Tissue-specific regulation of Igf2r/Airn imprinting during gastrulation
© Marcho et al.; licensee BioMed Central. 2015
Received: 21 November 2014
Accepted: 13 February 2015
Published: 14 March 2015
Appropriate epigenetic regulation of gene expression during lineage allocation and tissue differentiation is required for normal development. One example is genomic imprinting, which is defined as parent-of-origin mono-allelic gene expression. Imprinting is established largely due to epigenetic differences arriving in the zygote from sperm and egg haploid genomes. In the mouse, there are approximately 150 known imprinted genes, many of which occur in imprinted gene clusters that are regulated together. One imprinted cluster includes the maternally expressed Igf2r, Slc22a2, and Slc22a3 genes and the paternally expressed long non-coding RNA (lncRNA) Airn. Although it is known that Igf2r and Airn are reciprocally imprinted, the timing of imprinted expression and accompanying epigenetic changes have not been well characterized in vivo.
Here we show lineage- and temporal-specific regulation of DNA methylation and histone modifications at the Igf2r/Airn locus correlating with differential establishment of imprinted expression during gastrulation. Our results show that Igf2r is expressed from both alleles in the E6.5 epiblast. After gastrulation commences, the locus becomes imprinted in the embryonic lineage with the lncRNA Airn expressed from the paternal allele and Igf2r restricted to maternal allele expression. We document differentially enriched allele-specific histone modifications in extraembryonic and embryonic tissues. We also document for the first time allele-specific spreading of DNA methylation during gastrulation concurrent with establishment of imprinted expression of Igf2r. Importantly, we show that imprinted expression does not change in the extraembryonic lineage even though maternal DMR2 methylation spreading does occur, suggesting distinct mechanisms at play in embryonic and extraembryonic lineages.
These results indicate that similar to preimplantation, gastrulation represents a window of dynamic lineage-specific epigenetic regulation in vivo.
Genomic imprinting is an epigenetic phenomenon that results in mono-allelic gene expression in a parent-of-origin manner. Imprinted expression has been identified at approximately 150 mouse genes, which often occurs in clusters containing multiple imprinted transcripts [1,2]. Expression of imprinted genes is thought to be established in cis by allele-specific DNA methylation established at imprinting control regions (ICRs) in the gametes, thus arriving in the zygote as maternal and paternal specific information. A regulatory theme has emerged at many imprinted clusters in which a single long non-coding RNA (lncRNA) is thought to repressively regulate genes in cis through direct transcriptional blocking and/or recruitment of repressive chromatin remodeling complexes such as G9a and PRC2, resulting in differential allele-specific histone modifications [3,4].
One cluster on mouse chromosome 17 includes the maternally expressed Igf2r, Slc22a2, and Slc22a3 genes and the paternally expressed lncRNA Airn , and several non-imprinted genes (Slc22a1, Mas, and Plg). The Airn promoter lies in the second intron of Igf2r, and Airn transcription occurs from the opposite strand overlapping Igf2r exons 1 and 2 [5-7]. Paternal Airn expression may participate in imprinting of the maternally expressed genes by blocking access of the transcriptional machinery to the Igf2r start site , and transcription of Airn has been shown to be required for silencing of Igf2r [8,9]. Paternal allele silencing of the other imprinted genes in the cluster only occurs in extraembryonic lineages and may be a result of Airn recruitment of repressive complexes such as G9a to their promoters . Biallelic expression of Igf2r is observed in ES cells and only becomes imprinted upon differentiation in vitro . Although the expression of Igf2r and Airn has been documented in preimplantation and late stage embryos [10-12], lineage-specific expression dynamics have not been observed during gastrulation. Recent studies have focused on mechanisms in ES cell models [4,8,13], but the precise timing and mechanisms responsible for imprinting at Igf2r/Airn in vivo remain unknown. Here we characterize tissue-specific dynamics of expression and epigenetic modifications that occur at Igf2r/Airn during normal gastrulation. We show that significant epigenetic regulation occurs at imprinted loci during epiblast differentiation in vivo.
Results and discussion
Imprinted expression of Igf2r and Airn during gastrulation
To further understand the imprinted expression of Igf2r and Airn during gastrulation, we carried out allele-specific expression analysis of C57BL/6JxPWD/PhJ-F1 and C57BL/6J-Chr 17PWD/Ph/ForeJxC57BL/6J-F1 embryos (hereafter referred to as B × P and P × B F1 embryos, respectively). Single-strand confirmation polymorphism (SSCP) revealed that Igf2r is expressed from both alleles in the EPI of E6.5 embryos (Figure 1C, red box). In E6.5 VE, Igf2r is maternally expressed and paternally imprinted (Figure 1C). At E7.5, Igf2r is imprinted in both tissues (Figure 1C). Our results show that in the multipotent epiblast, Igf2r is expressed from both alleles, but once embryonic cells have adopted defined lineages at E7.5, Igf2r expression becomes imprinted. This correlation suggests a relationship between relative differentiation state in vivo and imprinted expression at the locus - consistent with ES cell models.
Since Airn is thought to establish imprinting of Igf2r , we also examined allele-specific Airn expression. In E6.5 EPI, Airn is not expressed (Figure 1B,D, red box), corresponding with biallelic Igf2r expression (Figure 1C). In the VE at E6.5, where Igf2r is imprinted, we observe reciprocal imprinting (paternal expression) of Airn (Figure 1E). At E7.5, Igf2r and Airn are imprinted in both embryonic and extraembryonic tissue (Figure 1E). This change in imprinted expression between EPI and EM also occurs in the reciprocal cross (P × B, Figure 1C,D,E), ruling out background-specific genetic differences. Airn has also been shown to regulate imprinting of Slc22a2 and Slc22a3 in extraembryonic lineages ; however, we could not detect these transcripts at appreciable levels during gastrulation (Figure 1B). The change in Igf2r and Airn expression indicate a lineage- and stage-specific establishment of imprinted expression during normal development. We therefore examined allele-specific epigenetic modifications at the locus.
DNA methylation spreads at DMR 2
DMR2 methylation has been shown to be present in oocytes , defining DMR2 as an ICR. Previous reports documented DMR2 by methylation-sensitive restriction enzymes, presenting the analysis of two specific CpG dinucleotides [13,17]. To gain a more comprehensive understanding of the methylation status, we designed two overlapping PCR amplicons for bisulfite sequencing at DMR2 (Figure 2A). With this approach, we fortuitously identified the precise 3′ boundary of ICR methylation present in oocytes [CpG at Chr17:12,742,488-12,742,489 (Figure 2A, red asterisk)]. In E6.5 EPI and VE, the precise ICR border was maintained on the maternal allele (Figure 2B). However, by E7.5 DNA methylation had spread in the 3′ direction in both embryonic and extraembryonic tissues (Figure 2B). These results indicate that although ICR methylation at DMR2 is established in the female germline , maternal allele-specific methylation increases/spreads in cells of all lineages coincident with the onset of gastrulation. It is particularly intriguing that the methylation spreading occurs in the extraembryonic tissue given that reciprocal imprinting of Igf2r and Airn is already established. It is also evident that the increase in DNA methylation is coincident with initiation of Airn expression in the epiblast, suggesting a tissue-specific mechanistic relationship. It may be of interest in the future to determine how far DNA methylation continues, if the spreading also occurs in the 5′ direction, and if the spreading is required for paternal silencing of Igf2r and activation of Airn.
Airn is progressively expressed during development
Since the Airn transcript is 108 kb, we designed amplicons along its length to assess if the entire lncRNA is detectable in embryonic lineages at various developmental stages (Figure 3E). Qualitative RT-PCR indicates that Airn transcripts increase in length with developmental progress (Figure 3F). At E7.0, only the 5′-most amplicons are detected while the 3′-most amplicons are also detected in older embryos (Figure 3F). By E9.5, all but the very 3’-most amplicon is detected, and the entire lncRNA is detectable in adult tissue (Figure 3F, +). Although qualitative, these results also suggest that total levels of Airn transcripts increase as development proceeds (Figure 3F, compare to ActB control).
These findings raise the possibility that maternal DNA methylation spreading is required to inhibit maternal Airn transcription. This could explain activation of only the paternal unmethylated allele in the embryo. Furthermore, the difference in Igf2r and Airn expression between EPI (biallelic Igf2r and no Airn) and VE (reciprocal imprinting) indicate a distinct mechanism during epiblast differentiation that activates Airn (since Airn is already expressed in VE but the methylation dynamics are the same in EPI-EM and VE-EX). Alternatively, there may be regulation on the paternal allele that initially inhibits Airn transcription in epiblast. Either scenario indicates that neither DNA methylation nor Airn expression is responsible for silencing paternal Igf2r in the epiblast. Taken together, the observation that DNA methylation dynamics are the same in embryonic and extraembryonic tissues but that allelic expression patterns are different, require an embryonic lineage-specific mechanism responsible for establishment of imprinted expression.
Ctcf binding at DMR2
Ctcf chromatin immunoprecipitation (ChIP) indicates Ctcf binds to DMR2 in both embryonic and extraembryonic E8.5 tissues (Figure 4B). Sequencing of ChIP-PCR products clearly shows allele-specific binding of Ctcf at DMR2 to the unmethylated paternal allele in extraembryonic tissues (Figure 4B). Surprisingly, both alleles are bound by Ctcf in embryonic lineage - although there is a detectable shift toward the paternal allele (compare input and ChIP, Figure 4B). Together, the lack of Ctcf in the epiblast at E6.5 and the biallelic binding of Ctcf at E8.5 suggest that while Ctcf may play a role in maintaining imprinted expression at later stages, it is not involved in the initiation of paternal Igf2r silencing or Airn activation in the embryonic lineage during gastrulation.
Differential histone enrichment at DMR2
Allele-specific histone modifications (HMODs) have been shown to correlate with Igf2r imprinting in the central nervous system [12,22]. We therefore performed ChIP to examine enrichment of H3K4me3, H3K9me3, and H3K27me3 at DMR1 and DMR2 (Figure 4C). In the embryonic tissue at DMR1, we observe maternal allele-specific enrichment of H3K4me3, as well as paternally biased H3K9me3 and H3K27me3. In the extraembryonic tissues, there is no allele-specific enrichment of H3K4me3 and a weak paternal bias of H3K9me3 and H3K27me3 (Figure 4D).
Similar to binding of Ctcf, we observe allele-specific enrichment of HMODs at DMR2 in extraembryonic tissues but not in chromatin derived from the embryo (Figure 4E). In extraembryonic tissues, the active H3K4me3 mark is greatly enriched on the paternal allele (which expresses Airn), and H3K9me3 is enriched on the maternal allele (where Airn is silent, Figure 4E). Surprisingly, PRC2-mediated H3K27me3 which has been shown to be required for imprinting at other loci  is not enriched on the silent Igf2r allele, suggesting that PRC2 does not participate in regulation of the Airn locus in extraembryonic cells. In the embryonic tissue however, we find maternal bias of repressive H3K9me3 and H3K27me3 (although not as highly enriched as extraembryonic cells).
Together these data indicate distinct lineage- and allele-specific enrichment of HMODs occur at DMR1 and DMR2. Strikingly, there is limited allele-specific enrichment in the extraembryonic tissue at DMR1 at E8.5, even though Igf2r is imprinted at least 2 days prior. This indicates differential methylation (Additional file 1: Figure S1), and HMODs at DMR1 (Figure 4D) may play a secondary role in the imprinting of Igf2r.
Dramatic allele-specific binding/enrichment of Ctcf, H3K4me3, and H3K9me3 is present at DMR2 in extraembryonic tissues suggesting that these chromatin modifications are established at an earlier stage. Although biased, DMR2 allele-specific chromatin modifications are not fully established in embryonic lineages by E8.5. While it is possible that multiple cell types of the E8.5 embryo contain distinct allele-specific enrichment, it is more likely that the allele-specific modifications are not yet fully established - particularly since imprinted expression of Igf2r and Airn is initiated only 24 h prior at E7.5.
All procedures were approved by the University of Massachusetts Amherst Institutional Animal Care and Use Committee. Embryos were derived from C57BL/6J (JAX 000664) and PWD/PhJ (JAX004660). Reciprocal F1 embryos were derived from female C75BL6/J Chr17PWD/PhJ/ForeJ (JAX 005267) and C57BL/6J (JAX 000664) males. Embryos were microdissected for DNA and mRNA extraction. MII oocytes were collected from superovulated B6D2F1 females to confirm ICR methylation.
Imprinted expression analysis
Total RNA was isolated using the Roche High Pure RNA Isolation Kit (Roche 11828665001, Roche, Basel, Switzerland). cDNA synthesis was performed using Bio-Rad iScript cDNA synthesis kit (170-8891) (Bio-Rad Laboratories, Inc., Hercules, USA). Primers for allele-specific expression and full-length Airn RT-PCR are shown in Additional file 3: Table S1. Airn restriction fragment length polymorphism (RFLP) was performed with AvaI. SSCP was performed on Igf2r PCR products with MDE polyacrylamide gel electrophoresis (Lonza 50620, Lonza Group, Basel, Switzerland). PCR products were visualized by ethidium bromide illumination and imaging.
E8.5 C75BL/6JxPWD/PhJ-F1 embryos were dissected, and embryonic and extraembryonic tissues were separated and immediately processed using instructions in either ChIP-IT High Sensitivity kit (Active Motif 53040, Active Motif, Carlsbad, USA) or Zymo-Spin ChIP kit (Zymo D5210, Zymo Research, Irvine, USA). Samples were kept on ice and either sonicated twice for 20 s with the Heat Systems Sonicator/Ultra Processor (output 3) or sonicated for 30 s on/20 s off for 3 min using a cup horn adaptor for the QSonica A500 (QSonica, Newtown, USA). After sonication, 1% of each sample was removed for input control. Immunoprecipitation was carried out using Active Motif Protein G agarose beads or magnetic Protein G Dynabeads (10003D, Life Technologies, Carlsbad, USA) and either anti-Ctcf (Santa Cruz sc-28198, Santa Cruz Biotechnology, Inc., Dallas, USA), anti-H3K4me3 (Abcam ab8580, Abcam, Cambridge, UK), anti-H3K9me3 (Abcam ab8898), or H3K27me3 (Millipore 07-449, Millipore, Billerica, USA) along with normal rabbit IgG. After antibody incubation, beads were washed and DNA was collected using manufacturer’s protocol. ChIP-PCR primers found in Additional file 3: Table S1 (Additional file 4: Supplemental Methods).
differentially methylated region
- ES cells:
embryonic stem cells
histone H3 lysine 27 tri-methylation
histone H3 lysine 4 tri-methylation
histone H3 lysine 9 tri-methylation
inner cell mass
imprinting control region
long non-coding RNA
Polycomb Repressive Complex 2
restriction fragment length polymorphism
single-strand conformation polymorphism
CM was supported by ICE-NSF-IGERT (Grant DGE-0654128). The work was supported in part by NSF RCN grant 1049849 and March of Dimes Grant 6-FY11-367 to JM.
- Barlow DP. Genomic imprinting: a mammalian epigenetic discovery model. Annu Rev Genet. 2011;45:379–403.View ArticlePubMedGoogle Scholar
- Bartolomei MS. Genomic imprinting: employing and avoiding epigenetic processes. Genes Dev. 2009;23:2124–33.View ArticlePubMed CentralPubMedGoogle Scholar
- Lindroth AM, Park YJ, McLean CM, Dokshin GA, Persson JM, Herman H, et al. Antagonism between DNA and H3K27 methylation at the imprinted Rasgrf1 locus. PLoS Genet. 2008;4:e1000145.View ArticlePubMed CentralPubMedGoogle Scholar
- Nagano T, Mitchell JA, Sanz LA, Pauler FM, Ferguson-Smith AC, Feil R, et al. The Air noncoding RNA epigenetically silences transcription by targeting G9a to chromatin. Science. 2008;322:1717–20.View ArticlePubMedGoogle Scholar
- Zwart R, Sleutels F, Wutz A, Schinkel AH, Barlow DP. Bidirectional action of the Igf2r imprint control element on upstream and downstream imprinted genes. Genes Dev. 2001;15:2361–6.View ArticlePubMed CentralPubMedGoogle Scholar
- Wutz A, Barlow DP. Imprinting of the mouse Igf2r gene depends on an intronic CpG island. Mol Cell Endocrinol. 1998;140:9–14.View ArticlePubMedGoogle Scholar
- Sleutels F, Barlow DP. Investigation of elements sufficient to imprint the mouse Air promoter. Mol Cell Biol. 2001;21:5008–17.View ArticlePubMed CentralPubMedGoogle Scholar
- Latos PA, Pauler FM, Koerner MV, Senergin HB, Hudson QJ, Stocsits RR, et al. Airn transcriptional overlap, but not its lncRNA products, induces imprinted Igf2r silencing. Science. 2012;338:1469–72.View ArticlePubMedGoogle Scholar
- Santoro F, Mayer D, Klement RM, Warczok KE, Stukalov A, Barlow DP, et al. Imprinted Igf2r silencing depends on continuous Airn lncRNA expression and is not restricted to a developmental window. Development. 2013;140:1184–95.View ArticlePubMedGoogle Scholar
- Lerchner W, Barlow DP. Paternal repression of the imprinted mouse Igf2r locus occurs during implantation and is stable in all tissues of the post-implantation mouse embryo. Mech Dev. 1997;61:141–9.View ArticlePubMedGoogle Scholar
- Szabo PE, Mann JR. Allele-specific expression and total expression levels of imprinted genes during early mouse development: implications for imprinting mechanisms. Genes Dev. 1995;9:3097–108.View ArticlePubMedGoogle Scholar
- Yamasaki Y, Kayashima T, Soejima H, Kinoshita A, Yoshiura K, Matsumoto N, et al. Neuron-specific relaxation of Igf2r imprinting is associated with neuron-specific histone modifications and lack of its antisense transcript Air. Hum Mol Genet. 2005;14:2511–20.View ArticlePubMedGoogle Scholar
- Latos PA, Stricker SH, Steenpass L, Pauler FM, Huang R, Senergin BH, et al. An in vitro ES cell imprinting model shows that imprinted expression of the Igf2r gene arises from an allele-specific expression bias. Development. 2009;136:437–48.View ArticlePubMed CentralPubMedGoogle Scholar
- Li E, Beard C, Jaenisch R. Role for DNA methylation in genomic imprinting. Nature. 1993;366:362–5.View ArticlePubMedGoogle Scholar
- Wutz A, Smrzka OW, Schweifer N, Schellander K, Wagner EF, Barlow DP. Imprinted expression of the Igf2r gene depends on an intronic CpG island. Nature. 1997;389:745–9.View ArticlePubMedGoogle Scholar
- Thorvaldsen JL, Duran KL, Bartolomei MS. Deletion of the H19 differentially methylated domain results in loss of imprinted expression of H19 and Igf2. Genes Dev. 1998;12:3693–702.View ArticlePubMed CentralPubMedGoogle Scholar
- Stoger R, Kubicka P, Liu CG, Kafri T, Razin A, Cedar H, et al. Maternal-specific methylation of the imprinted mouse Igf2r locus identifies the expressed locus as carrying the imprinting signal. Cell. 1993;73:61–71.View ArticlePubMedGoogle Scholar
- Murrell A, Heeson S, Reik W. Interaction between differentially methylated regions partitions the imprinted genes Igf2 and H19 into parent-specific chromatin loops. Nat Genet. 2004;36:889–93.View ArticlePubMedGoogle Scholar
- Engel N, Thorvaldsen JL, Bartolomei MS. CTCF binding sites promote transcription initiation and prevent DNA methylation on the maternal allele at the imprinted H19/Igf2 locus. Hum Mol Genet. 2006;15:2945–54.View ArticlePubMedGoogle Scholar
- Zhang H, Niu B, Hu JF, Ge S, Wang H, Li T, et al. Interruption of intrachromosomal looping by CCCTC binding factor decoy proteins abrogates genomic imprinting of human insulin-like growth factor II. J Cell Biol. 2011;193:475–87.View ArticlePubMed CentralPubMedGoogle Scholar
- Pant V, Mariano P, Kanduri C, Mattsson A, Lobanenkov V, Heuchel R, et al. The nucleotides responsible for the direct physical contact between the chromatin insulator protein CTCF and the H19 imprinting control region manifest parent of origin-specific long-distance insulation and methylation-free domains. Genes Dev. 2003;17:586–90.View ArticlePubMed CentralPubMedGoogle Scholar
- Fournier C, Goto Y, Ballestar E, Delaval K, Hever AM, Esteller M, et al. Allele-specific histone lysine methylation marks regulatory regions at imprinted mouse genes. Embo J. 2002;21:6560–70.View ArticlePubMed CentralPubMedGoogle Scholar
- Mager J, Montgomery ND, de Villena FP, Magnuson T. Genome imprinting regulated by the mouse Polycomb group protein Eed. Nat Genet. 2003;33:502–7.View ArticlePubMedGoogle Scholar
- Yang L, Froberg JE, Lee JT. Long noncoding RNAs: fresh perspectives into the RNA world. Trends Biochem Sci. 2014;39:35–43.View ArticlePubMed CentralPubMedGoogle Scholar
- Zhang K, Haversat JM, Mager J. CTR9/PAF1c regulates molecular lineage identity, histone H3K36 trimethylation and genomic imprinting during preimplantation development. Dev Biol. 2013;383:15–27.View ArticlePubMedGoogle Scholar
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.