolutionarily conserved from the fission yeast Schizosaccharomyces pombe (S. pombe) to humans (3). Fission yeast has two HP1 family proteins, Chromodomain-containing protein 2 (Chp2) and Switching 6 (Swi6), which form different complexes to drive histone deacetylation and methylation, respectively. Chp2 associates with the Snf2/Hdac-containing Repressor Complex (SHREC) histone deacetylase, which includes the chromatin remodeler Mit1. Swi6 associates with RNAi-dependent silencing machinery and recruits the histone H3K9 methyltransferase Cryptic loci regulator 4 (Clr4). Both Chp2 and Swi6 facilitate gene silencing.
A comparison of the amino acid sequences of HP1 and Polycomb, another protein involved in gene silencing, revealed the homologous sequence named the chromodomain (CD), which binds to the methylated lysine of histone H3K9, a heterochromatin marker. The chromo shadow domain (CSD) was subsequently detected, which does not bind to methylated lysine but instead self-dimerizes and recruits other proteins (4–5). HP1 protein families share the common structure of an N-terminal CD and a C-terminal CSD linked by a hinge region (Fig. 1). A heterochromatin marker H3K9me3 is recognized by the CD, and then a protein complex is formed at the site through the CSD. This coordination may explain how heterochromatin is established and maintained by HP1.
Model of HP1 binding to DNA and chromatin for its function. The hinge and N-terminal CSD regions of Chp2, the yeast homologue of HP1, cooperatively bind DNA to facilitate heterochromatin assembly and gene silencing. Basic amino acid residues (+) in both regions are critical for the DNA binding. For simplicity, the self-dimerization of Chp2 is omitted. It remains to be determined whether Chp2 binds to the DNA region wrapped around the nucleosome, as shown here, or to the linker DNA connecting nucleosomes in vivo.
An additional way for HP1 to associate with heterochromatin is nucleic acid binding, which is mediated by the hinge region that bridges the CD and CSD. The hinge region is less conserved and structurally disordered. Regardless of the low similarities, this region in different species commonly binds DNA/RNA without sequence specificity (8–14,). Studies of human HP1alpha have shown that a cluster of basic amino acid residues in the hinge region binds DNA (8, 14, 15). Yeast Chp2 also contains stretches of basic amino acids in this region.
Previously, Nishibuchi et al. demonstrated that nucleic acid binding in this region, in collaboration with the N-terminal region of HP1, contributes to its nucleosome-binding specificity (16,). Molecular simulations illustrated basic amino acids in the N-terminal region of the hinge region bind DNA (17).
A recent study by Rahayu et al. revealed the unprecedented binding mode of Chip2, involving the cooperative DNA binding of the hinge and CSD (Fig. 1) (18). The authors demonstrated, by biochemical fractionation, that Chp2 associates with chromatin independently of Clr4 and Mit1, implying that Chp2 binds chromatin differently from the other fission yeast HP1 protein, Swi6. Using electrophoretic mobility shift assays with recombinant proteins, the authors showed that a cluster of basic residues, including four arginines and one lysine in the hinge region, is critical for the direct DNA binding. The replacement of these positively charged amino acids with alanine diminished the binding ability. In the nearby CSD N-terminus, two consecutive lysines immediately follow a glutamine. Again, their alanine replacement disrupted the DNA binding, and the simultaneous alanine replacements of the basic residues in the hinge and CSD resulted in the complete loss of DNA binding.
An important question to be addressed is whether the DNA binding by Chp2 plays a role in gene silencing in vivo. Using a silencing assay of S. pombe cells, Rahayu et al. analysed the expression levels of a marker gene introduced into the mating-type (MAT) locus, which is known to be subject to heterochromatic silencing. The mutant S. pombe cells with alanine replacements in the hinge and the N-terminus of the CSD were defective in gene repression, implying the role of DNA binding in heterochromatic silencing.
The authors further showed that the DNA-binding activities of Chp2 are involved in its in vivo stability and stable association with heterochromatin, by subcellular fractionations and ChIP-qPCR analyses of representative heterochromatin loci, including centromere, telomere and MAT locus. These demonstrations clarified that Chp2 functions by DNA binding, in addition to epigenetic mark recognition and protein recruitment for heterochromatin assembly.
Several issues remain to be investigated. First, it is not clear which part of the nucleosome arrays HP1 binds to, in vivo. The structure of human HP1 bound to the H3K9me3-containing dinucleosome was previously determined by cryo-EM (19,), and showed that the H3K9me3 nucleosomes are connected by an HP1 dimer and assembled into a higher order chromatin structure. The linker DNA between the nucleosomes did not directly interact with HP1. It would be interesting to determine whether this is also the case in vivo. It would also be interesting to investigate whether the HP1 dimer binds to nucleosomal DNA instead of linker DNA, and whether self-dimerized CSD modulates the DNA-binding activity of the hinge region (14, 16).
Second, it is not known how the basic patch in HP1 functions in DNA compaction. Provocative observation is that HP1 exists in biomolecular condensates in the nucleus. Two previous studies showed that phase separation, a phenomenon for non-membranous organelle formation, drives heterochromatin formation in Drosophila and human cells (20, 21,). Proteins that undergo liquid–liquid phase separation often contain intrinsically disordered regions (IDRs). HP1 harbours an IDR in its hinge region, and the hinge promotes the formation of phase-separated liquid droplets. In line with this speculation, arginine-enriched domains are sequence features for the condensation of other nuclear bodies (22).
The HP1 hinge plays another role during mitosis. A population of HP1 binds the chromosome passenger complex (CPC), which is enriched at the inner centromere and functions in accurate chromosome segregation for chromosome stability. The CPC consists of the inner centromere protein (INCENP) as a scaffold, Aurora kinase B as a catalytic subunit and other components. INCENP interacts with the hinge region of HP1 (23, 24,) through its disordered sequence, leading to a structured form (25), suggesting the significance of the HP1 hinge in the protein complex assembly.
Since the discovery of HP1 in Drosophila polytene chromosomes, substantial amounts of knowledge have been accumulated. HP1 consists of the CD, which reads H3K9me3, and the CSD for self-dimerization and the recruitment of other factors, thus supporting the maintenance and propagation of the epigenetic markers on chromatin. In addition, the hinge is not merely a connecting moiety bridging the CD and CSD, but has its own function in nucleic acid binding and protein stability. To fully understand HP1, more detailed dissections and in vivo studies are necessary. Mutations in human genome sequences may dictate the role of HP1 under physiological and disease conditions (26–28).
Acknowledgements
We thank Dr. Jun-ichi Nakayama (National Institute for Basic Biology, Japan) and the members of Saitoh’s laboratory for fruitful discussions.
Funding
This work was supported by KAKENHI grants from the Japan Society for the Promotion of Science (JSPS) [JP22K19466, JP23H00411 and JP24H0,1382], and the Daiichi Sankyo Foundation of Life Science (to N.S.).
Author Contributions
H.T. and N.S. wrote the manuscript.
Conflict of Interest
None is declared.
