Logo: to the web site of Uppsala University

Epigenetic marks at cis acting imprinting control regions (ICRs) regulate parent of origin-specific expression of multiple genes in imprinted gene clusters. Epigenetic marks are acquired during gametogenesis and maintained faithfully thereafter. However, the mechanism by which differential epigenetic marks are established and maintained at ICRs is currently unclear. By using Kcnq1 ICR as a model system, we have investigated the functional role of genetic signatures in the acquisition and maintenance of epigenetic marks. Kcnq1 ICR is methylated on the maternal chromosome but remains unmethylated on the paternal chromosome. Here, we show that a paternal allele of Kcnq1 ICR lacking the Kcnq1 ot 1 promoter remains unmethylated during spermatogenesis; however, it becomes methylated specifically during pre-implantation development. Analysis of the chromatin structure at the paternal ICR in spermatogenic cells and in E13.5 embryonic tissues revealed that the ICRs of both wild type and mutant mice are enriched with H3K4me2 in spermatiogenic cells of the testicular compartment, but the mutant ICR lost H3K4me2 specifically in epididymal sperm and an increase in repressive marks was observed in embryonic tissues. Interestingly, we also detected a decrease in nucleosomal histone levels at the mutant ICR in comparison to the wild-type ICR in epididymal sperm. Taken together, these observations suggest that the Kcnq1 lot1 promoter plays a critical role in establishing an epigenetic memory in the male germline by ensuring that the paternal allele remains in an unmethylated state during pre-implantation development.

Establishment of silencing by noncoding RNAs (ncRNAs) via targeting of chromatin remodelers is relatively well investigated; however, their role in the maintenance of silencing is poorly understood. Here, we explored the functional role of the long ncRNA Kcnq1ot1 in the maintenance of transcriptional gene silencing in the one mega-base Kcnq1 imprinted domain in a transgenic mouse model. By conditionally deleting the Kcnq1ot1 ncRNA at different stages of mouse development, we suggest that Kcnq1ot1 ncRNA is required for the maintenance of the silencing of ubiquitously imprinted genes (UIGs) at all developmental stages. In addition, Kcnq1ot1 ncRNA is also involved in guiding and maintaining the CpG methylation at somatic differentially methylated regions flanking the UIGs, which is a hitherto unknown role for a long ncRNA. On the other hand, silencing of some of the placental-specific imprinted genes (PIGs) is maintained independently of Kcnq1ot1 ncRNA. Interestingly, the non-imprinted genes (NIGs) that escape RNA-mediated silencing are enriched with enhancer-specific modifications. Taken together, this study illustrates the gene-specific maintenance mechanisms operational at the Kcnq1 locus for tissue-specific transcriptional gene silencing and activation.

A cluster of imprinted genes at chromosome 11p15.5 is associated with the growth disorders, Silver-Russell syndrome (SRS) and Beckwith-Wiedemann syndrome (BWS). The cluster is divided into two domains with independent imprinting control regions (ICRs). We describe two maternal 11p15.5 microduplications with contrasting phenotypes. The first is an inverted and in cis duplication of the entire 11p15.5 cluster associated with the maintenance of genomic imprinting and with the SRS phenotype. The second is a 160 kb duplication also inverted and in cis, but resulting in the imprinting alteration of the centromeric domain. It includes the centromeric ICR (ICR2) and the most 5' 20 kb of the non-coding KCNQ1OT1 gene. Its maternal transmission is associated with ICR2 hypomethylation and the BWS phenotype. By excluding epigenetic mosaicism, cell clones analysis indicated that the two closely located ICR2 sequences resulting from the 160 kb duplication carried discordant DNA methylation on the maternal chromosome and supported the hypothesis that the ICR2 sequence is not sufficient for establishing imprinted methylation and some other property, possibly orientation-dependent, is needed. Furthermore, the 1.2 Mb duplication demonstrated that all features are present for correct imprinting at ICR2 when this is duplicated and inverted within the entire cluster. In the individuals maternally inheriting the 160 kb duplication, ICR2 hypomethylation led to the expression of a truncated KCNQ1OT1 transcript and to down-regulation of CDKN1C. We demonstrated by chromatin RNA immunopurification that the KCNQ1OT1 RNA interacts with chromatin through its most 5' 20 kb sequence, providing a mechanism likely mediating the silencing activity of this long non-coding RNA.

It is becoming increasingly evident that noncoding RNA (ncRNA) constitutes an important component of chromatin and that ncRNA has a critical role in organizing the chromatin architecture and epigenetic memory by acting as an interface with the chromatin modifying machinery. Xist is the only RNA that has been shown to regulate gene expression by modulating chromatin structure using a multilayered silencing pathway. Recent emerging evidence indicates that long ncRNAs such as Kcnq1ot1 and Air which map to the Kcnq1 and Igf2r imprinted gene clusters, respectively, mediate the transcriptional silencing of multiple genes by interacting with chromatin and recruiting the chromatin modifying machinery. Though there are some parallels in the mechanistic actions of Kcnq1ot1 and Air, they seem to differ greatly in the way they achieve the silencing of overlapping and nonoverlapping genes. By reviewing the latest available evidence, we propose that Kcnq1ot1 RNA itself seems to play a critical role in the bidirectional silencing of genes in the Kcnq1 domain, thus resembling the Xist RNA; whereas in the case of Air, the act of transcription plays a critical role in the silencing of the overlapping gene, whilst Air RNA itself mediates the silencing of nonoverlapping genes in a fashion similar to Kcnq1ot1 and Xist RNAs.

Genomic analyses have demonstrated that although less than 2% of the mammalian genome encodes proteins, at least two thirds is transcribed. Many nontranslated RNAs have now been characterized, and several long transcripts, ranging from 0.5 to over 100 kb, have been shown to regulate gene expression by modifying chromatin structure. Functions uncovered at a few well characterized loci demonstrate a wide diversity of mechanisms by which long noncoding RNAs can regulate chromatin over a single promoter, a gene cluster, or an entire chromosome, in order to activate or silence genes in cis or in trans. In reviewing the activities of these ncRNAs, we will look for common features in their interactions with the chromatin modifying machinery, and highlight new experimental approaches by which to address outstanding issues in ncRNA-dependent regulation of gene expression in development, disease and evolution.

The prevailing view has been that noncoding antisense RNA primarily regulates the transcriptional activity of its sense counterpart, however, a subset of long noncoding antisense RNAs, such as Kcnq1ot1 and Air, have been shown to regulate the transcriptional silencing of multiple genes spread over several hundred kilobases on either side of their promoters. It is however unknown how these long RNAs regulate the transcriptional silencing of multiple genes. Our recent work demonstrated through exploiting an episomal-based system that the Kcnq1ot1 RNA harbors a silencing domain at its 5' end which mediates transcriptional silencing through introducing repressive histone modifications, and by interacting with the chromatin and targeting the associated regions to the perinucleolar space. Based on our current findings as well as previous findings from other labs, we propose a model resembling that of Xist-mediated chromosomal silencing, wherein Kcnq1ot1 RNA initiates transcriptional silencing by coating the flanking chromatin regions and directing the heterochromatin machinery. Maintenance of silencing through subsequent cell divisions would then occur via the targeting of Kcnq1ot1-associated chromatin to the perinucleolar space.

Recent investigations have implicated long antisense noncoding RNAs in the epigenetic regulation of chromosomal domains. Here we show that Kcnq1ot1 is an RNA polymerase II-encoded, 91 kb-long, moderately stable nuclear transcript and that its stability is important for bidirectional silencing of genes in the Kcnq1 domain. Kcnq1ot1 interacts with chromatin and with the H3K9- and H3K27-specific histone methyltransferases G9a and the PRC2 complex in a lineage-specific manner. This interaction correlates with the presence of extended regions of chromatin enriched with H3K9me3 and H3K27me3 in the Kcnq1 domain in placenta, whereas fetal liver lacks both chromatin interactions and heterochromatin structures. In addition, the Kcnq1 domain is more often found in contact with the nucleolar compartment in placenta than in liver. Taken together, our data describe a mechanism whereby Kcnq1ot1 establishes lineage-specific transcriptional silencing patterns through recruitment of chromatin remodeling complexes and maintenance of these patterns through subsequent cell divisions occurs via targeting the associated regions to the perinucleolar compartment.

The Kcnq1ot1 antisense noncoding RNA has been implicated in long-range bidirectional silencing, but the underlying mechanisms remain enigmatic. Here we characterize a domain at the 5' end of the Kcnq1ot1 RNA that carries out transcriptional silencing of linked genes using an episomal vector system. The bidirectional silencing property of Kcnq1ot1 maps to a highly conserved repeat motif within the silencing domain, which directs transcriptional silencing by interaction with chromatin, resulting in histone H3 lysine 9 trimethylation. Intriguingly, the silencing domain is also required to target the episomal vector to the perinucleolar compartment during mid-S phase. Collectively, our data unfold a novel mechanism by which an antisense RNA mediates transcriptional gene silencing of chromosomal domains by targeting them to distinct nuclear compartments known to be rich in heterochromatic machinery.

Antisense transcription has been shown to be one of the hierarchies that control gene expression in eukaryotes. Recently, we have documented that the mouse Kcnq1 imprinting control region (ICR) harbors bidirectional silencing property, and this feature is linked to an antisense RNA, Kcnq1ot1. In this investigation, using genomic footprinting, we have identified three NF-Y transcription factor binding sites appearing in a methylation-sensitive manner in the Kcnq1ot1 promoter. By employing a dominant negative mutant to the NF-Y transcription factor, we have shown that the NF-Y transcription factor positively regulates antisense transcription. Selective mutation of the conserved nucleotides in the NF-Y binding sites resulted in the loss of antisense transcription. The loss of antisense transcription from the Kcnq1ot1 promoter coincides with an enrichment in the levels of deacetylation and methylation at the lysine 9 residue of histone H3 and DNA methylation at the CpG residues, implying a crucial role for the NF-Y transcription factor in organizing the parent of origin-specific chromatin conformation in the Kcnq1 ICR. Parallel to the loss of antisense transcription, the loss of silencing of the flanking reporter genes was observed, suggesting that NF-Y-mediated Kcnq1ot1 transcription is critical in the bidirectional silencing process of the Kcnq1 ICR. These data highlight the NF-Y transcription factor as a crucial regulator of antisense promoter-mediated bidirectional silencing and the parent of origin-specific epigenetic marks at the Kcnq1 ICR. More importantly, for the first time, we document that NF-Y is involved in maintaining the antisense promoter activity against strong silencing conditions.