The H19 imprinted gene is silenced when paternally inherited and active only when inherited maternally. This is thought to involve a cis-acting control region upstream of H19 that is responsible for regulating a number of functions including DNA methylation, asynchronous replication of parental chromosomes and an insulator. Here we report on the function of a 1.2 kb upstream element in the mouse, which was previously shown to function as a bi-directional silencer in Drosophila. The cre-loxP-mediated targeted deletion of the 1.2 kb region had no effect on the maternal allele. However, there was loss of silencing of the paternal allele in many endodermal and other tissues. The pattern of expression was very similar to the expression pattern conferred by the enhancer elements downstream of H19. We could not detect an effect on the expression of the neighbouring imprinted Igf2 gene, suggesting that the proposed boundary element insulating this gene from the downstream enhancers was unaffected. Despite derepression of the paternal H19 allele, the deletion surprisingly did not affect the differential DNA methylation of the locus, which displayed an appropriate epigenetic switch in the parental germlines. Furthermore, the characteristic asynchronous pattern of DNA replication at H19 was also not disrupted by the deletion, suggesting that the sequences that mediate this were also intact. The silencer is therefore part of a complex cis-regulatory region upstream of the H19 gene and acts specifically to ensure the repression of the paternal allele, without a predominant effect on the epigenetic switch in the germline.

Genomic imprinting is fundamental to normal mammalian development. It involves the selective expression of an imprinted gene from a single parental allele resulting in functional differences between the parental genomes (Reik and Walter, 1998; Tilghman, 1999). The mechanism responsible for silencing of the reciprocal parental allele is thought to involve chromatin modifications (Ferguson-Smith et al., 1993; Hark and Tilghman, 1998; Khosla et al., 1999) and DNA methylation (Li et al., 1993). Specific cis control elements are apparently required to regulate the imprinting process, possibly by initiation of appropriate germ-line restricted epigenetic modifications (Surani, 1998). However, the precise mechanism by which these elements act is largely unknown.

The distal mouse chromosome 7 contains at least 12 imprinted genes, including H19 and Igf2, within a 1 Mb domain (Beechey and Cattanach, http://www.mgu.har.mrc.ac.uk). A 130 kb YAC from this domain, containing only the maternally expressed H19 and paternally expressed Igf2 genes, can imprint appropriately at ectopic loci (Ainscough et al., 1997), indicating that all the cis-acting control elements critical for imprinting are present within this region. Data from targeted deletion studies and H19 transgenic experiments have indicated that an imprinting control element lies in the 4 kb region immediately upstream of H19 (Srivastava et al., 2000; Elson and Bartolomei, 1997; Leighton et al., 1995a; Ripoche et al., 1997). This region contains a 2 kb differentially methylated domain (DMD), located between −2 and −4 kb from the start of transcription, which is required for parental-origin-specific silencing in the mouse (Thorvaldsen et al., 1998; Tremblay et al., 1997). We previously showed that a 1.2 kb element from the DMD functions as a cis-acting silencer of transgenic reporter genes in Drosophila (Lyko et al., 1997) and mice (Brenton et al., 1999).

To elucidate the role of this element in the imprinting of H19 and Igf2 in the context of the endogenous locus, we generated a targeted deletion of the 1.2 kb element, located between −1.7 and −2.9 kb from the transcription start site of H19. We carried out comprehensive in situ analysis on the expression of the H19 and Igf2 genes to determine the extent to which expression was affected in specific tissues. The deletion results in a relaxation of H19 silencing after paternal inheritance, without disrupting expression of Igf2. Strikingly, the deletion does not alter the DNA methylation status or asynchronous pattern of DNA replication of the H19 locus.

H19 silencer targeting construct

Genomic fragments were isolated from a λ2001 mouse 129/Sv library (a kind gift of A. Smith). The targeting vector (Fig. 1) carries a 1177 bp deletion of the BspEI-BamHI fragment located between −2.9 kb and −1.7 kb from the H19 transcription start site. It consists of a left arm of homology of 1.4 kb (SphI-BspEI), the loxP-flanked PGK-tkneo cassette (Schwartz et al., 1991) and a right arm of homology, encompassing the H19 gene, of 9.0 kb (BamHI-XbaI). The vector was linearised by NotI digestion.

Fig. 1.

Targeting of a 1.2 kb region upstream of H19. (a)Genomic structure of the mouse H19 locus. The H19, Igf2 and Rpl23 genes (white boxes) and their allele-specific transcription are shown (maternal chromosome is indicated above the line and the paternal chromosome below the line). An enlarged view of the H19 gene and surrounding sequence are also shown. Individual methylation-sensitive CfoI restriction enzyme sites in the H19 promoter and upstream regions are shown as unmethylated on the maternal chromosome (open circles) or methylated on the paternal chromosome (filled circles). Vertical arrows indicate DNaseI hypersensitivity sites. Four clustered maternal chromosome-specific hypersensitivity sites are located in the H19 upstream region. The two known H19 enhancers are located downstream of the H19 gene (large black circles). The 2 kb differentially methylated domain (DMD) and 1.2 kb region deleted in this study are also shown. B, BamHI; Bs, BspEI; H, HindIII; R, EcoRI. (b) Targeting construct used to replace a 1.2 kb region, between 1.7 and 2.9 kb upstream of H19. The transcriptional orientation of the PGK-tkneo (Neo) and H19 genes are indicated by horizontal arrows. The loxP sites flanking the Neo cassette are directionally represented by triangles. Dashed lines join the regions of homology shared by the endogenous locus and the targeting construct. Positions of restriction enzyme sites used and the size of fragments generated to detect a targeted replacement event are indicated. PCR primers for genotyping are also shown. P, PstI; H, HindIII; M, MunI; (MfeI); Bs, BspEI; B, BamHI; X, XbaI; Bg, BglII. (c) Southern hybridisation to confirm H19SAS genotype (+/−). Genomic DNA was digested with PstI or MunI and hybridised to the external probes A or B, respectively. (d) Genotyping by Southern hybridisation to confirm H19SilK genotype after deletion of the PGK-tkneo cassette from wild-type (+/+) and heterozygous (+/−) ES cells. Genomic DNA was digested with PstI and hybridised to probe A, which detected a 5.8 kb wild-type fragment, a 4.6 kb targeted deletion (SilK) fragment and a 3.3 kb fragment from the targeted replacement allele (SAS).

Fig. 1.

Targeting of a 1.2 kb region upstream of H19. (a)Genomic structure of the mouse H19 locus. The H19, Igf2 and Rpl23 genes (white boxes) and their allele-specific transcription are shown (maternal chromosome is indicated above the line and the paternal chromosome below the line). An enlarged view of the H19 gene and surrounding sequence are also shown. Individual methylation-sensitive CfoI restriction enzyme sites in the H19 promoter and upstream regions are shown as unmethylated on the maternal chromosome (open circles) or methylated on the paternal chromosome (filled circles). Vertical arrows indicate DNaseI hypersensitivity sites. Four clustered maternal chromosome-specific hypersensitivity sites are located in the H19 upstream region. The two known H19 enhancers are located downstream of the H19 gene (large black circles). The 2 kb differentially methylated domain (DMD) and 1.2 kb region deleted in this study are also shown. B, BamHI; Bs, BspEI; H, HindIII; R, EcoRI. (b) Targeting construct used to replace a 1.2 kb region, between 1.7 and 2.9 kb upstream of H19. The transcriptional orientation of the PGK-tkneo (Neo) and H19 genes are indicated by horizontal arrows. The loxP sites flanking the Neo cassette are directionally represented by triangles. Dashed lines join the regions of homology shared by the endogenous locus and the targeting construct. Positions of restriction enzyme sites used and the size of fragments generated to detect a targeted replacement event are indicated. PCR primers for genotyping are also shown. P, PstI; H, HindIII; M, MunI; (MfeI); Bs, BspEI; B, BamHI; X, XbaI; Bg, BglII. (c) Southern hybridisation to confirm H19SAS genotype (+/−). Genomic DNA was digested with PstI or MunI and hybridised to the external probes A or B, respectively. (d) Genotyping by Southern hybridisation to confirm H19SilK genotype after deletion of the PGK-tkneo cassette from wild-type (+/+) and heterozygous (+/−) ES cells. Genomic DNA was digested with PstI and hybridised to probe A, which detected a 5.8 kb wild-type fragment, a 4.6 kb targeted deletion (SilK) fragment and a 3.3 kb fragment from the targeted replacement allele (SAS).

Targeting in ES cells

Transformation of ES cells was by electroporation with linearised targeting construct. Optimal transformation was achieved using 8×106 R1 ES cells (Nagy et al., 1993) and 90 μg of targeting vector DNA at 250 V, 500 μF (Biorad Gene Pulser apparatus). ES cells were grown on neomycin-resistant feeders and selected for neomycin resistance with G418 (250 μg/ml) for 8 days. The surviving clones were expanded and passaged in 24-well plates. One plate was frozen and the other harvested for genomic DNA preparation as previously described (Torres and Kuhn, 1997). Homologous recombination was analysed in resistant ES clones initially using a multiplex PCR reaction, with annealing at 60°C. The homologous recombination event was detected with primers 5𠌒 to the left arm of homology (RADKO3=5𠌒-GGCTCCCCTGGATGTTTCACTTCC-3𠌒) and at the 3𠌒 end of the neomycin gene (RADKO1=5𠌒-CTATCGCCTTCTTGACGAGTTC-TTC-3𠌒) generating a 2.2 kb product. The wild-type H19 allele was identified by amplification from RADKO3 and a primer hybridising to the 5𠌒 end of the deleted 1177 bp region (H19-NotI=5𠌒-GCCCTTGGACATTGTCATGG-3𠌒) generating a 1.9 kb product. Positive clones (designated H19SAS) were further screened using a 5𠌒 external probe (probe A – 473bp HindIII fragment) and a 3𠌒 external probe (probe B – 800 bp XbaI fragment) by Southern blot analyses (Fig. 1), as previously described (Ainscough et al., 1997). Hybridisation signals were analysed using a Fuji PhosphorImager or on Kodak XAR film.

cre-mediated deletion of the selection cassette

The plasmid PGK-cre (a kind gift of Sam Aparicio) was used to transiently transfect correctly targeted H19SAS ES cells. Supercoiled plasmid (30 μg) was electroporated into 5×106 ES cells as described earlier. The transfected cells were cultured without selection. Colonies were picked on day 4 postelectroporation and expanded for characterisation as previously described. Genomic DNAs were probed with the 5𠌒 external probe A (HindIII fragment) on a Southern blot, to identify cre-mediated recombination between the loxP sites flanking the selection cassette (designated H19SilK).

Derivation and genotyping of targeted mice

Mutant ES cells were used to generate chimeras by aggregation (Wood et al., 1993). MF1 morulae were co-cultured with ES cells, and embryos transferred to pseudopregnant (C57BL/6 × CBA/Ca) F1 females. The chimeras derived were 100% 129/Sv strain by coat colour and GPI analysis and were propagated on the 129/Sv inbred background. Typing of the H19SilK mutant mice was by PCR, using primers that amplified across the remaining loxP site (RADKO11=5𠌒-GTCATGGGCTTCATGAGGCCAG-3𠌒 and RADKO10=5𠌒-TCCTG-CTTCACTTCAAACTAAG-3𠌒) annealing at 58°C and generating a0.6 kb product, on 1 μl crude lysates from neonatal toe DNA obtained by incubation in 50 μl Tris-HCl (50 mM, pH 8), NaCl (2 mM), EDTA (1 mM), 0.5% Tween-20, 0.5% NP40 and Proteinase K (0.4 μg/μl) for 1-3 hours at 65°C, or by Southern analysis (Fig. 1d).

RNA and expression analysis

RNA was prepared from embryos at day 13.5 of gestation or from neonatal tissues and allele-specific RT-PCR analysis performed as previously described (Ainscough et al., 1997). PCRs were performed for 30 cycles. 25 μl of each reaction was digested to completion with MspI, the fragments separated on a 2% agarose gel, transferred to Hybond N+ and hybridised as described earlier.

In situ hybridisation

Embryos at day 13.5 of gestation were wax embedded, sectioned at 10 μm, fixed in 4% paraformaldehyde in PBS for 30 minutes and subjected to ISH as described (Wilkinson and Nieto, 1993). A 2 kb mouse H19 (Poirier et al., 1991) and a 139 bp mouse Igf2 exon 4 cDNA clone were used to generate sense and antisense probes by in vitro transcription using the digoxigenin RNA labelling kit (Boehringer Mannheim). Detection of the signal was with the BM Purple substrate (Boehringer Mannheim). Counterstaining was with 0.5% Eosin.

DNA and methylation analysis

DNA was prepared, using standard procedures, from: (a) decapitated embryos at day 13.5 of gestation, day 3 neonatal tissues or adult tissues for methylation analysis, (b) embryonic or neonatal heads for genotyping by Southern analysis and (c) embryonic or neonatal limbs for rapid PCR genotyping. For methylation analysis, 20 μg DNA was digested to completion with appropriate restriction enzymes, separated on a 1.0% agarose gel, transferred to a Hybond N+ membrane and hybridised with the appropriate probe. Bisulphite genomic sequencing was performed as described previously (Olek and Walter, 1997).

Fluorescence in situ hybridisation

FISH was performed on cultured splenocytes from age-matched adult mice as previously described (Boggs and Chinault, 1997). Cosmid DNA was labelled by nick translation with biotin (cAH) or digoxigenin (c17.2) and hybridisation performed as previously described (Boggs and Chinault, 1997). Biotinylated probes were detected with Neutravidin-Alexa™ 488 (Molecular Probes), amplified with biotinylated goat anti-avidin antibody (Vector Laboratories) and a further layer of Neutravidin-Alexa™ 488. Digoxigenin-labelled probes were detected with sheep anti-digoxigenin antibody (Boehringer Mannheim), followed by rabbit Texas Red-conjugated anti-sheep antibody (Vector Laboratories) and a final layer of Alexa™468 anti-rabbit antibody (Molecular Probes). Images were obtained using a Zeiss Axiophot epifluorescence microscope and SmartCapture VP (Vysis) digital imaging software.

Targeting the H19 upstream region

The targeting construct (Fig. 1 and Methods) was used to create a deletion between −1.7 kb and −2.9 kb relative to the H19 transcriptional start site by homologous recombination. The H19 gene and the region immediately 5𠌒 to the gene remained intact, including part of the differentially methylated region at the promoter (Bartolomei et al., 1993; Ferguson-Smith et al., 1993) and a G-rich repeat containing region (Tremblay et al., 1997). A region 5𠌒 to the deletion was also preserved, which forms part of the upstream DMD (Tremblay et al., 1997; Fig. 1a). A correct targeting event replaced the 1.2 kb region with a 2.9 kb loxP-flanked selection cassette (Schwartz et al., 1991; H19SAS locus) which was subsequently deleted in ES cells by transient expression of cre recombinase (Fig. 1). The resulting deletion, termed H19SilK, was transmitted into 129/Sv mice.

H19SilK deletion reactivates H19 expression after paternal inheritance

To analyse allele-specific expression of H19, we first used a MspI polymorphism in exon 1 that detects a 321bp BALB/c-specific fragment, which was not present in the H19SilK allele on the 129/Sv genetic background (Fig. 2a; Ainscough et al., 1997). The expression of the H19SilK paternal allele was therefore examined in (H19SilK × BALB/c) embryos. Compared to the wild-type controls (+/+) that expressed only the maternally inherited (BALB/c) allele of H19, expression of both parental alleles was detected in the embryos with a mutant locus (+/−; Fig. 2b). Although not strictly quantitative, Southern analysis combined with phosphorimager densitometry suggested that the paternally inherited H19SilK allele was expressed at a level of approximately 50% of the wild-type maternally inherited allele (Fig. 2b). Since the 321 bp probe used was internal to both of the allele-specific fragments, it should hybridise to each with equal affinity, which provides a reasonable measure of the relative amounts of each transcript. It was therefore possible that activation of the mutant paternal allele was either partial or restricted to a subset of cells (see below).

Fig. 2.

Expression of the H19SilK allele after paternal and maternal inheritance by RT-PCR. (a) An MspI polymorphism in exon 1 of H19 between 129/Sv and BALB/c strains was used to analyse gene expression. The H19SilK allele was from the 129/Sv strain (440 bp band), while the wild-type H19 allele was from BALB/c, which has the additional MspI site (321 bp band). Genotypes are indicated with the maternal allele listed first. RT-PCR analysis was performed on day 13.5 embryos and the probe used for Southern analysis is indicated by thick black bar. (b) Paternal transmission of the H19SilK allele. Heterozygous embryos (+/−) exhibited bi-allelic H19 expression. MspI digested reaction products were hybridised with the 321 bp probe shown in A. Wild-type littermates (+/+) only expressed the maternally inherited allele. Control lanes contain genomic DNA.

Fig. 2.

Expression of the H19SilK allele after paternal and maternal inheritance by RT-PCR. (a) An MspI polymorphism in exon 1 of H19 between 129/Sv and BALB/c strains was used to analyse gene expression. The H19SilK allele was from the 129/Sv strain (440 bp band), while the wild-type H19 allele was from BALB/c, which has the additional MspI site (321 bp band). Genotypes are indicated with the maternal allele listed first. RT-PCR analysis was performed on day 13.5 embryos and the probe used for Southern analysis is indicated by thick black bar. (b) Paternal transmission of the H19SilK allele. Heterozygous embryos (+/−) exhibited bi-allelic H19 expression. MspI digested reaction products were hybridised with the 321 bp probe shown in A. Wild-type littermates (+/+) only expressed the maternally inherited allele. Control lanes contain genomic DNA.

Maternal inheritance of the H19SilK deletion does not disrupt H19 expression

The reciprocal mating (Balb/c × H19SilK) was used to examine H19 expression after maternal inheritance of the targeted deletion allele using RT-PCR analysis on day 13.5 embryos. Expression was only detected from the maternal allele in both mutant (−/+) and wild-type (+/+) embryos (data not shown). This observation was confirmed by in situ hybridisation, which revealed a pattern of expression indistinguishable from a wild-type maternal locus allele (Fig. 3a,b).

Fig. 3.

Expression of mutant H19 alleles in day 13.5 embryos. Expression of H19 in sections of embryos at day 13.5 of gestation was examined by in situ hybridisation. (a) H19 expression in wild-type (+/+) embryo. (b) H19 Expression in embryos from the maternally inherited H19SilK allele in heterozygous (H19SilK/+) embryos; expression in these embryos is indistinguishable from the pattern in wild-type litter mates shown in a. (c) Expression from the H19 null (H19Δ3) allele inherited maternally (H19Δ3/+); there is no detectable expression of H19. (d) Expression from the paternally inherited H19SilK allele examined in embryos with a maternally inherited H19Δ3 null allele (H19Δ3/H19SilK); the expression of the H19SilK paternal allele is restricted to specific tissues, unlike the maternally inherited allele shown in b. G, gut; H, heart; L, liver; Lu, lung; Sc, Sclerotome; T, tongue; Th, thymus. Scale bar in a = 1 mm.

Fig. 3.

Expression of mutant H19 alleles in day 13.5 embryos. Expression of H19 in sections of embryos at day 13.5 of gestation was examined by in situ hybridisation. (a) H19 expression in wild-type (+/+) embryo. (b) H19 Expression in embryos from the maternally inherited H19SilK allele in heterozygous (H19SilK/+) embryos; expression in these embryos is indistinguishable from the pattern in wild-type litter mates shown in a. (c) Expression from the H19 null (H19Δ3) allele inherited maternally (H19Δ3/+); there is no detectable expression of H19. (d) Expression from the paternally inherited H19SilK allele examined in embryos with a maternally inherited H19Δ3 null allele (H19Δ3/H19SilK); the expression of the H19SilK paternal allele is restricted to specific tissues, unlike the maternally inherited allele shown in b. G, gut; H, heart; L, liver; Lu, lung; Sc, Sclerotome; T, tongue; Th, thymus. Scale bar in a = 1 mm.

Paternal H19 reactivation is restricted

To gain further insight into the nature of the derepression of the paternal H19SilK locus, we also used in situ hybridisation analysis in embryos carrying a null H19 allele from H19Δ3 females (Ripoche et al., 1997). As expected, almost no expression was detected from the wild-type paternal H19 locus (Fig. 3c), except in some embryos where expression was detected from a small subset of cells in the liver, as previously described (Jouvenot et al., 1999). By contrast, paternal inheritance of the H19SilK locus showed widespread, but not full, H19 expression (Fig. 3d). Detailed examination of this restricted reactivated expression revealed that, in many tissues, H19 was only expressed in a subset of cells, when compared to the normal expression pattern of the wild-type or H19SilK maternally inherited alleles.

In the gut, for example, H19 is normally expressed in both the endodermally derived gastric epithelium and the mesodermally derived smooth muscle cells (Fig. 4a). However, the paternal H19SilK allele exhibited strong expression in the epithelial cells only (Fig. 4b). Similarly, in the stomach, there was no expression from the paternal H19SilK allele in the smooth muscle (Fig. 4c,d). Restricted expression was also observed in the epithelial cells of bronchi in the lung, but not in the surrounding mesenchymal cells (Fig. 4e,f). Endoderm-derived epithelial cell expression was also seen in the mesonephric tubules in the developing kidney, in the oesophagus and in the olfactory cavity (data not shown). Subsets of cells within the liver, tongue and the developing oral cavity also all showed expression. H19 appeared to be active in the muscle cells of the tongue, but absent from the interspersed adipose and neural tissues and Meckel’s cartilage in the developing lower jaw (Fig. 4h). High-level expression was also detected in the sclerotome. This expression pattern was consistently observed in all embryos without evidence of variegated expression. These observations revealed that the mutant H19SilK locus was expressed in specific cell types and tissues. The observations argue against an equal overall lower level reactivation in all tissues. Furthermore, the restricted reactivated expression pattern is similar to the expression conferred by the two previously characterised enhancers downstream of H19 (Leighton et al., 1995b; Brenton et al., 1999; see Discussion).

Fig. 4.

Detailed expression pattern of the H19SilK maternal and paternal alleles. Expression from the maternally inherited H19SilK allele in (H19SilK/+) heterozygous embryos (a,c,e,g); this expression is comparable to that from the wild-type maternal H19 allele (see Fig. 3). Expression from the paternally inherited H19SilK allele in embryos with the maternally inherited H19Δ3 null allele in (H19Δ3/H19SilK) heterozygous embryos (b,d,f,h). The panels show expression in the gut (a,b), stomach (c,d), lung (e,f) and tongue (g,h). C, Meckel’s cartilage; E, epithelium; H, heart; M, muscle; T, tongue. Scale bar in f = 0.1 mm for a-f; in h = 1.0 mm for g,h.

Fig. 4.

Detailed expression pattern of the H19SilK maternal and paternal alleles. Expression from the maternally inherited H19SilK allele in (H19SilK/+) heterozygous embryos (a,c,e,g); this expression is comparable to that from the wild-type maternal H19 allele (see Fig. 3). Expression from the paternally inherited H19SilK allele in embryos with the maternally inherited H19Δ3 null allele in (H19Δ3/H19SilK) heterozygous embryos (b,d,f,h). The panels show expression in the gut (a,b), stomach (c,d), lung (e,f) and tongue (g,h). C, Meckel’s cartilage; E, epithelium; H, heart; M, muscle; T, tongue. Scale bar in f = 0.1 mm for a-f; in h = 1.0 mm for g,h.

The effect of the H19SilK deletion on growth and Igf2 expression

It has been shown previously that expression of the closely associated and reciprocally imprinted Igf2 gene is virtually identical to H19 as both genes share the same enhancers (Leighton et al., 1995b; Webber et al., 1998) and that targeted deletions at the H19/Igf2 domain can have an effect on the growth of mice (Leighton et al., 1995a,b; Ripoche et al., 1997; Thorvaldsen et al., 1998). We therefore examined the effect of the H19SilK deletion on growth and Igf2 expression. A study of the mass of animals carrying the silencer deletion allele, revealed a weight phenotype on the inbred 129/Sv genetic background. Mice inheriting the mutant allele were weighed 2 days after birth and then approximately every 30 days up to 3 months of age. When compared to wild-type littermates, mice inheriting the mutation maternally were on average 12% heavier (±s.e.m. 1.26%, n=44, P<0.003) while mice inheriting the mutant allele paternally were 12% smaller (±s.e.m. 2.09%, n=90, P<0.05).

In order to determine whether this growth phenotype was a result of changes in Igf2 expression, we examined Igf2 expression from the H19SilK locus by in situ hybridisation and northern analysis in day 13.5 embryos, using a probe specific to exon 4 of Igf2. To ensure that we only detected expression from the Igf2 gene in cis to the H19SilK allele, heterozygous H19SilK mice were crossed to Igf2−?− mice (DeChiara et al., 1990). There was no detectable effect on Igf2 expression in cis to the H19SilK allele after maternal or paternal transmission (data not shown). The phenotypic growth effect that we observe could be a result of a very minor, and hence undetectable, modulation of Igf2 levels. Indeed, in one targeted deletion of the H19 gene, an increase in Igf2 levels of only 17% specifically in skeletal muscle was associated with an overgrowth phenotype (Schmidt et al., 1999). It is, however, of interest to note that in other models in which changes in Igf2 levels are associated with growth phenotypes, H19 transcript levels are also altered (Jones et al., 1998; Leighton et al., 1995a; Ripoche et al., 1997; Thorvaldsen et al., 1998). Our data may therefore raise the possibility of a functional role for the H19 RNA in an unknown growth pathway.

H19SilK deletion does not affect differential methylation

One of the key aspects concerning imprinting is how the imprint is initiated and maintained. We therefore examined whether the 1.2 kb deletion affected this process. It has been demonstrated that the CpG residues in the promoter region and upstream of the H19 transcription unit are differentially methylated on parental chromosomes throughout development (Bartolomei et al., 1993; Ferguson-Smith et al., 1993; Tremblay et al., 1995, 1997). The upstream region (termed DMD) extends from −2 to −4 kb upstream of H19. Therefore, in the H19SilK locus, some 1.1 kb of the DMD remains (Fig. 1a). We analysed these regions using the methylation sensitive restriction enzyme CfoI (Fig. 5a). We found that the methylation pattern of the paternal H19SilK locus was in fact similar to the wild-type locus, both in tissues where H19 expression is high (liver) as well as in tissues where the expression is confined to very few cells (brain; Fig. 5b,c). Maternal inheritance of the H19SilK allele revealed a consistent and appropriate hypomethylated epigenotype, which was comparable to the wild-type maternal locus (Fig. 5b,c).

Fig. 5.

Methylation analysis of the maternally and paternally inherited H19SilK allele. (a) The H19 locus with the location of CfoI sites (open circles) known to show differential methylation between parental alleles. H19 transcription start site is represented by a horizontal arrow and the two probes used are indicated by bold lines (RBs and RX). The positions of restriction enzyme sites used to generate probes are indicated. The dashed line represents the deleted region in the H19SilK allele and the black triangle the remaining loxP site. Restriction endonuclease recognition sites are abbreviated as follows: R, EcoRI; Bs, BspEI; B, BamHI; X, XbaI. Scale bar indicates 1 kb. (b) Methylation state of the maternally (Mat) or paternally (Pat) inherited H19SilK allele, and in wild-type (WT) day 13.5 embryos in the upstream DMD region (RBs probe). Methylation analysis in neonatal or adult tissues were; liver (L), brain (B), heart (H), skeletal muscle (M) and sperm (S). The DNAs were digested with EcoRI (−) or EcoRI and the methylation sensitive enzyme CfoI (+). Two independent samples are shown for digests on whole embryos and sperm. The wild-type allele generates a methylated band at 3781 bp and the methylated H19SilK allele a band at 2604 bp. A completely unmethylated state in this region would allow digestion by CfoI to produce bands at 386 bp, 376 bp and 180 bp in size. (c) Methylation analysis of the promoter region of H19 (RX probe). The same size bands as in B are generated for methylated alleles. A completely unmethylated state would produce CfoI generated bands of 959 bp and 79 bp (not detectable on blot). Quantitation of single bands was performed using Scion Image software and the percentage of DNA in each lane which is found in the demethylated band of 959 bp compared to the methylated H19SilK allele band is indicated. For maternal inheritance a fully demethylated allele would therefore give a value of 100% and for paternal inheritance a fully methylated allele gives a value of 50%. (d) Bisulphite genomic sequencing analysis of upstream DMD region. The region analysed in detail in this study (grey box) extends from −4000 to −3440 relative to the H19 transcriptional start site (horizontal arrow). CpG methylation patterns of individual chromosomes derived from sequencing are shown as single lines. Methylated and unmethylated CpG dinucleotides are represented by filled and open circles, respectively. CpGs that were not read unambiguously (dashed lines) were omitted from the figure. Methylation patterns of chromosomes from wild-type (WT), paternal H19SilK heterozygotes (Pat) and maternal H19SilK heterozygotes (Mat) are shown.

Fig. 5.

Methylation analysis of the maternally and paternally inherited H19SilK allele. (a) The H19 locus with the location of CfoI sites (open circles) known to show differential methylation between parental alleles. H19 transcription start site is represented by a horizontal arrow and the two probes used are indicated by bold lines (RBs and RX). The positions of restriction enzyme sites used to generate probes are indicated. The dashed line represents the deleted region in the H19SilK allele and the black triangle the remaining loxP site. Restriction endonuclease recognition sites are abbreviated as follows: R, EcoRI; Bs, BspEI; B, BamHI; X, XbaI. Scale bar indicates 1 kb. (b) Methylation state of the maternally (Mat) or paternally (Pat) inherited H19SilK allele, and in wild-type (WT) day 13.5 embryos in the upstream DMD region (RBs probe). Methylation analysis in neonatal or adult tissues were; liver (L), brain (B), heart (H), skeletal muscle (M) and sperm (S). The DNAs were digested with EcoRI (−) or EcoRI and the methylation sensitive enzyme CfoI (+). Two independent samples are shown for digests on whole embryos and sperm. The wild-type allele generates a methylated band at 3781 bp and the methylated H19SilK allele a band at 2604 bp. A completely unmethylated state in this region would allow digestion by CfoI to produce bands at 386 bp, 376 bp and 180 bp in size. (c) Methylation analysis of the promoter region of H19 (RX probe). The same size bands as in B are generated for methylated alleles. A completely unmethylated state would produce CfoI generated bands of 959 bp and 79 bp (not detectable on blot). Quantitation of single bands was performed using Scion Image software and the percentage of DNA in each lane which is found in the demethylated band of 959 bp compared to the methylated H19SilK allele band is indicated. For maternal inheritance a fully demethylated allele would therefore give a value of 100% and for paternal inheritance a fully methylated allele gives a value of 50%. (d) Bisulphite genomic sequencing analysis of upstream DMD region. The region analysed in detail in this study (grey box) extends from −4000 to −3440 relative to the H19 transcriptional start site (horizontal arrow). CpG methylation patterns of individual chromosomes derived from sequencing are shown as single lines. Methylated and unmethylated CpG dinucleotides are represented by filled and open circles, respectively. CpGs that were not read unambiguously (dashed lines) were omitted from the figure. Methylation patterns of chromosomes from wild-type (WT), paternal H19SilK heterozygotes (Pat) and maternal H19SilK heterozygotes (Mat) are shown.

To confirm the ability of the hypomethylated H19SilK allele to undergo re-methylation in the male germline, we examined adult male heterozygotes that had inherited the deletion through the female germline. As expected, the H19SilK allele in this animal was predominantly hypomethylated in all somatic tissues examined (Fig. 5b,c). However, in mature sperm, both the upstream DMD and promoter-proximal regions were methylated (Fig. 5b,c). The combined evidence showed that the mutant locus was therefore capable of an appropriate epigenetic switch as demonstrated by its hypomethylated state after transmission through the female germline and the acquisition of methylation in the male germline.

The germline-specific changes in methylation were confirmed by a study of differential methylation on individual chromosomes by bisulphite genomic sequencing of a 560 bp region of the remaining DMD (Fig. 5d). This region harbours 16 CpG dinucleotides that have been previously shown to exhibit extensive differential methylation (Olek and Walter, 1997; Tremblay et al., 1997). As expected, individual chromosomes from this region were either predominantly hypermethylated or hypomethylated in wild-type embryos (Fig. 5d). No variation from this pattern was detected on chromosomes from embryos inheriting the paternal H19SilK or maternal H19SilK locus (Fig. 5d), confirming that the H19SilK allele was methylated after paternal transmission and demethylated after maternal transmission.

Asynchronous DNA replication is not disrupted at the H19SilK locus

Another characteristic epigenetic feature of imprinted domains is asynchronous DNA replication, with early replication of the paternal chromosome (Kitsberg et al., 1993; Simon et al., 1999). At the wild-type H19 locus, a cosmid covering the H19 gene and upstream region detects an asynchronous pattern of DNA replication between the two chromosomes (Greally et al., 1998; Fig. 6a). Conversely, a cosmid that maps to the Rpl23 gene approximately 30 kb downstream of H19 and outside of the imprinted domain, detects a synchronous DNA replication pattern (Fig. 6a). The asynchronous pattern of replication is disrupted when the entire 10 kb region upstream of H19 is deleted (Greally et al., 1998). In contrast, we found that our 1.2 kb deletion did not affect the asynchronous replication pattern at H19 after paternal or maternal inheritance of the H19SilK locus (Fig. 6b), demonstrating that the epigenetic mark regulating asynchronous replication was unaffected.

Fig. 6.

DNA replication patterns at wild-type and mutant alleles. (a) Genomic map showing the location of cosmids, cAH and c17.2, used to analyse the timing of DNA replication at the H19 and Rpl23 genes, respectively. Scale bar indicates 10 kb. (b) The percentage of cells showing each pattern of hybridization, single/single (SS), single/double (SD) or double/double (DD), is illustrated for wild-type (black), paternal H19SilK heterozygous (white) and maternal H19SilK heterozygous (grey) cells, expressed as a percentage of BrdU-positive cells. The cAH cosmid upstream from H19 detected asynchronous replication in wild-type and mutant cells, manifested by proportions of cells with single/double patterns of 24% or greater and cells with double/double patterns of less than 19%. Thiscontrasts with the pattern for the downstream Rpl23-containing c17.2 cosmid which only detects a background rate of 15-16% single/double patterns and a proportion with double/double patterns of greater than 23% in both wild-type and mutant cells. This indicates that the transition in replication patterns between H19 and Rpl23 is not disrupted at the mutant H19SilK allele. Numbers of cells examined were; H19; WT (n=520), paternal H19SilK (n=287), maternal H19SilK (n=384). Rpl23; WT (n=260), paternal H19SilK (n=106), maternal H19SilK (n=110)

Fig. 6.

DNA replication patterns at wild-type and mutant alleles. (a) Genomic map showing the location of cosmids, cAH and c17.2, used to analyse the timing of DNA replication at the H19 and Rpl23 genes, respectively. Scale bar indicates 10 kb. (b) The percentage of cells showing each pattern of hybridization, single/single (SS), single/double (SD) or double/double (DD), is illustrated for wild-type (black), paternal H19SilK heterozygous (white) and maternal H19SilK heterozygous (grey) cells, expressed as a percentage of BrdU-positive cells. The cAH cosmid upstream from H19 detected asynchronous replication in wild-type and mutant cells, manifested by proportions of cells with single/double patterns of 24% or greater and cells with double/double patterns of less than 19%. Thiscontrasts with the pattern for the downstream Rpl23-containing c17.2 cosmid which only detects a background rate of 15-16% single/double patterns and a proportion with double/double patterns of greater than 23% in both wild-type and mutant cells. This indicates that the transition in replication patterns between H19 and Rpl23 is not disrupted at the mutant H19SilK allele. Numbers of cells examined were; H19; WT (n=520), paternal H19SilK (n=287), maternal H19SilK (n=384). Rpl23; WT (n=260), paternal H19SilK (n=106), maternal H19SilK (n=110)

We have addressed the function of a 1.2 kb region located between 1.7 kb and 2.9 kb upstream of the imprinted H19 gene, by targeted deletion at the endogenous murine locus. This element was originally identified as a silencer in Drosophila (Lyko et al., 1997), but its function in imprinting was unknown. Using extensive in situ analysis at the level of single cells, we demonstrated that its deletion in the mouse results in a tissue-specific relaxation of silencing after paternal transmission, but it has no effect after maternal inheritance. Strikingly, this loss of H19 silencing occurred without a detectable change in the methylation status or asynchronous DNA replication pattern of the locus. This suggests that the imprinted epigenotype of H19 may be initiated by a different cis-acting element to that responsible for transcriptional silencing of the gene.

Identification of a cis-acting silencer at H19

The deletion of a 1.2 kb element from the region upstream of the imprinted H19 gene resulted in reactivation of the normally silent paternally inherited H19 allele. One possibility was that reactivation of the paternal H19 gene occurs in all tissues to the same extent. However, close examination revealed that the loss of silencing appears to be restricted to specific cells and tissues. Furthermore, the pattern of expression resembles that conferred by the two known enhancers downstream of H19 (Ainscough et al., 1997; Brenton et al., 1999; Leighton et al., 1995b), as well as by some unknown enhancers that are also present on a 130 kb YAC (Ainscough et al., 2000). The two H19 enhancers downstream of H19 were previously shown to have an open chromatin configuration regardless of their parental origin, indicating that they have the potential to function on both parental chromosomes (Bartolomei et al., 1993). In the absence of the silencer, these enhancers are apparently able to activate expression of the paternal H19 allele. The apparent tissue-specific effect of the silencer and its potential interaction with only a subset of enhancers is intriguing. The identification of a mesoderm-specific silencer at the Igf2 gene suggests that tissue-specific expression of imprinted genes may be regulated by specific interactions between silencers and enhancers (M. Constancia and W. Reik, personal communication). The mechanism by which such interactions could occur is unknown at present.

Loss of silencing is independent of DNA methylation

DNA methylation is closely associated with silencing of the paternally inherited wild-type H19 allele. Examination of the methylation state of the H19SilK allele after paternal inheritance revealed that methylation remained unchanged both upstream of the deletion and at the promoter-proximal region of H19, even in tissues where H19 was reactivated. The 1.2 kb deleted region constitutes a significant part of the H19 upstream DMD. However, its deletion apparently had no effect on many of the key functions of this region. Most striking amongst these was the ability of the mutant locus to undergo an appropriate epigenetic switch in the germline as demonstrated by differential DNA methylation, indicating that the signal for both the initiation and propagation of the imprint was unaffected. This observation dissociates the commonly held view of an intrinsic link between differential methylation and transcriptional regulation at imprinted loci. There is a precedent for allele-specific expression of an imprinted gene independent of methylation at the mouse Mash2 locus (Caspary et al., 1998), suggesting that differences in DNA methylation at some imprinted loci may not be essential to the imprinting process.

Another characteristic epigenetic mark established in the germline is the asynchronous replication of imprinted domains, including the H19 region (Kitsberg et al., 1993; Simon et al., 1999). The sequences upstream of H19 are essential for the maintenance of this replication pattern (Greally et al., 1998). However, the 1.2 kb deletion at the H19SilK locus did not disrupt this asynchronous replication, suggesting that the critical sequences for this epigenetic modification are also located elsewhere in the region. It would be of interest to examine the effect of the larger deletion of the DMD (Thorvaldsen et al., 1998) on the DNA replication pattern at the H19 locus, which may help to pinpoint the sequences required for asynchronous replication of the region.

The fate of the insulator upstream of H19

The weight phenotype that we observed in mice carrying the 1.2 kb silencer deletion possibly suggests an effect on the Igf2 gene. However, despite examining expression using a number of different methods, we could not detect a change in Igf2 expression. It is possible that a small, but undetectable modulation in Igf2 levels was sufficient to induce the growth phenotype, as has been shown previously (Schmidt et al., 1999). This is particularly likely in the inbred 129/Sv background mice that we have used here. However, the absence of a marked and detectable effect on Igf2 suggests that the proposed insulator or boundary element upstream of the H19 gene is unaffected in the H19SilK allele (Greally et al., 1998; Khosla et al., 1999; Webber et al., 1998). As a result, there is no large-scale activation of the Igf2 gene on the maternal chromosome by the enhancers downstream of H19. There are nuclease hypersensitivity sites within the DMD on the maternal chromosome of unknown function (Hark and Tilghman, 1998; Khosla et al., 1999; Fig. 1a). The1.2 kb deletion leaves 1.1 kb of the DMD intact (Fig. 1a), including some hypersensitive sites that may constitute the insulator. Indeed, recent studies have identified a number of CTCF protein binding sites that map close to some of these hypersensitivity sites, suggesting that at the endogenous locus these sequences are important for the insulator function (Szabo et al., 2000; Bell and Felsenfeld, 2000; Hark et al., 2000).

Evidence for multifunctional cis elements upstream of H19

The 5𠌒 1.1 kb region of the DMD may also be required for at least some of the critical imprinting functions, such as the initiation and/or propagation of the imprint and the asynchronous replication of the region. This notion is supported by comparing the 1.6 kb deletion of the DMD previously reported (Thorvaldsen et al., 1998), which apparently resulted in a loss of the appropriate epigenetic switch of the locus in the germline. This larger deletion also had a very marked effect on Igf2 expression, suggesting that the insulator may also have been deleted. The precise role of the region upstream of the 1.2 kb silencer element needs to be addressed by deleting it from the endogenous locus. In this way, it should be possible to dissect the multiple functions of the DMD region. We have also recently identified three transcripts at −10.5, −8.5 and −3.0 kb upstream of the H19 gene, but the role of these sequences in the imprinting mechanism is as yet unknown.

There is evidence suggesting that control elements are often complex and can display both functional redundancy as well as divergent functions within control regions (Zhou and Levine, 1999). A multifunctional role for the upstream region of H19, including a silencer, imprint initiation and insulator, is very reminiscent of regulatory systems in other organisms, such as the mating type locus silencing in yeast (Donze et al., 1999; Fourel et al., 1999) and Hox gene regulation in Drosophila (Hagstrom et al., 1997; Mihaly et al., 1997). We are particularly interested in the mechanism by which silencer elements from imprinting control regions also function as silencers in Drosophila. Such an activity has not only been detected for the 1.2 kb H19 upstream region, but also the imprinting centre from the SNRPN gene (Lyko et al., 1998), suggesting an evolutionary conserved epigenetic silencing mechanism in flies and mice.

R. A. D. was funded by a Wellcome Trust Prize Studentship and Wellcome Trust Prize Fellowship. J. D. B. received a Cancer Research Campaign Research Fellowship for a Clinician [CRC]. K. L. A. was funded by an Elmore Research Studentship. The work was supported by a grant from the Wellcome Trust to M. A. S. We thank W. Reik and M. Constancia for discussion and comments on the manuscript. We are grateful to A. Efstratiadis for providing the Igf2 deletion mice, J. Greally for the c17.2 cosmid and the laboratory of J. Walter for providing PCR primers for bisulphite genomic sequencing.

Ainscough
,
J. F.
,
Koide
,
T.
,
Tada
,
M.
,
Barton
,
S.
and
Surani
,
M. A.
(
1997
).
Imprinting of Igf2 and H19 from a 130 kb YAC transgene
.
Development
124
,
3621
3632
.
Ainscough
,
J. F.-X.
,
Dandolo
,
L.
and
Surani
,
M. A. H.
(
2000
).
Appropriate expression of the mouse H19 gene utilises three or more distinct enhancer regions spread over more than 130 kb
.
Mech. Dev.
91
,
368
372
.
Bartolomei
,
M. S.
,
Webber
,
A. L.
,
Brunkow
,
M. E.
and
Tilghman
,
S. M.
(
1993
).
Epigenetic mechanisms underlying the imprinting of the mouse H19 gene
.
Genes Dev.
7
,
1663
1673
.
Bell
,
A. C.
and
Felsenfeld
,
G.
(
2000
).
Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene
.
Nature
405
,
482
485
.
Boggs
,
B. A.
and
Chinault
,
A. C.
(
1997
).
Analysis of DNA replication by fluorescence in situ hybridization
.
Methods
13
,
259
270
.
Brenton
,
J. D.
,
Drewell
,
R. A.
,
Viville
,
S.
,
Hilton
,
K. J.
,
Barton
,
S. C.
,
Ainscough
,
J. F.-X.
and
Surani
,
M. A.
(
1999
).
A silencer element identified in Drosophila is required for imprinting of H19 reporter transgenes in mice
.
Proc. Natl Acad. Sci. USA
96
,
9242
9247
.
Caspary
,
T.
,
Cleary
,
M. A.
,
Baker
,
C. C.
,
Guan
,
X. J.
and
Tilghman
,
S. M.
(
1998
).
Multiple mechanisms regulate imprinting of the mouse distal chromosome 7 gene cluster
.
Mol. Cell. Biol.
18
,
3466
3474
.
DeChiara
,
T. M.
,
Efstratiadis
,
A.
and
Robertson
,
E. J.
(
1990
).
A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting
.
Nature
345
,
78
80
.
Donze
,
D.
,
Adams
,
C. R.
,
Rine
,
J.
and
Kamakaka
,
R. T.
(
1999
).
The boundaries of the silenced HMR domain in Saccharomyces cerevisiae
.
Genes Dev.
13
,
698
708
.
Elson
,
D. A.
and
Bartolomei
,
M. S.
(
1997
).
A 5𠌒 differentially methylated sequence and the 3𠌒-flanking region are necessary for H19 transgene imprinting
.
Mol. Cell. Biol.
17
,
309
317
.
Ferguson-Smith
,
A. C.
,
Sasaki
,
H.
,
Cattanach
,
B. M.
and
Surani
,
M. A.
(
1993
).
Parental-origin-specific modification of the mouse H19 gene
.
Nature
362
,
751
755
.
Fourel
,
G.
,
Revardel
,
E.
,
Koering
,
C. E.
and
Gilson
,
E.
(
1999
).
Cohabitation of insulators and silencing elements in yeast subtelomeric regions
.
EMBO J.
18
,
2522
2537
.
Greally
,
J. M.
,
Starr
,
D. J.
,
Hwang
,
S.
,
Song
,
L.
,
Jaarola
,
M.
and
Zemel
,
S.
(
1998
).
The mouse H19 locus mediates a transition between imprinted and non-imprinted DNA replication patterns
.
Hum. Mol. Genet.
7
,
91
96
.
Hagstrom
,
K.
,
Muller
,
M.
and
Schedl
,
P.
(
1997
).
A Polycomb and GAGA dependent silencer adjoins the Fab-7 boundary in the Drosophila bithorax complex
.
Genetics
146
,
1365
1380
.
Hark
,
A. T.
,
Schoenherr
,
C. J.
,
Katz
,
D. J.
,
Ingram
,
R. S.
,
Levorse
,
J. M.
and
Tilghman
,
S. M.
(
2000
).
CTCF mediates methylation-sensitive enhancer-blocking activity at the H19/Igf2 locus
.
Nature
405
,
486
489
.
Hark
,
A. T.
and
Tilghman
,
S. M.
(
1998
).
Chromatin conformation of the H19 epigenetic mark
.
Hum. Mol. Gen.
7
,
1979
1985
.
Jones
,
B. K.
,
Levorse
,
J. M.
and
Tilghman
,
S. M.
(
1998
).
Igf2 imprinting does not require its own DNA methylation or H19 RNA
.
Genes Dev.
12
,
2200
2207
.
Jouvenot
,
Y.
,
Poirier
,
F.
,
Jami
,
J.
,
Paldi
,
A.
(
1999
).
Biallelic transcription of Igf2 and H19 in individual cells suggests a post-transcriptional contribution to genomic imprinting
.
Curr. Biol.
9
,
1199
1202
.
Khosla
,
S.
,
Aitchison
,
A.
,
Gregory
,
R.
,
Allen
,
N. D.
and
Feil
,
R.
(
1999
).
Parental allele-specific chromatin configuration in a boundary-imprinting-control element upstream of the mouse H19 gene
.
Mol. Cell. Biol.
19
,
2556
2566
.
Kitsberg
,
D.
,
Selig
,
S.
,
Brandeis
,
M.
,
Simon
,
I.
,
Keshet
,
I.
,
Driscoll
,
D. J.
,
Nicholls
,
R. D.
and
Cedar
,
H.
(
1993
).
Allele-specific replication timing of imprinted gene regions
.
Nature
364
,
459
463
.
Leighton
,
P. A.
,
Ingram
,
R. S.
,
Eggenschwiler
,
J.
,
Efstratiadis
,
A.
and
Tilghman
,
S. M.
(
1995a
).
Disruption of imprinting caused by deletion of the H19 gene region in mice
.
Nature
375
,
34
39
.
Leighton
,
P. A.
,
Saam
,
J. R.
,
Ingram
,
R. S.
,
Stewart
,
C. L.
and
Tilghman
,
S. M.
(
1995b
).
An enhancer deletion affects both H19 and Igf2 expression
.
Gene Dev.
9
,
2079
2089
.
Li
,
E.
,
Beard
,
C.
and
Jaenisch
,
R.
(
1993
).
Role of DNA methylation in genomic imprinting
.
Nature
366
,
362
365
.
Lyko
,
F.
,
Brenton
,
J. D.
,
Surani
,
M. A.
and
Paro
,
R.
(
1997
).
An imprinting element from the mouse H19 locus functions as a silencer in Drosophila
.
Nat. Genet.
16
,
171
173
.
Lyko
,
F.
,
Buiting
,
K.
,
Horsthemke
,
B.
and
Paro
,
R.
(
1998
).
Identification of a silencing element in the human 15q11-q13 imprinting center by using transgenic Drosophila
.
Proc. Natl Acad. Sci. USA
95
,
1698
1702
.
Mihaly
,
J.
,
Hogga
,
I.
,
Gausz
,
J.
,
Gyurkovics
,
H.
and
Karch
,
F.
(
1997
).
In situ dissection of the Fab-7 region of the bithorax complex into a chromatin domain boundary and a Polycomb-response element
.
Development
124
,
1809
1820
.
Nagy
,
A.
,
Rossant
,
J.
,
Nagy
,
R.
,
Abramow Newerly
,
W.
and
Roder
,
J. C.
(
1993
).
Derivation of completely cell culture-derived mice from early-passage embryonic stem cells
.
Proc. Natl Acad. Sci. USA
90
,
8424
8428
.
Olek
,
A.
and
Walter
,
J.
(
1997
).
The pre-implantation ontogeny of the H19 methylation imprint [letter]
.
Nature Genetics
17
,
275
276
.
Poirier
,
F.
,
Chan
,
C.-T. J.
,
Timmons
,
P. M.
,
Robertson
,
E. J.
,
Evans
,
M. J.
and
Rigby
,
P. W. J.
(
1991
).
The murine H19 gene is activated during embryonic stem cell differentiation in vitro and at the time of implantation in the developing embryo
.
Development
113
,
1105
1114
.
Reik
,
W.
and
Walter
,
J.
(
1998
).
Imprinting mechanisms in mammals
.
Curr. Opin. Gen. Dev.
8
,
154
164
.
Ripoche
,
M. A.
,
Kress
,
C.
,
Poirier
,
F.
and
Dandolo
,
L.
(
1997
).
Deletion of the H19 transcription unit reveals the existence of a putative imprinting control element
.
Gene Dev.
11
,
1596
1604
.
Schmidt
,
J. V.
,
Levorse
,
J. M.
and
Tilghman
,
S. M.
(
1999
).
Enhancer competition between H19 and Igf2 does not mediate their imprinting
.
Proc. Natl Acad. Sci. USA
96
,
9733
9738
.
Schwartz
,
F.
,
Maeda
,
N.
,
Smithies
,
O.
,
Hickey
,
R.
,
Edelmann
,
W.
,
Skoultchi
,
A.
and
Kucherlapati
,
R.
(
1991
).
A dominant positive and negative selectable gene for use in mammalian cells
.
Proc. Natl Acad. Sci. USA
88
,
10416
10420
.
Simon
,
I.
,
Tenzen
,
T.
,
Reubinoff
,
B. E.
,
Hillman
,
D.
,
McCarrey
,
J. R.
and
Cedar
,
H.
(
1999
).
Asynchronous replication of imprinted genes is established in the gametes and maintained during development
.
Nature
401
,
929
932
.
Srivastava
,
M.
,
Hsieh
,
S.
,
Grinberg
,
A.
,
Williams-Simons
,
L.
,
Huang
,
S.-P.
and
Pfeifer
,
K.
(
2000
).
H19 and Igf2 monoallelic expression is regulated in two distinct ways by a shared cis acting regulatory region upstream of H19
.
Genes Dev.
14
,
1186
1195
.
Surani
,
M. A.
(
1998
).
Imprinting and the initiation of gene silencing in the germ line
.
Cell
93
,
309
312
.
Szabo
,
P. E.
,
Tang
,
S.-H. E.
,
Rentsendorj
,
A.
,
Pfeifer
,
G. P.
and
Mann
,
J. R.
(
2000
).
Maternal-specific footprints at putative CTCF sites in the H19 imprinting control region give evidence for insulator function
.
Curr. Biol.
10
,
607
610
.
Thorvaldsen
,
J. L.
,
Duran
,
K. L.
and
Bartolomei
,
M. S.
(
1998
).
Deletion of the H19 differentially methylated domain results in loss of imprinted expression of H19 and Igf2
.
Genes Dev.
12
,
3693
3702
.
Tilghman
,
S. M.
(
1999
).
The sins of the fathers and mothers: genomic imprinting in mammalian development
.
Cell
96
,
185
193
.
Torres
,
R. M.
and
Kuhn
,
R.
(
1997
).
Laboratory Protocols for Conditional Gene Targeting.
Oxford
:
Oxford University Press
.
Tremblay
,
K. D.
,
Duran
,
K. L.
and
Bartolomei
,
M. S.
(
1997
).
A 5𠌒 2-kilobase-pair region of the imprinted mouse H19 gene exhibits exclusivepaternal methylation throughout development
.
Mol. Cell. Biol.
17
,
4322
4329
.
Tremblay
,
K. D.
,
Saam
,
J. R.
,
Ingram
,
R. S.
,
Tilghman
,
S. M.
and
Bartolomei
,
M. S.
(
1995
).
A paternal-specific methylation imprint marks the alleles of the mouse H19 gene
.
Nat. Genet.
9
,
407
413
.
Webber
,
A. L.
,
Ingram
,
R. S.
,
Levorse
,
J. M.
and
Tilghman
,
S. M.
(
1998
).
Location of enhancers is essential for the imprinting of H19 and Igf2 genes
.
Nature
391
,
711
715
.
Wilkinson
,
D. G.
and
Nieto
,
M. A.
(
1993
).
Detection of message RNA by in situ hybridisation to tissue sections and whole mounts
.
Methods Enzymol.
225
,
361
373
.
Wood
,
S. A.
,
Pascoe
,
W. S.
,
Schmidt
,
C.
,
Kemler
,
R.
,
Evans
,
M. J.
and
Allen
,
N. D.
(
1993
).
Simple and efficient production of embryonic stem cell-embryo chimeras by coculture
.
Proc. Natl Acad. Sci. USA
90
,
4582
4585
.
Zhou
,
J.
and
Levine
,
M.
(
1999
).
A novel cis-regulatory element, the PTS, mediates an anti-insulator activity in the Drosophila embryo
.
Cell
99
,
567
575
.