Abstract
Here we investigate the role of epigenetics in the formation, transcription regulation, maintenance and termination of several non-canonical chromatin structures. Using two examples, we demonstrate how studying non-canonical structures may reveal underlying mechanisms with implications for disease and propose intriguing epigenetic avenues for further exploration.
Human DNA, approximately 3 billion base pairs in length, is organized into chromosomes. To be compactly stored in the nucleus, DNA is spooled by histone proteins to form a nucleoprotein complex. The nucleosome is the fundamental structural unit of chromatin, and comprises about 147 base pairs of DNA wrapped around an octamer consisting of two copies each of H2A, H2B, H3 and H4 histones. The accessible histone tails undergo post-translational modification (PTM), forming packed signaling platforms that dynamically regulate gene transcription and other DNA-templates processes. These histone PTMs can be enzymatically or non-enzymatically added or removed and are recognized by specific protein readers that jointly influence chromatin structure and function.
The canonical organization of chromatin comprises a hierarchical structure that extends from individual nucleosomes to chromosomal territories. However, seminal findings from the past few decades have shown that this organization is far from monolithic, and chromatin can adopt several alternative conformations. Various non-canonical chromatin structures have been identified, with different implications in pathology from viral infection to tumor initiation, development and metastasis. Although some genomic features of these structures have been characterized, we are only beginning to unravel the precise mechanisms by which they are formed and maintained, as well as how they direct gene expression, particularly in terms of their epigenetic contributions.
The epigenetic landscape of non-canonical chromatin structures
Hints about the potential presence of epigenetic abnormalities in non-canonical chromatin structures were gleaned from several reports that chromosomes (or fragments of chromosomes) trapped in micronuclei, as a consequence of chromosomal instability, undergo epigenetic dysregulation. This is particularly evident at the level of histone PTMs and chromatin accessibility. Such changes have been shown to be heritable and long-lasting and continue to affect the structure of the impacted chromosomes long after the micronuclei are reincorporated into the nucleus1,2. The micronucleation of a chromosome isolates it within a distinct compartment from its counterparts in the nucleus. As it loses its compartmental context, changes in histone PTM and chromatin accessibility begin to occur1,2. Arguably, similar processes take place in non-canonical chromatin structures, such as covalently closed circular DNA (cccDNA) and extrachromosomal circular DNA (eccDNA) including extrachromosomal DNA (ecDNA; also known as double minutes), which also lose their context within the higher-order chromatin structure, even when remaining within the primary nucleus. Further evidence suggests that ecDNA can become encapsulated and sequestered within micronuclei, or even expulsed from the cell3.
Non-canonical chromatin structures are also encountered in exosomes. The emerging exosomal biomarker small extracellular vesicle (sEV)-DNA is thought to facilitate horizontal gene transfer between cells. Although exosomes possess an abundant quantity of histones, whether they interact with the DNA within it is unknown. Neutrophil extracellular traps (NETs), a scaffold consisting of chromatin and cytoplasmic proteins, are also released by neutrophils to degrade virulence factors and kill bacteria, and specific histone PTMs in NETs may contribute to autoimmune disease, such as lupus erymathosus. Table 1 summarizes the non-canonical chromatin species identified and the diseases associated with their presence, absence or imbalance in cells.
Table 1 |.
Non-canonical chromatin species and their associated diseases
| Non-canonical chromatin species | Associated disease(s) |
|---|---|
| Micronuclei | Various cancer types |
| cccDNA | Hepatitis B (HBV), Human papillomavirus (HPV), Human immunodeficiency virus (HIV), Herpes simplex virus (HSV), Epstein–Barr virus (EBV) |
| ecDNA | Various cancer types |
| Small polydisperse circular DNA and micro-DNA | Various cancer types |
| c- and t-circles | Various cancer types |
| rDNA circles | Mostly abundant in healthy tissues, but imbalance may cause gene loss. |
| sEV-DNA (extracellular) | Various cancer types |
| NETs (extracellular) | Cystic fibrosis, asthma, chronic obstructive pulmonary disease (COPD), autoimmune diseases |
Together, these findings illustrate that chromatin is susceptible to abnormal epigenetic alterations when it loses its higher-order structure. Consequently, exploring epigenetic changes in non-canonical chromatin species represents a promising frontier, given their notable implications in various diseases. Below, we describe two primary examples of non-canonical chromatin structures that have been rigorously investigated genomically, and consider proposed methods to unravel their complete epigenetic landscape.
Example 1: ecDNA
Normal cells contain diploid chromosomes that are compartmentalized into chromosome territories within the nucleus. However, cancer cells can produce small circular DNA fragments, known as eccDNA, that do not associate with any of the nuclear chromosomes4. Typically, eccDNA structures are not sufficiently large to contain entire genes. By contrast, ecDNA can reach megabase-pair resolution in size and may include several genes, often oncogenes such as EGFR and MYC. The amplification of these oncogenes is the most common gain-of-function change in cancerous cells and is reflected by copy number changes in whole-genome sequencing analyses. These findings indicate that some copy number changes are focal, rather than occurring across entire chromosomes or fragments of chromosome arms5. Seminal work by the Mischel, Regenberg, Chang, Henssen and Bafna groups, among others, has shown that focal amplification of oncogenes most commonly occur through ecDNA, and that these events are detected in at least half of human cancers across different tumor types5–7. Oncogene amplification by ecDNA correlates with poor disease outcome and a higher likelihood of patients developing therapeutic resistance, as this increases the heterogeneity of intratumoral copy number4. The combined efforts of the Chang, Ren, Bafna and Mischel groups have shown that the physical structure of ecDNA is circular and that it lacks higher-order compaction8. The same study found that ecDNA is decorated by active histone PTMs that confer enhanced chromatin accessibility8. Meanwhile, work by the Whetstine group and others showed that aberrant regulation of the epigenome can contribute to oncogene amplification by ecDNA9. Together, these provide potential epigenetic mechanisms to how oncogenes within ecDNA can be rapidly transcribed and translated, unleashing the oncogenes that they contain.
These findings underscore the importance of studying the epigenetic landscape of non-canonical structures in more detail. It would be insightful to comprehensively profile the composition and distribution of histone variants and PTMs within these structures in an unbiased fashion to deconvolute how it may govern ecDNA transcriptional regulation. By contrast, efforts should be directed towards deciphering the epigenetic abnormalities that give rise to ecDNA. Promising experimental approaches could involve high-throughput screening of ecDNA generation after genetic or pharmacological perturbations to the epigenome. Unraveling the dynamics of histone deposition and its associated proteins will also yield important insights into how transcription is regulated in ecDNA. Towards this goal, reliable systems for in vitro and in vivo ecDNA isolation and construction should be established. Although some elegant methods such as CRISPR-CATCH have been developed to isolate ecDNA10, they are currently limited to the isolation of non-chromatinized DNA. Techniques such as CRISPR-C to create eccDNA or using the Cre-lox system to create ecDNA in a temporal manner provide potential opportunities to establish a useful system for studying eccDNA and ecDNA from an epigenetic perspective11,12.
Example 2: cccDNA
Viruses can hijack the host system used by endogenous species of eccDNA to create non-canonical chromatin species comprised of their own genetic material. For example, cccDNA represents a vital viral minichromosome structure that assembles in the nucleus of hepatocytes infected with hepatitis B virus (HBV). At its core, this structure comprises histones derived from host cells and the viral genome13. The host histones involved in cccDNA formation undergo PTMs reminiscent of canonical cellular euchromatin14. Although it is unclear how chromatinization is driven, including which histone variants or modifications are installed and where, the mature cccDNA is essential to facilitate replication of HBV, making it a crucial step in the virus’s life cycle13.
We recently discovered that chromatinization of cccDNA a an important role in regulating the transcriptional state of the hepatitis B viral protein X gene, which encodes the key HBx infection protein. However, a knowledge gap remains concerning how histone PTMs within this structure influence viral gene transcription. By reconstituting cccDNA in vitro, we were able to obtain a deeper understanding of how chromatinization regulates the initiation of viral infection15. Together with the identification of active transcription histone PTMs, this suggests that epigenetic and conformational changes of the chromatin structure of cccDNA may facilitate the transcription of viral genes. Moreover, this powerful platform can now be used to study epigenetic mechanisms of HBV regulation in more detail. Further understanding of the HBV epigenetic landscape and other viral non-canonical chromatin structures will lead to a deeper appreciation of how viral genomes are amplified and how they evade current treatment regimens. Future research that focuses on how histone PTMs affect viral transcription can reveal fundamental mechanisms that will determine whether modifying epigenetic enzymes can modulate this effect, thereby opening new potential modes of treatment using epigenetic or epigenetic-adjacent drugs. The powerful in vitro platforms used to study the epigenetic regulation of non-canonical chromatin structures provide one of the cornerstones to begin unravelling this tantalizing possibility.
Outlook
Non-canonical chromatin species are implicated in a wide range of diseases. Although substantial efforts have been made to understand their genomic features, investigation of the state of their epigenetic landscape, including the composition of histone variants and PTMs, that govern transcription within these structures remains underexplored. Future endeavors to profile the resident histones, their distribution, and their regulation by epigenetic enzymes should provide insights into how transcriptional regulation is achieved in these unique chromatin environments. Better systems need to be developed to capture these structures in the most physiologically relevant state possible, as well as to reconstitute them in vitro, which will enable researchers to investigate them mechanistically. By enhancing our understanding of this matter, more effective therapeutic strategies can be designed to mitigate the pathophysiological consequences associated with these intriguing, yet potentially threatening, chromatin structures.
Footnotes
Competing interests
The authors declare no competing financial and/or non-financial interests in relation to the work described.
References
- 1.Papathanasiou S et al. Nature 619, 184–192 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Agustinus AS et al. Nature 619, 176–183 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Carroll SM et al. Mol Cell Biol 8, 1525–1533 (1988). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Verhaak RGW, Bafna V & Mischel PS Nat. Rev. Cancer 19, 283–288 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Turner KM et al. Nature 543, 122–125 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kim H et al. Nat. Genet 52, 891–897 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Deshpande V et al. Nat. Comm 10, 1–14 (2019). [Google Scholar]
- 8.Wu S et al. Nature 575, 699–703 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gray ZH et al. Cell 186, 4528–4545.e18 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hung KL et al. Nat. Genet 54, 1746–1754 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Møller HD Nucleic Acids Res 46, e131–e131 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Pradella D et al. Preprint at bioRxiv 10.1101/2023.06.25.546239 (2023). [DOI] [Google Scholar]
- 13.Allweiss L, Dandri M, Protzer U & Nassal M Viruses 9, 156 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Belloni L et al. Proc. Natl Acad. Sci. USA 106, 19975–19979 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Prescott NA et al. Preprint at bioRxiv 10.1101/2023.03.03.531011 (2023). [DOI] [Google Scholar]