Enhancer-driven chromatin interactions during development promote escape from silencing by a long non-coding RNA
© Korostowski et al; licensee BioMed Central Ltd. 2011
Received: 16 September 2011
Accepted: 15 November 2011
Published: 15 November 2011
Gene regulation in eukaryotes is a complex process entailing the establishment of transcriptionally silent chromatin domains interspersed with regions of active transcription. Imprinted domains consist of clusters of genes, some of which exhibit parent-of-origin dependent monoallelic expression, while others are biallelic. The Kcnq1 imprinted domain illustrates the complexities of long-range regulation that coexists with local exceptions. A paternally expressed repressive non-coding RNA, Kcnq1ot1, regulates a domain of up to 750 kb, encompassing 14 genes. We study how the Kcnq1 gene, initially silenced by Kcnq1ot1, undergoes tissue-specific escape from imprinting during development. Specifically, we uncover the role of chromosome conformation during these events.
We show that Kcnq1 transitions from monoallelic to biallelic expression during mid gestation in the developing heart. This transition is not associated with the loss of methylation on the Kcnq1 promoter. However, by exploiting chromosome conformation capture (3C) technology, we find tissue-specific and stage-specific chromatin loops between the Kcnq1 promoter and newly identified DNA regulatory elements. These regulatory elements showed in vitro activity in a luciferase assay and in vivo activity in transgenic embryos.
By exploring the spatial organization of the Kcnq1 locus, our results reveal a novel mechanism by which local activation of genes can override the regional silencing effects of non-coding RNAs.
Genomic imprinting is a transcriptional regulatory mechanism that results in parental-specific gene expression. Over the past two decades, many mechanistic insights have emerged from the study of such loci as the H19/Igf2 domain . However, most imprinted loci are much more complex and exhibit tissue-specific as well as stage-specific imprinting. Many significant questions remain concerning the regulatory mechanisms governing such extended domains. For example, how genes that are monoallelic can coexist interspersed with others that exhibit partial or full biallelic expression is still not understood. The prevalent hypotheses for the appearance of imprinting imply that, although advantageous in some respects, it imposed a burden on bystander genes that came under its influence. Thus, it is likely that mechanisms emerged to bypass the effects of allelic silencing.
Recent experiments have strongly suggested that the silencing mechanism of Kcnq1ot1 RNA involves a spreading activity in cis, recruitment of Polycomb group proteins and physical compaction of the whole domain [4–6]. However, the presence of the ncRNA is not uniform throughout the region , leaving open the question of how this relates to biallelic expression of some genes.
Genes such as Trpm5 are constitutively biallelic both in the embryo and placenta. The Kcnq1 gene has a more complex pattern: it is monoallelic and ubiquitously expressed in early embryos, but the paternal copy is activated in conjunction with the acquisition of a tissue-restricted expression pattern established by mid gestation, that is, in the heart, kidney and brain . Thus, the Kcnq1 domain illustrates the regulatory challenges that must be met in a complex imprinted domain.
How do genes such as Kcnq1 achieve tissue-specific escape from imprinting? One possible explanation is that the Kcnq1 promoter is exceptionally strong and that as tissue-specific factors become expressed and bind to it, the silencing effect of Kcnq1ot1 is overcome. Alternatively, tissue-specific enhancers that become active may override the effects of ncRNAs. An additional possibility is that boundary elements serving as barriers to the spread of the ncRNA may exist within the Kcnq1 gene. These proposed mechanisms are not mutually exclusive.
None of the regulatory elements that account for the complex patterns of gene expression in this region have been identified. We used an optimized approach for identifying novel regulatory DNA elements and to determine their role in promoting escape from silencing. We investigated the mechanism of Kcnq1 reactivation by determining the in vivo spatial organization of the Kcnq1 domain with chromosome conformation capture (3C) assays [8, 9]. We present evidence for the role of specific DNA interactions by chromatin looping as a mechanism for acquisition of tissue-specific expression. These contacts occur during the developmental window in which the paternal allele of Kcnq1 escapes silencing by Kcnq1ot1. To determine if the regions contacted predominantly by the Kcnq1 promoter are regulatory elements, we integrated comparative genomics data and publicly available genome-wide chromatin immunoprecipitation (ChIP) profiles and identified DNA sequences that are candidates for modulating gene activity. Candidates were tested in vitro and in vivo and several elements were identified as active transcriptional enhancers. Their role in overriding RNA-mediated repression is discussed.
Kcnq1 RNA transitions from monoallelic to biallelic expression during development
One possible explanation for escape of silencing by the paternal Kcnq1 allele is that expression is initiated from an alternative start site. However, we did not observe any alternative transcripts after the transition from monoallelic to biallelic expression, thus excluding this hypothesis (data not shown). We then speculated that, because the transcriptional start site (TSS) at exon 1 lies within a strong CpG island, it could be paternally methylated in the early embryo and become demethylated upon activation. To determine if the CpG island was methylated during spermatogenesis or early embryogenesis, we performed methylated DNA immunoprecipitation (MeDIP) on sperm and embryos at 7.5 days post coitum (dpc) (Figure 2D). No DNA methylation was detected on the paternal allele, suggesting that the promoter is protected from primary (gametic) and secondary (post fertilization) methylation, even though it is silent at these stages. Furthermore, as opposed to the secondary methylation of the upstream Cdkn1c gene orchestrated by Kcnq1ot1, the silencing effect of the ncRNA on Kcnq1 is not mediated by DNA methylation.
Kcnq1 exhibits tissue-specific changes in chromatin conformation
To see if this profile was tissue specific, we performed 3C in neonatal brain, a tissue in which Kcnq1 is highly expressed and also biallelic. The interaction profile observed in the brain was distinct from that of the heart. The Kcnq1 promoter strongly associated instead with fragment 2 (Figure 3B, blue line, peak at -12 kb) and showed no contact with fragment 12 in the intron. These data show that there are distinct three-dimensional conformations that regulate expression in each tissue and identify the contact regions as putative tissue-specific enhancers.
Onset of loop formation between the Kcnq1 promoter and putative enhancers correlates with the monoallelic to biallelic switch
Distinct developmental history of potential regulatory elements at the Kcnq1 locus
To expand the epigenetic profile of Kcnq1 to earlier developmental stages, we performed locus-wide 3C assays on mouse ES cells (Figure 5B, blue line). None of the heart-specific contacts between the Kcnq1 promoter and regions upstream were detected, but the interaction with fragment 11 in intron 1 was already present. Thus, 3C assays show an interaction between the Kcnq1 promoter and a region in intron 1 that is already apparent in early stages of embryogenesis and becomes progressively stronger as development ensues. Additionally, the Kcnq1 promoter associates with upstream elements concomitantly with the transition from monoallelic to biallelic expression. These upstream elements partner with the Kcnq1 promoter in a tissue-specific fashion.
Sequences identified by contact analysis have enhancer activity in vitro and in vivo
To determine whether these sequences had enhancer activity in vitro, we cloned the DNA fragments upstream of a luciferase reporter gene and tested enhancer activity by transient transfection into NIH-3T3 cells (Figure 6C). ECR 4, contained in fragment 4 that showed frequent colocalization with the Kcnq1 promoter in neonatal heart, had the most pronounced enhancer effect in vitro. ECR 2, which associated with the Kcnq1 promoter in the neonatal brain, did not show significant transcriptional activity in this assay, suggesting that fibroblasts may not contain the factors required for brain-specific activity.
We next cloned several candidate sequences into an enhancer reporter vector and transgenic mouse embryos were generated. Several independent embryos were assessed for LacZ activity at 9.5 and 13.5 dpc. Reproducible staining in hindbrain, forebrain and thoracic region was observed with ECR4 at 13.5 dpc (Figure 6D), but none at 9.5 dpc (data not shown), recapitulating the developmental expression pattern of Kcnq1. ECR2 directed expression to hindbrain and forebrain in the 13.5 dpc embryos.
Our data reveal a novel role for transcriptional enhancers, namely overcoming regional imprinting effects to permit escape from silencing and leading to biallelic expression. Thus, the evolution of enhancers in the mammalian genome to enable biallelic expression of some genes located near imprinted genes appears to be an essential adaptation to accommodate imprinting as a regulatory mechanism. By mapping the architecture adopted by a specific genomic region during development, we have found dynamic changes that counter the regional repressive effect of a long non-coding RNA, Kcnq1ot1. Specifically, escape from silencing of the Kcnq1 gene is accomplished by the activation of tissue-specific enhancers and involves local reorganization of higher order structure.
The biallelic expression of genes within the Kcnq1 domain is reminiscent of certain genes on the × chromosome that escape inactivation [18, 19], but differs in significant ways. Kcnq1 is initially silenced paternally but transitions to biallelic expression at a specific time point during development. In theory, tissue-specific enhancer activity could override silencing by Kcnq1ot1 by physically contacting the promoter and sequestering it from condensing factors recruited by the ncRNA. In fact, the data from the 3C assays that we present are highly supportive of such a model by showing that stage and tissue-specific conformations of chromatin bring the Kcnq1 promoter into physical contact with DNA sequences that are highly conserved and exhibit enhancer activity in vivo.
In fact, we could distinguish two types of contacts. Promoter interaction with an element in intron 1 (mm183) is already apparent in early stages of embryogenesis and becomes progressively stronger and broader as development ensues. This association foreshadows a later event in which the Kcnq1 promoter contacts tissue-specific upstream elements concomitantly with the reactivation of the paternal allele and the restriction of expression to specific organs. Indeed, previous studies on the chromatin marks in the region showed that, in contrast to other neighboring genes, the Kcnq1 promoter does not exhibit differential histone modifications associated with repression [20, 21], and our data indicate that it is also devoid of DNA methylation on the paternal allele. We speculate that the intragenic enhancer may be marked in early embryogenesis by pioneer factors that maintain an open, poised chromatin status  and prevent assembly of repressive factors associated with the spreading of Kcnq1ot1[4, 5].
In summary, we have explored the organization of the Kcnq1 region in three-dimensional nuclear space during development and have identified DNA sequence elements that achieve the dual purpose of reactivating a gene silenced by a non-coding RNA and directing tissue-specific expression. Our studies underscore the importance of providing a three-dimensional context to the information on DNA methylation and chromatin modifications in order to fully grasp the dynamics of gene regulation in different cell lineages during development.
This study was carried out on mice in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was approved by the Temple University Animal Care and Use Committee (Protocol 3294).
Heart tissues for total RNA extraction were collected at appropriate days of gestation using F1 hybrid mice from C57BL/6J × B6(CAST7) crosses from embryonic to neonatal stages. RNA was extracted using TRIzol Reagent (Invitrogen, Carlsbad, CA no. 15596-018) and following manufacturer's protocol for RNA extraction from tissues. All RNA samples were subjected to DNase treatment using Turbo DNA-free (Ambion, Austin, TX no. AM1907) using the rigorous DNase treatment protocol. Three to five biological samples were collected for each days of gestation.
Following the manufacturer's instructions, complementary DNA synthesis was performed on total RNA using SuperScript II Reverse Transcriptase (Invitrogen, Carlsbad CA, no. 18064-014). A reverse transcriptase negative control was used to ensure there was no DNA contamination.
Transcript levels of Kcnq1 and β-actin were analyzed on the ABI Prism 7000 system (ABI, Foster City, CA). Reactions were conducted using the SYBR Green PCR Master Mix (ABI, Foster City, CA no. 4309155). On a 96-well plate, a reduced reaction of 20 μl was used instead of the suggested 50 μl indicated in the manufacturer's protocol. This included 10 μl of the SYBR Green reaction mix, a final concentration of 25 μM of each primer, the cDNA template and water to bring the final volume up to 20 μl. The PCR was performed under the following conditions: an initial denaturing step for 10 min at 95°C, an amplification step for 45 cycles of 95°C for 20 s, 55°C for 30 s and 72°C for 30 s, the final elongation step was at 72°C for 2 min. The Kcnq1 transcript was detected using the following left and right primers: 5'-CAAAGACCGTGGCAGTAAC-3' and 5'-CCTTCATTGCTGGCTACAAC-3'. The Kcnq1 transcript was normalized to β-actin using the following left and right primers: 5'-TGTTACCAACTGGGACGACA-3' and 5'-CCATCACAATGCCTGTGGTA-3'. Each qRT-PCR reaction was performed in triplicate with three biological replicates along with a no template negative control. The CT value of the Kcnq1 transcript was normalized to the CT value of the β-actin transcript. The standard deviation for each ratio was determined and the error bars represent 1 SD away from the average ratio.
Allele-specific RT-PCR and quantification
The Kcnq1 transcript was amplified using Ruby Taq Master Mix (Affymetrix, Santa Clara, CA no. 71191) in a reduced 15 μl reaction using the following primers: 5'-CATCGGTGCCCGTCTGAACAGG-3' and 5'-TTGCTGGGTAGGAAGAGCTCAG-3'. PCR reactions were performed with experimental and control templates in parallel. Following the PCR, a restriction digest was performed with the Nla III (New England Biolabs, Ipswich, MA no. R0125) for 1 h at 37°C. PCR and digestion products were run on 7% polyacrylamide gels and quantified using the Kodak Gel Logic 2000 imaging system (Kodak, Rochester, NY). The relative paternal to maternal band intensities were calculated.
3C was performed as described previously with the following modifications: tissues were homogenized and crosslinked with 1% formaldehyde. Before restriction enzyme digestion, all samples were subjected to RNAse treatment with a combination of RNAse A and RNAse H at 37°C for 20 minutes. Restriction digests were carried out with Afl III/Nco I. Digestion efficiency at each restriction site was determined by comparing amplification with primers spanning the restriction site to amplification with primers immediately downstream on the digested template. No restriction bias was observed across the region assayed by 3C (data not shown). PCR efficiency of each combination of primers was assessed with a control template prepared from two bacterial artificial chromosomes (BACs) encompassing the Kcnq1ot1 region analyzed, as described previously. The linear range for each primer pair was determined by serial dilution. PCR reactions were performed with experimental and control templates in parallel and PCR products were run on 7% polyacrylamide gels and quantified using the Kodak Gel Logic 2000 imaging system (Kodak, Rochester, NY). Crosslinking frequencies were calculated from duplicate PCR analyses of three independent 3C preparations. Interaction frequencies for Kcnq1 and Kcnq1ot1 were normalized to a control interaction at the H19 locus to allow comparisons between different tissues and cell types. A full list of primers is available upon request from the authors.
Luciferase reporter assays
Enhancer candidates were amplified from mouse genomic DNA and subcloned into the pGL3-promoter vector (Promega, Madison, WI) upstream of the luciferase reporter. All clones were confirmed by sequencing. Constructs were screened by transfection of 100 ng pGL3 promoter vector and 10 ng pHRL basic vector/well (24-well plate) into NIH 3T3 cells with 1 μl Lipofectamine 2000/well (Invitrogen, Carlsbad, CA). Cell lysate was harvested the following day and firefly and Renilla luciferase activities were measured in 10 μl of each lysate using a Dual-Luciferase Reporter Assay System kit (Promega, Madison, WI) and firefly activity was normalized to Renilla. All transfections were performed in triplicate.
Transgenic embryo assays
Enhancer candidate constructs were amplified from mouse genomic DNA, subcloned into the Hsp68-LacZ vector and transgenic mouse embryos were generated and stained (Cyagen Biosciences, Sunvale, CA). Images were obtained with an Olympus MVX10 stereoscope (Olympus, Center Valley, PA), cropped and levels adjusted with Adobe Photoshop (Adobe, San Jose, CA).
DNA from pooled mouse 7.5 dpc embryos and from mouse sperm was sonicated to obtain fragments ranging from 200 to 1,000 bp. An aliquot of DNA was saved for an input positive control. Me-DIP was performed using the Methylamp Methylated DNA Capture Kit (Epigentek, Farmingdale, NY)) and following the manufacturer's protocol. The Kcnq1 CpG Island was amplified using the following: 5'-CTGAGGGCAGCACGGTCTAT-3' and 5'-CTCCTGAGTCTCTCTTGTCACAACT-3'. H19 and Kcnq1ot1 promoters were used to control the quality of the immunoprecipitated substrates.
We thank Natalie Sedlak and May Truong-Cao for technical support, and Keith Latham for comments on the manuscript. This work was supported by National Institutes of Health (grant K22CA140361-01).
- Bartolomei MS: Genomic imprinting: employing and avoiding epigenetic processes. Genes Dev. 2009, 23: 2124-2133. 10.1101/gad.1841409.PubMed CentralView ArticlePubMedGoogle Scholar
- Wan LB, Bartolomei MS: Regulation of imprinting in clusters: noncoding RNAs versus insulators. Adv Genet. 2008, 61: 207-223.View ArticlePubMedGoogle Scholar
- Smilinich NJ, Day CD, Fitzpatrick GV, Caldwell GM, Lossie AC, Cooper PR, Smallwood AC, Joyce JA, Schofield PN, Reik W, Nicholls RD, Weksberg R, Driscoll DJ, Maher ER, Shows TB, Higgins MJ: A maternally methylated CpG island in KvLQT1 is associated with an antisense paternal transcript and loss of imprinting in Beckwith-Wiedemann syndrome. Proc Natl Acad Sci USA. 1999, 96: 8064-8069. 10.1073/pnas.96.14.8064.PubMed CentralView ArticlePubMedGoogle Scholar
- Pandey RR, Mondal T, Mohammad F, Enroth S, Redrup L, Komorowski J, Nagano T, Mancini-Dinardo D, Kanduri C: Kcnq1ot1 antisense noncoding RNA mediates lineage-specific transcriptional silencing through chromatin-level regulation. Mol Cell. 2008, 32: 232-246. 10.1016/j.molcel.2008.08.022.View ArticlePubMedGoogle Scholar
- Terranova R, Yokobayashi S, Stadler MB, Otte AP, van Lohuizen M, Orkin SH, Peters AH: Polycomb group proteins Ezh2 and Rnf2 direct genomic contraction and imprinted repression in early mouse embryos. Dev Cell. 2008, 15: 668-679. 10.1016/j.devcel.2008.08.015.View ArticlePubMedGoogle Scholar
- Redrup L, Branco MR, Perdeaux ER, Krueger C, Lewis A, Santos F, Nagano T, Cobb BS, Fraser P, Reik W: The long noncoding RNA Kcnq1ot1 organises a lineage-specific nuclear domain for epigenetic gene silencing. Development. 2009, 136: 525-530. 10.1242/dev.031328.PubMed CentralView ArticlePubMedGoogle Scholar
- Gould TD, Pfeifer K: Imprinting of mouse Kvlqt1 is developmentally regulated. Hum Mol Genet. 1998, 7: 483-487. 10.1093/hmg/7.3.483.View ArticlePubMedGoogle Scholar
- Dekker J, Rippe K, Dekker M, Kleckner N: Capturing chromosome conformation. Science. 2002, 295: 1306-1311. 10.1126/science.1067799.View ArticlePubMedGoogle Scholar
- Tolhuis B, Palstra RJ, Splinter E, Grosveld F, de Laat W: Looping and interaction between hypersensitive sites in the active beta-globin locus. Mol Cell. 2002, 10: 1453-1465. 10.1016/S1097-2765(02)00781-5.View ArticlePubMedGoogle Scholar
- Bokil NJ, Baisden JM, Radford DJ, Summers KM: Molecular genetics of long QT syndrome. Mol Genet Metab. 2010, 101: 1-8. 10.1016/j.ymgme.2010.05.011.View ArticlePubMedGoogle Scholar
- Kadauke S, Blobel GA: Chromatin loops in gene regulation. Biochim Biophys Acta. 2009, 1789: 17-25.PubMed CentralView ArticlePubMedGoogle Scholar
- Tsai CL, Rowntree RK, Cohen DE, Lee JT: Higher order chromatin structure at the X-inactivation center via looping DNA. Dev Biol. 2008, 319: 416-425. 10.1016/j.ydbio.2008.04.010.PubMed CentralView ArticlePubMedGoogle Scholar
- Engel N, Raval AK, Thorvaldsen JL, Bartolomei SM: Three-dimensional conformation at the H19/Igf2 locus supports a model of enhancer tracking. Hum Mol Genet. 2008, 17: 3021-3029. 10.1093/hmg/ddn200.PubMed CentralView ArticlePubMedGoogle Scholar
- Visel A, Minovitsky S, Dubchak I, Pennacchio LA: VISTA Enhancer Browser--a database of tissue-specific human enhancers. Nucleic Acids Res. 2007, 35: D88-92. 10.1093/nar/gkl822.PubMed CentralView ArticlePubMedGoogle Scholar
- Ovcharenko I, Nobrega MA, Loots GG, Stubbs L: ECR Browser: a tool for visualizing and accessing data from comparisons of multiple vertebrate genomes. Nucleic Acids Res. 2004, 32: W280-286. 10.1093/nar/gkh355.PubMed CentralView ArticlePubMedGoogle Scholar
- Mikkelsen TS, Ku M, Jaffe DB, Issac B, Lieberman E, Giannoukos G, Alvarez P, Brockman W, Kim TK, Koche RP, Lee W, Mendenhall E, O'Donovan A, Presser A, Russ C, Xie X, Meissner A, Wernig M, Jaenisch R, Nusbaum C, Lander ES, Bernstein BE: Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature. 2007, 448: 553-560. 10.1038/nature06008.PubMed CentralView ArticlePubMedGoogle Scholar
- Visel A, Blow MJ, Li Z, Zhang T, Akiyama JA, Holt A, Plajzer-Frick I, Shoukry M, Wright C, Chen F, Afzal V, Ren B, Rubin EM, Pennacchio LA: ChIP-seq accurately predicts tissue-specific activity of enhancers. Nature. 2009, 457: 854-858. 10.1038/nature07730.PubMed CentralView ArticlePubMedGoogle Scholar
- Filippova GN, Cheng MK, Moore JM, Truong JP, Hu YJ, Nguyen DK, Tsuchiya KD, Disteche CM: Boundaries between chromosomal domains of × inactivation and escape bind CTCF and lack CpG methylation during early development. Dev Cell. 2005, 8: 31-42. 10.1016/j.devcel.2004.10.018.View ArticlePubMedGoogle Scholar
- Carrel L, Park C, Tyekucheva S, Dunn J, Chiaromonte F, Makova KD: Genomic environment predicts expression patterns on the human inactive × chromosome. PLoS Genet. 2006, 2: e151-10.1371/journal.pgen.0020151.PubMed CentralView ArticlePubMedGoogle Scholar
- Lewis A, Mitsuya K, Umlauf D, Smith P, Dean W, Walter J, Higgins M, Feil R, Reik W: Imprinting on distal chromosome 7 in the placenta involves repressive histone methylation independent of DNA methylation. Nat Genet. 2004, 36: 1291-1295. 10.1038/ng1468.View ArticlePubMedGoogle Scholar
- Umlauf D, Goto Y, Cao R, Cerqueira F, Wagschal A, Zhang Y, Feil R: Imprinting along the Kcnq1 domain on mouse chromosome 7 involves repressive histone methylation and recruitment of Polycomb group complexes. Nat Genet. 2004, 36: 1296-1300. 10.1038/ng1467.View ArticlePubMedGoogle Scholar
- Xu J, Watts JA, Pope SD, Gadue P, Kamps M, Plath K, Zaret KS, Smale ST: Transcriptional competence and the active marking of tissue-specific enhancers by defined transcription factors in embryonic and induced pluripotent stem cells. Genes Dev. 2009, 23: 2824-2838. 10.1101/gad.1861209.PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.