EHMT2 suppresses the variation of transcriptional switches in the mouse embryo.

EHMT2 is the main euchromatic H3K9 methyltransferase. Embryos with zygotic, or maternal mutation in the Ehmt2 gene exhibit variable developmental delay. To understand how EHMT2 prevents variable developmental delay we performed RNA sequencing of mutant and somite stage-matched normal embryos at 8.5-9.5 days of gestation. Using four-way comparisons between delayed and normal embryos we clarified what it takes to be normal and what it takes to develop. We identified differentially expressed genes, for example Hox genes that simply reflected the difference in developmental progression of wild type and the delayed mutant uterus-mate embryos. By comparing wild type and zygotic mutant embryos along the same developmental window we detected a role of EHMT2 in suppressing variation in the transcriptional switches. We identified transcription changes where precise switching during development occurred only in the normal but not in the mutant embryo. At the 6-somite stage, gastrulation-specific genes were not precisely switched off in the Ehmt2-/- zygotic mutant embryos, while genes involved in organ growth, connective tissue development, striated muscle development, muscle differentiation, and cartilage development were not precisely switched on. The Ehmt2mat-/+ maternal mutant embryos displayed high transcriptional variation consistent with their variable survival. Variable derepression of transcripts occurred dominantly in the maternally inherited allele. Transcription was normal in the parental haploinsufficient wild type embryos despite their delay, consistent with their good prospects. Global profiling of transposable elements revealed EHMT2 targeted DNA methylation and suppression at LTR repeats, mostly ERVKs. In Ehmt2-/- embryos, transcription over very long distances initiated from such misregulated 'driver' ERVK repeats, encompassing a multitude of misexpressed 'passenger' repeats. In summary, EHMT2 reduced transcriptional variation of developmental switch genes and developmentally switching repeat elements at the six-somite stage embryos. These findings establish EHMT2 as a suppressor of transcriptional and developmental variation at the transition between gastrulation and organ specification.

The remodeling of Z-DNA in the mammalian germ line.

We recently discovered a novel biological process, the scheduled remodeling of Z-DNA structures in the developing fetal mouse male germ cells [Nat. Cell Biol. 24, 1141-1153]. This process affects purine/pyrimidine dinucleotide repeat (PPR) rich sequences, which can form stable left-handed Z-DNA structures. The protein that carries out this function is identified as ZBTB43, member of a large family of ZBTB proteins. Z-DNA remodeling by ZBTB43 not only coincides with global remodeling of DNA methylation and chromatin events in the male germ line, but it also is a prerequisite for de novo DNA methylation. When ZBTB43 changes DNA structure from the left-handed zigzag shaped Z-DNA to the regular smooth right-handed B-DNA, it also generates a suitable substrate for the de novo DNA methyltransferase, DNMT3A. By instructing de novo DNA methylation at PPRs in prospermatogonia, ZBTB43 safeguards epigenomic integrity of the male gamete. PPRs are fragile sequences, sites of large deletions and rearrangements in mammalian cells, and this fragility is thought to be due to Z-DNA structure formation rather than the sequence itself. This idea is now supported by the in vivo finding that DNA double strand breaks accumulate in mutant prospermatogonia which lack ZBTB43-dependent Z-DNA remodeling. If unrepaired, double stranded DNA breaks can lead to germ line mutations. Therefore, by preventing such breaks ZBTB43 is critical for guarding genome stability between generations. Here, we discuss the significance and implications of these findings in more detail.

EHMT2 and SETDB1 protect the maternal pronucleus from 5mC oxidation.

Genome-wide DNA "demethylation" in the zygote involves global TET3-mediated oxidation of 5-methylcytosine (5mC) to 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxylcytosine (5caC) in the paternal pronucleus. Asymmetrically enriched histone H3K9 methylation in the maternal pronucleus was suggested to protect the underlying DNA from 5mC conversion. We hypothesized that an H3K9 methyltransferase enzyme, either EHMT2 or SETDB1, must be expressed in the oocyte to specify the asymmetry of 5mC oxidation. To test these possibilities, we genetically deleted the catalytic domain of either EHMT2 or SETDB1 in growing oocytes and achieved significant reduction of global H3K9me2 or H3K9me3 levels, respectively, in the maternal pronucleus. We found that the asymmetry of global 5mC oxidation was significantly reduced in the zygotes that carried maternal mutation of either the Ehmt2 or Setdb1 genes. Whereas the levels of 5hmC, 5fC, and 5caC increased, 5mC levels decreased in the mutant maternal pronuclei. H3K9me3-rich rings around the nucleolar-like bodies retained 5mC in the maternal mutant zygotes, suggesting that the pericentromeric heterochromatin regions are protected from DNA demethylation independently of EHMT2 and SETDB1. We observed that the maternal pronuclei expanded in size in the mutant zygotes and contained a significantly increased number of nucleolar-like bodies compared with normal zygotes. These findings suggest that oocyte-derived EHMT2 and SETDB1 enzymes have roles in regulating 5mC oxidation and in the structural aspects of zygote development.

The role of Tet3 DNA dioxygenase in epigenetic reprogramming by oocytes.

Sperm and eggs carry distinctive epigenetic modifications that are adjusted by reprogramming after fertilization. The paternal genome in a zygote undergoes active DNA demethylation before the first mitosis. The biological significance and mechanisms of this paternal epigenome remodelling have remained unclear. Here we report that, within mouse zygotes, oxidation of 5-methylcytosine (5mC) occurs on the paternal genome, changing 5mC into 5-hydroxymethylcytosine (5hmC). Furthermore, we demonstrate that the dioxygenase Tet3 (ref. 5) is enriched specifically in the male pronucleus. In Tet3-deficient zygotes from conditional knockout mice, paternal-genome conversion of 5mC into 5hmC fails to occur and the level of 5mC remains constant. Deficiency of Tet3 also impedes the demethylation process of the paternal Oct4 and Nanog genes and delays the subsequent activation of a paternally derived Oct4 transgene in early embryos. Female mice depleted of Tet3 in the germ line show severely reduced fecundity and their heterozygous mutant offspring lacking maternal Tet3 suffer an increased incidence of developmental failure. Oocytes lacking Tet3 also seem to have a reduced ability to reprogram the injected nuclei from somatic cells. Therefore, Tet3-mediated DNA hydroxylation is involved in epigenetic reprogramming of the zygotic paternal DNA following natural fertilization and may also contribute to somatic cell nuclear reprogramming during animal cloning.

Characterization of the imprinting signature of mouse embryo fibroblasts by RNA deep sequencing.

Mouse embryo fibroblasts (MEFs) are convenient sources for biochemical studies when cell number in mouse embryos is limiting. To derive the imprinting signature of MEFs and potentially detect novel imprinted genes we performed strand- and allele-specific RNA deep sequencing. We used sequenom allelotyping in embryo and adult organs to verify parental allele-specific expression. Thirty-two known ubiquitously imprinted genes displayed correct parental allele-specific transcripts in MEFs. Our analysis did not reveal any novel imprinted genes, but detected extended parental allele-specific transcripts in several known imprinted domains: maternal allele-specific transcripts downstream of Grb10 and downstream of Meg3, Rtl1as and Rian in the Dlk1-Dio3 cluster, an imprinted domain implicated in development and pluripotency. We detected paternal allele-specific transcripts downstream of Nespas, Peg3, Peg12 and Snurf/Snrpn. These imprinted transcript extensions were not unique to MEFs, but were also present in other somatic cells. The 5' end points of the imprinted transcript extensions did not carry opposing chromatin marks or parental allele-specific DNA methylation, suggesting that their parental allele-specific transcription is under the control of the extended imprinted genes. Based on the imprinting signature of MEFs, these cells provide valid models for understanding the biochemical aspects of genomic imprinting.

5-Hydroxymethylcytosine: a stable or transient DNA modification?

The DNA base 5-hydroxymethylcytosine (5hmC) is produced by enzymatic oxidation of 5-methylcytosine (5mC) by 5mC oxidases (the Tet proteins). Since 5hmC is recognized poorly by DNA methyltransferases, DNA methylation may be lost at 5hmC sites during DNA replication. In addition, 5hmC can be oxidized further by Tet proteins and converted to 5-formylcytosine and 5-carboxylcytosine, two bases that can be removed from DNA by base excision repair. The completed pathway represents a replication-independent DNA demethylation cycle. However, the DNA base 5hmC is also known to be rather stable and occurs at substantial levels, for example in the brain, suggesting that it represents an epigenetic mark by itself that may regulate chromatin structure and transcription. Focusing on a few well-studied tissues and developmental stages, we discuss the opposing views of 5hmC as a transient intermediate in DNA demethylation and as a modified DNA base with an instructive role.

Reprogramming of the paternal genome upon fertilization involves genome-wide oxidation of 5-methylcytosine.

Genome-wide erasure of DNA cytosine-5 methylation has been reported to occur along the paternal pronucleus in fertilized oocytes in an apparently replication-independent manner, but the mechanism of this reprogramming process has remained enigmatic. Recently, considerable amounts of 5-hydroxymethylcytosine (5hmC), most likely derived from enzymatic oxidation of 5-methylcytosine (5mC) by TET proteins, have been detected in certain mammalian tissues. 5hmC has been proposed as a potential intermediate in active DNA demethylation. Here, we show that in advanced pronuclear-stage zygotes the paternal pronucleus contains substantial amounts of 5hmC but lacks 5mC. The converse is true for the maternal pronucleus, which retains 5mC but shows little or no 5hmC signal. Importantly, 5hmC persists into mitotic one-cell, two-cell, and later cleavage-stage embryos, suggesting that 5mC oxidation is not followed immediately by genome-wide removal of 5hmC through excision repair pathways or other mechanisms. This conclusion is supported by bisulfite sequencing data, which shows only limited conversion of modified cytosines to cytosines at several gene loci. It is likely that 5mC oxidation is carried out by the Tet3 oxidase. Tet3, but not Tet1 or Tet2, was expressed at high levels in oocytes and zygotes, with rapidly declining levels at the two-cell stage. Our results show that 5mC oxidation is part of the early life cycle of mammals.

Role of transcription in imprint establishment in the male and female germ lines.

The authors highlight an area of research that focuses on the establishment of genomic imprints: how the female and male germlines set up opposite instructions for imprinted genes in the maternally and paternally inherited chromosomes. Mouse genetics studies have solidified the role of transcription across the germline differentially methylated regions in the establishment of maternal genomic imprinting. One work now reveals that such transcription is also important in paternal imprinting establishment. This allows the authors to propose a unifying mechanism, in the form of transcription across germline differentially methylated regions, that specifies DNA methylation imprint establishment. Differences in the timing, genomic location and nature of such transcription events in the male versus female germlines in turn explain the difference between paternal and maternal imprints.

Postnatal survival of mice with maternal duplication of distal chromosome 7 induced by a Igf2/H19 imprinting control region lacking insulator function.

The misexpressed imprinted genes causing developmental failure of mouse parthenogenones are poorly defined. To obtain further insight, we investigated misexpressions that could cause the pronounced growth deficiency and death of fetuses with maternal duplication of distal chromosome (Chr) 7 (MatDup.dist7). Their small size could involve inactivity of Igf2, encoding a growth factor, with some contribution by over-expression of Cdkn1c, encoding a negative growth regulator. Mice lacking Igf2 expression are usually viable, and MatDup.dist7 death has been attributed to the misexpression of Cdkn1c or other imprinted genes. To examine the role of misexpressions determined by two maternal copies of the Igf2/H19 imprinting control region (ICR)-a chromatin insulator, we introduced a mutant ICR (ICR(Delta)) into MatDup.dist7 fetuses. This activated Igf2, with correction of H19 expression and other imprinted transcripts expected. Substantial growth enhancement and full postnatal viability was obtained, demonstrating that the aberrant MatDup.dist7 phenotype is highly dependent on the presence of two unmethylated maternal Igf2/H19 ICRs. Activation of Igf2 is likely the predominant correction that rescued growth and viability. Further experiments involved the introduction of a null allele of Cdkn1c to alleviate its over-expression. Results were not consistent with the possibility that this misexpression alone, or in combination with Igf2 inactivity, mediates MatDup.dist7 death. Rather, a network of misexpressions derived from dist7 is probably involved. Our results are consistent with the idea that reduced expression of IGF2 plays a role in the aetiology of the human imprinting-related growth-deficit disorder, Silver-Russell syndrome.

CTCF-dependent chromatin bias constitutes transient epigenetic memory of the mother at the H19-Igf2 imprinting control region in prospermatogonia.

Genomic imprints-parental allele-specific DNA methylation marks at the differentially methylated regions (DMRs) of imprinted genes-are erased and reestablished in germ cells according to the individual's sex. Imprint establishment at paternally methylated germ line DMRs occurs in fetal male germ cells. In prospermatogonia, the two unmethylated alleles exhibit different rates of de novo methylation at the H19/Igf2 imprinting control region (ICR) depending on parental origin. We investigated the nature of this epigenetic memory using bisulfite sequencing and allele-specific ChIP-SNuPE assays. We found that the chromatin composition in fetal germ cells was biased at the ICR between the two alleles with the maternally inherited allele exhibiting more H3K4me3 and less H3K9me3 than the paternally inherited allele. We determined genetically that the chromatin bias, and also the delayed methylation establishment in the maternal allele, depended on functional CTCF insulator binding sites in the ICR. Our data suggest that, in primordial germ cells, maternally inherited allele-specific CTCF binding sets up allele-specific chromatin differences at the ICR. The erasure of these allele-specific chromatin marks is not complete before the process of de novo methylation imprint establishment begins. CTCF-dependent allele-specific chromatin composition imposes a maternal allele-specific delay on de novo methylation imprint establishment at the H19/Igf2 ICR in prospermatogonia.

Establishment of paternal methylation imprint at the H19/Igf2 imprinting control region.

The insulator model explains the workings of the H19 and Igf2 imprinted domain in the soma, where insulation of the Igf2 promoter from its enhancers occurs by CTCF in the maternally inherited unmethylated chromosome but not the paternally inherited methylated allele. The molecular mechanism that targets paternal methylation imprint establishment to the imprinting control region (ICR) in the male germline is unknown. We tested the function of prospermatogonia-specific broad low-level transcription in this process using mouse genetics. Paternal imprint establishment was abnormal when transcription was stopped at the entry point to the ICR. The germline epimutation persisted into the paternal allele of the soma, resulting in reduced Igf2 in fetal organs and reduced fetal growth, consistent with the insulator model and insulin-like growth factor 2 (IGF2)'s role as fetal growth factor. These results collectively support the role of broad low-level transcription through the H19/Igf2 ICR in the establishment of its paternal methylation imprint in the male germ line, with implications for Silver-Russell syndrome.

Coordinated allele-specific histone acetylation at the differentially methylated regions of imprinted genes.

Genomic imprinting is an epigenetic inheritance system characterized by parental allele-specific gene expression. Allele-specific DNA methylation and chromatin composition are two epigenetic modification systems that control imprinted gene expression. To get a general assessment of histone lysine acetylation at imprinted genes we measured allele-specific acetylation of a wide range of lysine residues, H3K4, H3K18, H3K27, H3K36, H3K79, H3K64, H4K5, H4K8, H4K12, H2AK5, H2BK12, H2BK16 and H2BK46 at 11 differentially methylated regions (DMRs) in reciprocal mouse crosses using multiplex chromatin immunoprecipitation SNuPE assays. Histone acetylation marks generally distinguished the methylation-free alleles from methylated alleles at DMRs in mouse embryo fibroblasts and embryos. Acetylated lysines that are typically found at transcription start sites exhibited stronger allelic bias than acetylated histone residues in general. Maternally methylated DMRs, that usually overlap with promoters exhibited higher levels of acetylation and a 10% stronger allele-specific bias than paternally methylated DMRs that reside in intergenic regions. Along the H19/Igf2 imprinted domain, allele-specific acetylation at each lysine residue depended on functional CTCF binding sites in the imprinting control region. Our results suggest that many different histone acetyltransferase and histone deacetylase enzymes must act in concert in setting up and maintaining reciprocal parental allelic histone acetylation at DMRs.

Concordance of hydrogen peroxide-induced 8-oxo-guanine patterns with two cancer mutation signatures of upper GI tract tumors.

Oxidative DNA damage has been linked to inflammation, cancer, and aging. Here, we have mapped two types of oxidative DNA damage, oxidized guanines produced by hydrogen peroxide and oxidized thymines created by potassium permanganate, at a single-base resolution. 8-Oxo-guanine occurs strictly dependent on the G/C sequence context and shows a pronounced peak at transcription start sites (TSSs). We determined the trinucleotide sequence pattern of guanine oxidation. This pattern shows high similarity to the cancer-associated single-base substitution signatures SBS18 and SBS36. SBS36 is found in colorectal cancers that carry mutations in MUTYH, encoding a repair enzyme that operates on 8-oxo-guanine mispairs. SBS18 is common in inflammation-associated upper gastrointestinal tract tumors including esophageal and gastric adenocarcinomas. Oxidized thymines induced by permanganate occur with a distinct dinucleotide specificity, 5'T-A/C, and are depleted at the TSS. Our data suggest that two cancer mutational signatures, SBS18 and SBS36, are caused by reactive oxygen species.

Z-DNA is remodelled by ZBTB43 in prospermatogonia to safeguard the germline genome and epigenome.

Mutagenic purine-pyrimidine repeats can adopt the left-handed Z-DNA conformation. DNA breaks at potential Z-DNA sites can lead to somatic mutations in cancer or to germline mutations that are transmitted to the next generation. It is not known whether any mechanism exists in the germ line to control Z-DNA structure and DNA breaks at purine-pyrimidine repeats. Here we provide genetic, epigenomic and biochemical evidence for the existence of a biological process that erases Z-DNA specifically in germ cells of the mouse male foetus. We show that a previously uncharacterized zinc finger protein, ZBTB43, binds to and removes Z-DNA, preventing the formation of DNA double-strand breaks. By removing Z-DNA, ZBTB43 also promotes de novo DNA methylation at CG-containing purine-pyrimidine repeats in prospermatogonia. Therefore, the genomic and epigenomic integrity of the species is safeguarded by remodelling DNA structure in the mammalian germ line during a critical window of germline epigenome reprogramming.

DNA methylation dynamics and dysregulation delineated by high-throughput profiling in the mouse.

We have developed a mouse DNA methylation array that contains 296,070 probes representing the diversity of mouse DNA methylation biology. We present a mouse methylation atlas as a rich reference resource of 1,239 DNA samples encompassing distinct tissues, strains, ages, sexes, and pathologies. We describe applications for comparative epigenomics, genomic imprinting, epigenetic inhibitors, patient-derived xenograft assessment, backcross tracing, and epigenetic clocks. We dissect DNA methylation processes associated with differentiation, aging, and tumorigenesis. Notably, we find that tissue-specific methylation signatures localize to binding sites for transcription factors controlling the corresponding tissue development. Age-associated hypermethylation is enriched at regions of Polycomb repression, while hypomethylation is enhanced at regions bound by cohesin complex members. Apc Min/+ polyp-associated hypermethylation affects enhancers regulating intestinal differentiation, while hypomethylation targets AP-1 binding sites. This Infinium Mouse Methylation BeadChip (version MM285) is widely accessible to the research community and will accelerate high-sample-throughput studies in this important model organism.

Immunochemical Detection of Modified Cytosine Species in Mammalian Preimplantation Embryos.

DNA methylation undergoes dynamic changes at the genome-wide scale during the early steps of mammalian embryo development. Immunochemical detection of 5-methylcytosine (5mC) in the zygote has led to the discovery that a global loss of DNA methylation takes place soon after fertilization, occurring rapidly in the paternal pronucleus. Using the same method employed above, which detects modified bases in the denatured single stranded DNA, we showed that this active DNA "demethylation" in the paternal pronucleus involves oxidation of 5mC to 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC), and 5-carboxycytosine (5caC) by the TET3 enzyme. By immunostaining of genetically altered zygotes we revealed that the maternal pronucleus is protected from TET3-mediated oxidation by histone H3K9 methyltransferase enzymes, EHMT2 and SETDB1. The same assays are also applicable for visualizing the temporal and spatial distribution of the modified cytosine residues in preimplantation embryos. Here, we provide a detailed protocol for detecting 5mC, 5hmC, 5fC, and 5caC in mouse zygotes and preimplantation-stage embryos using antibodies raised against modified cytosine species.

H3K9 methyltransferase EHMT2/G9a controls ERVK-driven noncanonical imprinted genes.

Aim: Paternal allele-specific expression of noncanonical imprinted genes in the extraembryonic lineages depends on an H3K27me3-based imprint in the oocyte, which is not a lasting mark. We hypothesized that EHMT2, the main euchromatic H3K9 dimethyltransferase, also has a role in controlling noncanonical imprinting. Methods: We carried out allele-specific total RNA-seq analysis in the ectoplacental cone of somite-matched 8.5 days post coitum embryos using reciprocal mouse crosses. Results: We found that the maternal allele of noncanonical imprinted genes was derepressed from its ERVK promoter in the Ehmt2-/- ectoplacental cone. In Ehmt2-/- embryos, loss of DNA methylation accompanied biallelic derepression of the ERVK promoters. Canonical imprinting and imprinted X chromosome inactivation were generally undisturbed. Conclusion: EHMT2 is essential for repressing the maternal allele in noncanonical imprinting.

High-Resolution Analysis of 5-Hydroxymethylcytosine by TET-Assisted Bisulfite Sequencing.

DNA cytosine modification is an important epigenetic mechanism that serves critical functions in a variety of biological processes in development and disease. 5-Methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) are the two most common epigenetic marks found in the mammalian genome. 5hmC is generated from 5mC by the ten-eleven translocation (TET) family of dioxygenase enzymes. This modification can reach substantial levels in certain cell types such as embryonic stem cells and neurons. Standard bisulfite sequencing techniques cannot distinguish between 5mC and 5hmC. Therefore, the method of TET-assisted bisulfite sequencing has been developed for detecting 5hmC specifically. The method is based on protection of 5hmC by glycosylation followed by complete oxidation of both 5mC and 5fC to 5caC, which converts to uracil after bisulfite treatment leaving only 5hmC remaining as a cytosine signal after PCR and sequencing. The method requires a highly active TET protein for the conversion steps. Here, we present an efficient TET protein purification method and a streamlined TAB-sequencing protocol for 5hmC analysis at single base resolution.

Prenatal correction of IGF2 to rescue the growth phenotypes in mouse models of Beckwith-Wiedemann and Silver-Russell syndromes.

Beckwith-Wiedemann syndrome (BWS) and Silver-Russell syndrome (SRS) are imprinting disorders manifesting as aberrant fetal growth and severe postnatal-growth-related complications. Based on the insulator model, one-third of BWS cases and two-thirds of SRS cases are consistent with misexpression of insulin-like growth factor 2 (IGF2), an important facilitator of fetal growth. We propose that the IGF2-dependent BWS and SRS cases can be identified by prenatal diagnosis and can be prevented by prenatal intervention targeting IGF2. We test this hypothesis using our mouse models of IGF2-dependent BWS and SRS. We find that genetically normalizing IGF2 levels in a double rescue experiment corrects the fetal overgrowth phenotype in the BWS model and the growth retardation in the SRS model. In addition, we pharmacologically rescue the BWS growth phenotype by reducing IGF2 signaling during late gestation. This animal study encourages clinical investigations to target IGF2 for prenatal diagnosis and prenatal prevention in human BWS and SRS.

Maternal DOT1L is dispensable for mouse development.

A battery of chromatin modifying enzymes play essential roles in remodeling the epigenome in the zygote and cleavage stage embryos, when the maternal genome is the sole contributor. Here we identify an exemption. DOT1L methylates lysine 79 in the globular domain of histone H3 (H3K79). Dot1l is an essential gene, as homozygous null mutant mouse embryos exhibit multiple developmental abnormalities and die before 11.5 days of gestation. To test if maternally deposited DOT1L is required for embryo development, we carried out a conditional Dot1l knockout in growing oocytes using the Zona pellucida 3-Cre (Zp3-Cre) transgenic mice. We found that the resulting maternal mutant Dot1lmat-/+ offspring displayed normal development and fertility, suggesting that the expression of the paternally inherited copy of Dot1l in the embryo is sufficient to support development. In addition, Dot1l maternal deletion did not affect the parental allele-specific expression of imprinted genes, indicating that DOT1L is not needed for imprint establishment in the oocyte or imprint protection in the zygote. In summary, uniquely and as opposed to other histone methyltransferases and histone marks, maternal DOT1L deposition and H3K79 methylation in the zygote and in the preimplantation stage embryo is dispensable for mouse development.