Publications > Journals > Exploratory Research and Hypothesis in Medicine > Article Full Text


Role of Epigenetic Modification of N6-methyladenosine in Phase Separation

  • Si-Yu Liu1,
  • Yi Feng2,
  • Zi-Li Sun1,
  • Ming-Li Zou1,
  • Jun-Jie Wu2,
  • Zheng-Dong Yuan2 and
  • Feng-Lai Yuan1,2,3,*
 Author information
Exploratory Research and Hypothesis in Medicine 2020;():-

DOI: 10.14218/ERHM.2019.00028


In recent years, the development of biophysical analysis methods has crossed with macromolecular condensates in cells. Researchers are interested in membrane-less organelles assembling into biomolecule ‘aggregates’ with similar liquid-like properties of phase separation. Cell biologists now think that many of the membrane-less organelles observed in cells are formed by phase separation caused by interactions between proteins and nucleic acids. Phase separation, thus, becomes a major player in the control of a variety of biological functions. Nevertheless, the biophysical regulation of these cells is still poorly understood. Here, we reviewed the current literature that collectively reveals the roles of epigenetic modification of N6-methyladenosine (m6A) in phase separation.


Methylation, Phase separation, m6A, YTHDF protein


Phase transition/phase separation of biological macromolecules in cytoplasm has developed rapidly into a hot research area in recent years. Cells consist of many different organelles (or compartments) that separate cell functions. Organelles can be divided into compartments enveloped by lipid membranes (such as the nucleus, vacuoles, endoplasmic reticulum, and mitochondria) and organelles not enveloped by lipid membranes (such as P bodies, nucleoli, pericentric substances, stress granules, and reproductive particles). The distribution and assembly dynamics of non-membrane-bound compartments have aroused great interest over the last ten years and for good reason.13 Accumulating evidence has indicated that these non-membrane-bound organelles, such as Cajal bodies, nucleoli, stress granules, synaptic cytoskeleton, and microRNA-induced silencing complex, behave as fluid droplets, which undergo phase separation phenomenon.4,5

Through the study of these non-membrane-bound organelles, scientists have gained a deep understanding of the molecular mechanism of different diseases. Earlier work in this area focused on stress granules involved in cell survival and its relationship with amyotrophic lateral sclerosis. Recently, several studies provided us with fresh insight into the dynamic of RNA modifications and RNA-protein interactions in contributing to phase separation and non-membrane-bound organelles’ formation in cells.6 However, it has never been clarified what triggers the aggregation of some macromolecules rather than others in the same droplet.

Thus, we reviewed the current literature on the roles of the modification of N6-methyladenosine (m6A) in phase separation, with a focus on recent research findings regarding m6A regulation of phase separation.

Phase separation regulates m6A-modified mRNAs

A recent report shows that the modification of m6A could stimulate the formation of phase separation of mRNA and YTH N6-methyladenosine RNA binding proteins (YTHDFs),3 and subsequently affect the fate of those cells. m6A modification occurs via a methyltransferase complex (referred to as dedicated writers) mainly consisting of methyltransferase like protein 3 (METTL3), methyltransferase like protein 14 (METTL14) and Wilms tumor 1-associated protein (WTAP). This modification can be reversed by the demethylases (referred to as dedicated erasers) like fat mass and obesity-associated protein (FTO) and AlkB homolog 5 (ALKBH5). Moreover, the YTH domain containing several m6A-binding proteins, including YTHDF1, YTHDF2 and YTHDF3 in the cytoplasm, and the nuclear YTH domain containing protein 1 (YTHDC1) have been identified to be the ‘readers’ of m6A and to be able to regulate mRNA stability and translation, thereby mediating downstream effects (Fig. 1).79 Recent studies indicated that mRNA metabolism needs the active involvement of the modification of m6A, in particular modulating mRNA stability, determining cell fate, affecting transcriptional control of cell, and playing an important role in regulating the fate determination of various cell types,10 lipid metabolism,11 and immunity.12 Scientists selected three of the major m6A-binding YTH domain family proteins YTHDF1,2,3 as research subjects, trying to reveal the regulation mechanism of m6A on mRNA stability and translation efficiency.13 Primary protein sequence analysis has demonstrated that YTHDF protein consists of about 15 kDa of m6A-binding YTH domain and a low complexity region of about 40 kDa.14 It is speculated that the three genes may undergo liquid-liquid phase separation (LLPS) according to this structural arrangement (Fig. 2). YTHDF2 protein was the highest content of YTHDF paralogue purified from cells. Several studies have characterized that the factors regulating strength of intermolecular interactions, including temperature, pH and ionic strength, will affect the ultimate phase separation behavior.15 The gene solution of YTHDF2 was transparent at the low-temperature of 4 °C; however, when the temperature raised to 37 °C, the YTHDF2 gene solution became turbid; the solution became clear again at the low-temperature of 4 °C. This finding served to confirm that the YTHDF2 phenomenon was LLPS. Meanwhile, Ries et al.3 found that YTHDF2 protein at the physiological concentration (about 5 µM) increased sufficiently for phase separation occurring. Those authors additionally examined whether phase separation enhancement of YTHDF2 protein was regulated by m6A modification in vitro. They confirmed that a 65 nucleotide-long transcriptome regions containing 10 modified m6A caused a significant increase of the phase separation in YTHDF proteins. Consequently, the authors speculated that adjacent m6A moieties on RNA promote ‘multivalent interactions’ between the intrinsically disordered regions of YTHDF proteins that were bound to m6A through their YTH domain, thus facilitating phase separation.

m<sup>6</sup>A modification is dynamically regulated by writer, eraser, and reader protein.
Fig. 1  m6A modification is dynamically regulated by writer, eraser, and reader protein.

Writer protein is a methyltransferase, like WTAP, METTL3, and METTL14. Eraser protein is a demethylase, like FTO and ALKBH5 localized primarily in the nucleus. Reader protein is a binding methylated protein, which is divided into in cytoplasm and in nucleus. In the nucleus, m6A can be regulated by m6A readers YTHDC1, HNRNPA2B1, and IGF2BP1/2/3, while in the cytoplasm, m6A can be regulated by m6A readers YTHDF1/2/3, YTHDC2, and elF3. WTAP, Wilms tumor 1-associated protein; METTL, methyltransferase like protein; FTO, fat mass and obesity-associated protein; ALKBH5, AlkB homolog 5; YTHDC, YTH domain containing protein; HNRNPA2B1, heterogeneous nuclear ribonucleoprotein A2B1; IGF2BP1/2/3, insulin-like growth factor 2 mRNA binding protein 1/2/3; YTHDF1/2/3, YTH domain family proteins 1/2/3; Prrc2a, proline-rich coiled-coil 2A; HNRNPC, heterogeneous nuclear ribonucleoprotein C; elF3, eukaryotic initiation factor 3; SRSF2, serine and arginine rich splicing factor 2.

m<sup>6</sup>A is an intermediate linker between mRNA and YTHDF protein, which promotes YTHDF-mediated phase separation.
Fig. 2  m6A is an intermediate linker between mRNA and YTHDF protein, which promotes YTHDF-mediated phase separation.

To address whether m6A can influence the phase separation of YTHDF2 gene, Ries et al.3 confirmed that endogenous stress granules were formed after exposure to heat shock in m6A methyltransferase14-knockout mouse embryonic stem cells; however, the localization of YTHDF2 genes in the stress granules was obviously reduced. Thus, the localization of YTHDF2 to stress granules under stress conditions relies on its modification with m6A; that is, the subcellular localization of YTHDF2 gene is regulated by m6A-modified mRNAs. The translation efficiency of m6A-modified mRNA in methyltransferase14-knockout cells was obviously inhibited when cells are exposed to heat shock. Summarizing the above findings, Ries et al.3 affirmed that the RNA modification by m6A, especially polymethylated m6A-mRNAs, could recruit multiple YTHDF2 genes, and obviously enhanced the phase separation of YTHDF gene. Conversely, the phase separation of YTHDF2 was found to regulate the translation efficiency of polymethylated mRNA. Ultimately, these findings proposed a novel theory for the precision regulation of m6A modification on cellular biological processes.

In addition, Fu et al.16 recently shown that when cells were under oxidative stress, a large number of m6A-modified mRNAs were localized to stress granules. Additional studies indicated that YTHDF gene was very important for the formation of stress particles under oxidative stress. After knocking down of YTHDF by small interfering RNA, the level of stress granules decreased markedly. Interestingly, an article published in Nature proved that the knock-out of METTL14 would not affect the stress granules’ formation under heat shock conditions.3 While the article from bioRxiv indicated that the binding of YTHDF and m6A was hindered,16 and stress particle formation was markedly reduced. In other words, stress granules’ formation under thermal and oxidative stress may be triggered by dissimilar genes; the underlying precise regulatory mechanisms require further analysis.4 The two articles drew the reasonable conclusion from different perspectives; that is, m6A modification can promote the phase separation process of specific RNA-binding proteins.

Recently, Gao et al.17 found that multivalent m6A-containing RNA can increase the phase separation YTHDF genes in HEK293 cells. Their data indicated that the m6A promotion of phase separation of YTHDF proteins is related to the response of cells to stress. YTHDF1/2/3 comprise a family of m6A readers. The researchers first analyzed the sequence of these three proteins and found a large number of low complexity domains. Then, they verified that both the low complexity domains and full-length proteins of YTHDF exhibited phase separation. m6A has a key role in diverse biological processes by binding to the relevant YTH domain-containing genes, as well as binding to other translation initiation factors. Multivalent m6A-driven phase separation of YTHDF genes may contribute to the physiology of a cell in the stress response.

These studies provide an experimental basis for the regulation of m6A phase separation and gene expression regulation and also provide an understanding for the regulation of key gene expression during cell fate determination and disease development.1820 At present, the structures of membrane-less organelles of phase separation are very clear, and a subsequent major objective is to investigate the distinct properties of phase separation. Concentrated states of DNA, protein, and RNA are a primary aspect of intracellular compartments.21 It has become commonly accepted that large amounts of these condensates form via LLPS, resulting in liquid-like states of intracellular matter. Thus, intracellular liquid condensates can drive richer structural assembly than uniform droplets from multilayer structures, such as nucleoli and stress particles, to liquid crystal assemblies, such as spindle-shaped and actin condensates.21,22

Physical and chemical properties of phase separation will be an increasingly important subject for future challenges. Employing m6A-modified mRNAs governed by phase separation as a target to regulate various cellular and physiological processes, such as transcriptional control, stress exposure, signal transduction, etc. will also provide a new insight for m6A-regulated phase separation in different research fields. Research on the relationship between phase separation and disease is progressing rapidly. The current research is primarily focusing on the molecular pathogenesis of neurodegenerative diseases.23 Some gene point mutations, such as FUS protein, can increase the transition from a solid phase to a liquid solvent and may be the reason for the formation of insoluble protein accumulation which is commonly found in neurodegenerative disorders.24 Moreover, Nuclear transport has been shown to decrease nucleic acid binding protein concentration in the cytoplasm, dissolve the LLPS structure, and prevent neurotoxicity formation attributed to solid aggregation.25


Recent studies have indicated that the modification of m6A is involved in the regulation process of mRNA metabolism, in particular in modulating mRNA stability, determining cell fate, affecting different cellular biological systems, and taking a responsible role in cell fate determination, immunity, and lipid metabolism. Scientists have tried to explain the regulation mechanisms of m6A to the translation efficiency and stability of mRNA based on three of the major m6A-binding proteins. Primary gene sequence analysis showed that the YTHDF genes consisted of an m6A-binding YTH location of about 15 kDa and a low complexity location of about 40 kDa. Thus, the authors speculated that the YTHDF genes may undergo LLPS based on this structural arrangement (Fig. 2).


Phase separation is increasingly recognized as key regulatory biological macromolecules in cytoplasm. It is time to expand the research focus to phase separation for numerous biological functions. Studying the m6A-mediated phase separation mechanism, especially finding or screening key proteins will provide a theoretical basis for verifying the effectiveness of target macromolecules.





liquid-liquid phase separation


YTH N6-methyladenosine RNA binding protein


Wilms tumor 1-associated protein


methyltransferase like protein


fat mass and obesity-associated protein


AlkB homolog 5


YTH domain containing protein


heterogeneous nuclear ribonucleoprotein A2B1


insulin-like growth factor 2 mRNA binding protein 1/2/3


proline-rich coiled-coil 2A


heterogeneous nuclear ribonucleoprotein C


eukaryotic initiation factor 3


serine and arginine rich splicing factor 2





Funding for this study was provided by the Natural Science Foundation of China (81770876).

Conflict of interest

The authors have no conflicts of interest related to this publication.

Authors’ contributions

Designing, reviewing and revising the article (YFL), drafting of the manuscript (LSY), reviewing and drawing (FY, WJJ), collecting materials and revising the manuscript (YZD, ZML, SZL), approved the final version submitted (all the authors).


  1. Brangwynne CP, Eckmann CR, Courson DS, Rybarska A, Hoege C, Gharakhani J. Germline P granules are liquid droplets that localize by controlled dissolution/condensation. Science 2009;324:1729-1732 View Article
  2. Marnik EA, Updike DL. Membraneless organelles: P granules in Caenorhabditis elegans. Traffic 2019;20:373-379 View Article
  3. Ries RJ, Zaccara S, Klein P, Olarerin-George A, Namkoong S, Pickering BF. m(6)A enhances the phase separation potential of mRNA. Nature 2019;571:424-428 View Article
  4. Wang JH, Wang LY, Diao JB, Shi YG, Shi Y, Ma HH. Binding to m(6)A RNA promotes YTHDF2-mediated phase separation. Protein Cell 2019 View Article
  5. Shin Y, Chang YC, Lee DSW, Berry J, Sanders DW, Ronceray P. Liquid nuclear condensates mechanically sense and restructure the genome. Cell 2019;176:1518 View Article
  6. Alberti S, Gladfelter A, Mittag T. Considerations and challenges in studying liquid-liquid phase separation and biomolecular condensates. Cell 2019;176:419-434 View Article
  7. Liu SY, Feng Y, Wu JJ, Zou ML, Sun ZL, Li X. m(6)A facilitates YTHDF-independent phase separation. J Cell Mol Med 2019;24:2070-2072 View Article
  8. Wang Y, Mao J, Wang X, Lin Y, Hou G, Zhu J. Genome-wide screening of altered m6A-tagged transcript profiles in the hippocampus after traumatic brain injury in mice. Epigenomics 2019;11:805-819 View Article
  9. Luo GZ, MacQueen A, Zheng G, Duan H, Dore LC, Lu Z. Unique features of the m6A methylome in Arabidopsis thaliana. Nat Commun 2014;5:5630 View Article
  10. Wu Y, Xie L, Wang M, Xiong Q, Guo Y, Liang Y. Mettl3-mediated m(6)A RNA methylation regulates the fate of bone marrow mesenchymal stem cells and osteoporosis. Nat Commun 2018;9:4772 View Article
  11. Zhong X, Yu J, Frazier K, Weng X, Li Y, Cham CM. Circadian clock regulation of hepatic lipid metabolism by modulation of m(6)A mRNA methylation. Cell Rep 2018;25:1816-1828.e4 View Article
  12. Han D, Liu J, Chen C, Dong L, Liu Y, Chang R. Anti-tumour immunity controlled through mRNA m(6)A methylation and YTHDF1 in dendritic cells. Nature 2019;566:270-274 View Article
  13. Lu W, Tirumuru N, St Gelais C, Koneru PC, Liu C, Kvaratskhelia M. N(6)-Methyladenosine-binding proteins suppress HIV-1 infectivity and viral production. J Biol Chem 2018;293:12992-13005 View Article
  14. Zhuang M, Li X, Zhu J, Zhang J, Niu F, Liang F. The m6A reader YTHDF1 regulates axon guidance through translational control of Robo3.1 expression. Nucleic Acids Res 2019;47:4765-4777 View Article
  15. Ceballos AV, McDonald CJ, Elbaum-Garfinkle S. Methods and strategies to quantify phase separation of disordered proteins. Methods Enzymol 2018;611:31-50 View Article
  16. Fu Y, Zhuang X. m6A-binding YTHDF proteins promote stress granule formation by modulating phase separation of stress granule proteins. bioRxiv 2019:694455 View Article
  17. Gao Y, Pei G, Li D, Li R, Shao Y, Zhang QC. Multivalent m6A motifs promote phase separation of YTHDF proteins. Cell Res 2019;29:767-769 View Article
  18. Ji P, Wang X, Xie N, Li Y. N6-methyladenosine in RNA and DNA: an epitranscriptomic and epigenetic player implicated in determination of stem cell fate. Stem Cells Int 2018;2018:3256524 View Article
  19. Wang H, Xu B, Shi J. N6-methyladenosine METTL3 promotes the breast cancer progression via targeting Bcl-2. Gene 2020;722:144076 View Article
  20. Li T, Hu PS, Zuo Z, Lin JF, Li X, Wu QN. METTL3 facilitates tumor progression via an m(6)A-IGF2BP2-dependent mechanism in colorectal carcinoma. Mol Cancer 2019;18:112 View Article
  21. Shin Y, Brangwynne CP. Liquid phase condensation in cell physiology and disease. Science 2017;357:eaaf4382 View Article
  22. Weirich KL, Banerjee S, Dasbiswas K, Witten TA, Vaikuntanathan S, Gardel ML. Liquid behavior of cross-linked actin bundles. Proc Natl Acad Sci 2017;114:2131-2136 View Article
  23. Ambadipudi S, Biernat J, Riedel D, Mandelkow E, Zweckstetter M. Liquid-liquid phase separation of the microtubule-binding repeats of the Alzheimer-related protein Tau. Nat Commun 2017;8:275 View Article
  24. Qamar S, Wang G, Randle SJ, Ruggeri FS, Varela JA, Lin JQ. FUS phase separation is modulated by a molecular chaperone and methylation of arginine cation-pi interactions. Cell 2018;173:720-734.e15 View Article
  25. Duan Y, Du A, Gu J, Duan G, Wang C, Gui X. PARylation regulates stress granule dynamics, phase separation, and neurotoxicity of disease-related RNA-binding proteins. Cell Res 2019;29:233-247 View Article
  • Exploratory Research and Hypothesis in Medicine
  • eISSN 2472-0712