Repression of germline genes by PRC1.6 and SETDB1 in the early embryo precedes DNA methylation-mediated silencing

doi:10.21203/rs.3.rs-400263/v1

Download PDF

Article

Repression of germline genes by PRC1.6 and SETDB1 in the early embryo precedes DNA methylation-mediated silencing

https://doi.org/10.21203/rs.3.rs-400263/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 01 Dec, 2021

Read the published version in Nature Communications →

Version 1

posted

You are reading this latest preprint version

Silencing of a subset of germline genes is dependent upon DNA methylation (DNAme) post-implantation. However, these genes are generally hypomethylated in the ICM, implicating alternative silencing pathways before implantation. Indeed, in embryonic stem cells (ESCs), an overlapping set of genes, including germline “genome-defence” (GGD) genes, are upregulated following deletion of the H3K9 methyltransferase SETDB1 or subunits of the non-canonical PRC1 complex PRC1.6. Here, we show that in pre-implantation embryos and naïve ESCs (nESCs), hypomethylated promoters of germline genes bound by the PRC1.6 DNA-binding subunits MGA/MAX/E2F6 are enriched for RING1B-dependent H2AK119ub1 and H3K9me3. Accordingly, silencing of these genes in nESCs shows a greater dependence on PRC1.6 than DNAme. In contrast, GGD genes are hypermethylated in epiblast-like cells and their silencing is dependent upon SETDB1, PRC1.6/RING1B and DNAme, with H3K9me3 and DNAme establishment dependent upon MGA binding. Thus, GGD genes are initially repressed by PRC1.6, with DNAme subsequently engaged in post-implantation embryos.

Epigenetics & Genomics

embryology

genetics

silencing

DNA methylation (DNAme) plays an essential role in transcriptional regulation in development and disease¹. This epigenetic mark is highly dynamic during embryonic development, with a genome-wide wave of demethylation occurring after fertilization, mediated at least in part by dioxygenases ten-eleven translocation (TET) protein family². Subsequently, a wave of global de novo DNAme occurs during post-implantation epiblast formation, mediated by DNMT3A and DNMT3B^3,4. While DNAme is generally maintained at high levels in somatic lineages⁵, another wave of demethylation occurs early in germ cell development in both males and females⁶. During this period, many genes, including those essential for meiosis and gametogenesis, are upregulated. Strict temporal regulation of such germline genes is clearly critical for normal development, as their ectopic expression in somatic cells can contribute to tumorigenesis^7,8. Furthermore, premature activation of germline genes in early germ cells, and in turn precocious induction of meiosis/gametogenesis pathways, leads to reduced germ cell numbers and infertility^9–11. The requirement for faithful regulation of germline genes likely explains the engagement of several distinct chromatin marks in their repression, including DNAme¹², H3K9me3¹³, and H2AK119ub1¹⁴.

In contrast to the vast majority of CpG island (CGI) promoters, a subset of germline genes harbor CGI promoters that are de novo methylated between embryonic day (E) 4.5 and E6.5, yielding a high level of DNAme in the epiblast and early embryo^3,4. While the CGI promoter regions of these genes are demethylated in PGCs coincident with the wave of genome-wide DNA demethylation, they remain hypermethylated in somatic tissues^15–17. Indeed, 137 germline genes that acquire dense promoter DNAme in wildtype (WT) embryos by E6.5 are de-repressed in Dnmt3a/3b-double knock-out (DKO) embryos at E8.5¹². This list of DNAme-sensitive (DMS) germline genes includes many that encode proteins involved in the repression of transposable elements (TEs), such as piRNA biogenesis factors. A number of these germline “genome-defence” (GGD) genes, including Dazl, Ddx4, Mael, Mov10l1, Piwil2, Slc25a31, Taf7l and Tex19.1, were also previously reported to be regulated by promoter DNAme in E9.5 murine embryonic fibroblasts (MEFs)¹⁸. Similarly, Hill et al. identified an overlapping set of “germline reprogramming-responsive” genes that are upregulated in developing PGCs coincident with their demethylation¹⁹. Notably, a subset of these genes harbor E2F consensus (TCCCGC) motifs in their promoter regions and are bound by E2F6 in somatic cells^20,21. While it has been speculated that DNMT3B recruitment through E2F6 transcriptional repressor mediates germline gene silencing in mouse somatic tissues²², DNMT3A and DNMT3B act in a redundant manner to methylate the CGI promoters of such genes in mouse peri-implantation embryos⁴.

While DNAme in the promoter region of a subset of germline genes, including GGD genes, clearly plays a role in their repression post-implantation, many of the same genes are marked by H3K9me3. As these genes show loss of H3K9me3 and are upregulated in primed ESCs (pESCs) deficient in the lysine methyltransferase (KMTase) SETDB1, this histone post-translational modification (PTM) is also clearly involved in their silencing¹³. Indeed, H3K9me3 is recognized and bound by HP1 proteins and triple KO (TKO) of Cbx1, Cbx3 and Cbx5 genes, which encode the three mammalian HP1 isoforms HP1β, HP1γ and HP1α, also leads to de-repression of germline genes in pESCs²³. While SETDB1-mediated silencing of TEs is dependent upon its canonical binding partner TRIM28/KAP1 and DNA-binding proteins of the KRAB-ZFP superfamily^24–28, evidence for a non-canonical SETDB1 recruitment pathway dependent upon the DNA-binding factor MAX was recently reported in pESCs²⁹. MAX binds to E-box sites (CACGTG) when heterodimerized with its partner MGA³⁰, via an interaction between their basic helix-loop-helix leucine-zipper (bHLHZ) domains. Binding of this heterodimer represses the transcription of genes harboring the E-box motif, including many germline genes^31,32. Consistent with the importance of the bHLHZ domain, MGA/MAX-mediated silencing of germline genes is alleviated in male germ cells at least in part by the expression prior to the onset of meiosis of a truncated form of MGA that lacks this domain³³. Notably, the promoters of many germline genes harbor both E-box and E2F consensus motifs^32,34, indicating that the H3K9me3 and DNAme silencing pathways may converge at these genes. However, binding of MGA/MAX is inhibited by DNAme^35–37, suggesting that MGA/MAX-mediated silencing may occur prior to the wave of global de novo DNAme that occurs at implantation.

In addition to their reported roles in promoting H3K9me3 and DNAme, E2F6, MAX and MGA are core subunits of the non-canonical PRC1 repressor complex PRC1.6^31,38−40, which is essential for embryonic viability^{20,31,34,41−46}. PRC1.6 is abundant in pESCs⁴⁷ and recruited to germline genes, dependent upon MGA and/or E2F6^{14,20,31,32,45,48−52}. Deletion of Mga, Max or E2f6 leads to derepression of an overlapping set of germline genes in pESCs or somatic cells^20,49,51, with loss of MGA in particular leading to dramatic upregulation^31,32. Deletion of PRC1.6 subunits L3mbtl2 or Rybp in ESCs also leads to upregulation of an overlapping set of germline genes^45,50. Furthermore, deletion of the genes encoding PRC1.6 catalytic subunits RING1A/B, which ubiquitylate H2A at lysine 119 (H2AK119ub1)¹⁴, or PCGF6, which promotes RING1A/B catalytic activity^31,52, leads to upregulation of many PRC1.6 bound germline genes, including GGD genes. However, as pESCs are a population of mixed cellular differentiation states^53–55, it is difficult to determine the temporal order in which these silencing marks act or the role of interactions between them in mediating gene silencing. Of note, several of these germline genes are robustly de-repressed in adult liver or brain in Pcgf6 KO mice³⁴, indicating that PRC1.6 is required for silencing of such genes even in tissues in which their promoter regions are hypermethylated.

These observations indicate that repression of germline genes during mouse embryonic development is mediated at least in part by three epigenetic marks: DNAme, deposited by DNMT3A/3B; H3K9me3, deposited by SETDB1; and H2AK119ub1, deposited by PRC1.6 subunits RING1A/B. However, a number of questions concerning the specific developmental stages in which these silencing complexes are engaged and the interplay between them remain to be addressed. Furthermore, although the PRC2 complex, including JARID2, EED and the H3K27 KMTase EZH1/2, is recruited to PRC1-targeted genes, the role H3K27me3 plays in the silencing of germline genes has not been studied in detail^56,57.

Here, we characterize the dynamics of DNAme, H3K9me3, H3K27me3 and H2AK119ub1 in the early embryo, as well as in vitro using the naïve ESCs (nESCs; cultured in 2i; inhibitors PD0325901 and CHIR99021) to epiblast-like cell (EpiLC) to PGC-like cell (PGCLC) culture system. Consistent with the observations that MGA/MAX both recruit PRC1.6 and associate with H3K9me3^58,59, we find that H2AK119ub1 and H3K9me3 mark the promoter regions of GGD genes during pre-implantation stages, prior to their de novo DNAme. Furthermore, KO of core/catalytic subunits of each of the previously implicated repressive complexes in nESCs and EpiLCs reveals that PRC1.6 subunits MGA and PCGF6 are required for SETDB1-mediated H3K9me3 at bound GGD genes. Finally, consistent with their dynamics in vivo, we demonstrate that H2AK119ub1 and H3K9me3 are generally more important for silencing of GGD genes than DNAme in nESCs, while DNAme plays a more pronounced role in EpiLCs. Taken together, these observations suggest that PRC1.6 and SETDB1 are essential for silencing of GGD genes during pre-implantation embryonic development, prior to the establishment of DNAme, while DNAme is engaged in silencing only after implantation.

Roles of DNAme, H3K9me3 and H2AK119ub1 in repression of germline genes in pESCs. Three distinct epigenetic marks, including DNAme, H3K9me3 and H2AK119ub1, deposited by de novo DNMT3A/3B, SETDB1 and PRC1.6, respectively, have been previously implicated in silencing of germline genes in pESCs and/or somatic cells in mice (Fig. 1A). To determine whether the 137 DMS germline genes upregulated in E8.5 Dnmt3a/3b DKO embryos¹² (Supplementary Data 1) are hypermethylated as well as bound by SETDB1 and PRC1.6 in pESCs, we reanalyzed a series of published whole-genome bisulphite sequencing (WGBS) and ChIP-seq datasets (Supplementary Data 2). As expected, the CGI promoter regions of most DMS germline genes, including all GGD genes¹⁸, are methylated in E8.5 embryos as well as pESCs (Fig. 1B-C and Supplementary Fig. 1A-B). However, not all of the 137 DMS germline genes are enriched for MGA/MAX and/or E2F6 (Fig. 1B- C, Supplementary Fig. 1C), consistent with the fact that 82 and 97 of these genes include E-box or E2F motifs, respectively, in their promoter regions, with only 61 including both motifs (Supplementary Fig. 1D and Supplementary Data 1). Manual curation of the function of the 40 genes in ESCs enriched for both MGA (RPKM > 2) and E2F6 (RPKM > 1) and showing > 30% DNAme in the promoter region revealed 18 associated with meiosis and 16 associated with genome-defence/piRNA biogenesis (Supplementary Data 3), including all GGD genes. Intriguingly, the DNAme level of these MGA and E2F6 bound genes is generally lower than that of the DMS germline genes that are not bound by these transcription factors (TFs), a trend also observed in E8.5 embryos (Fig. 1B and Supplementary Fig. 1B).

Many of the DMS germline and all GGD genes are also enriched for H3K9me3, SETDB1 and HP1β (Fig. 1C-D and Supplementary Fig. 1E), as well as H2AK119ub1 and PRC1.6 complex subunits PCGF6 and RING1B (Fig. 1C, Fig. 1E and Supplementary Fig. 1F). These regions are also enriched for H3K27me3 and the PRC2 complex subunit EED (Supplementary Fig. 1G), likely reflecting H2AK119ub1-dependent recruitment of PRC2^14,60. TRIM28, on the other hand, showed only low levels of enrichment at DMS germline gene promoters, including those with high H3K9me3 (Fig. 1C and Supplementary Fig. 1E), consistent with a non-canonical mechanism of SETDB1 recruitment to these loci. Indeed, co-enrichment of H2AK119ub1 and H3K9me3 is relatively rare at genic promoters (Supplementary Fig. 1H). Thus, PRC1.6 bound genes, including GGD genes, are marked by an unusual combination of repressive histone PTMs and intermediate levels of DNAme in pESCs.

To determine whether these repressive complexes play a role in silencing of DMS germline (and in particular GGD) genes in pESCs, we next reanalyzed a series of published RNA-seq data (Supplementary Data 2), where key subunits of each of the complexes described above were depleted. Strikingly, DMS GGD genes as well as other methylated germline genes bound by MGA and/or enriched for H3K9me3 or H2AK119ub1 were de-repressed upon KO or knock-down (KD) not only of Mga, Max and Dnmt3a/3b, but also of Setdb1, Cbx1/3/5 and Pcgf6 (Fig. 2A-D and Supplementary Fig. 2A-C). These observations are consistent with previous studies showing that a subset of the DMS germline genes, including GGD genes, are de-repressed in pESCs following KO of L3mbtl2, another core PRC1.6 subunit^40,45,61. To determine whether H2AK119ub1 is required for silencing, we generated a Ring1b conditional (c)KO line. GGD genes in particular were upregulated following Ring1b deletion in pESCs (Fig. 2E), to levels similar to those observed in Pcgf6 cKO pESCs (Fig. 2D and Supplementary Fig. 2C). In contrast, only modest upregulation of GGD genes was observed in Trim28 cKO ESCs (Supplementary Fig. 2B), consistent with the absence of KAP1 binding at these genes (Fig. 1C and Supplementary Fig. 1D). Similarly, no upregulation of DMS germline genes was observed in Eed cKO ESCs (Supplementary Fig. 2C), indicating that H3K27me3, while enriched at PRC1.6 bound promoters, is not required for silencing of these genes. These results suggest that in pESCs, repression of a subset of DMS germline genes is dependent not only upon DNAme, but also on MGA/MAX, SETDB1 and the canonical PRC1.6 complex. However, as pESCs are a mixed population of cells at distinct differentiation states^53–55, it is not possible to deduce either the temporal order in which these silencing complexes act during development, or their inter-dependencies in mediating such silencing.

H2AK119ub1, H3K9me3 and H3K27me3 are sequentially deposited prior to de novo DNAme at DMS germline genes To investigate the dynamics of these histone PTMs as well as DNAme at DMS germline genes during early embryonic/PGC development, we analyzed previously published genome-wide datasets from 2-cell, morula, inner cell mass (ICM), epiblast, and E9.5 PGCs and carried out H3K9me3 ULI-ChIP-seq⁶² for E9.5 PGCs (Fig. 3A). In the promoter regions of many DMS germline gene loci, H2AK119ub1, H3K9me3 and H3K27me3 are already present in the ICM or earlier stages (Fig. 3B, Supplementary Fig. 3A), with H2AK119ub1 deposited at DMS germline and GGD genes as early as the 2-cell stage, followed by H3K9me3 in morula and ICM and H3K27me3 in ICM and E6.5 epiblast^63,64 (Supplementary Fig. 3A-B). In contrast, DNAme is not deposited at these loci until the epiblast stage, as expected (Fig. 3B). As observed in pESCs, MGA and E2F6 bound genes show a lower level of DNAme in E6.5 epiblast than most other DMS germline genes (Supplementary Fig. 3C). While H2AK119ub1, H3K9me3 and H3K27me3 persist at DMS germline genes following implantation, H3K9me3 and H3K27me3 progressively decrease at these genes in PGCs, coincident with their DNA demethylation and upregulation in E13.5 PGCs (Fig. 3B-C). Thus, H2AK119ub1 and H3K9me3 are deposited prior to de novo DNAme at many DMS germline gene promoters, including all GGD genes, implicating DNA methylation-independent repression of these loci during pre-implantation development.

To further characterize the roles of these chromatin marks, we employed the PGCLC culture system⁶⁵, in which nESCs, EpiLCs, day (d)4–6 PGCLCs and d4 plus 7 days of culture (d4c7) PGCLCs represent the in vitro counterparts of E3.5 ICM, E6.5 epiblast, E9.5-11.5 PGCs and E13.5 PGCs, respectively (Fig. 3A). As expected, while DNAme levels in nESCs are relatively low in the promoter regions of DMS germline genes, H2AK119ub1, H3K9me3 and H3K27me3 are clearly enriched at a subset of these genes, including GGD genes (Fig. 3B and Supplementary Fig. 3A-B). Consistent with PRC1.6 binding at their promoter regions, the enrichment of H2AK119ub1 in nESCs and EpiLCs is positively correlated with the enrichment levels of MGA and E2F6, as measured in pESCs and nESCs, respectively (Supplementary Fig. 3C).

Following EpiLC induction, these loci are de novo methylated and retain each of these histone PTMs, with enrichment modestly increased relative to nESCs. Mirroring the E6.5 epiblast, DNAme levels in EpiLCs are still lower at MGA and E2F6 bound genes (including GGD genes) relative to most other genes, including DMS germline genes that are not bound by these TFs (Supplementary Fig. 3D). These observations indicate that PRC1.6 bound promoters are refractory to the wave of de novo DNAme that occurs during epiblast formation. H2AK119ub1, H3K9me3 and H3K27me3 are subsequently reduced in d4-6 PGCLCs, while DNAme persists (Fig. 3B and Supplementary Fig. 3A-B). Despite the increase in repressive marks in EpiLCs, expression of the majority of the GGD genes transiently increases in EpiLC relative to nESCs, perhaps due to the simultaneous upregulation of positive regulatory factors that bind to their CGI promoters (Fig. 3C). Regardless, these genes are then downregulated in d4 PGCLCs, before their upregulation again in d4c7 PGCLCs, as expected. Based on these observations, we subsequently employed the nESC/EpiLC/PGCLC system to model the relative importance of and crosstalk between each repressive pathway during pre- and post-implantation development.

DNAme plays a more important role in silencing of DMS germline genes in EpiLCs than nESCs. Analysis of previously generated WGBS data from DNMT deficient E6.5 epiblast cells⁶⁶ revealed a dramatic reduction in DNAme at DMS germline genes in both Dnmt1 KO and Dnmt3a/3b DKO embryos (Fig. 4A-B), confirming the importance of de novo DNAme at this stage. As observed in the ICM to epiblast transition, EpiLCs also show much higher levels of DNAme in the promoter regions of DMS germline genes than the nESCs from which they are derived^67,68. However, nESCs show a modestly higher level of DNAme in these regions than E3.5 ICM (Fig. 4A-B). Notably, pESCs express DNMT3A as well as TET1, which mediates oxidation of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC) in an active DNA demethylation pathway⁶⁹. As bisulphite sequencing cannot discriminate between 5mC and 5hmC, we surmised that the low level of DNAme observed in DMS germline genes in nESCs may reflect the presence of both 5mC and 5hmC. Analysis of previously published 5hmCIP-seq data⁷⁰ revealed that relative to all other genes in nESCs, the promoters of DMS germline genes, including GGD genes, are enriched for 5hmC (Supplementary Fig. 4A-B). Surprisingly, the levels of 5hmC enrichment are similar between nESCs and EpiLCs, and analysis of previously published ChIPseq data also reveals similar levels of TET1 enrichment between these cell types^70,71. Taken together with the fact that WGBS indicates significantly higher levels of DNAme in EpiLCs, these data suggest that the ratio of 5hmC to 5mC is higher in nESCs. Closer inspection of these loci reveals that 5hmC and TET1 peaks overlap with each other and with MGA/MAX and E2F6 in the promoter region, corresponding to regions of reduced DNAme relative to flanking CpGs (Supplementary Fig. 4C). Thus, the low levels of DNAme observed in the promoter regions of GGD genes in nESCs represents a combination of 5mC and 5hmC, likely reflecting reiterative de novo DNAme followed by oxidation of this mark under these culture conditions.

To directly address the role of DNAme in silencing of DMS germline genes in nESCs, we employed a line harboring floxed alleles of Dnmt1, Dnmt3a and Dnmt3b⁷². Following the addition of 4-hydroxytamoxifen (4OHT) for four days, RNA-seq was conducted on control or Dnmt1/3a/3b conditional triple (cT)KO nESCs, or EpiLCs (isolated after two days of differentiation) (Fig. 4C). As expected, given the low levels of DNAme in nESCs, only 14/137 DMS germline genes showed > 30% promoter DNAme in control nESCs and were upregulated in Dnmt cTKO nESCs (Fig. 4D). The observed upregulation of the GGD genes Dazl and Piwil2, as well as the control gene Hoxb1 (Fig. 4D-E), is likely an indirect effect, as each of these genes show very low levels of DNAme in nESCs. In contrast, 74/137 DMS germline genes showed > 30% promoter DNAme in control EpiLCs and were upregulated in Dnmt cTKO EpiLCs (Fig. 4D). These include GGD genes Slc25a31, Taf7l and Tex19.1, which were only modestly upregulated in DNAme deficient nESCs (Fig. 4E). Intriguingly, the remaining GGD genes Ddx4, Mael and Mov10l1, which ranged in promoter DNAme levels from 5.5–21.3% and 19.5–67.3% in WT nESCs and EpiLCs, respectively, remained repressed following Dnmt cTKO in both cultures. Thus, consistent with the wave of global de novo DNAme that takes place with EpiLC induction, DNAme play a more important role in silencing of DMS germline genes in EpiLCs than nESCs, including several GGD genes. However, a subset of GGD genes is silenced independent of DNAme even in EpiLCs.

PRC1.6 plays a more important role than DNAme in silencing of GGD genes in nESCs. To investigate the potential roles of H3K9me3 and H2AK119ub1 in the regulation of DMS germline genes, we carried out ChIP-seq on a line that expresses a truncation mutant of MGA from the native locus that lacks the bHLHZ domain⁷³, as well as Setdb1- and Pcgf6 cKO nESCs and EpiLCs. As conditional KO of Setdb1 in nESCs leads to a dramatic reduction in viability by day three in culture⁷⁴, cKO lines were harvested following two days of 4OHT treatment (Supplementary Fig. 5A). Strikingly, H3K9me3 levels were reduced at DMS germline and GGD genes in both nESCs and EpiLCs of all three mutant lines (Fig. 5A), with the effect more pronounced for Mga-ΔHLH and Pcgf6 cKO in EpiLCs. Thus, both the bHLHZ DNA-binding domain of MGA and the PCGF6 core subunit of PRC1.6 are required for SETDB1 recruitment and/or activity at these loci. While H2AK119ub1 at DMS germline and GGD genes was also significantly reduced in Pcgf6 cKO nESCs and EpiLCs, as expected, this mark was unaffected in the Mga-ΔHLH and Setdb1 cKO lines (Fig. 5B). This suggests that neither MGA bHLHZ-dependent E-box-binding nor SETDB1 are required for persistence of H2AK119ub1 at these loci.

RNA-seq analyses of these KO nESCs and EpiLCs revealed varying levels of de-repression of DMS germline genes (Fig. 5C and Supplementary Data 4). Mga-ΔHLH showed the most consistent upregulation of GGD genes in nESCs and EpiLCs (Fig. 5D). Dnmt cTKO in nESCs resulted in only modest upregulation of most GGD genes when compared to Mga-ΔHLH and Pcgf6 cKO nESCs. For several of these genes, DNAme is apparently dispensable for silencing in EpiLCs as well, whereas PRC1.6 is required for transcriptional repression in both nESCs and EpiLCs. However, for 3 of the GGD genes, namely Slc25a31, Taf7l and Tex19.1, a more robust upregulation is clearly evident in Dnmt cTKO EpiLCs than nESCs. Consistent with this trend, DESeq2-based TCC (Tag Count Comparison) analysis⁷⁵ of Dnmt cTKO EpiLCs yielded 70 upregulated DMS germline genes, versus only 28 in Dnmt cTKO nESCs (Supplementary Data 4). The increased level of DNAme in the CGI promoter regions of these genes (Fig. 4E) is likely responsible for the greater impact of Dnmt cTKO in EpiLCs. Furthermore, the engagement of DNAme-mediated repression may explain the reduced impact of Pcgf6 deletion on expression of all of the GGD genes in EpiLCs relative to nESCs (Fig. 5D).

As Pcgf6 deletion disrupted both H2AK119ub1 and H3K9me3 deposition, we next wished to determine the relative importance of these histone PTMs in transcriptional repression. We carried out RNA-seq on Setdb1 cKO, Ring1b cKO as well as Setdb1/Ring1b (Set/Ring) cDKO nESCs and EpiLCs, as above (Supplementary Fig. 5A). DESeq2-based TCC analysis yielded a similar number of upregulated DMS germline genes in Ring1b cKO versus Pcgf6 cKO nESCs or EpiLCs (Supplementary Fig. 5B and Supplementary Data 4). However, DMS germline and GGD genes showed a greater level of de-repression in Setdb1/Ring1b cDKO than individual Ring1b or Setdb1 cKO nESCs and EpiLCs (Fig. 5C-D and Supplementary Fig. 5C-D), with the fold-increase in expression for many GGD genes in Setdb1/Ring1b cDKO nESCs similar to that observed in Pcgf6 cKO nESCs (Fig. 5D and Supplementary Fig. 5D). These results suggest that RING1B and SETDB1 act in a combinatorial manner to repress PRC1.6 bound genes and are consistent with the observation that loss of PCGF6 disrupts both H2AK119ub1 and H3K9me3 deposition. However, unlike the Pcgf6 cKO, Setdb1/Ring1b cDKO EpiLCs show a level of de-repression that does not decrease relative to nESCs, indicating that SETDB1 is still essential for robust silencing of these genes even in the presence of a higher level of DNAme in their promoter regions.

Intriguingly, the fold-change in expression of most GGD genes in Mga-ΔHLH EpiLCs far exceeds that observed in EpiLCs generated from the other KO lines (Fig. 5D and Supplementary Fig. 5D), indicating that MGA/MAX may act at the nexus of PRC1.6, SETDB1 and de novo DNAme. To determine whether MGA is indeed required for de novo DNAme, we isolated genomic DNA from Mga-ΔHLH EpiLCs and conducted bisulphite sequencing on the CGI promoter regions of Mael and Mov10l1. Relative to control EpiLCs, DNAme levels decreased significantly at both genes, as observed in Dnmt cTKO EpiLCs (Fig. 5E, Supplementary Fig. 6A-B). Since these genes are not de-repressed in Dnmt cTKO EpiLCs (Fig. 4E, Fig. 5D), these observations indicate that even when present, DNAme is not necessarily required for silencing of GGD genes in EpiLCs. Only a moderate decrease or no change in DNAme was observed in these regions in Pcgf6-, Setdb1- or Ring1b cKO EpiLCs (Fig. 5D and Supplementary Fig. 6A-B), indicating that H3K9me3 and H2AK119ub1 may play redundant roles in potentiating de novo DNAme of these loci. To directly address whether PRC1.6 is indeed required for de novo DNAme of DMS germline gene promoter regions in vivo, we reanalyzed published WGBS data from Ring1b or L3mbtl2 KO E6.5 epiblast⁶⁶. A subset of these DMS germline genes showed a modest decrease in DNAme in both KOs relative to controls, though not to the extent observed in Dnmt3a/3b DKO E6.5 epiblast (Fig. 5F). This includes the promoter regions of all GGD genes, with the exception of Dazl, which retains high levels of DNAme in both mutants (Fig. 5G). Taken together, these observations suggest that MGA/MAX binding potentiates de novo DNAme at DMS germline genes during epiblast formation, likely via establishment of a repressed chromatin state.

By integrating numerous previously published datasets from pre-implantation stages, we show that H2AK119ub1, H3K9me3 and H3K27me3 sequentially accumulate in the promoter regions of DMS germline genes bound by MGA/MAX and E2F6 in ESCs, followed by de novo DNAme in the epiblast (Fig. 6). The deposition of these histone PTMs prior to the establishment of DNAme is consistent with roles for PRC1.6 and SETDB1 prior to and independent of DNAme-mediated silencing. To study the relative importance of these repressive marks, we utilized the nESC-EpiLC-PGCLC in vitro differentiation system. As expected given their hypomethylated state, most DMS germline genes were not de-repressed in Dnmt cTKO nESCs. On the other hand, many of these genes were upregulated in nESCs expressing an MGA mutant incapable of binding to E-box sites, or following deletion of Pcgf6 or Ring1b, confirming the critical role that MGA/PRC1.6 plays in silencing of bound germline genes even in the presence of low levels of DNAme (Fig. 6). Of note however, E-box-mediated binding of MGA/MAX at a subset of germline genes is dispensable for the persistence of H2AK119ub1, implying that E2F6 binding may be sufficient for recruitment of PRC1.6 and in turn H2AK119ub1 deposition at these loci⁴⁸.

Consistent with the global increase in DNAme in epiblast, most of these germline genes showed a dramatic increase in DNAme in WT EpiLCs, and a subset were de-repressed in Dnmt cTKO EpiLCs. This includes several GGD genes that are also upregulated in both nESCs and EpiLC following deletion of PRC1.6 subunits, indicating that both DNAme and H2AK119ub1 may be simultaneously engaged in silencing of germline genes in some developmental contexts, such as in epiblast cells. As germline genes are also upregulated in adult liver and brain in Pcgf6 KO mice³⁴, DNAme-dependent silencing clearly does not always supersede PRC1.6-mediated silencing following gastrulation. Additional studies will be required to determine the relative importance of these epigenetic marks in silencing of germline genes in somatic tissues.

We show that de novo DNAme of the CGI promoters of several germline genes in EpiLCs is dependent on the MGA bHLHZ domain (Fig. 6). This observation is consistent with our recent report showing that a few dozen germline genes, including Ddx4 and Dazl, show reduced DNAme following Max KD in pESCs²⁹. Thus, the establishment of DNAme at the peri-implantation stage may depend upon MGA/MAX binding to the E-box motif and subsequent recruitment of PRC1.6 and SETDB1. Indeed, the promoter regions of most GGD genes also show reduced DNAme in Ring1b or L3mbtl2 KO E6.5 epiblast⁶⁶, albeit to a lesser extent than observed in Mga-ΔHLH EpiLCs. However, as DNAme of the E-box motif inhibits MGA/MAX binding^35–37, active DNA demethylation may be essential for MGA/MAX recruitment in tissues expressing relatively high levels of the de novo DNMTs.

As the promoter regions of many germline genes that lack E-box or E2F motifs are efficiently de novo methylated, the presence of H2AK119ub1 and H3K9me3 is clearly not a prerequisite for the establishment of DNAme. Indeed, we find that the subset of DMS germline genes bound by MGA/MAX and E2F6 actually show delayed DNAme both in vitro and in vivo relative to the DMS germline genes not enriched for PRC1.6. While H2AK119ub1 and/or H3K9me3 may play a direct role in recruitment of the de novo DNMTs, inhibition of H3K4me3 may also play a critical role in potentiating their de novo DNAme. Alternative repressive pathways that inhibit H3K4 methylation likely potentiate de novo DNAme of the DMS germline genes not targeted by PRC1.6. As H3K36 di- and tri-methylation also promotes de novo DNAme^76–78, this histone PTM may also play a role at germline genes, but this remains to be determined.

Recent reports reveal that H2AK119ub1 is deposited in 2-cell stage embryos as early as E1.5, with enrichment of H3K27me3 only clearly evident at most Polycomb targets, in E3.5 ICM^63,64. The germline genes studied here show similar kinetics of accumulation of these marks, indicating that PRC1.6, like other non-canonical PRC1 complexes, is likely bound to its target regions very early in embryonic development. Deposition of these and other repressive marks at this stage may compensate for the permissive chromatin state associated with the global DNA demethylation that follows fertilization⁶. Embryonic acquisition of H2AK119ub1 is also observed in zebrafish embryos just before ZGA⁷⁹, indicating that early deposition of this mark may be conserved across species. While PRC1.6 and H2AK119ub1 play a critical role in silencing of MGA/MAX bound germline genes in pESCs, deletion of Eed had no effect. As this PRC2 subunit is essential for H3K27me3 deposition, H2AK119ub1 is clearly the more important Polycomb mark for silencing of these loci. This observation is consistent with a recent study showing that deletion of Eed in pESCs abolishes H3K27me3 deposition but fails to recapitulate the transcriptional defects observed in the absence of RING1A/B¹⁴.

Although the roles of MGA/MAX and/or E2F6 in the recruitment of PRC1.6 or SETDB1 during early embryogenesis has not been established, mice lacking the gene encoding PRC1.6 subunits or Setdb1 exhibit early embryonic lethality^{31,41,43,45,46,80}, which may be due to the precocious expression of specific genes, including DMS germline genes. Of note, Polycomb-dependent repression of germline genes is also observed in species that do not utilize CpG methylation. For example, mutations in l(3)mbt, a homolog of L3mbtl2, result in ectopic expression of germline genes in larval brain in D. melanogaster⁸¹. Similarly, LIN-61/L3MBTL2 plays a role in silencing of germline genes in somatic cells in C. elegans⁸². Notably, MET-2, the SETDB1 homolog in C. elegans, is also required for repression of germline genes in somatic tissues, via deposition of H3K9me2 (rather than H3K9me3) at their promoters⁸³. Thus, PcG proteins and H3K9 methylation may play an ancestral role in silencing of germline genes in metazoans, with DNAme engaged in the process in mammals only after implantation.

Here, we show that PCGF6 is required for H3K9me3 deposition at DMS germline genes in nESCs and EpiLCs. Consistent with this observation, mass spectrometry analyses revealed that PCGF6 interacts with SETDB1 in pESCs⁸⁴ and with H3K9me3 in several mouse tissues⁵⁸. However, as deletion of Pcgf6 in ESCs also compromises MAX as well as RING1B chromatin loading⁴⁰, PCGF6 may play an indirect role in SETDB1 recruitment, via stabilization of the PRC1.6 complex at its target sites. Destabilization of PRC1.6 following deletion of PCGF6 likely explains why the fold-change in expression of MGA/MAX bound germline genes in Pcgf6 KO EpiLCs is similar to that observed following Setdb1/Ring1b DKO, as deletion of Ring1b does not have the same destabilizing effect⁴⁰. Consistent with a direct role for MGA/MAX-mediated SETDB1 recruitment, MAX was recently shown to interact with SETDB1 via co-immunoprecipitation, and Ddx4 and Dazl promoters showed reduced H3K9me3 in Max KD pESCs²⁹. However, a recent study suggests that SETDB1 may indirectly interact with PRC1.6, via its known binding partner ATF7IP⁸⁵. ATF7IP contains a C-terminal fibronectin type-III (FNIII) domain, which was shown to interact with MGA in pESCs via mass spectrometry. This mechanism of SETDB1 recruitment would explain why TRIM28 is dispensable for SETDB1-mediated silencing of DMS germline genes, while ATF7IP is required for their silencing⁸⁵. Regardless, our analyses reveal that MGA/MAX bound GGD genes are generally upregulated in Setdb1 KO nESC and EpiLCs to a lesser extent than observed in the PRC1.6 subunit mutants, indicating that H3K9me3 plays an important but ancillary role in their silencing.

In summary, we show that H2AK119ub1 and H3K9me3 mark the hypomethylated CGI promoter regions of PRC1.6-bound DMS germline genes during pre-implantation stages, with de novo DNAme occurring at these regions only after implantation. Consistent with their kinetics in vivo, PRC1.6 and to a lesser extent SETDB1 play important roles in silencing of these genes in nESCs and EpiLCs, suggesting that H2AK119ub1 and H3K9me3 regulate DMS germline genes during pre-implantation embryonic development, prior to the establishment of DNAme. The atypical combination of these histone PTMs, which generally mark distinct genomic regions, and DNAme, likely serves to prevent precocious expression of germline genes during early embryonic development, in the early germline and in somatic tissues, where expression of these genes would otherwise be deleterious.

Mice. The Oct4::EGFP transgenic mice⁸⁶ were maintained in a C57BL/6 genetic background. Mouse experiments were approved by the Animal Care Committee at the University of British Columbia (UBC) under certificate numbers A16-0230, A16-0269, A20-0229 and A20-0230, with the guidelines from the national Canadian Council on Animal Care (CCAC).

Isolation of PGCs. Embryos were obtained from the mating of wildtype C57BL/6 mice with C57BL/6 mice carrying an Oct4::EGFP transgene at appropriate embryonic days (noon of the day when a copulation plug was identified was designated as embryonic day 0.5 [E0.5]). E9.5 embryos were collected and dissected in DMEM supplemented with 10% fetal bovine serum (FBS). The hindgut endoderm regions that contained primordial germ cells (PGCs) were dissected from the embryos. Tissue fragments containing PGCs were incubated at 37°C for 5–8 minutes in 0.25% trypsin/0.5 mM EDTA/PBS with serial pipetting to obtain single-cell suspensions. After quenching the reaction with fetal bovine serum (FBS), cells were washed once with FACS buffer (0.1% BSA/PBS), counterstained with Propidium Iodide (PI) and passed through a 35 µm cell strainer (Falcon). FACS was carried out on a BD Influx. Following removal of doublet/triplet cells, OCT4::EGFP+ (PI-) cells were sorted into a 1.5 ml low-retention tube containing 500 µl of FACS buffer. Following centrifugation, the supernatant was removed, and the cells were snap-frozen in liquid nitrogen and kept at -80°C until use.

Mouse ESC lines. ESC lines were generated from respective blastocyst embryos as described previously⁸⁷. A published Ring1b conditional allele⁸⁸ was used to establish Ring1b cKO ESCs. This Ring1b conditional allele and a published Setdb1 conditional allele²⁵ were used to establish Setdb1/Ring1b cDKO ESCs. EBRTcH3 ESCs (WT) ⁸⁹ and their derivative Mga-ΔHLH ESCs⁷³, Setdb1 cKO ESCs²⁵, Pcgf6 cKO ESCs³¹ and Dnmt1/3a/3b cTKO ESCs⁷² were described previously. Male ESCs were used in this study. All cKO ESC lines express the Cre-oestrogen receptor fusion protein Cre-ERT2 or MER-Cre-MER, and the addition of 4-OHT eliminates each targeted gene.

Cell culture. EpiLC-PGCLC induction was performed essentially as described previously⁶⁵ with minor modifications. Briefly, pESCs were adapted to 2i (PD0325901, CHIR99021) + LIF feeder-free culture conditions to transit to nESCs. The EpiLCs were induced by plating nESCs in a well coated with human plasma fibronectin (Millipore) in N2B27 medium containing Activin A (R&D), bFGF (Gibco) and KnockOut Serum Replacement (KSR; Gibco). For cKO cells, 4-hydroxytamoxifen (4OHT; Sigma, dissolved in ethanol) at a final concentration of 100 nM was added to culture medium to induce Cre recombinase activity. RT–qPCR was performed to demonstrate depletion of mRNAs of the targeted genes in each cKO cell line upon treatment with 4-OHT as described below. The PGCLCs were induced under floating aggregation culture conditions by plating EpiLCs in a well of a low-cell-binding U-bottom 96-well plate (Nunc) in GMEM-based serum-free medium in the presence of the cytokines BMP4 (R&D), LIF (ESGRO; Chemicon), SCF (R&D) and EGF (R&D). Cell aggregates including d4 PGCLCs were dissociated with TrypLE (10 minutes, 37°C) and incubated with anti-SSEA1 antibody (BioLegend) and anti-CD61 antibody (BioLegend) conjugated with Phycoerythrin (PE) and Alexa Fluor 647, respectively. Cells were washed with FACS buffer and passed through a 35 µm cell strainer. FACS was carried out on a BD Influx. Following removal of doublet/triplet cells, double positive PGCLCs were sorted into a 1.5 ml low-retention tube containing 500 µl of FACS buffer. Following centrifugation, the supernatant was removed, and the cells were snap-frozen in liquid nitrogen and kept at -80°C until use. Ring1b cKO pESCs were also cultured in DMEM (Kohjin-bio) with 20% fetal bovine serum (Sigma), MEM nonessential amino acids, L-glutamine, 2-mercaptoethanol, and LIF on mitomycin C-treated (Sigma) primary MEF feeder layers. For cKO, 4OHT was added to the medium to a final concentration of 800 nM.

RT-qPCR analysis. After extraction and purification, total RNA was reverse transcribed by Superscript IV (Thermo Fisher Scientific), and cDNA was used for quantitative PCR with the PowerUP SYBR Green PCR Master Mix (Applied Biosystems). PCR signal was detected by QuantStudio 3 (Applied Biosystems). The sequences of the PCR primers used in this study are listed in Supplementary Data 5.

Strand-specific total RNA-sequencing and transcriptome analysis. Total RNA was extracted from ten thousand nESCs or EpiLCs using the RNeasy Micro kit (Qiagen) with on-column DNaseⅠ treatment. rRNA was depleted using the NEBNext rRNA Depletion Kit (NEB) and cDNA was synthesized using the NEBNext UltraII RNA First Strand Synthesis Module (NEB) and NEBNext UltraII Directional RNA Second Strand Synthesis Module (NEB). Following purification of double-stranded DNA by AMPure XP beads, libraries were constructed as described previously⁶², with minor modifications. Briefly, end repair and A-tailing were performed using NEBNext UltraII End Repair/dA-Tailing Module (NEB) and ligation was carried out using NEBNext Ultra II Ligation Module (NEB) with an in-house adaptor which is compatible with Illumina sequencers. The sequences of the adaptor oligo are listed in Supplementary Data 5. After purifying adaptor-ligated products by AMPure XP beads (Beckman Coulter), libraries were amplified with unique indexing primers for 12 cycles. Following purification of the amplified libraries by AMPure XP beads, libraries were pooled and run on a 2% E-Gel EX Gel (ThermoFisher Scientific). Fragments from 200–600 bp were excised and purified by Zymoclean Gel DNA Recovery Kit (Zymo Research). The concentration of pooled libraries was determined by qPCR using NEBNext Library Quant Kit for Illumina (NEB), adjusted to 4 nM and then sequenced on the NextSeq 500 platform with 76 paired-end high-output mode according to the manufacturer’s instructions. Sequenced reads were then aligned to the mouse genome (mm10) and transcripts aligned using STAR⁹⁰ with --sjdbGTFfile and --twopassMode Basic options and then read counts and RPKM (reads per kilobase mapped reads) values for each of the genes were calculated using the RSEM program with default parameter⁹¹. For analyses of differential expression genes (DEGs) between WT and KO, we used TCC (Tag Count Comparison)-GUI, a robust DEG normalization program (https://github.com/swsoyee/TCC-GUI)⁷⁵ with default parameters (Number of Iteration = 3, FDR Cut-off = 0.1, Elimination of Potential DEGs = 0.05) using DESeq2 estimation⁹². Gene ontology analyses were performed with ShinyGO, a graphical gene-set enrichment tool (http://bioinformatics.sdstate.edu/go/)⁹³. For Ring1b cKO vs WT pESCs, total RNA was prepared from cultured cells using the Allprep DNA/RNA mini kit (Qiagen) following vendor’s instructions. Each RNA-seq library was prepared from 1 µg of total RNA, using the NEBNext Ultra Directional RNA Library Prep Kit for Illumina (NEB), following the supplier’s instructions. Libraries were quantified with the KAPA Library Quantification Kit (Illumina). Samples were multiplexed and adjusted to 2 nM concentration, followed by Illumina HiSeq2000 sequencing. Reads were generated by 50 bp single end sequencing.

Ultra-low-input native chromatin immunoprecipitation (ULI-N-ChIP)-sequencing. The ULI-N-ChIP-seq protocol⁶² was used to generate ChIP-seq libraries, with minor modifications. Briefly, chromatin was fragmented with 3.33 units of Micrococcal Nuclease (NEB), diluted in native ChIP buffer (20 mM Tris-HCl pH 8.0; 2 mM EDTA; 150 mM NaCl; 0.1% Triton X-100) containing 1 mM PMSF and EDTA-free protease inhibitor cocktail (Roche) and sonicated three times with 5-second cycles with low power (Diagenode Bioruptor). Chromatin fragments were then split into two aliquots (~ 1000 cells/ChIP) and incubated with each of the histone methylation antibodies at 4°C overnight. Antibodies used for ChIP experiments were H3K9me3 (39161, Active Motif) and H2AK119ub1 (8240, Cell Signaling). Following purification of ChIPed DNA by AMPure XP beads, libraries were constructed, pooled and sequenced as described above. ChIP-seq reads were aligned to the mouse genome (mm10) using bwa⁹⁴. PCR duplicates were removed by Picard-tools (http://broadinstitute.github.io/picard) and reads with low mapping quality (MAPQ < 5) were excluded by Samtools (http://www.htslib.org/). Processed reads were then used for calculating RPKM values by VisR⁹⁵.

Bisulphite sequencing. Using the EZ DNA Methylation-Direct Kit (Zymo Research), bisulphite reactions were performed directly on ten thousand EpiLCs without DNA extraction. The sequences of the PCR primers designed with MethPrimer (http://www.urogene.org/methprimer/index1.html)⁹⁶. The sequences of the PCR primers are listed in Supplementary Data 5. The target regions were amplified by PCR with the EpiTaq HS DNA Polymerase (Takara) using the following conditions: after the first denaturation (3 minutes at 95°C), 40 cycles of 30 seconds at 94°C, 30 seconds at 60°C, 60 seconds at 72°C followed by 10 minutes at 72°C. The PCR products were cloned by TA cloning in the pGEM T-Easy vector (Promega) and 8–16 clones were sequenced. Sequence data were analyzed with the QUantification tool for Methylation Analysis (QUMA; http://quma.cdb.riken.jp/top/index.html)⁹⁷.

Analysis of publicly available data. All publicly available data listed in Supplementary Data 2 were reprocessed for consistency of data analyses. The sra format file downloaded from the Sequence Read Archive (SRA) was converted into fastq file format using sratoolkit (https://github.com/ncbi/sra-tools). Read quality was then assessed by FastQC and the reads were processed according to the experimental application, as described above.

Visualization. bedGraphToBigWig (UCSC) was used to convert bedgraph data to bigwig format for visualization. Graphs were generated by VisR⁹⁵ or Microsoft Excel. Genome browser screenshots were generated using IGV⁹⁸ and Venn diagrams were generated using DeepVenn (https://www.deepvenn.com).

Statistics. Two-tailed t-tests of two samples assuming unequal variances were performed when comparing levels of H3K9me3, H2AK119ub1 or DNAme at DMS germline genes, including GGD genes or their expression levels between KO and WT. Mann-Whitney U-tests were performed between KO and WT when comparing DNAme levels at the CGI promoters of GGD genes. Violin plots show the median (line inside the Violin), where 50% of the data are distributed.

Data availability

The statistics of datasets generated in this study are summarized in Supplementary Data 6. ChIP-seq and RNA-seq data have been deposited in the Gene Expression Omnibus database under accession number GSE######.

Acknowledgements

We thank J. Brind’Amour, S. Janssen, J. Richard Albert and K. Jensen for helpful discussions and technical assistance; L. Lefebvre (UBC) for critical reading of the manuscript; and T. Stach, R. Vander Werff, Y. Chung and Y. Zhao (BRC-seq, UBC) for deep sequencing and J. Wong and A. Johnson (ubcFLOW, UBC) for cell sorting. M.C.L. was supported by CIHR grants PJT-153049 and PJT-166170. K.M. was a recipient of an Uehara Memorial Foundation postdoctoral fellowship, a Nakatani Foundation technology exchange grant, a MEXT Fund for the Promotion of Joint International Research 17KK0185, a Mother & Child Health Foundation grant R01-1 and a Sumitomo Foundation grant 200304.

Author contributions

K.M. and M.C.L. conceived the experiments and wrote the paper with the help of K.S. and A.B. J.S. and H.K. generated the Dnmt cTKO, Pcgf6-, Ring1b cKO and Setdb1/Ring1b cDKO ESC lines and performed RNA-seq in Ring1b cKO pESCs. K.U., A.S. and A.O. generated the Mga-ΔHLH ESC line. K.M. performed all other experiments and analyses.

Smith, Z. D. & Meissner, A. DNA methylation: roles in mammalian development. Nat. Rev. Genet. 14, 204–220 (2013).
Greenberg, M. V. C. & his, D. B. X. The diverse roles of DNA methylation in mammalian development and disease. Nat. Rev. Mol. Cell Biol. 20, 590–607 (2019).
Borgel, J. et al. Targets and dynamics of promoter DNA methylation during early mouse development. Nat. Genet. 42, 1093–1100 (2010).
Auclair, G., Guibert, S., Bender, A. & Weber, M. Ontogeny of CpG island methylation and specificity of DNMT3 methyltransferases during embryonic development in the mouse. Genome Biol. 15, 545–16 (2014).
Hon, G. C. et al. Epigenetic memory at embryonic enhancers identified in DNA methylation maps from adult mouse tissues. Nat. Genet. 45, 1198–1206 (2013).
Lee, H. J., Hore, T. A. & Reik, W. Reprogramming the methylome: erasing memory and creating diversity. Cell Stem Cell 14, 710–719 (2014).
Simpson, A. J. G., Caballero, O. L., Jungbluth, A., Chen, Y.-T. & Old, L. J. Cancer/testis antigens, gametogenesis and cancer. Nat Rev Cancer 5, 615–625 (2005).
Gantchev, J. et al. The ectopic expression of meiCT genes promotes meiomitosis and may facilitate carcinogenesis. Cell Cycle 19, 837–854 (2020).
Matson, C. K. et al. The mammalian doublesex homolog DMRT1 is a transcriptional gatekeeper that controls the mitosis versus meiosis decision in male germ cells. Developmental Cell 19, 612–624 (2010).
Yokobayashi, S. et al. PRC1 coordinates timing of sexual differentiation of female primordial germ cells. Nature 495, 236–240 (2013).
Hargan-Calvopina, J. et al. Stage-Specific Demethylation in Primordial Germ Cells Safeguards against Precocious Differentiation. Developmental Cell 39, 1–12 (2016).
Dahlet, T. et al. Genome-wide analysis in the mouse embryo reveals the importance of DNA methylation for transcription integrity. Nature Communications 11, 3153–14 (2020).
Karimi, M. M. et al. DNA methylation and SETDB1/H3K9me3 regulate predominantly distinct sets of genes, retroelements, and chimeric transcripts in mESCs. Cell Stem Cell 8, 676–687 (2011).
Tamburri, S. et al. Histone H2AK119 Mono-Ubiquitination Is Essential for Polycomb-Mediated Transcriptional Repression. Mol. Cell 77, 840–856.e5 (2020).
Guibert, S., Forné, T. & Weber, M. Global profiling of DNA methylation erasure in mouse primordial germ cells. Genome Res. 22, 633–641 (2012).
Seisenberger, S. et al. The dynamics of genome-wide DNA methylation reprogramming in mouse primordial germ cells. Mol. Cell 48, 849–862 (2012).
Kobayashi, H. et al. High-resolution DNA methylome analysis of primordial germ cells identifies gender-specific reprogramming in mice. Genome Res. 23, 616–627 (2013).
Hackett, J. A. et al. Promoter DNA methylation couples genome-defence mechanisms to epigenetic reprogramming in the mouse germline. Development 139, 3623–3632 (2012).
Hill, P. W. S. et al. Epigenetic reprogramming enables the transition from primordial germ cell to gonocyte. Nature 13, 781 (2018).
Pohlers, M. et al. A role for E2F6 in the restriction of male-germ-cell-specific gene expression. Curr. Biol. 15, 1051–1057 (2005).
Kehoe, S. M. et al. A conserved E2F6-binding element in murine meiosis-specific gene promoters. Biol. Reprod. 79, 921–930 (2008).
Velasco, G. et al. Dnmt3b recruitment through E2F6 transcriptional repressor mediates germ-line gene silencing in murine somatic tissues. Proc. Natl. Acad. Sci. U.S.A. 107, 9281–9286 (2010).
Ostapcuk, V. et al. Activity-dependent neuroprotective protein recruits HP1 and CHD4 to control lineage-specifying genes. Nature 557, 739–743 (2018).
Rowe, H. M. et al. KAP1 controls endogenous retroviruses in embryonic stem cells. Nature 463, 237–240 (2010).
Matsui, T. et al. Proviral silencing in embryonic stem cells requires the histone methyltransferase ESET. Nature 464, 927–931 (2010).
Elsässer, S. J., Noh, K.-M., Diaz, N., Allis, C. D. & Banaszynski, L. A. Histone H3.3 is required for endogenous retroviral element silencing in embryonic stem cells. Nature 522, 240–244 (2015).
Sadic, D. et al. Atrx promotes heterochromatin formation at retrotransposons. EMBO Rep. 16, 836–850 (2015).
Hoelper, D., Huang, H., Jain, A. Y., Patel, D. J. & Lewis, P. W. Structural and mechanistic insights into ATRX-dependent and -independent functions of the histone chaperone DAXX. Nature Communications 8, 1193–13 (2017).
Tatsumi, D. et al. DNMTs and SETDB1 function as co-repressors in MAX-mediated repression of germ cell–related genes in mouse embryonic stem cells. PLoS ONE 13, e0205969–23 (2018).
Hurlin, P. J., Steingrìmsson, E., Copeland, N. G., Jenkins, N. A. & Eisenman, R. N. Mga, a dual-specificity transcription factor that interacts with Max and contains a T-domain DNA-binding motif. EMBO J. 18, 7019–7028 (1999).
Endoh, M. et al. PCGF6-PRC1 suppresses premature differentiation of mouse embryonic stem cells by regulating germ cell-related genes. Elife 6, 1513 (2017).
Stielow, B., Finkernagel, F., Stiewe, T., Nist, A. & Suske, G. MGA, L3MBTL2 and E2F6 determine genomic binding of the non-canonical Polycomb repressive complex PRC1.6. PLoS Genet 14, e1007193–29 (2018).
Kitamura, Y. et al. Identification of germ cell-specific Mga variant mRNA that promotes meiotic entry via impediment of a non-canonical PRC1. Preprint at https://www.biorxiv.org/content/10.1101/2020.09.03.282368v1 (2020).
Liu, M. et al. The polycomb group protein PCGF6 mediates germline gene silencing by recruiting histone-modifying proteins to target gene promoters. J. Biol. Chem. 295, 9712–9724 (2020).
Bartke, T. et al. Nucleosome-interacting proteins regulated by DNA and histone methylation. Cell 143, 470–484 (2010).
Spruijt, C. G. et al. Dynamic readers for 5-(hydroxy)methylcytosine and its oxidized derivatives. Cell 152, 1146–1159 (2013).
Wang, D. et al. MAX is an epigenetic sensor of 5-carboxylcytosine and is altered in multiple myeloma. Nucleic Acids Research 45, 2396–2407 (2016).
Gao, Z. et al. PCGF homologs, CBX proteins, and RYBP define functionally distinct PRC1 family complexes. Mol. Cell 45, 344–356 (2012).
Hauri, S. et al. A High-Density Map for Navigating the Human Polycomb Complexome. CellReports 17, 583–595 (2016).
Huang, Y. et al. Combinatorial Control of Recruitment of a Variant PRC1.6 Complex in Embryonic Stem Cells. CellReports 22, 3032–3043 (2018).
Shen-Li, H. et al. Essential role for Max in early embryonic growth and development. Genes Dev. 14, 17–22 (2000).
Storre, J. et al. Homeotic transformations of the axial skeleton that accompany a targeted deletion of E2f6. EMBO Rep. 3, 695–700 (2002).
Voncken, J. W. et al. Rnf2 (Ring1b) deficiency causes gastrulation arrest and cell cycle inhibition. Proc Natl Acad Sci USA 100, 2468–2473 (2003).
Pirity, M. K., Locker, J. & Schreiber-Agus, N. Rybp/DEDAF is required for early postimplantation and for central nervous system development. Molecular and Cellular Biology 25, 7193–7202 (2005).
Qin, J. et al. The Polycomb Group Protein L3mbtl2 Assembles an Atypical PRC1-Family Complex that Is Essential in Pluripotent Stem Cells and Early Development. Cell Stem Cell 11, 319–332 (2012).
Washkowitz, A. J. et al. Mga is essential for the survival of pluripotent cells during peri-implantation development. Development 142, 31–40 (2015).
Kloet, S. L. et al. The dynamic interactome and genomic targets of Polycomb complexes during stem-cell differentiation. Nat. Struct. Mol. Biol. 23, 682–690 (2016).
Scelfo, A. et al. Functional Landscape of PCGF Proteins Reveals Both RING1A/B-Dependent-and RING1A/B- Independent-Specific Activities. Mol. Cell 74, 1037–1052.e7 (2019).
Storre, J. et al. Silencing of the meiotic genes SMC1beta and STAG3 in somatic cells by E2F6. J. Biol. Chem. 280, 41380–41386 (2005).
Ujhelly, O. et al. Lack of Rybp in Mouse Embryonic Stem Cells Impairs Cardiac Differentiation. Stem Cells Dev 24, 2193–2205 (2015).
Suzuki, A. et al. Loss of MAX results in meiotic entry in mouse embryonic and germline stem cells. Nature Communications 7, 11056 (2016).
Zhao, W. et al. Essential Role for Polycomb Group Protein Pcgf6 in Embryonic Stem Cell Maintenance and a Noncanonical Polycomb Repressive Complex 1 (PRC1) Integrity. J. Biol. Chem. 292, 2773–2784 (2017).
Hayashi, K., de Sousa Lopes, S. M. C., Tang, F., Lao, K. & Surani, M. A. Dynamic equilibrium and heterogeneity of mouse pluripotent stem cells with distinct functional and epigenetic states. Cell Stem Cell 3, 391–401 (2008).
Morgani, S. M. et al. Totipotent embryonic stem cells arise in ground-state culture conditions. CellReports 3, 1945–1957 (2013).
Kumar, R. M. et al. Deconstructing transcriptional heterogeneity in pluripotent stem cells. Nature 516, 56–61 (2014).
van Mierlo, G. et al. Integrative Proteomic Profiling Reveals PRC2-Dependent Epigenetic Crosstalk Maintains Ground-State Pluripotency. Cell Stem Cell 24, 123–137.e8 (2019).
Lavarone, E., Barbieri, C. M. & Pasini, D. Dissecting the role of H3K27 acetylation and methylation in PRC2 mediated control of cellular identity. Nature Communications 10, 1679–16 (2019).
Eberl, H. C., Spruijt, C. G., Kelstrup, C. D., Vermeulen, M. & Mann, M. A Map of General and Specialized Chromatin Readers in Mouse Tissues Generated by Label-free Interaction Proteomics. Mol. Cell 49, 368–378 (2013).
Kunowska, N., Rotival, M., Yu, L., Choudhary, J. & Dillon, N. Identification of protein complexes that bind to histone H3 combinatorial modifications using super-SILAC and weighted correlation network analysis. Nucleic Acids Research 43, 1418–1432 (2015).
Blackledge, N. P. et al. Variant PRC1 complex-dependent H2A ubiquitylation drives PRC2 recruitment and polycomb domain formation. Cell 157, 1445–1459 (2014).
Qin, J. et al. Mga safeguards embryonic stem cells from acquiring extraembryonic endoderm fates. Sci. Adv. 7, eabe5689. (2021).
Brind'Amour, J. et al. An ultra-low-input native ChIP-seq protocol for genome-wide profiling of rare cell populations. Nature Communications 6, 6033 (2015).
Chen, Z., Djekidel, M. N. & Zhang, Y. Distinct dynamics and functions of H2AK119ub1 and H3K27me3 in mouse preimplantation embryos. Nat. Genet. 1–34 (2021). doi:10.1038/s41588-021-00821-2
Mei, H. et al. H2AK119ub1 guides maternal inheritance and zygotic deposition of H3K27me3 in mouse embryos. Nat. Genet. 1–35 (2021). doi:10.1038/s41588-021-00820-3
Hayashi, K., Ohta, H., Kurimoto, K., Aramaki, S. & Saitou, M. Reconstitution of the mouse germ cell specification pathway in culture by pluripotent stem cells. Cell 146, 519–532 (2011).
Grosswendt, S. et al. Epigenetic regulator function through mouse gastrulation. Nature 584, 1–32 (2020).
Shirane, K. et al. Global Landscape and Regulatory Principles of DNA Methylation Reprogramming for Germ Cell Specification by Mouse Pluripotent Stem Cells. Developmental Cell 39, 87–103 (2016).
Habibi, E. et al. Whole-genome bisulfite sequencing of two distinct interconvertible DNA methylomes of mouse embryonic stem cells. Cell Stem Cell 13, 360–369 (2013).
Gu, T. et al. DNMT3A and TET1 cooperate to regulate promoter epigenetic landscapes in mouse embryonic stem cells. Genome Biol. 19, 88–15 (2018).
Khoueiry, R. et al. Lineage-specific functions of TET1 in the postimplantation mouse embryo. Nat. Genet. 49, 1061–1072 (2017).
Xiong, J. et al. Cooperative Action between SALL4A and TET Proteins in Stepwise Oxidation of 5-Methylcytosine. Mol. Cell 64, 913–925 (2016).
Choi, J. et al. Prolonged Mek1/2 suppression impairs the developmental potential of embryonic stem cells. Nature 548, 219–223 (2017).
Uranishi, K. et al. Two DNA binding domains of Mga act in combination to suppress ectopic activation of meiosis-related genes in mouse embryonic stem cells. Preprint at https://www.biorxiv.org/content/10.1101/2020.07.21.215079v1 (2020).
Wu, K. et al. SETDB1-Mediated Cell Fate Transition between 2C-Like and Pluripotent States. CellReports 30, 25–36.e6 (2020).
Su, W., Sun, J., Shimizu, K. & Kadota, K. TCC-GUI: a Shiny-based application for differential expression analysis of RNA-Seq count data. BMC Res Notes 12, 133–6 (2019).
Weinberg, D. N. et al. The histone mark H3K36me2 recruits DNMT3A and shapes the intergenic DNA methylation landscape. Nature 573, 281–286 (2019).
Xu, Q. et al. SETD2 regulates the maternal epigenome, genomic imprinting and embryonic development. Nat. Genet. 51, 844–856 (2019).
Shirane, K., Miura, F., Ito, T. & Lorincz, M. C. NSD1-deposited H3K36me2 directs de novo methylation in the mouse male germline and counteracts Polycomb-associated silencing. Nat. Genet. 14, 1–31 (2020).
Hickey, G. J. M. et al. Establishment of Developmental Gene Silencing by Ordered Polycomb Complex Recruitment in Early Zebrafish Embryos. Preprint at https://www.biorxiv.org/content/10.1101/2021.03.16.435592v1 (2021).
Dodge, J. E., Kang, Y. K., Beppu, H., Lei, H. & Li, E. Histone H3-K9 Methyltransferase ESET Is Essential for Early Development. Molecular and Cellular Biology 24, 2478–2486 (2004).
Janic, A., Mendizabal, L., Llamazares, S., Rossell, D. & Gonzalez, C. Ectopic expression of germline genes drives malignant brain tumor growth in Drosophila. Science 330, 1824–1827 (2010).
Wu, X., Shi, Z., Cui, M., Han, M. & Ruvkun, G. Repression of germline RNAi pathways in somatic cells by retinoblastoma pathway chromatin complexes. PLoS Genet 8, e1002542 (2012).
Rechtsteiner, A. et al. Repression of Germline Genes in Caenorhabditis elegansSomatic Tissues by H3K9 Dimethylation of Their Promoters. Genetics 212, 125–140 (2019).
Yang, C.-S., Chang, K.-Y., Dang, J. & Rana, T. M. Polycomb Group Protein Pcgf6 Acts as a Master Regulator to Maintain Embryonic Stem Cell Identity. Scientific Reports 6, 26899 (2016).
Tsusaka, T., Fukuda, K., Shimura, C., Kato, M. & Shinkai, Y. The fibronectin type-III (FNIII) domain of ATF7IP contributes to efficient transcriptional silencing mediated by the SETDB1 complex. Epigenetics & Chromatin 13, 52–15 (2020).
Szabó, P. E., Hübner, K., Schöler, H. & Mann, J. R. Allele-specific expression of imprinted genes in mouse migratory primordial germ cells. Mechanisms of Development 115, 157–160 (2002).
Sharif, J. et al. Activation of Endogenous Retroviruses in Dnmt1(-/-) ESCs Involves Disruption of SETDB1-Mediated Repression by NP95 Binding to Hemimethylated DNA. Cell Stem Cell 19, 81–94 (2016).
Endoh, M. et al. Polycomb group proteins Ring1A/B are functionally linked to the core transcriptional regulatory circuitry to maintain ES cell identity. Development 135, 1513–1524 (2008).
Masui, S. et al. An efficient system to establish multiple embryonic stem cell lines carrying an inducible expression unit. Nucleic Acids Research 33, e43 (2005).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Li, B. & Dewey, C. N. RSEM: accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinformatics 12, 323–16 (2011).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, 550–21 (2014).
Ge, S. X., Jung, D. & Yao, R. ShinyGO: a graphical gene-set enrichment tool for animals and plants. Bioinformatics 36, 2628–2629 (2020).
Li, H. & Durbin, R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics 25, 1754–1760 (2009).
Younesy, H., Möller, T., Lorincz, M. C., Karimi, M. M. & Jones, S. J. M. VisRseq: R-based visual framework for analysis of sequencing data. BMC Bioinformatics 16 Suppl 11, S2–14 (2015).
Li, L.-C. & Dahiya, R. MethPrimer: designing primers for methylation PCRs. Bioinformatics 18, 1427–1431 (2002).
Kumaki, Y., Oda, M. & Okano, M. QUMA: quantification tool for methylation analysis. Nucleic Acids Research 36, W170–5 (2008).
Robinson, J. T. et al. Integrative genomics viewer. Nat Biotechnol 29, 24–26 (2011).

There is NO Competing Interest.

SupplementaryData1.xlsx
Supplementary Data 1
SupplementaryData2.xlsx
Supplementary Data 2
SupplementaryData3.xlsx
Supplementary Data 3
SupplementaryData4.xlsx
Supplementary Data 4
SupplementaryData5.xlsx
Supplementary Data 5
SupplementaryData6.xlsx
Supplementary Data 6
DescriptionofAdditionalSupplementaryFiles.pdf
Description of Additional Supplementary Files
Mochizukietal.SUPPLEMENTARYINFONCOMMS2113135.pdf
Supplementary Information

Download PDF

Journal Publication

published 01 Dec, 2021

Read the published version in Nature Communications →

Version 1

posted

You are reading this latest preprint version

Repression of germline genes by PRC1.6 and SETDB1 in the early embryo precedes DNA methylation-mediated silencing

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Discussion

Methods

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1