Blocked transcription through KvDMR1 results in absence of ...Blocked transcription through KvDMR1 results in absence of methylation and gene silencing resembling Beckwith-Wiedemann

RESEARCH ARTICLE

Blocked transcription through KvDMR1 results in absence ofmethylation and gene silencing resembling Beckwith-WiedemannsyndromeVir B. Singh1,*, Sirinapa Sribenja1, Kayla E. Wilson2, Kristopher M. Attwood3, Joanna C. Hillman1,Shilpa Pathak1,‡ and Michael J. Higgins1,§

ABSTRACTThe maternally methylated KvDMR1 ICR regulates imprintedexpression of a cluster of maternally expressed genes on humanchromosome 11p15.5. Disruption of imprinting leads to Beckwith-Wiedemann syndrome (BWS), an overgrowth and cancerpredisposition condition. In the majority of individuals with BWS,maternal-specific methylation at KvDMR1 is absent and genes underits control are repressed. We analyzed a mouse model carrying apoly(A) truncation cassette inserted to prevent RNA transcripts fromelongation through KvDMR1. Maternal inheritance of this mutationresulted in absence of DNA methylation at KvDMR1, which led tobiallelic expression of Kcnq1ot1 and suppression of maternallyexpressed genes. This study provides further evidence thattranscription is required for establishment of methylation at maternalgametic DMRs. More importantly, this mouse model recapitulates themolecular phenotypic characteristics of the most common form ofBWS, including loss of methylation at KvDMR1 and biallelicrepression of Cdkn1c, suggesting that deficiency of maternaltranscription through KvDMR1 may be an underlying cause ofsome BWS cases.

KEY WORDS: KvDMR1, Genomic imprinting, Imprinting controlregion, Kcnq1ot1, Methylation, Beckwith-Wiedemann syndrome,Transcription

INTRODUCTIONGenomic imprinting plays an essential role in mammaliandevelopment, regulating the preferential expression of certaingenes in a parent of origin-dependent manner. Imprinted genesare expressed exclusively or preferentially from only one of the twoparental alleles, tend to occur in clusters (imprinted domains), andare frequently involved in growth regulation and development, withpaternally expressed genes often coding for growth promoters andmaternally expressed genes often coding for growth suppressors(Barlow and Bartolomei, 2014; Bartolomei and Ferguson-Smith,2011; Li and Sasaki, 2011; Tycko and Morison, 2002). Monoallelic

expression of genes present in an imprinted domain is controlled bycis-acting short DNA sequences known as imprinting controlregions (ICRs) or imprinting centers (ICs) that can function overlong genomic distances (Spahn and Barlow, 2003; Verona et al.,2003). As first demonstrated for the H19/Igf2 (Thorvaldsen et al.,1998), the Igf2r/Air (Sleutels et al., 2002) and KvDMR1 imprinteddomains (Fitzpatrick et al., 2002), targeted deletion of thesesequences in the mouse results in disruption of imprinting. To date,two mechanisms have been described by which an ICR functions tosilence genes in somatic cells, through chromatin insulation and vialong noncoding RNAs (lncRNAs) (Barlow and Bartolomei, 2014;Pauler and Barlow, 2006; Wan and Bartolomei, 2008). The crucialfeature that determines whether an ICR is active is DNAmethylation status. ICRs contain CpG island (CGI)-like elementsthat are methylated (or not) in developing oocytes and sperm; thisgermline-specific methylation is manifested as differentiallymethylated regions (DMRs) in somatic cells, which aremaintained as a memory of parental origin and mediatemonoallelic silencing of imprinted genes. The large majority ofcharacterized imprinted domains contain ICR associated DMRs thatbecome methylated specifically in the female germline. To date,only four imprinted domains (H19/Igf2, Rasgrf1, Dlk1-Gtl2 andZdbf2) have been shown to contain DMRs methylated in the malegermline (Barlow and Bartolomei, 2014; Bartolomei and Ferguson-Smith, 2011; MacDonald and Mann, 2014).

Disruption of imprinted expression in humans leads to severaldevelopmental disorders and may contribute to cancer (Lee andBartolomei, 2013; Soellner et al., 2016; Tomizawa and Sasaki,2012; Wilkins and Úbeda, 2011). One of the earliest examples of animprinting disorder is Beckwith-Wiedemann syndrome (BWS), ageneralized overgrowth and cancer predisposition condition(Weksberg et al., 2010). Multiple genetic and epigeneticmechanisms give rise to BWS but all have the net effect ofdisrupting expression of imprinted genes in human chromosome11p15.5, with IGF2 andCDKN1C thought to play major roles in theovergrowth and cancer susceptibility (Choufani et al., 2013;Demars and Gicquel, 2012; Soejima and Higashimoto, 2013).The 11p15.5 imprinted domain in human, and its orthologouscounterpart in mouse distal chromosome 7, comprises twoindependently regulated subdomains, with the H19/Igf2 ICR (alsoknown as IC1) regulating imprinted expression of theH19 and IGF2genes, and the KvDMR1 ICR (IC2) controlling a cluster ofmaternally expressed genes, including CDKN1C (cyclin dependentkinase inhibitor 1C), a negative regulator of cell proliferation(Fitzpatrick et al., 2002). The KvDMR1 ICR includes the promoterfor the lncRNA KCNQ1OT1 and is normally methylated on thematernally inherited chromosome, thereby repressing expression ofthe lncRNA from this allele; on the paternally-inherited allele,Received 27 September 2016; Accepted 23 March 2017

1Departments of Molecular and Cellular Biology, Roswell Park Cancer Institute,Buffalo, NY 14263, USA. 2Cancer Genetics, Roswell Park Cancer Institute, Buffalo,NY 14263, USA. 3Biostatistics and Bioinformatics, Roswell Park Cancer Institute,Buffalo, NY 14263, USA.*Present address: Department of Microbiology and Immunology, University ofRochester Medical Center, Rochester, NY 14642, USA. ‡Present address: BattelleCenter for Mathematical Medicine and Department of Neurology, The ResearchInstitute at Nationwide Children’s Hospital, Columbus, OH 43215, USA.

§Author for correspondence ([email protected])

M.J.H., 0000-0002-3571-018X

1820

© 2017. Published by The Company of Biologists Ltd | Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

KvDMR1 is unmethylated and promotes expression ofKCNQ1OT1(Lee et al., 1999; Smilinich et al., 1999) (Fig. 1). In mouse models,expression of Kcnq1ot1 from the paternal allele has been shown to beinstrumental in subdomain-wide silencing of maternal-specific geneson the paternal chromosome (Mancini-Dinardo et al., 2006; Shinet al., 2008). More than half of individuals with BWS have anepimutation (absence of DNA methylation at the maternal allele) atKvDMR1 (Bliek et al., 2001; DeBaun et al., 2002; Engel et al., 2000;Weksberg et al., 2001), which results in biallelic expression ofKCNQ1OT1 (Lee et al., 1999; Smilinich et al., 1999) and associatedsilencing of CDKN1C and other maternally expressed genes in thisimprinted domain (Diaz-Meyer et al., 2003). The cause for theabsence or loss of methylation at KvDMR1 is not clear butpresumably results from improper establishment and/or maintenanceduring oogenesis or failure of maintenance after fertilization.KvDMR1 may be particularly susceptible to epigenetic disruption,as evidenced by frequent disruption of methylation following assisted-reproductive technologies (Amor and Halliday, 2008;Weksberg et al.,2007; White et al., 2015).Establishment of methylation imprints in the female mouse

germline occurs postnatally in growing oocytes arrested in meiosis I

(Hiura et al., 2006; Lucifero et al., 2004). Oocyte-specificmethylation is carried out by cooperative activities of DNAmethyltransferase 3A (DNMT3A) and its nonenzymatic co-factorDNA methyltransferase 3-like (DNMT3L) (Bourc’his et al., 2001;Hata et al., 2002; Kaneda et al., 2004). Recently, seeminglydisparate pieces of evidence have suggested a role for transcriptionin the establishment of the methylation mark at germline DMRs(reviewed by Veselovska et al., 2015). Most known maternallymethylated DMRs are located within transcription units and, in day10 oocytes (the time when de novo methylation is first detected),methylated intragenic CGIs are more likely to map to activetranscription units compared with unmethylated intragenic CGIs.Interestingly, trimethylation of H3K36, a histone modificationassociated with transcriptional elongation, can recruit DNMT3A(Dhayalan et al., 2010; Zhang et al., 2010). Furthermore, promoterCGIs found to be methylated at this stage frequently overlappedwith an adjacent transcription unit (Smallwood et al., 2011). Directdemonstration of a link between transcription and DNAmethylationat ICRs was first provided by Kelsey and colleagues for the Gnaslocus, where it was shown that disruption of the Nesp transcript,which initiates further upstream in this imprinting domain,

Fig. 1. Map of the KvDMR1-regulated imprinted domain in distal chromosome 7 and targeting strategy.Maternally expressed (paternally silenced) genesare indicated as black boxes, and non-imprinted genes as white boxes; the imprinted status of genes shown as gray boxes is uncertain. The sole paternallyexpressed (maternally silenced) lncRNA gene (Kcnq1ot1) is shown as a white box with diagonal black lines. Arrows above or below the genes indicate thedirection and status of transcription of the respective gene. tel, telomere; cen, centromere. The lower diagram is an enlarged view of the KvDMR1 domain showingthe two CpG islands (CGIs) and the position of the poly(A) cassette insertions; the Kcnq1ot1 transcription start site (TSS) is also indicated. The poly(A) cassettewas inserted in two orientations; the forward orientation shown previously to truncate the Kcnq1ot1 transcript (blue) (Shin et al., 2008) following paternaltransmission is indicated as YJ69, and the reverse orientation predicted to affect the Kcnq1 primary transcript (magenta) is indicated as YJ11.

1821

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

precludes normal establishment of methylation of the femalegermline DMR (Chotalia et al., 2009). More recently, using deepsequencing and de novo assembly of the mouse oocytetranscriptome, the same group demonstrated that transcription wasa major determinant of oocyte methylome; moreover, deletion of anoocyte-specific promoter of the Zac1 (Plagl1) gene results in thelack of DNAmethylation at the Zac1 ICR (also called Zac1 igDMR)(Veselovska et al., 2015). Importantly, these authors providedevidence that oocyte-specific transcription exists through othermaternally methylated ICRs, including KvDMR1 (Chotalia et al.,2009; Veselovska et al., 2015). In the case of KvDMR1, thistranscript is co-linear with the Kcnq1 primary transcript and arisesthrough activation of several oocyte-specific transcription start sites(TSS) close to the somatic TSS of Kcnq1 (Chotalia et al., 2009).Further support for transcription-mediated methylation duringoogenesis comes from studies of mouse transgenes demonstratingthat Snrpn DMR methylation in oocytes depended on activity of anupstream promoter (Smith et al., 2011). Whether transcriptionthrough an ICR is a common mechanism that also functions at theKvDMR1 ICR remains to be shown. If it is a common mechanism,then the etiology of some individuals with BWS may involvedefective transcription through the KvDMR1 locus, leading to theabsence of methylation.We tested this hypothesis by showing that disruption of Kcnq1

transcription before it reaches KvDMR1 prevents normalestablishment and/or maintenance of methylation at KvDMR1. Asa consequence, Kcnq1ot1 lncRNAwas expressed from the maternalallele, resulting in silencing of maternally expressed genes acrossthe domain. Thus, we provide further evidence that transcriptionacross DMRs is essential for establishment of methylation atmaternally methylated gametic DMRs.Moreover, this mouse modelpartially recapitulates the most common form of BWS, as it lacksmaternal methylation at KvDMR1 and has biallelic silencing ofCdkn1c, a crucial growth regulatory gene. This is the first reportsuggesting impairment of transcription through KvDMR1 as apossible cause of BWS.

RESULTSTargeted truncation of Kcnq1 transcriptionPreviously, we described a mouse line with a polyadenylation[poly(A)] truncation cassette inserted downstream of KvDMR1 toterminate elongation of Kcnq1ot1 (pA or YJ69, Fig. 1) (Shin et al.,2008). As a control, we inserted the same truncation cassette in the‘reverse’ orientation (Ap or YJ11). Following paternal transmissionof the YJ69 allele,Kcnq1ot1 transcription was blocked and imprintedexpression of genes under the control of KvDMR1 was lost,demonstrating the requirement for Kcnq1ot1 transcription in theimprinted silencing of this domain. By contrast, paternal inheritanceof the YJ11 had no observable effect on imprinted expression (Shinet al., 2008) (also see Fig. S1). Reports demonstrating the requirementof transcription through the ICRs at theGnas, Snrpn and Zac1 loci forproper methylation during oogenesis (Chotalia et al., 2009; Smithet al., 2011; Veselovska et al., 2015) prompted us to use the YJ11mutant line to determine whether transcription of Kcnq1 [whichoverlaps and transcribes antisense to Kcnq1ot1 (Fig. 1)] throughKvDMR1, is required for the establishment of maternal methylationat this ICR. The YJ69 allele should not truncate RNAs transcribed inthis direction and therefore serves as a control (Fig. S1). The effect ofKcnq1 truncation was first tested by crossing female miceheterozygous for the YJ11 allele with wild-type males; no pupscarrying the Kcnq1 truncation (YJ11) were observed in over 10litters from this cross, suggesting lethality in utero. Embryos from

additional crosses of the same type were analyzed at embryonic day(E) 9.5 and ∼50% of the conceptuses carried the YJ11 allele. Basedon previous literature, blocking transcription through the maternalKvDMR1 ICR during oogenesis is predicted to prevent itsmethylation, therefore allowing maternal expression of Kcnq1ot1,and consequent repression of maternally expressed genes andimpaired development, which likely explains the observed lethality.To test this possibility, female mice heterozygous for the YJ11 allelewere crossed with males carrying a deletion of KvDMR1 that resultsin expression of the maternal-specific genes from the paternalchromosome (Fitzpatrick et al., 2002). Significantly, embryoniclethality observed following maternal transmission of the YJ11 allelewas rescued (Table S1) supporting this contention.

To directly show the effect of the YJ11 allele on Kcnq1transcription, real-time RT-PCR was performed on RNA isolatedfrom kidney tissue from 3- to 4-week-old mice and from oocytescollected from 15 days post-partum (dpp) females. As YJ11/+ micecannot be obtained due to embryonic lethality, these studies werecarried out using YJ11/KvDMR1Δmice (i.e. offspring of YJ11−/+

females and KvDMR1Δ+/− males). To circumvent possibleconfounding effects due to insertion of foreign DNA (i.e. the pAtermination cassette) into the locus, kidney and oocyte RNAisolated from similarly aged YJ69/KvDMR1Δ mice was used as acontrol (Fig. S1). cDNA was synthesized using Kcnq1 transcript-specific primers located upstream (Kcnq1URT8) or downstream(Int-ls-RT3) of the poly(A) cassette (Ap or pA) with respect to thedirection of Kcnq1 transcription (Fig. S2A), and used as template inPCR assays using primers immediately upstream of the two Kcnq1-specific primers; because the primers corresponding to thedownstream assay are located within the deleted region in theKvDMR1Δ allele, this assay only detected transcripts generatedfrom the maternally derived YJ11 and YJ69 alleles (Fig. S2A). Asneither the YJ11 nor YJ69 truncation cassettes are predicted toaffect transcription upstream of the insertion site, upstream Ctvalues were used as a normalizing factor for the calculation of ΔCtvalues. Duplicate experiments analyzing kidney RNA from threeYJ69/KvDMR1Δ and three YJ11/KvDMR1Δ animals showed asignificant difference in the mean ΔCt (−0.16 for YJ69 and −1.08for YJ11, P=0.003) corresponding to a 47% decrease in Kcnq1transcript levels downstream of the YJ11 cassette compared with theYJ69 termination cassette (Fig. S2B). A similar reduction of Kcnq1primary transcript level was observed in oocytes collected from 15-day-old YJ11/KvDMR1Δ females (Fig. S2C). These resultssuggested that maternal inheritance of the YJ11 allele onlypartially impedes Kcnq1 elongation towards KvDMR1. However,upon further investigation, this surprisingly ‘inefficient’ terminationof Kcnq1 transcription was found to likely be an artifact due toprimer-independent reverse transcription (see Discussion).

To circumvent the apparent problem with RT-PCR at this locus,an RNA in situ hybridization approach was adopted usingRNAscope probes, which are specific to regions of the primaryKcnq1 transcript upstream and downstream of the pA andAp insertionsites (Fig. 2A,B); thus, the downstream probe is designed tohybridize only with RNA transcripts that have elongated through thepA or Ap termination cassettes. The two probes were co-hybridizedto sections of ovaries removed from four 15 dpp YJ69/KvDMR1Δand four YJ11/KvDMR1Δ females. At this developmental stage,chromosomes have been replicated, homologous chromosomespaired and oocytes arrested in the diplotene stage of meioticprophase 1 (reviewed by Hu et al., 2012; Sánchez and Smitz, 2012).When sectioned oocytes fromYJ69/KvDMR1Δ ovaries werehybridized, signals were detected for both upstream (turquoise)

1822

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

and downstream (magenta) RNAscope probes, indicatingelongation of Kcnq1 transcript through the poly(A) cassette(Fig. 2C). By contrast, only upstream (turquoise) signals weredetected following hybridization of both probes to sectionedoocytes from YJ11/KvDMR1Δ ovaries. These results wereconsistent, and statistically significant, across all scorable oocytesections fromYJ69/KvDMR1Δ (n=9) and YJ11/KvDMR1Δ (n=6)ovaries, with no downstream (magenta) signals observed in YJ11/KvDMR1Δ oocytes (Fig. 2D,E). Therefore, in contrast to RT-PCRanalysis of oocytes, which suggested only partial blockage ofKcnq1 transcription by the YJ11 allele, these results suggest thatblockage is virtually complete. Consistent with the notion that theYJ11 truncation cassette effectively blocks elongation of the Kcnq1primary transcript was the finding that it resulted in 40% decrease inthe steady state level of spliced Kcnq1 mRNA in kidney from 3- to4-week-old mice (Fig. S3).

Maternal transmission of the Kcnq1 truncation results inabsence of methylation at KvDMR1Methylation analysis at KvDMR1 was first carried out on genomicDNA extracted from embryonic tissues of two wild-type (+/SD7)and two mutant (YJ11/SD7) E9.5 conceptuses. KvDMR1 containstwo annotated CpG islands (CGI) separated by 250 bp (Fig. S4).First, region C, which is located near the end of CGI 2, wasanalyzed by COBRA. As expected for an imprinted differentiallymethylated DMR, approximately half of the PCR productsgenerated using bisulfite-treated DNA from wild-type littermatesDNA was digested by BstUI, indicating that roughly 50% of CpGdinucleotides in this restriction site were methylated; by contrast,only a small proportion of PCR product generated using DNA inmutant embryos was cleaved, suggesting little or no methylation atthis site (Fig. S4). As a control, COBRA assays were performed onbisulfite-treated DNA from the YJ69 line. Equivalent degrees of

Fig. 2. Targeted truncation of Kcnq1 transcript results in loss of transcription through KvDMR1. Kcnq1 transcription in 15 dpp oocytes following maternalinheritance of the YJ11 (Ap) or YJ69 (pA) allele was assessed by RNA in situ hybridization using RNAscope. (A,B) Schematic representation of the replicatedand paired homologous chromosomes (centromeres, solid gray circles) in oocyte nuclei from YJ69/KvDMR1Δ (A) and YJ11/KvDMR1Δ (B) ovaries. Locations ofthe upstream (turquoise) and downstream (magenta) hybridization probes are shown with respect to the poly(A) truncation cassette (triangle), the KvDMR1 ICRand the direction of theKcnq1 transcript (blue arrow). The downstream (magenta) probe is located within the KvDMR1 deletion and therefore only detects transcriptsgenerated from the maternally derived alleles (YJ69 and YJ11). (C,D) Representative sections of oocytes from YJ69/KvDMR1Δ (C) and YJ11/KvDMR1Δ(D) ovaries. (E) Summary of RNAscope results. The difference in the average number of magenta hybridization signals between YJ69/KvDMR1Δ and YJ11/KvDMR1Δ oocytes was statistically significant (**P<0.01, Mann-Whitney U-test). As the two RNAscope probes are very close to each other (red asterisk in A), andthe chromogenic reactions are performed sequentially with magenta done first, the downstream magenta signal deposit masks the upstream turquoise deposit(Xiao-Jun Ma, Advanced Cell Diagnostics, personal communication). This explains the lack of turquoise signal on the two YJ69 chromatids in C and the average oftwo signals per scorable oocyte rather than four (E). By contrast, as there is no hybridization of the downstream probe and consequent magenta signal developmentin the YJ11/KvDMR1Δ oocyte, all four chromatids carry turquoise signals (D,E).

1823

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

cleavage in wild-type andmutant DNA, at three different restrictionsites (two BstUI and one TaqI), suggested no significant changes inmethylation at KvDMR1 following maternal transmission of theYJ69 allele (Fig. S4). More comprehensive methylation analysisof KvDMR1was carried out using bisulfite sequencing; in total, 10CpG sites in CGI 1 and 16 CpG sites in CGI 2 were analyzed.Following maternal inheritance of the Kcnq1 truncation (YJ11),methylation was essentially absent at KvDMR1 in mutant embryos,whereas wild-type embryos had ∼50% methylation, as expected(Fig. 3, middle panel).In addition to failure in methylation establishment during

oogenesis, the absence of maternal methylation at KvDMR1 inYJ11 mutant embryos could have resulted from impairedmaintenance of DNA methylation at this locus following the waveof demethylation after fertilization. To determine whether the YJ11allele was methylated prior to fertilization, oocytes, collected fromtwo 15 dpp compound heterozygotes (YJ11/KvDMRΔ) and two‘wild type’ (+/KvDMRΔ) littermates were analyzed by bisulfitesequencing. As the PCR primers used were located within the regiondeleted in the KvDMRΔ (Fitzpatrick et al., 2002), PCR products

originating from oocyte DNA isolated from these mice originateonly from the maternal allele. Oocyte DNA from wild-typelittermates showed predominantly methylated alleles (37/52sequenced molecules) as expected (Fig. 3, lower panel). Theunmethylated alleles in the (+/KvDMRΔ) samples likely reflectheterogeneity in the developmental stage of the oocyte population in15 ddp ovaries (Hiura et al., 2006). By contrast, and consistent withresults obtained from E9.5 day embryos, methylation at KvDMR1was almost completely absent in oocytes following Kcnq1truncation (YJ11). These results suggest that, in the absence oftranscription through KvDMR1, methylation was either notestablished or was lost later during oocyte maturation;furthermore, the aberrantly unmethylated YJ11 allele remainedunmethylated during global post-fertilization methylation events.These data strongly suggest that, similar to the Gnas, Snrpn andZac1 loci (Chotalia et al., 2009; Smith et al., 2011; Veselovskaet al., 2015), transcription through the KvDMR1 is a prerequisite toestablish and/or maintain the methylation mark at the maternalallele, and confirm this transcription-associated mechanism at athird endogenous locus.

Fig. 3.Kcnq1 truncation in thematernal germline results in the lack of methylation atKvDMR1. The top panel shows the position of two CGIs present withinKvDMR1 and the location of themajor transcription start site forKcnq1ot1 lncRNA. Three amplicons (A-C) of∼200 bp eachwere analyzed by bisulfite sequencingperformed on embryos (n=2 for each genotype) collected at E9.5 (middle panel) as well as on oocytes (n=2 for each genotype) collected from 15 dpp compoundheterozygotes (YJ11/KvDMR1Δ) and wild-type (+/KvDMR1Δ) littermates (lower panel). Following PCR and cloning, 10 clones for each amplicon weresequenced; each row represents CpG sites of an individual sequenced clone. Each circle represents one CpG site, with filled circles depicting methylated andopen circles depicting unmethylated sites. *YJ11=YJ11/SD7, WT=+/SD7; **YJ11=YJ11/ KvDMR1Δ, WT=+/KvDMR1Δ.

1824

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

Blockage of transcription through KvDMR1 results inbiallelic expression of Kcnq1ot1 and repression ofmaternally expressed genesIn the KvDMR1 domain, imprinted protein-coding genes areexpressed exclusively or predominately from maternal alleles,whereas the Kcnq1ot1 lncRNA is repressed on the maternalmethylated KvDMR1 allele and expressed from the unmethylatedpaternal allele. To determine whether truncation of the Kcnq1primary transcript and consequent absence of methylation atKvDMR1 affected gene expression across the imprinted domain,analysis was carried out on E9.5 conceptuses from crosses betweenfemales heterozygous for the YJ11 or YJ69 (control) allele and SD7males (congenic for distal chromosome 7 ofM. spretus in a C57BL/6J background); SD7 males were used to provide polymorphismsfor allele-specific expression analysis. The imprinted status of theKcnq1ot1 lncRNA was first tested using primers whose ampliconspans an MwoI RFLP between 129SvJae and SD7 mice (Lewiset al., 2006). Both wild-type placenta and embryo proper expressedonly the MwoI-cleaved (SD7) paternal allele, while biallelicexpression of Kcnq1ot1 occurred in placental and embryonictissues with the Kcnq1 truncation (Fig. 4A). This ‘loss ofimprinting’ was reflected as a two- to threefold increase inKcnq1ot1 expression in placental and embryonic tissues with amaternally inherited YJ11 allele, whereas the steady-state level ofKcnq1ot1 lncRNA was not significantly changed in the YJ69control line (Fig. 4B). Thus, following the maternal Kcnq1truncation, Kcnq1ot1 becomes biallelically expressed.

In wild-type mice, Kcnq1ot1 is expressed only from the paternalallele and silences (in cis) at least eight maternal-specific genes on thepaternal chromosome (Fitzpatrick et al., 2002; Mancini-Dinardoet al., 2006; Shin et al., 2008). Thus, maternal expression ofKcnq1ot1is also anticipated to silence the same set of protein coding genes onthe maternal chromosome. Allele-specific primers were used in qRT-PCR assays to analyze expression of the ubiquitously imprintedgenes Cdkn1c and Phlda2, as well as the Ascl2 gene, which isexpressed primarily in the placenta. In wild-type controls, expressionin wild-type placenta and embryo proper was almost exclusively fromthe maternal allele; however, following maternal transmission of theYJ11 allele, the maternal allele of each of these genes was partially orcompletely silenced (Fig. S5, YJ11 X SD7 panels). In E9.5 embryos,maternal transmission of the YJ69 allele (control) had little or noeffect on gene expression across the imprinted domain, with theexception of Cdkn1c (Fig. S5, YJ69×SD7 panels) where, in contrastto E9.5 placenta (where expression was unchanged), Cdkn1cexpression in E9.5 embryos was almost twice as high in thechromosome carrying the YJ69 termination cassette compared withthe wild-type chromosome. This result almost reached statisticalsignificance (P=0.07) (Fig. S5, middle panel, far right) and suggeststhat the insertion may have disrupted a regulatory element for thisgene. The expression of the Kcnq1 and Slc22a18 genes was analyzedusing primers that measure overall gene expression levels. These twogenes also showed a significant reduction in expression in placentaland embryonic tissues, suggesting silencing of the normallyexpressed maternal allele (Fig. S6). Although, maternal inheritance

Fig. 4. Absence of KvDMR1 methylation results in biallelic expression of Kcnq1ot1 lncRNA. (A) Imprinted expression of Kcnq1ot1 was assessed in E9.5conceptuses using a MwoI RFLP (present in the SD7 allele, absent in 129SvJae). A 250 bp RT-PCR product was amplified and subjected to MwoI digestion.Although the wild-type (WT) littermates (+/SD7) showed expression from only the paternal (SD7) allele, YJ11 mutants (YJ11/SD7) also had expression from thematernal allele. Left panel shows maternal and paternal allele amplified from 129SvJae and SD7 mouse DNA as a control. (B) Total expression levels of thelncRNA were measured using a real-time RT-PCR assay. Expression was assessed in E9.5 placental and embryonic tissues in mouse lines (YJ11/SD7 andYJ69/SD7). Compared with wild-type littermates, and consistent with biallelic expression in YJ11/SD7 tissues above, the mean±s.e.m. expression level ofKcnq1ot1was two to three times higher in tissues from the YJ11/SD7 line; expression ofKcnq1ot1 in the control YJ69 linewas not significantly different comparedwith controls. For each genotype, n=3. P values were obtained from F-tests corresponding to the general linear models: ***P<0.001, **P=0.05-0.07.

1825

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

of the YJ69 allele showed no statistically significant changes in thegene expression of Slc22a18, the expression of Kcnq1 in placentawas unexpectedly increased by ∼40% (P<0.001, Fig. S6), againsuggesting potential disruption of a regulatory element. In summary,gene expression analysis suggests that the absence of methylation atKvDMR1 was associated with blocked elongation of a ‘Kcnq1’transcript through KvDMR1 and allows Kcnq1ot1 to be expressedbiallelically. Similar to the situation of the paternal allele, expressionof Kcnq1ot1 on the maternal allele directly or indirectly repressesexpression of maternal-specific genes present in this imprinteddomain. Although some inconsistencies are observed[e.g. increased expression of Cdkn1c and Kcnq1 in control YJ69tissues compared with wild type (Figs S5 and S6), see Discussion],the majority of the expression data described above supports thenotion that the net effect of the YJ11 Kcnq1 truncation model is thatboth parental alleles acquire similar epigenetic states. These findingsfurther support the notion that transcription through an ICR is acommon requirement for the establishment and/or maintenance ofDNA methylation at maternal gDMRs and regulation of imprintedgene expression. Furthermore, this Kcnq1 truncation mouse modelrecapitulates the ‘loss of methylation’ molecular phenotypefrequently observed in individuals with BWS (Bliek et al., 2004;DeBaun et al., 2002; Engel et al., 2000; Smilinich et al., 1999;Weksberg et al., 2001).

DISCUSSIONPrevious studies have demonstrated that transcription is required forestablishment of methylation at maternal gDMRs (Veselovska et al.,2015). These same studies also demonstrated the existence of RNAtranscripts in the vicinity of other maternally methylated DMRs,including KvDMR1. Interestingly, a large domain of virtuallycontiguous hypermethylation was observed across the Kcnq1 gene,beginning just after exon 1 and extending across the entire locus(>300 kb) (Veselovska et al., 2015). Given that, in the majority ofindividuals with BWS, KvDMR1 is devoid of methylation and genesunder its control are repressed, one mechanism by which thisanomaly could result is through absence of transcription via thisregulatory element. RNA in situ hybridization showed that maternaltransmission of the YJ11 allele blocked Kcnq1 primary transcriptelongation from reaching KvDMR1. Moreover, the lack of thesetranscripts correlated with the absence or very low level ofmethylation at KvDMR1 in E9.5 conceptuses and 15 dpp oocytes.This lack of methylation is likely due to defective establishment,rather than to its developmental postponement because a delay inestablishment would not result in the absence of methylation in E9.5conceptuses (Fig. 3). Thus, similar to other ICRs (Chotalia et al.,2009; Smith et al., 2011; Veselovska et al., 2015), transcriptionthrough KvDMR1 appears to be essential for its proper methylation.These results are consistent with the finding that a 260 kb BACtransgene, encompassing the KvDMR1 ICR and most of the Kcnq1gene, but excluding the promoter region, did not undergo properimprinting following maternal transmission (John et al., 2001).RT-PCR analysis of Kcnq1 primary transcription downstream of

the YJ11 truncation cassette suggested only partial blockage oftranscription, an observation inconsistent with our RNA in situhybridization analysis, which provided evidence for efficienttermination, at least below the level of detection of the assay.However, further investigation has shown that the apparent less thancomplete blockage of the Kcnq1 primary transcript, as determinedby RT-PCR, is due to primer-independent reverse transcription; i.e.some PCR product is generated from cDNA reactions that do notcontain a primer for first-strand synthesis. It should be noted that no

amplification product was observed in –RT controls, indicating thatthis product was not amplified from contaminating genomic DNAin the RNA. As this region of the Kcnq1 primary transcript istranscribed from the same genomic region as the Kcnq1ot1 lncRNAtranscript, it may be Kcnq1ot1 RNA degradation products that areaberrantly priming cDNA synthesis.

Until recently, the prevailing models of germline-specificmethylation of gDMRs presumed targeting of de novo DNMTasesto specific loci, perhaps by a combination of localized DNAcomposition and/or sequence and chromatin structure (reviewed byBartolomei and Ferguson-Smith, 2011). However, genome-widemethylation studies of the mouse germline have promptedre-evaluation of this notion. First, reduced representation bisulfitesequencing (RRBS) showed that CGIs methylated in oocytes werepredominately found within active transcription units rather than attheir 5′ends; importantly, these included most maternal gDMRs(Smallwood et al., 2011). Moreover, whole-genome bisulfitesequencing coupled with RNA-seq demonstrated that mostmethylation in oocytes is within gene bodies and that thismethylation was highly correlated with expression levels of thosegenes (Kobayashi et al., 2012). Thus, the developing paradigmproposes that, at least in the female germline, the DNMT3A/DNMT3L methylation machinery is not specifically targeted togDMR sequences. Instead gDMRmethylation takes place as part oflarger more-general gene body methylation, and specific gDMRsbecome methylated by virtue of their intragenic location (Kelseyand Feil, 2013; Smallwood et al., 2011). Although generally notresulting from studies using oocytes, several lines of evidencesuggest that transcription through gene bodies and associatedintragenic CGI and gDMRs brings about a specific chromatinstructure that promotes the binding and action of the DMNT3A/DMNT3L complex. The model suggests that transcription-coupledrecruitment of histone modifying enzymes, such as the H3K4me1/2demethylase KDM1B and the H3K36 methylase SETD2, yields achromatin template with unmethylated H3K4 and trimethylatedH3K36, histone modifications that are ‘preferred’ by the DNMT3A/DNMT3L complex (reviewed by Kelsey and Feil, 2013; Stewartet al., 2015). The number of transcription units and intragenic CGIsthat are methylated throughout their gene body in oocytes is fargreater than the number of known imprinted gDMRs (Kobayashiet al., 2012; Smallwood et al., 2011). Consistently, genome-wideChIP-seq experiments have demonstrated a relative depletion of‘protective’ H3K4me2 and H3K4me3, as well as enrichment of‘permissive’H3K36me3marks at the majority of the CGI destined tobecome methylated during oocyte growth and maturation (Stewartet al., 2015). In the present case, it is noteworthy that KvDMR1, likemost maternallymethylated ICRs, shows a dramatic decrease inDNAmethylation in oocytes isolated from KDM1B knockout mice(Stewart et al., 2015). Thus, the emerging model for how allele-specific methylation occurs at only several tens of loci, includinggDMRs is one of ‘selective protection from demethylation’ duringearly embryonic development (Kelsey and Feil, 2013; Smallwoodet al., 2011; Stewart et al., 2015; Veselovska et al., 2015), perhaps byproteins such as ZPF57 (Li et al., 2008; Mackay et al., 2008;Quenneville et al., 2011) and NLRP2 (Meyer et al., 2009).

We propose that the reduction or absence of transcriptionalelongation prior to reaching KvDMR1 prevents formation of achromatin state permissive to DNA methylation by de novo DNAmethyltransferases (Fig. 5). Further studies are required to confirmthe exact chromatin conformation at KvDMR1 following thematernal inheritance of the YJ11 truncation allele as well as thetemporal association and interactions of the transcription process

1826

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

and specific factors that affect de novo methylation (Ciccone et al.,2009; Hiura et al., 2006).Our observations suggest that maternal transmission of the YJ11

allele causes embryonic lethality (later than E9.5) due to thesilencing of one or more maternal-specific genes in the domain. Themost likely culprit is Ascl2, expression of which is crucial forplacental development, and whose deletion results in embryonicdeath at E10.5 (Guillemot et al., 1994, 1995; Tunster et al., 2016).Surprisingly, maternal transmission of the YJ69 allele (identicalsequence to YJ11 but inserted in opposite orientation) wasassociated with significantly increased expression of Cdkn1c inE9.5 embryos (Fig. S5) and of Kcnq1 in E9.5 placenta (Fig. S6).This could indicate that insertion of the truncation cassette disrupteda currently undefined negative regulatory element for these twogenes; in this regard, it is notable that recent evidence supports theexistence of multiple enhancer elements in the KvDMR1/Kcnq1ot1region (Korostowski et al., 2011; Schultz et al., 2015). Disruption ofa gene repressor/silencer at this insertion site may also explain theapparently less efficient silencing effect that insertion of the YJ11cassette has on Cdkn1c and Kcnq1 (Figs S5 and S6). Despite theapparent complexity of this genomic region, it is unlikely thatthe absence of methylation at KvDMR1 following the insertion ofthe YJ11 allele is due simply to insertional mutagenesis, as the sameeffect is not observed with the YJ69 allele (Fig. S4).

Roughly half of individuals with BWS have an epimutation(absence of DNA methylation at the maternal allele) at KvDMR1(Bliek et al., 2001; DeBaun et al., 2002; Engel et al., 2000;Weksberg et al., 2001), which results in biallelic expression ofKCNQ1OT1 (Lee et al., 1999; Smilinich et al., 1999) and associatedsilencing of CDKN1C and other maternally expressed genes in thisimprinted domain (Diaz-Meyer et al., 2003). Mouse models withmutations in Cdkn1c have been described and in some cases exhibitfetal overgrowth that is reminiscent of BWS but are lethal at birth(Tunster et al., 2011; Yan et al., 1997; Zhang et al., 1997). However,these models do not precisely mirror the majority of cases of BWSas most patients do not have mutations in CDKN1C. The presentmouse model results in the absence of methylation at KvDMR1,biallelic expression of the Kcnq1ot1 lncRNA and biallelic silencingof genes under the control of this ICR, closely recapitulating themolecular phenotype observed in the majority of individuals withBWS. The lethality observed in the YJ11 mouse is not seen in BWSbecause, unless tumors develop, individuals with this conditionusually survive into adulthood. The embryonic viability of humanembryos may be due to the apparent lack of ASCL2 imprinting inhuman placenta (Miyamoto et al., 2002) as well as somewhat‘leaky’ imprinting ofCDKN1C (Chung et al., 1996;Matsuoka et al.,1996). Although it is widely thought that maternal expression ofKCNQ1OT1 in BWS is responsible for silencing of maternal-

Fig. 5. Model for the role of transcription for the establishment of methylation in oocytes at KvDMR1. In normal mouse oocytes (upper), transcriptionproceeds along the Kcnq1 transcription unit, including KvDMR1. The elongating form of RNA polymerase II complexes, together with a ‘KDM1B-like’H3K4me2/3demethylase and an H3K36 methyltransferase, modify the chromatin and increase binding affinity for DNMT3A and DNMT3L. Following fertilization, the embryo(right) does not express Kcnq1ot1 from the maternal allele, allowing for the expression of protein-coding genes. Absence of methylation at the paternal allele ofKvDMR1 permits Kcnq1ot1 expression, which, in turn, silences of genes in cis. Pink and blue circles represent maternal and paternal centromeres, respectively.Blockage or reduction of Kcnq1 elongation prior to reaching KvDMR1 (lower) precludes appropriate histone modifications required for binding of DNMT3A andDNMT3L, resulting in an absence of DNA methylation on the maternal allele. In the embryo, both parental alleles of Kcnq1ot1 are expressed, leading to biallelicsilencing of protein-coding genes. In the case of the mouse, silencing of Ascl2, and perhaps of other genes, results in embryonic lethality.

1827

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

specific genes in the domain, it remained a formal possibility thatthese aberrant expression patterns were not directly related. Thepresent model provides a direct mechanistic link between theabsence of methylation at KvDMR1, maternal expression ofKcnq1ot1 and maternal silencing of the ICR regulated genes. It isnot known whether any individuals with BWS and an absence ofKvDMR1 methylation have a defect in oocyte-related KCNQ1transcription (Fig. S7). Nevertheless, microdeletions specificallyaffecting an oocyte-specific promoter may comprise a subset offamilial BWS cases that do not have mutations in CDKN1C. Thisstudy also predicts that individuals with BWS who have achromosome translocation that breaks between an oocyte-specificpromoter and KvDMR1 would also exhibit absence of methylationat KvDMR1. This may be the case for a patient described in a recentreport who had both BWS and long QT syndrome (Kaltenbachet al., 2013). Epigenetic silencing of an oocyte-specific promoter isanother possible mechanism by which DNA methylation is notestablished in this domain, thus leading to BWS. Further analysis ofindividuals with BWS is necessary to determine to what extentvarious mechanisms that disrupt transcription through KvDMR1during oogenesis are responsible for its aberrant methylation.

MATERIALS AND METHODSGeneration of Kcnq1 truncation miceMouse studies were approved by the Roswell Park Cancer Institute (RPCI)Institutional Animal Care and Use Committee (IACUC). The generation ofthe YJ11 and YJ69 mutant lines has been described previously (Shin et al.,2008). Female offspring heterozygous for the YJ11 or YJ69 alleles werecrossed with either SD7 (congenic for distal chromosome 7 ofM. spretus ina C57BL/6J background, kindly provided by Wolf Reik, BabrahamInstitute, Cambridge, UK) or KvDMR1Δmales, depending on the analysis.

Isolation of growing oocytesOocytes were collected from 15 dpp YJ11/KvDMR1Δ, YJ69/KvDMR1Δ or+/KvDMR1Δ control females to allow exclusive PCR amplification of thematernal KvDMR1 allele (YJ11) in the absence of the paternal allele inoocytes (see Results). Oocytes were collected (Chotalia et al., 2009)(protocol kindly provided by Jiahao Huang and Gavin Kelsey, BabrahamInstitute, Cambridge, UK). Ovaries were dissected in PBS containing2 mg/ml collagenase (Sigma, C2674), 0.025% trypsin (Sigma, 93615) andincubated for 30 min at 37°C in a Thermomixer R (Eppendorf ) rotating at600 rpm, with intermittent mixing by pipetting. The digested ovaries werespread out onto a 10 cm cell culture plate and an equal volume of M2medium was added. Individual oocytes were serially transferred throughseveral droplets of M2 medium to dilute out somatic cells.

RNA in situ hybridizationRNA in situ hybridization was carried out using RNAscope on ovariescollected from 15 dpp YJ69/KvDMR1Δ and YJ11/KvDMR1Δ. Ovaries werefixed in 4% paraformaldehyde, dehydrated and embedded in paraffin wax.Sections (5 μm) were processed for RNA using the RNAscope chromogenickit according to the manufacturer’s instructions (Advanced Cell Diagnostics;ACD). Custom probes designed and synthesized by ACD to detect Kcnq1primary transcripts were used for hybridization for 5 h at 40°C. Following amultistep signal amplification, microscopic evaluation of the ovaries wasperformed by scoring the turquoise and magenta spots in oocyte nuclei.

Methylation analysisMethylation at KvDMR1 was assessed by combined bisulfite restrictionanalysis (COBRA) (Xiong and Laird, 1997) and bisulfite sequencing (Clarket al., 1994). Genomic DNA (1 µg) was bisulfite converted using EZ DNAmethylation kit (ZYMO Research). Three fragments within KvDMR1 wereamplified by PCR using primers listed in Table S2. For COBRA, PCRproducts were purified and digested with BstU1 or TaqI restriction enzymes.For bisulfite sequencing, PCR products were cloned into pCRII-TOPO

cloning vector and 10 clones for each region were sequenced using M13primers. Sequencing data was quantified by web-based QUMAmethylationanalysis tool (http://quma.cdb.riken.jp/).

Gene expression analysisRNAwas extracted using Trizol, digested with RNase-free DNase (Ambion)and used as template to synthesize cDNA using the SuperScript III FirstStrand cDNA synthesis Kit or Thermoscript (for analysis of Kcnq1 primarytranscript, Fig. S2B,C) (ThermoFisher). cDNA syntheses were carried outusing random primers, except in Fig. S2B,C; in this case, Kcnq1 strand-specific RT primers (Int-ls-RT3 & Kcnq1URT8) were used. The imprintedexpression of the Kcnq1ot1 lncRNAwas analyzed by RFLP RT-PCR usingan MwoI polymorphism between 129SvJae and SD7 (Lewis et al., 2006).Expression levels of Kcnq1, Kcnq1ot1, Slc22a18, Cdkn1c, Phlda2 andAscl2 were determined using real-time quantitative RT-PCR in a 20 µlreaction using iQSYBER Green Supermix (Bio-Rad) PCR was carried outusing a BIO-RAD CFX96 instrument. Expression levels were analyzedusing the ΔΔCt method. In some cases, allele-specific primers were used todistinguish between maternal (C57Bl/6J) and paternal (SD7) alleles(Mohammad et al., 2010). Alternatively, where allele-specific primers didnot perform well, RT-PCR was carried out using primers that detect totalexpression of genes. Except for allele-specific primers (see Mohammadet al., 2010), primer information is given in Table S2.

Statistical analysisThe expression levels in Fig. 4, Figs S2, S3, S5 and S6 were modeled as afunction of mutant type (YJ11 and YJ69), tissue type (kidney, placenta andembryo, oocyte), type (WT-mat, WT-pat, YJ11/69-mat and YJ11/69-mat) andreplicate using general linear models. The specific variables included in a givenmodel depend on the experiment being evaluated. The mean expression levelswere compared between groups of interest using F-tests about the appropriatelinear contrasts of model estimates. All model assumptions were verifiedgraphically using quantile-quantile and residual plots, and transformationswereapplied as appropriate using theBox-Coxmethod. All analyses were conductedin SAS v9.4 (Cary, NC) at a significance level of 0.05. P-values less than 0.05therefore denote statistically significant differences.

AcknowledgementsThe authors thank Dr Jong-Yeon Shin for construction of the YJ11 and YJ69 mouselines, and Debbie Tabaczynski for animal husbandry and dissections.

Competing interestsThe authors declare no competing or financial interests.

Author contributionsConceptualization: V.B.S., M.J.H.; Methodology: V.B.S., S.S., K.E.W., K.M.A.,J.C.H., S.P., M.J.H.; Validation: S.S., K.M.A.; Formal analysis: V.B.S., S.S., K.E.W.,K.M.A., S.P., M.J.H.; Investigation: M.J.H.; Resources: M.J.H.; Writing - originaldraft: V.B.S., M.J.H.; Writing - review & editing: V.B.S., S.S., K.M.A., J.C.H., S.P.,M.J.H.; Supervision: M.J.H.; Project administration: M.J.H.; Funding acquisition:M.J.H.

FundingThis study was supported by National Institutes of Health (NIH) (RO1 CA89426 toM.J.H.) and use core facilities supported in part by the Roswell Park Cancer InstituteNational Cancer Institute (NCI)-funded Cancer Center Support Grant (P30CA016056-27). Deposited in PMC for release after 12 months.

Data availability

Supplementary informationSupplementary information available online athttp://dev.biologists.org/lookup/doi/10.1242/dev.145136.supplemental

ReferencesAmor, D. J. and Halliday, J. (2008). A review of known imprinting syndromes and

their association with assisted reproduction technologies. Hum. Reprod. 23,2826-2834.

Barlow, D. P. and Bartolomei, M. S. (2014). Genomic imprinting in mammals. ColdSpring Harb. Perspect. Biol. 6, 1-20.

1828

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

Bartolomei, M. S. and Ferguson-Smith, A. C. (2011). Mammalian genomicimprinting. Cold Spring Harb. Perspect. Biol. 3, 1-17.

Bliek, J., Maas, S. M., Ruijter, J. M., Hennekam, R. C., Alders, M., Westerveld, A.and Mannens, M. M. (2001). Increased tumour risk for BWS patients correlateswith aberrant H19 and not KCNQ1OT1 methylation: occurrence of KCNQ1OT1hypomethylation in familial cases of BWS. Hum. Mol. Genet. 10, 467-476.

Bliek, J., Gicquel, C., Maas, S., Gaston, V., Le Bouc, Y. and Mannens, M. (2004).Epigenotyping as a tool for the prediction of tumor risk and tumor type in patientswith Beckwith-Wiedemann syndrome (BWS). J. Pediatr. 145, 796-799.

Bourc’his, D., Xu, G. L., Lin, C. S., Bollman, B. and Bestor, T. H. (2001). Dnmt3Land the establishment of maternal genomic imprints. Science 294, 2536-2539.

Chotalia, M., Smallwood, S. A., Ruf, N., Dawson, C., Lucifero, D., Frontera, M.,James, K., Dean, W. and Kelsey, G. (2009). Transcription is required forestablishment of germline methylation marks at imprinted genes. Genes Dev. 23,105-117.

Choufani, S., Shuman, C. and Weksberg, R. (2013). Molecular findings inBeckwith-Wiedemann syndrome. Am. J. Med. Genet. C Semin. Med. Genet. 163,131-140.

Chung, W.-Y., Yuan, L., Feng, L., Hensle, T. and Tycko, B. (1996). Chromosome11p15.5 regional imprinting: comparative analysis of KIP2 and H19 in humantissues and Wilms’ tumors. Hum. Mol. Genet. 5, 1101-1108.

Ciccone, D. N., Su, H., Hevi, S., Gay, F., Lei, H., Bajko, J., Xu, G., Li, E. and Chen,T. (2009). KDM1B is a histone H3K4 demethylase required to establish maternalgenomic imprints. Nature 461, 415-418.

Clark, S. J. I., Harrison, J., Paul, C. L. and Frommer, M. (1994). High sensitivitymapping of methylated cytosines. Nucleic Acids Res. 22, 2990-2997.

DeBaun, M. R., Niemitz, E. L., McNeil, D. E., Brandenburg, S. A., Lee, M. P. andFeinberg, A. P. (2002). Epigenetic alterations of H19 and LIT1 distinguishpatients with Beckwith-Wiedemann syndrome with cancer and birth defects.Am. J. Hum. Genet. 70, 604-611.

Demars, J. and Gicquel, C. (2012). Epigenetic and genetic disturbance of theimprinted 11p15 region in Beckwith-Wiedemann and Silver-Russell syndromes.Clin. Genet. 81, 350-361.

Dhayalan, A., Rajavelu, A., Rathert, P., Tamas, R., Jurkowska, R. Z., Ragozin, S.and Jeltsch, A. (2010). The Dnmt3a PWWP domain reads histone 3 lysine 36trimethylation and guides DNA methylation. J. Biol. Chem. 285, 26114-26120.

Diaz-Meyer, N., Day, C. D., Khatod, K., Maher, E. R., Cooper, W., Reik, W.,Junien, C., Graham, G., Algar, E., Der Kaloustian, V. M. et al. (2003). Silencingof CDKN1C (p57(KIP2)) is associated with hypomethylation at KvDMR1 inBeckwith-Wiedemann syndrome. J. Med. Genet. 40, 797-801.

Engel, J. R., Smallwood, A., Harper, A., Higgins, M. J., Oshimura, M., Reik, W.,Schofield, P. N. andMaher, E. R. (2000). Epigenotype-phenotype correlations inBeckwith-Wiedemann syndrome. J. Med. Genet. 37, 921-926.

Fitzpatrick, G. V., Soloway, P. D. and Higgins, M. J. (2002). Regional loss ofimprinting and growth deficiency in mice with a targeted deletion of KvDMR1. Nat.Genet. 32, 426-431.

Guillemot, F., Nagy, A., Auerbach, A., Rossant, J. and Joyner, A. L. (1994).Essential role of Mash-2 in extraembryonic development. Nature 371, 333-336.

Guillemot, F., Caspary, T., Tilghman, S. M., Copeland, N. G., Gilbert, D. J.,Jenkins, N. A., Anderson, D. J., Joyner, A. L., Rossant, J. andNagy, A. (1995).Genomic imprinting of Mash2, a mouse gene required for trophoblastdevelopment. Nat. Genet. 9, 235-242.

Hata, K., Okano, M., Lei, H. and Li, E. (2002). Dnmt3L cooperates with the Dnmt3family of de novo DNA methyltransferases to establish maternal imprints in mice.Development 129, 1983-1993.

Hiura, H., Obata, Y., Komiyama, J., Shirai, M. and Kono, T. (2006). Oocytegrowth-dependent progression of maternal imprinting in mice. Genes Cells 11,353-361.

Hu, M.-W., Wang, Z.-B., Schatten, H. and Sun, Q.-Y. (2012). New understandingson folliculogenesis/oogenesis regulation in mouse as revealed by conditionalknockout. J. Genet. Genomics 39, 61-68.

John, R. M., Ainscough, J. F., Barton, S. C. and Surani, M. A. (2001). Distant cis-elements regulate imprinted expression of the mouse p57(Kip2) (Cdkn1c) gene:implications for the human disorder, Beckwith–Wiedemann syndrome. Hum. Mol.Genet. 10, 1601-1609.

Kaltenbach, S., Capri, Y., Rossignol, S., Denjoy, I., Soudee, S., Aboura, A.,Baumann, C. and Verloes, A. (2013). Beckwith-Wiedemann syndrome and longQT syndrome due to familial-balanced translocation t(11;17)(p15.5;q21.3)involving the KCNQ1 gene. Clin. Genet. 84, 78-81.

Kaneda, M., Okano, M., Hata, K., Sado, T., Tsujimoto, N., Li, E. and Sasaki, H.(2004). Essential role for de novo DNA methyltransferase Dnmt3a in paternal andmaternal imprinting. Nature 429, 900-903.

Kelsey, G. and Feil, R. (2013). New insights into establishment and maintenance ofDNA methylation imprints in mammals. Philos. Trans. R. Soc. Lond. B Biol. Sci.368, 20110336.

Kobayashi, H., Sakurai, T., Imai, M., Takahashi, N., Fukuda, A., Yayoi, O., Sato,S., Nakabayashi, K., Hata, K., Sotomaru, Y. et al. (2012). Contribution ofintragenic DNA methylation in mouse gametic DNA methylomes to establishoocyte-specific heritable marks. PLoS Genet. 8, e1002440.

Korostowski, L., Raval, A., Breuer, G. and Engel, N. (2011). Enhancer-drivenchromatin interactions during development promote escape from silencing by along non-coding RNA. Epigenetics Chromatin 4, 21.

Lee, J. T. and Bartolomei, M. S. (2013). X-inactivation, imprinting, and longnoncoding RNAs in health and disease. Cell 152, 1308-1323.

Lee, M. P., DeBaun, M. R., Mitsuya, K., Galonek, H. L., Brandenburg, S.,Oshimura, M. and Feinberg, A. P. (1999). Loss of imprinting of a paternallyexpressed transcript, with antisense orientation to KVLQT1, occurs frequently inBeckwith-Wiedemann syndrome and is independent of insulin-like growth factor IIimprinting. Proc. Natl. Acad. Sci. USA 96, 5203-5208.

Lewis, A., Green, K., Dawson, C., Redrup, L., Huynh, K. D., Lee, J. T.,Hemberger, M. andReik,W. (2006). Epigenetic dynamics of the Kcnq1 imprinteddomain in the early embryo. Development 133, 4203-4210.

Li, Y. and Sasaki, H. (2011). Genomic imprinting in mammals: its life cycle,molecular mechanisms and reprogramming. Cell Res. 21, 466-473.

Li, X., Ito, M., Zhou, F., Youngson, N., Zuo, X., Leder, P. and Ferguson-Smith,A. C. (2008). A maternal-zygotic effect gene, Zfp57, maintains both maternal andpaternal imprints. Dev. Cell 15, 547-557.

Lucifero, D., Mann, M. R. W., Bartolomei, M. S. and Trasler, J. M. (2004). Gene-specific timing and epigenetic memory in oocyte imprinting. Hum. Mol. Genet. 13,839-849.

MacDonald, W. A. and Mann, M. R. W. (2014). Epigenetic regulation of genomicimprinting from germ line to preimplantation. Mol. Reprod. Dev. 81, 126-140.

Mackay, D. J. G., Callaway, J. L. A., Marks, S. M., White, H. E., Acerini, C. L.,Boonen, S. E., Dayanikli, P., Firth, H. V., Goodship, J. A., Haemers, A. P. et al.(2008). Hypomethylation of multiple imprinted loci in individuals with transientneonatal diabetes is associated with mutations in ZFP57. Nat. Genet. 40,949-951.

Mancini-Dinardo, D., Steele, S. J., Levorse, J. M., Ingram, R. S. and Tilghman,S. M. (2006). Elongation of the Kcnq1ot1 transcript is required for genomicimprinting of neighboring genes. Genes Dev. 20, 1268-1282.

Matsuoka, S., Thompson, J. S., Edwards, M. C., Barletta, J. M., Grundy, P.,Kalikin, L. M., Harper, J. W., Elledge, S. J. and Feinberg, A. P. (1996).Imprinting of the gene encoding a human cyclin-dependent kinase inhibitor, p57-KIP2, on chromosome 11p15. Proc. Natl. Acad. Sci. USA 93, 3026-3030.

Meyer, E., Lim, D., Pasha, S., Tee, L. J., Rahman, F., Yates, J. R. W., Woods,C. G., Reik, W. andMaher, E. R. (2009). Germline mutation in NLRP2 (NALP2) ina familial imprinting disorder (Beckwith-Wiedemann Syndrome). PLoS Genet. 5,e1000423.

Miyamoto, T., Hasuike, S., Jinno, Y., Soejima, H., Yun, K., Miura, K., Ishikawa,M. and Niikawa, N. (2002). The human ASCL2 gene escaping genomicimprinting and its expression pattern. J. Assist. Reprod. Genet. 19, 240-244.

Mohammad, F., Mondal, T., Guseva, N., Pandey, G. K. and Kanduri, C. (2010).Kcnq1ot1 noncoding RNA mediates transcriptional gene silencing by interactingwith Dnmt1. Development 137, 2493-2499.

Pauler, F. M. and Barlow, D. P. (2006). Imprinting mechanisms–it only takes two.Genes Dev. 20, 1203-1206.

Quenneville, S., Verde, G., Corsinotti, A., Kapopoulou, A., Jakobsson, J.,Offner, S., Baglivo, I., Pedone, P. V., Grimaldi, G., Riccio, A. et al. (2011). Inembryonic stem cells, ZFP57/KAP1 recognize a methylated hexanucleotide toaffect chromatin and DNA methylation of imprinting control regions. Mol. Cell 44,361-372.

Sanchez, F. and Smitz, J. (2012). Molecular control of oogenesis. Biochim.Biophys. Acta 1822, 1896-1912.

Schultz, B. M., Gallicio, G. A., Cesaroni, M., Lupey, L. N. and Engel, N. (2015).Enhancers compete with a long non-coding RNA for regulation of the Kcnq1domain. Nucleic Acids Res. 43, 745-759.

Shin, J.-Y., Fitzpatrick, G. V. and Higgins, M. J. (2008). Two distinct mechanismsof silencing by the KvDMR1 imprinting control region. EMBO J. 27, 168-178.

Sleutels, F., Zwart, R. and Barlow, D. P. (2002). The non-coding Air RNA isrequired for silencing autosomal imprinted genes. Nature 415, 810-813.

Smallwood, S. A., Tomizawa, S., Krueger, F., Ruf, N., Carli, N., Segonds-Pichon, A., Sato, S., Hata, K., Andrews, S. R. and Kelsey, G. (2011). DynamicCpG island methylation landscape in oocytes and preimplantation embryos. Nat.Genet. 43, 811-814.

Smilinich, N. J., Day, C. D., Fitzpatrick, G. V., Caldwell, G. M., Lossie, A. C.,Cooper, P. R., Smallwood, A. C., Joyce, J. A., Schofield, P. N., Reik, W. et al.(1999). A maternally methylated CpG island in KvLQT1 is associated with anantisense paternal transcript and loss of imprinting in Beckwith- Wiedemannsyndrome. Proc. Natl. Acad. Sci. USA 96, 8064-8069.

Smith, E. Y., Futtner, C. R., Chamberlain, S. J., Johnstone, K. A. and Resnick,J. L. (2011). Transcription is required to establish maternal imprinting at thePrader-Willi syndrome and Angelman syndrome locus. PLoS Genet. 7,e1002422.

Soejima, H. and Higashimoto, K. (2013). Epigenetic and genetic alterations of theimprinting disorder Beckwith-Wiedemann syndrome and related disorders.J. Hum. Genet 58, 402-409.

Soellner, L., Begemann, M., Mackay, D. J., Gronskov, K., Tumer, Z., Maher,E. R., Karen Temple, I., Monk, D., Riccio, A., Linglart, A. et al. (2016). Recentadvances in imprinting disorders. Clin. Genet. 91, 3-13.

1829

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT

Spahn, L. and Barlow, D. P. (2003). An ICE pattern crystallizes. Nat. Genet. 35,11-12.

Stewart, K. R., Veselovska, L., Kim, J., Huang, J., Saadeh, H., Tomizawa, S.,Smallwood, S. A., Chen, T. and Kelsey, G. (2015). Dynamic changes in histonemodifications precede de novo DNA methylation in oocytes. Genes Dev.. 29,2449-2462.

Thorvaldsen, J. L., Duran, K. L. and Bartolomei, M. S. (1998). Deletion of the H19differentially methylated domain results in loss of imprinted expression of H19 andigf2. Genes Dev. 12, 3693-3702.

Tomizawa, S. and Sasaki, H. (2012). Genomic imprinting and its relevance tocongenital disease, infertility, molar pregnancy and induced pluripotent stem cell.J. Hum. Genet. 57, 84-91.

Tunster, S. J., Van de Pette, M. and John, R. M. (2011). Fetal overgrowth in theCdkn1c mouse model of Beckwith-Wiedemann syndrome. Dis. Model. Mech. 4,814-821.

Tunster, S. J., McNamara, G. I., Creeth, H. D. and John, R. M. (2016).Increaseddosage of the imprinted Ascl2 gene restrains two key endocrinelineages of the mouse placenta. Dev. Biol.

Tycko, B. and Morison, I. M. (2002). Physiological functions of imprinted genes.J. Cell. Physiol. 192, 245-258.

Verona, R. I., Mann, M. R. W. and Bartolomei, M. S. (2003). Genomic imprinting:intricacies of epigenetic regulation in clusters. Annu. Rev. Cell Dev. Biol. 19,237-259.

Veselovska, L., Smallwood, S. A., Saadeh, H., Stewart, K. R., Krueger, F.,Maupetit-Mehouas, S., Arnaud, P., Tomizawa, S., Andrews, S. and Kelsey, G.(2015). Deep sequencing and de novo assembly of the mouse oocytetranscriptome define the contribution of transcription to the DNA methylationlandscape. Genome Biol. 16, 209.

Wan, L.-B. and Bartolomei, M. (2008). Regulation of imprinting in clusters:noncoding RNAs versus insulators. Adv in Genetics. Long-RangeControl of Gene

Expression, Vol. 61 (ed. V. van Heyningen and R. Hill), pp. 207-223. Cambridge,MA: Elsevier.

Weksberg, R., Nishikawa, J., Caluseriu, O., Fei, Y. L., Shuman, C., Wei, C.,Steele, L., Cameron, J., Smith, A., Ambus, I. et al. (2001). Tumor developmentin the Beckwith-Wiedemann syndrome is associated with a variety ofconstitutional molecular 11p15 alterations including imprinting defects ofKCNQ1OT1. Hum. Mol. Genet. 10, 2989-3000.

Weksberg, R., Shuman, C., Wilkins-Haug, L., Mann, M., Croughan, M., Stewart,D., Rakowsky, C., Leader, A., Hall, J., Friedman, J. M. et al. (2007). Workshopreport: evaluation of genetic and epigenetic risks associated with assistedreproductive technologies and infertility. Fertil. Steril. 88, 27-31.

Weksberg, R., Shuman, C. and Beckwith, J. B. (2010). Beckwith-Wiedemannsyndrome. Eur. J. Hum. Genet. 18, 8-14.

White, C. R., Denomme, M. M., Tekpetey, F. R., Feyles, V., Power, S. G. A. andMann, M. R. W. (2015). High frequency of imprinted methylation errors in humanpreimplantation embryos. Sci. Rep. 5, 17311.

Wilkins, J. F. and Ubeda, F. (2011). Diseases associated with genomic imprinting.Prog. Mol. Biol. Transl. Sci. 101, 401-445.

Xiong, Z. and Laird, P. W. (1997). COBRA: a sensitive and quantitative DNAmethylation assay. Nucleic Acids Res. 25, 2532-2534.

Yan, Y., Frisen, J., Lee, M.-H., Massague, J. and Barbacid, M. (1997). Ablation ofthe CDK inhibitor p57KIP2 results in increased apoptosis and delayeddifferentiation during mouse development. Genes Dev. 11, 973-983.

Zhang, P., Liegeois, N., Wong, C., Finegold, M., Hou, H., Thompson, J. C.,Silverman, A., Harper, J. W., DePinho, R. A. et al. (1997). Altered celldifferentiation and proliferation in mice lacking p57KIP2 indicates a role for inBeckwith-Wiedemann syndrome. Nature 387, 151-158.

Zhang, Y., Jurkowska, R., Soeroes, S., Rajavelu, A., Dhayalan, A., Bock, I.,Rathert, P., Brandt, O., Reinhardt, R., Fischle, W. et al. (2010). Chromatinmethylation activity of Dnmt3a and Dnmt3a/3L is guided by interaction of the ADDdomain with the histone H3 tail. Nucleic Acids Res. 38, 4246-4253.

1830

RESEARCH ARTICLE Development (2017) 144, 1820-1830 doi:10.1242/dev.145136

DEVELO

PM

ENT