Polycomb repressive complexes 1 and 2 (PRC1/2) maintain transcriptional silencing of developmental genes largely by catalyzing the formation of mono-ubiquitinated histone H2A at lysine 119 (H2AK119ub1) and trimethylated histone H3 at lysine 27 (H3K27me3), respectively. How Polycomb domains are reprogrammed during mammalian preimplantation development remains largely unclear. Here we show that, although H2AK119ub1 and H3K27me3 are highly colocalized in gametes, they undergo differential reprogramming dynamics following fertilization. H3K27me3 maintains thousands of maternally biased domains until the blastocyst stage, whereas maternally biased H2AK119ub1 distribution in zygotes is largely equalized at the two-cell stage. Notably, while maternal PRC2 depletion has a limited effect on global H2AK119ub1 in early embryos, it disrupts allelic H2AK119ub1 at H3K27me3 imprinting loci including Xist. By contrast, acute H2AK119ub1 depletion in zygotes does not affect H3K27me3 imprinting maintenance, at least by the four-cell stage. Importantly, loss of H2AK119ub1, but not H3K27me3, causes premature activation of developmental genes during zygotic genome activation (ZGA) and subsequent embryonic arrest. Thus, our study reveals distinct dynamics and functions of H3K27me3 and H2AK119ub1 in mouse preimplantation embryos.
This is a preview of subscription content, access via your institution
Open Access articles citing this article.
Scientific Reports Open Access 16 August 2022
Nature Communications Open Access 25 July 2022
Epigenetics & Chromatin Open Access 18 July 2022
Subscribe to Nature+
Get immediate online access to Nature and 55 other Nature journal
Subscribe to Journal
Get full journal access for 1 year
only $6.58 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Get time limited or full article access on ReadCube.
All prices are NET prices.
All sequencing data were deposited in the Gene Expression Omnibus under accession number GSE153531.
The code for this study is available at https://github.com/YiZhang-lab/H2Aub_K27me3_preimplantation_dynamics.
de Napoles, M. et al. Polycomb group proteins Ring1A/B link ubiquitylation of histone H2A to heritable gene silencing and X inactivation. Dev. Cell 7, 663–676 (2004).
Wang, H. et al. Role of histone H2A ubiquitination in Polycomb silencing. Nature 431, 873–878 (2004).
Gao, Z. et al. PCGF homologs, CBX proteins, and RYBP define functionally distinct PRC1 family complexes. Mol. Cell 45, 344–356 (2012).
Tavares, L. et al. RYBP–PRC1 complexes mediate H2A ubiquitylation at Polycomb target sites independently of PRC2 and H3K27me3. Cell 148, 664–678 (2012).
Yu, J. R., Lee, C. H., Oksuz, O., Stafford, J. M. & Reinberg, D. PRC2 is high maintenance. Genes Dev. 33, 903–935 (2019).
Blackledge, N. P., Rose, N. R. & Klose, R. J. Targeting Polycomb systems to regulate gene expression: modifications to a complex story. Nat. Rev. Mol. Cell Biol. 16, 643–649 (2015).
Cao, R. et al. Role of histone H3 lysine 27 methylation in Polycomb-group silencing. Science 298, 1039–1043 (2002).
Min, J., Zhang, Y. & Xu, R. M. Structural basis for specific binding of Polycomb chromodomain to histone H3 methylated at Lys 27. Genes Dev. 17, 1823–1828 (2003).
Wang, L. et al. Hierarchical recruitment of Polycomb group silencing complexes. Mol. Cell 14, 637–646 (2004).
Blackledge, N. P. et al. Variant PRC1 complex-dependent H2A ubiquitylation drives PRC2 recruitment and Polycomb domain formation. Cell 157, 1445–1459 (2014).
Cooper, S. et al. Targeting Polycomb to pericentric heterochromatin in embryonic stem cells reveals a role for H2AK119u1 in PRC2 recruitment. Cell Rep. 7, 1456–1470 (2014).
Cooper, S. et al. Jarid2 binds mono-ubiquitylated H2A lysine 119 to mediate crosstalk between Polycomb complexes PRC1 and PRC2. Nat. Commun. 7, 13661 (2016).
Kalb, R. et al. Histone H2A monoubiquitination promotes histone H3 methylation in Polycomb repression. Nat. Struct. Mol. Biol. 21, 569–571 (2014).
Laugesen, A., Hojfeldt, J. W. & Helin, K. Molecular mechanisms directing PRC2 recruitment and H3K27 methylation. Mol. Cell 74, 8–18 (2019).
Cohen, I., Bar, C. & Ezhkova, E. Activity of PRC1 and histone H2AK119 monoubiquitination: revising popular misconceptions. Bioessays 42, e1900192 (2020).
Dahl, J. A. et al. Broad histone H3K4me3 domains in mouse oocytes modulate maternal-to-zygotic transition. Nature 537, 548–552 (2016).
Liu, X. et al. Distinct features of H3K4me3 and H3K27me3 chromatin domains in pre-implantation embryos. Nature 537, 558–562 (2016).
Zhang, B. et al. Allelic reprogramming of the histone modification H3K4me3 in early mammalian development. Nature 537, 553–557 (2016).
Inoue, A., Jiang, L., Lu, F. & Zhang, Y. Genomic imprinting of Xist by maternal H3K27me3. Genes Dev. 31, 1927–1932 (2017).
Inoue, A., Jiang, L., Lu, F., Suzuki, T. & Zhang, Y. Maternal H3K27me3 controls DNA methylation-independent imprinting. Nature 547, 419–424 (2017).
Chen, Z. & Zhang, Y. Maternal H3K27me3-dependent autosomal and X chromosome imprinting. Nat. Rev. Genet. 22, 555–571 (2020).
Zheng, H. et al. Resetting epigenetic memory by reprogramming of histone modifications in mammals. Mol. Cell 63, 1066–1079 (2016).
Skene, P. J. & Henikoff, S. An efficient targeted nuclease strategy for high-resolution mapping of DNA binding sites. eLife 6, e21856 (2017).
Fursova, N. A. et al. Synergy between variant PRC1 complexes defines Polycomb-mediated gene repression. Mol. Cell 74, 1020–1036 (2019).
Ernst, J. & Kellis, M. ChromHMM: automating chromatin-state discovery and characterization. Nat. Methods 9, 215–216 (2012).
Blackledge, N. P. et al. PRC1 catalytic activity is central to Polycomb system function. Mol. Cell 77, 857–874 (2020).
Tamburri, S. et al. Histone H2AK119 mono-ubiquitination is essential for Polycomb-mediated transcriptional repression. Mol. Cell 77, 840–856 (2020).
Schmitges, F. W. et al. Histone methylation by PRC2 is inhibited by active chromatin marks. Mol. Cell 42, 330–341 (2011).
Yuan, W. et al. H3K36 methylation antagonizes PRC2-mediated H3K27 methylation. J. Biol. Chem. 286, 7983–7989 (2011).
Yuan, G. et al. Histone H2A ubiquitination inhibits the enzymatic activity of H3 lysine 36 methyltransferases. J. Biol. Chem. 288, 30832–30842 (2013).
Matoba, S. et al. Loss of H3K27me3 imprinting in somatic cell nuclear transfer embryos disrupts post-implantation development. Cell Stem Cell 23, 343–354 (2018).
Borensztein, M. et al. Xist-dependent imprinted X inactivation and the early developmental consequences of its failure. Nat. Struct. Mol. Biol. 24, 226–233 (2017).
Inoue, A., Chen, Z., Yin, Q. & Zhang, Y. Maternal Eed knockout causes loss of H3K27me3 imprinting and random X inactivation in the extraembryonic cells. Genes Dev. 32, 1525–1536 (2018).
Chen, Z., Yin, Q., Inoue, A., Zhang, C. & Zhang, Y. Allelic H3K27me3 to allelic DNA methylation switch maintains noncanonical imprinting in extraembryonic cells. Sci. Adv. 5, eaay7246 (2019).
Almeida, M. et al. PCGF3/5-PRC1 initiates Polycomb recruitment in X chromosome inactivation. Science 356, 1081–1084 (2017).
Posfai, E. et al. Polycomb function during oogenesis is required for mouse embryonic development. Genes Dev. 26, 920–932 (2012).
Scheuermann, J. C. et al. Histone H2A deubiquitinase activity of the Polycomb repressive complex PR-DUB. Nature 465, 243–247 (2010).
Balasubramani, A. et al. Cancer-associated ASXL1 mutations may act as gain-of-function mutations of the ASXL1–BAP1 complex. Nat. Commun. 6, 7307 (2015).
Sahtoe, D. D., van Dijk, W. J., Ekkebus, R., Ovaa, H. & Sixma, T. K. BAP1/ASXL1 recruitment and activation for H2A deubiquitination. Nat. Commun. 7, 10292 (2016).
Dobrinić, P., Szczurek, A. & Klose, R. PRC1 drives Polycomb-mediated gene repression by controlling transcription initiation and burst frequency. Preprint at bioRxiv https://doi.org/10.1101/2020.10.09.333294 (2020).
Harris, C. et al. Conversion of random X-inactivation to imprinted X-inactivation by maternal PRC2. eLife 8, e44258 (2019).
Cohen, I. et al. PRC1 fine-tunes gene repression and activation to safeguard skin development and stem cell specification. Cell Stem Cell 22, 726–739 (2018).
Du, Z. et al. Polycomb group proteins regulate chromatin architecture in mouse oocytes and early embryos. Mol. Cell 77, 825–839 (2019).
Prokopuk, L. et al. Loss of maternal EED results in postnatal overgrowth. Clin. Epigenetics 10, 95 (2018).
Zenk, F. et al. Germ line-inherited H3K27me3 restricts enhancer function during maternal-to-zygotic transition. Science 357, 212–216 (2017).
Xia, W. et al. Resetting histone modifications during human parental-to-zygotic transition. Science 365, 353–360 (2019).
Wang, C. et al. Reprogramming of H3K9me3-dependent heterochromatin during mammalian embryo development. Nat. Cell Biol. 20, 620–631 (2018).
Xu, Q. et al. SETD2 regulates the maternal epigenome, genomic imprinting and embryonic development. Nat. Genet. 51, 844–856 (2019).
Weinberg, D. N. et al. The histone mark H3K36me2 recruits DNMT3A and shapes the intergenic DNA methylation landscape. Nature 573, 281–286 (2019).
Zhang, C. et al. The chromatin remodeler Snf2h is essential for oocyte meiotic cell cycle progression. Genes Dev. 34, 166–178 (2020).
Chen, Z. & Zhang, Y. Loss of DUX causes minor defects in zygotic genome activation and is compatible with mouse development. Nat. Genet. 51, 947–951 (2019).
Lu, F. et al. Establishing chromatin regulatory landscape during mouse preimplantation development. Cell 165, 1375–1388 (2016).
Sugimoto, M. et al. A simple and robust method for establishing homogeneous mouse epiblast stem cell lines by Wnt inhibition. Stem Cell Rep. 4, 744–757 (2015).
Inoue, A. & Zhang, Y. Nucleosome assembly is required for nuclear pore complex assembly in mouse zygotes. Nat. Struct. Mol. Biol. 21, 609–616 (2014).
Erkek, S. et al. Molecular determinants of nucleosome retention at CpG-rich sequences in mouse spermatozoa. Nat. Struct. Mol. Biol. 20, 868–875 (2013).
Langmead, B., Trapnell, C., Pop, M. & Salzberg, S. L. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome. Genome Biol. 10, R25 (2009).
Tarasov, A., Vilella, A. J., Cuppen, E., Nijman, I. J. & Prins, P. Sambamba: fast processing of NGS alignment formats. Bioinformatics 31, 2032–2034 (2015).
Krueger, F. & Andrews, S. R. SNPsplit: allele-specific splitting of alignments between genomes with known SNP genotypes. F1000Res 5, 1479 (2016).
Ramirez, F., Dundar, F., Diehl, S., Gruning, B. A. & Manke, T. deepTools: a flexible platform for exploring deep-sequencing data. Nucleic Acids Res. 42, W187–W191 (2014).
Meers, M. P., Bryson, T. D., Henikoff, J. G. & Henikoff, S. Improved CUT&RUN chromatin profiling tools. eLife 8, e46314 (2019).
Kim, D., Langmead, B. & Salzberg, S. L. HISAT: a fast spliced aligner with low memory requirements. Nat. Methods 12, 357–360 (2015).
Pertea, M. et al. StringTie enables improved reconstruction of a transcriptome from RNA-seq reads. Nat. Biotechnol. 33, 290–295 (2015).
Jin, Y., Tam, O. H., Paniagua, E. & Hammell, M. TEtranscripts: a package for including transposable elements in differential expression analysis of RNA-seq datasets. Bioinformatics 31, 3593–3599 (2015).
Anders, S. & Huber, W. Differential expression analysis for sequence count data. Genome Biol. 11, R106 (2010).
Yu, G., Wang, L. G., Han, Y. & He, Q. Y. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 16, 284–287 (2012).
Connell, J. & Højsgaard, S. Hidden semi Markov models for multiple observation sequences: the mhsmm package for R. J. Stat. Softw. 39, 1–22 (2011).
Lun, A. T. & Smyth, G. K. csaw: a Bioconductor package for differential binding analysis of ChIP-seq data using sliding windows. Nucleic Acids Res. 44, e45 (2016).
Lawrence, M., Gentleman, R. & Carey, V. rtracklayer: an R package for interfacing with genome browsers. Bioinformatics 25, 1841–1842 (2009).
Lawrence, M. et al. Software for computing and annotating genomic ranges. PLoS Comput. Biol. 9, e1003118 (2013).
Huber, W. et al. Orchestrating high-throughput genomic analysis with Bioconductor. Nat. Methods 12, 115–121 (2015).
Gu, Z., Eils, R. & Schlesner, M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 32, 2847–2849 (2016).
Gu, Z., Eils, R., Schlesner, M. & Ishaque, N. EnrichedHeatmap: an R/Bioconductor package for comprehensive visualization of genomic signal associations. BMC Genomics 19, 234 (2018).
Kent, W. J. et al. The human genome browser at UCSC. Genome Res. 12, 996–1006 (2002).
Wu, J. et al. The landscape of accessible chromatin in mammalian preimplantation embryos. Nature 534, 652–657 (2016).
Andergassen, D. et al. Mapping the mouse Allelome reveals tissue-specific regulation of allelic expression. eLife 6, e25125 (2017).
We thank C. Zhang and W. Zhang for critical reading of the manuscript. This project was supported by the NIH (R01HD092465) and the HHMI. Y.Z. is an Investigator of the Howard Hughes Medical Institute.
The authors declare no competing interests.
Peer review information Nature Genetics thanks Maxim Greenberg and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
a, Genome browser view of H2Aub profiles in oocytes and early embryos. The H2Aub ChIP-seq data of mESC were obtained from24. b, Bar plots showing the reproducibility of the biological replicates of H2Aub CUT&RUN datasets. The H2Aub enrichment was calculated as reads per kilobase per million reads (RPKM) by 5-kb bin. The Pearson correlation co-efficient are also shown. c, Heatmap with hierarchical clustering showing the correlation of global H2Aub enrichment between oocytes and early embryos. d, Bar plot showing the percentage of genomic regions occupied by H2Aub domains (regions with H2Aub signal determined using ChromHMM) at different developmental stages. e, Bar plot showing the percentage of H2Aub CUT&RUN reads that are mapped to each ChromHMM type at different developmental stages.
a, Genome browser views comparing our sperm H3K27me3 STAR ChIP-seq data with publicly available datasets22,55 b, Venn diagram showing that most H3K27me3 domains identified are common between Zheng et al., and our data. c, Genome browser view showing sperm H2Aub STAR ChIP-seq. d, Scatter plot showing the reproducibility of the biological replicates of H2Aub sperm ChIP-seq datasets. The H2Aub enrichment was calculated as reads per kilobase per million reads (RPKM) by 5-kb bin. The Pearson correlation co-efficient is also shown. e, Bar plot showing the number of allelically biased H2Aub regions at different developmental stages. Maternally (Mat) and paternally (Pat) biased H2Aub regions were identified by merging adjacent allelically biased 5kb bins meeting cutoff: fold change (FC) > 2 and adjusted p-value (p-adj) < 0.01. P values were calculated using a binomial exact test of the null hypothesis that both alleles are equally enriched and adjusted using Benjamin & Hochberg method. f, Box plot showing the expression levels of genes within allelically biased H2Aub regions at 1-cell stage. The number of genes within maternally or paternally biased H2Aub domains at 1-cell were 1113 and 15, respectively. The middle lines in the boxes represent medians. Box hinges indicate the twenty-fifth and seventy-fifth percentiles, the whiskers indicate the hinge ± 1.5 × interquartile range, and the dots represent outliers. RNA-seq data were obtained from47. Mat: maternal allele; Pat: paternal allele. g, Heatmaps showing allelic dynamics of the H2Aub regions that are not allelically biased at 1-cell during preimplantation development. The number of regions for each category are indicated. h, Genome browser view of allelic dynamics of the H2Aub regions that are not allelically biased at 1-cell during preimplantation development.
Extended Data Fig. 3 Distinct non-allelic dynamics of H3K27me3 and H2Aub in preimplantation embryos.
a, Genome browser view of H3K27me3, H2Aub and H3K36me348 dynamics in oocytes and preimplantation embryos. b, Venn diagrams showing overlaps of H3K27me3 and H2Aub domains at different developmental stages. The H3K27me3 ChIP-seq data in panels A and B were from17,22. c, Heatmap showing enrichment of H3K27me3, H2Aub, and H3K36me3 at H2Aub domains identified in GV oocytes. The GV H2Aub domains were classified into six groups based on dynamics of these histone modifications during development.
a, Genome browser view of allelic H3K27me3 and H2Aub dynamics at the indicated embryonic stages. Note that maternally biased H2Aub at 1-cell becomes equalized at 2-cell stage by gaining H2Aub on the paternal allele. However, H3K27me3 maintains maternally biased from 1-cell to morula stage. The H3K27me3 ChIP-seq/CUT&RUN data were from22,33. M: maternal; P: paternal. b, Genome browser view of allelic H3K27me3 and H2Aub at Jade1 (also known as Phf17) and Sfmbt2 locus. Note that maternal-biased H2Aub is lost at the imprinted loci in 2-cell embryos then regain during later development. In contrast, maternal biased H3K27me3 is maintained from 1-cell to morula stage. c, Box plots showing allelic bias of H2Aub at gene bodies of the 76 putative maternal H3K27me3-dependent imprinted genes. The left panel represents genes showing transient loss and regain of H2Aub allelic bias in early development, whereas the right panel shows genes that loss of H2Aub allelic bias after 2-cell stage. The middle lines in the boxes represent medians. Box hinges indicate the twenty-fifth and seventy-fifth percentiles, the whiskers indicate the hinge ± 1.5 × interquartile range, and the dots show outliers. Mat: maternal; Pat: paternal.
Extended Data Fig. 5 Loss of maternal Eed has limited effect on global H2Aub except the maternal H3K27me3-dependent imprinted loci.
a, Relative intensity of the H3K27me3 and H2Aub signal in 1-cell, 2-cell, and 4-cell embryos. For 1-cell embryos, only signals on the maternal pronuclei were measured. The average signal intensity of CTR embryos was set as 1.0. The total number of embryos analyzed were 9 (CTR) and 7 (matKO) for 1-cell embryos, 6 (CTR) and 6 (matKO) for 2-cell embryos, and 7 (CTR) and 8 (matKO) for 4-cell embryos, respectively. Center dot and error bars indicate mean and standard deviation, respectively. matKO: maternal knockout. b, Scatter plots showing the reproducibility between biological replicates of H2Aub CUT&RUN datasets. The H2Aub enrichment was calculated as RPKM in 5-kb bins. The Pearson correlation co-efficient are also shown. CKO: conditional knockout. c, Hierarchical clustering of maternal (M) and paternal (P) alleles of preimplantation embryo and GV oocytes based on their global H2Aub enrichment. d, Genome browser view showing comparable enrichment of H2Aub between CTR and CKO in GV oocytes and between CTR and matKO in preimplantation embryos. e, Genome browser view showing H2Aub enrichment at Gab1 and Sfmbt2 loci in CTR and Eed CKO GV oocytes and matKO morula embryos.
a, Representative zygotes immunostaining images. M: maternal pronucleus; P: paternal pronucleus; h: human; CTR: control; OE: over-expression; PR-DUB: Polycomb repressive deubiquitinase; WT: wild type. scale bar: 20 µm. The quantifications are in panel B) Relative H2Aub signal intensity of zygotes. Only signal on the maternal pronuclei (matPN) was measured. The number of embryos analyzed were 6 for non-injected, 11 for hBAP1 (C91A), 20 for CTR, 10 for hBAP1 (WT), and 22 for OE, respectively. c, Relative H2Aub signal intensity of 2-cell and 4-cell embryos. The number of embryos analyzed were 2 for non-injected 2-cell, 3 for CTR 2-cell, 3 for OE 2-cell, 9 for non-injected 4-cell, 10 for CTR 4-cell, and 9 for OE 4-cell, respectively. For Panel B-C), center dot and error bars indicate mean and standard deviation, respectively. d, Scatter plot comparing H2Aub enrichment between non-injected and CTR 4-cell embryos. Pearson correlation co-efficient is shown. e, Stacked bar plot showing the overlap between the top 1000 bins and RING1B-binding sites identified in mESCs24. f, Box plots showing RING1B signals in mESCs and the H2Aub levels in CTR and OE embryos at rank ordered RING1B-binding sites (n = 8833). The bottom 5 (Bot5%, n = 442), top 5 (Top5%, n = 442), and middle 50 percentile (Mid50%, n = 4416) RING1B sites are indicated. The middle lines in the boxes represent medians. Box hinges indicate the twenty-fifth and seventy-fifth percentiles, the whiskers indicate the hinge ± 1.5 × interquartile range. P-value was calculated by Wilcoxon test (two-sided). g, Bar plots showing reproducibility of the biological replicates of H3K27me3 CUT&RUN datasets. The Pearson correlation co-efficient are shown. h, Genome browser view of H2Aub and H3K27me3 enrichment at the Jade1 and Sfmbt2 loci in 4-cell embryos.
a, Heatmap showing the correlation between replicates of PR-DUB RNA-seq samples. b, Left panel shows representative immunostaining images of zygotes. M: maternal pronucleus; P: paternal pronucleus; Pb: polar body. scale bar: 20 µm. Right panel shows quantification of the H3K27me3 signal intensity. Only signal on the maternal pronuclei (matPN) was measured. The number of embryos analyzed were 8 for non-injected, 4 for KDM6b (CTR), and 8 for KDM6b (OE), respectively. Center dot and error bars indicate mean and standard deviation, respectively. C) Heatmap showing the correlation between replicates of Kdm6b and Eed RNA-seq samples. d, Scatter plot comparing gene expression levels of KDM6B (CTR) and (OE) 2-cell and 4-cell embryos. fold change (FC) >2, P-value < 0.05, FPKM > 1. e, Scatter plot comparing gene expression levels of Eed (CTR) and (matKO) 2-cell embryos. fold change (FC) >2, P-value < 0.05, FPKM > 1. f, Left panel, heatmaps of the genes up-regulated by H2Aub removal in 2-cell embryos; Right panel, heatmaps showing expression dynamics of the genes on the left panel at different developmental stages. RNA-seq data in right panel were from17,20,74,75. G) Box plots comparing the expression levels of ZGA and maternal decay genes. The middle lines in the boxes represent medians. Box hinges indicate the twenty-fifth and seventy-fifth percentiles, the whiskers indicate the hinge ± 1.5 × interquartile range. The ZGA (n = 3090) and maternal decay (n = 2343) genes were identified by comparing 2-cell and MII oocytes total RNA-seq data (Supplementary Table 1). Genes up- and down- regulated in 2-cell versus MII oocyte were defined as ZGA and maternal decay genes, respectively. fold change > 5, p-value < 0.05, and RPKM > 1.
About this article
Cite this article
Chen, Z., Djekidel, M.N. & Zhang, Y. Distinct dynamics and functions of H2AK119ub1 and H3K27me3 in mouse preimplantation embryos. Nat Genet 53, 551–563 (2021). https://doi.org/10.1038/s41588-021-00821-2
This article is cited by
Nature Reviews Molecular Cell Biology (2023)
Epigenetics & Chromatin (2022)
Scientific Reports (2022)
Nature Communications (2022)
Nature Reviews Molecular Cell Biology (2022)