Parental allele-specific expression of imprinted genes is mediated by imprinting control regions (ICRs) that are constitutively marked by DNA methylation imprints on the maternal or paternal allele. Mono-allelic DNA methylation is strictly required for the process of imprinting and has to be faithfully maintained during the entire life-span. While the regulation of DNA methylation itself is well understood, the mechanisms whereby the opposite allele remains unmethylated are unclear. Here, we show that in the mouse, at maternally methylated ICRs, the paternal allele, which is constitutively associated with H3K4me2/3, is marked by default by H3K27me3 when these ICRs are transcriptionally inactive, leading to the formation of a bivalent chromatin signature. Our data suggest that at ICRs, chromatin bivalency has a protective role by ensuring that DNA on the paternal allele remains unmethylated and protected against spurious and unscheduled gene expression. Moreover, they provide the proof of concept that, beside pluripotent cells, chromatin bivalency is the default state of transcriptionally inactive CpG island promoters, regardless of the developmental stage, thereby contributing to protect cell identity.

5-Hydroxymethylcytosine (5hmC), converted from 5-methylcytosine (5mC) by ten-eleven translocation (Tet) enzymes, has recently drawn attention as the "sixth base" of DNA since it is considered an intermediate of the demethylation pathway. Nonetheless, it remains to be addressed how 5hmC is linked to the development of human imprinting disorders. In this regard, conventional bisulfite (BS) treatment is unable to differentiate 5hmC from 5mC. It is thus hypothesized that BS conversion-derived "hypermethylation" at imprinting control regions (ICRs), which may cause imprinting disorders, would in fact be attributable to excessively increased levels of 5hmC as well as 5mC. To test this hypothesis, we applied the newly developed oxidative BS (oxBS) treatment to detect 5hmC in blood samples from Kagami-Ogata syndrome (KOS14) patients caused by an epimutation (hypermethylation) of two differentially methylated regions (DMRs) functioning as ICRs, namely, IG-DMR and MEG3-DMR.

Familial recurrent hydatidiform mole (RHM) is a maternal-effect autosomal recessive disorder usually associated with mutations of the NLRP7 gene. It is characterized by HM with excessive trophoblastic proliferation, which mimics the appearance of androgenetic molar conceptuses despite their diploid biparental constitution. It has been proposed that the phenotypes of both types of mole are associated with aberrant genomic imprinting. However no systematic analyses for imprinting defects have been reported. Here, we present the genome-wide methylation profiles of both spontaneous androgenetic and biparental NLRP7 defective molar tissues. We observe total paternalization of all ubiquitous and placenta-specific differentially methylated regions (DMRs) in four androgenetic moles; namely gain of methylation at paternally methylated loci and absence of methylation at maternally methylated regions. The methylation defects observed in five RHM biopsies from NLRP7 defective patients are restricted to lack-of-methylation at maternal DMRs. Surprisingly RHMs from two sisters with the same missense mutations, as well as consecutive RHMs from one affected female show subtle allelic methylation differences, suggesting inter-RHM variation. These epigenotypes are consistent with NLRP7 being a maternal-effect gene and involved in imprint acquisition in the oocyte. In addition, bioinformatic screening of the resulting methylation datasets identified over sixty loci with methylation profiles consistent with imprinting in the placenta, of which we confirm 22 as novel maternally methylated loci. These observations strongly suggest that the molar phenotypes are due to defective placenta-specific imprinting and over-expression of paternally expressed transcripts, highlighting that maternal-effect mutations of NLRP7 are associated with the most severe form of multi-locus imprinting defects in humans.

Human 15q11-13 is responsible for Prader-Willi syndrome (PWS) and Angelman syndrome (AS) and includes several imprinted genes together with bipartite elements named AS-IC (imprinting center) and PWS-IC. These concertedly confer allele specificity on 15q11-13. Here, we report DNA methylation status of 15q11-13 and other autosomal imprinted differentially methylated regions (iDMRs) in cases with various deletions within the PWS/AS-responsible region.

Cancer is as much an epigenetic disease as a genetic one; however, the interplay between these two processes is unclear. Recently, it has been shown that a large proportion of DNA methylation variability can be explained by allele-specific methylation (ASM), either at classical imprinted loci or those regulated by underlying genetic variants. During a recent screen for imprinted differentially methylated regions, we identified the genomic interval overlapping the non-coding nc886 RNA (previously known as vtRNA2-1) as an atypical ASM that shows variable levels of methylation, predominantly on the maternal allele in many tissues. Here we show that the nc886 interval is the first example of a polymorphic imprinted DMR in humans. Further analysis of the region suggests that the interval subjected to ASM is approximately 2 kb in size and somatically acquired. An in depth analysis of this region in primary cancer samples with matching normal adjacent tissue from the Cancer Genome Atlas revealed that aberrant methylation in bladder, breast, colon and lung tumors occurred in approximately 27% of cases. Hypermethylation occurred more frequently than hypomethylation. Using additional normal-tumor paired samples we show that on rare occasions the aberrant methylation profile is due to loss-of-heterozygosity. This work therefore suggests that the nc886 locus is subject to variable allelic methylation that undergoes cancer-associated epigenetic changes in solid tumors.

Elucidating how and to what extent CpG islands (CGIs) are methylated in germ cells is essential to understand genomic imprinting and epigenetic reprogramming. Here we present, to our knowledge, the first integrated epigenomic analysis of mammalian oocytes, identifying over a thousand CGIs methylated in mature oocytes. We show that these CGIs depend on DNMT3A and DNMT3L but are not distinct at the sequence level, including in CpG periodicity. They are preferentially located within active transcription units and are relatively depleted in H3K4me3, supporting a general transcription-dependent mechanism of methylation. Very few methylated CGIs are fully protected from post-fertilization reprogramming but, notably, the majority show incomplete demethylation in embryonic day (E) 3.5 blastocysts. Our study shows that CGI methylation in gametes is not entirely related to genomic imprinting but is a strong factor in determining methylation status in preimplantation embryos, suggesting a need to reassess mechanisms of post-fertilization demethylation.

Genome-wide dynamic changes in DNA methylation are indispensable for germline development and genomic imprinting in mammals. Here, we report single-base resolution DNA methylome and transcriptome maps of mouse germ cells, generated using whole-genome shotgun bisulfite sequencing and cDNA sequencing (mRNA-seq). Oocyte genomes showed a significant positive correlation between mRNA transcript levels and methylation of the transcribed region. Sperm genomes had nearly complete coverage of methylation, except in the CpG-rich regions, and showed a significant negative correlation between gene expression and promoter methylation. Thus, these methylome maps revealed that oocytes and sperms are widely different in the extent and distribution of DNA methylation. Furthermore, a comparison of oocyte and sperm methylomes identified more than 1,600 CpG islands differentially methylated in oocytes and sperm (germline differentially methylated regions, gDMRs), in addition to the known imprinting control regions (ICRs). About half of these differentially methylated DNA sequences appear to be at least partially resistant to the global DNA demethylation that occurs during preimplantation development. In the absence of Dnmt3L, neither methylation of most oocyte-methylated gDMRs nor intragenic methylation was observed. There was also genome-wide hypomethylation, and partial methylation at particular retrotransposons, while maintaining global gene expression, in oocytes. Along with the identification of the many Dnmt3L-dependent gDMRs at intragenic regions, the present results suggest that oocyte methylation can be divided into 2 types: Dnmt3L-dependent methylation, which is required for maternal methylation imprinting, and Dnmt3L-independent methylation, which might be essential for endogenous retroviral DNA silencing. The present data provide entirely new perspectives on the evaluation of epigenetic markers in germline cells.

Imprinting disorders (IDs) show overlapping phenotypes, particularly in Silver-Russell syndrome (SRS), Temple syndrome (TS14), and Prader-Willi syndrome (PWS). These three IDs include fetal and postnatal growth failure, feeding difficulty, and muscular hypotonia as major clinical features. However, the mechanism that causes overlapping phenotypes has not been clarified. To investigate the presence or absence of methylation signatures associated with overlapping phenotypes, we performed genome-wide methylation analysis (GWMA).

Genomic imprinting is a complex epigenetic mechanism of transcriptional control that utilizes DNA methylation and histone modifications to bring about parent-of-origin specific monoallelic expression in mammals. Genes subject to imprinting are often organised in clusters associated with large non-coding RNAs (ncRNAs), some of which have cis-regulatory functions. Here we have undertaken a detailed allelic expression analysis of an imprinted domain on mouse proximal chromosome 10 comprising the paternally expressed Plagl1 gene. We identified three novel Plagl1 transcripts, only one of which contains protein-coding exons. In addition, we characterised two unspliced ncRNAs, Hymai, the mouse orthologue of HYMAI, and Plagl1it (Plagl1 intronic transcript), a transcript located in intron 5 of Plagl1. Imprinted expression of these novel ncRNAs requires DNMT3L-mediated maternal DNA methylation, which is also indispensable for establishing the correct chromatin profile at the Plagl1 DMR. Significantly, the two ncRNAs are retained in the nucleus, consistent with a potential regulatory function at the imprinted domain. Analysis with catRAPID, a protein-ncRNA association prediction algorithm, suggests that Hymai and Plagl1it RNAs both have potentially high affinity for Trithorax chromatin regulators. The two ncRNAs could therefore help to protect the paternal allele from DNA methylation by attracting Trithorax proteins that mediate H3 lysine-4 methylation.

It is only partially understood how constitutive allelic methylation at imprinting control regions (ICRs) interacts with other regulation levels to drive timely parental allele-specific expression along large imprinted domains. The Peg13-Kcnk9 domain is an imprinted domain with important brain functions. To gain insights into its regulation during neural commitment, we performed an integrative analysis of its allele-specific epigenetic, transcriptomic, and cis-spatial organization using a mouse stem cell-based corticogenesis model that recapitulates the control of imprinted gene expression during neurodevelopment. We found that, despite an allelic higher-order chromatin structure associated with the paternally CTCF-bound Peg13 ICR, enhancer-Kcnk9 promoter contacts occurred on both alleles, although they were productive only on the maternal allele. This observation challenges the canonical model in which CTCF binding isolates the enhancer and its target gene on either side and suggests a more nuanced role for allelic CTCF binding at some ICRs.

Maintaining a single active X-chromosome by repressing Xist is crucial for embryonic development in mice. Although the Xist activator RNF12/RLIM is present as a maternal factor, maternal Xist (Xm-Xist) is repressed during preimplantation phases to establish imprinted X-chromosome inactivation (XCI). Here we show, using a highly reproducible chromatin immunoprecipitation method that facilitates chromatin analysis of preimplantation embryos, that H3K9me3 is enriched at the Xist promoter region, preventing Xm-Xist activation by RNF12. The high levels of H3K9me3 at the Xist promoter region are lost in embryonic stem (ES) cells, and ES-cloned embryos show RNF12-dependent Xist expression. Moreover, lack of Xm-XCI in the trophectoderm, rather than loss of paternally expressed imprinted genes, is the primary cause of embryonic lethality in 70-80% of parthenogenotes immediately after implantation. This study reveals that H3K9me3 is involved in the imprinting that silences Xm-Xist. Our findings highlight the role of maternal-specific H3K9me3 modification in embryo development.