Mouse/human sequence divergence in a region with a paternal-specific methylation imprint at the human H19 locus
Mouse/human sequence divergence in a region with a paternal-specific methylation imprint at the human H19 locusYoshihiro Jinno1,*, Kazuo Sengoku2, Mitsuyoshi Nakao3, Kenichi Tamate2, Toshinobu Miyamoto1,2, Tetsuo Matsuzaka4, James S. Sutcliffe5, Tadashi Anan3,6, Naoyuki Takuma2, Kunihiko Nishiwaki2, Yuichiro Ikeda7, Tadayuki Ishimaru7, Mutsuo Ishikawa2 and Norio Niikawa1
1Department of Human Genetics, Nagasaki University School of Medicine, Nagasaki 852, Japan, 2Department of Obstetrics and Gynecology, Asahikawa Medical College, Asahikawa 078, Japan, 3Department of Tumor Genetics and Biology, Kumamoto University School of Medicine, Kumamoto 860, Japan, 4Department of Pediatrics, Nagasaki University School of Medicine, Nagasaki 852, Japan, 5Howard Hughes Medical Institute and Department of Molecular and Human Genetics, Baylor College of Medicine, Houston TX77030, USA, 6Department of Pediatrics, Kumamoto University School of Medicine, Kumamoto 860, Japan and 7Department of Obstetrics and Gynecology, Nagasaki University School of Medicine, Nagasaki 852, Japan
Received April 8, 1996;Revised and Accepted May 31, 1996
We have identified a region with characteristics of a paternal-specific methylation imprint at the human H19 locus. This region, extending from -2.0 kb upstream to the start of transcription, is heavily methylated in sperm and on the paternal allele in somatic cells. This methylation was preserved during pre-implantation. Structural analysis revealed the presence of CpG islands and a large direct repeat with a 400 bp sequence reiterated several times, but no significant sequence homology to the corresponding region of the mouse H19 gene. These findings could suggest a role for secondary DNA structure in genomic imprinting across the species, and they also present a puzzling aspect of the evolution of the H19 regulatory region in human and mouse.
It is a firmly established concept that normal mammalian development requires both a maternal and a paternal genome. This effect is attributed to genes which are expressed in parent-specific manner, a phenomenon known as genomic imprinting. It is evident that imprinted genes are often clustered in regions to form `imprinted domains', although the evolutionary significance of this and how it relates to control of imprinted expression is not entirely clear (1 ,2 ). While imprinted expression of genes is generally conserved between human and mouse, maternal-specific expression of Igf2r is seen in mouse, but not in most of the human population (3 -5 ). IGF2 and H19, which map in close proximity on chromosome 11 in human and chromosome 7 in mouse, represent a paradigm for the study of genomic imprinting in mammals (2 ). IGF2 is expressed from the paternal allele in both species (6 -8 ), while expression of H19 is exclusively from the maternal allele (9 ,10 ). However, we found biallelic expression of H19 in the human placenta at an early stage, which contrasts with consistent monoallelic expression in mouse placenta (11 ,12 ).
In spite of great progress in the identification and characterization of imprinted genes, the mechanisms by which each allele of a gene is distinguished are largely unknown. It has been assumed that the alleles must be marked differently and that the distinction must be reversible. DNA methylation is a promising candidate for the distinguishing mark. The methylation state of several imprinted genes has been established (13 ). A paternal-specific methylation imprint was recently identified at the mouse H19 locus (12 ,14 ). Although the inactive paternal allele is usually heavily methylated in the human H19 gene as in mice and although there were observations of abnormal methylation linked to reduced expression of H19 in Wilms' tumours (15 -17 ), our previous work indicated that DNA methylation at sites studied previously did not meet the criteria to be the distinguishing characteristic, because they were found to be hypomethylated in placenta at an early gestational stage (11 ).
In order to explore the apparent difference in monoallelic versus biallelic expression of H19 in human placenta compared to mouse, and likely mechanisms for control of expression, we characterized the region further 5' to the transcription start site and searched for a possible paternal-specific methylation imprint at the human H19 locus. These studies revealed that methylation over a 1.3 kb sequence extending from 2 kb upstream to the start of transcription was paternal-specific using newly identified polymorphisms for AvaI and HhaI (18 ). The nucleotide sequences of the region share features of CpG islands (19 ), and demonstrate a striking 400 bp repeated sequence. No significant organizational or sequence similarity to the corresponding mouse H19 region was identified.
As the previous study indicated that the methylation in the promoter and 3' regions of the human H19 gene could not be a primary imprint, we extended a methylation analysis to the far 5' region of H19 to find regions with a possible gametic methylation imprint. For this purpose, we carried out Southern blot hybridization and quantitative analyses on methylation sensitive restriction enzyme digests based on an assumption that methylation signals are regional rather than restricted to a specific sequence, although these analyses do not distinguish the methylation status of individual sites. Taking into account the limitations, we carefully chose enzymes which cleaved within the regions analyzed, and examined various control samples such as peripheral blood leucocytes and hydatidiform moles which represent 100% paternal DNA. Signal intensities between different regions and between different samples in the same blot were compared to obtain good assessments of relative methylation status. Each of two HpaII sites on the 469 bp PstI-PvuII fragment just upstream of the promoter seemed still hypomethylated in placentae (20 to 30% methylation) as indicated by appearance of expected HpaII (plus PstI and PvuII)-restricted bands with the same sizes (236, 188 and 147 bp) and the similar pattern as MspI-restricted bands, while those were relatively heavily methylated in the hydatidiform mole (70-80%). Sperm DNA also showed hypomethylation, but the pattern of HpaII-restricted bands was slightly different from those of the placentae. This may indicate that one of the two sites is more heavily methylated than the other. Peripheral blood leucocytes and fetal liver DNA showed a 50% methylation, which is consistent with results previously reported (15 ,16 ) (Fig. 1 B). Further upstream a PstI fragment contains 16 HpaII sites, and BglII divides them almost equally. Reduction of signal intensity of the 3' BglII-PstI fragment (1.2 kb) after HpaII digestion is more evident than that of the 5' fragment (1.3 kb) in the placenta (Fig. 1 C). Radioactive intensities indicated a 40-50% methylation of the sites on the 5' PstI-BglII fragment in the placenta, peripheral blood leukocytes and fetal liver DNA. Sperm and hydatidiform moles were heavily methylated but did not reach 100% methylation (70-90%). This would be interpreted that at least one of the HpaII sites in this region was not methylated in about 20% cell populations.
These results are summarized in a histogram (Fig. 1 D). Although the degree of methylation of the upstream HpaII sequences was variable depending on their locations, no significant variation of methylation at the same sites was observed between placentae regardless of H19 biallelic/monoallelic expression or of gestational ages, which remarkably contrasts to the progressive methylation pattern in the 3' portion of the gene (11 ).
Methylation analyses suggested that HpaII sequences might be differentially methylated in the most upstream PstI-BglII region. We next examined allele-specificity of methylation in this region by means of newly identified AvaI and HhaI restriction fragments length polymorphisms (RFLPs). As recognition sequences of AvaI and HhaI contain a CpG dinucleotides sequence within themselves and methyl-CpG in the sequences may obscure hybridization results, we adopted a PCR-based strategy to determine the methylation state on each allele. Locations of polymorphic AvaI and HhaI sites and primers for PCR amplification are indicated in Fig. 1 A. Undigested DNA and DNA digested with methylation-sensitive enzymes (HpaII or HpaII plus HhaI) were amplified with combinations of primers I-III and II-III for detection of the AvaI and HhaI RFLPs, respectively. For the AvaI RFLP, the 1.5 kb PCR product was digested with BglI and a 405 bp fragment containing the polymorphic sequence was purified prior to AvaI digestion. Whereas the PCR of undigested DNA heterozygous for the AvaI RFLP yielded a 64 bp constant band and each of a 342 bp and 270 bp plus 72 bp polymorphic bands, HpaII digestion resulted in amplification of only one of polymorphic bands, which indicates monoallelic methylation (Fig. 2 A). Amplified alleles were proved to be paternal in two of three samples analyzed. Thus, all of the HpaII sequences on the paternal allele in this region seemed to be methylated in the majority of cells. In addition, it was likely that all the HhaI sequences on the paternal allele were also methylated because HpaII plus HhaI digestion gave the same result as the HpaII digestion alone (data not shown). The paternal-allele-specific methylation at HpaII sites in the region was further confirmed using the HhaI RFLP. Again, only the paternal allele was amplified after HpaII digestion (Fig. 2 B).
In order to be a candidate for the imprinting distinguishing characteristic the methylation observed in sperm and on the paternal allele in somatic cells must be preserved during pre-implantation. This was tested on early embryos at the 8 to 32 cell-stage by PCR-based methylation analysis principally according to Tremblay et al. (12 ). We chose a region between primers II and III as the target and the promoter region (IV-V, Fig. 1 A) as a control of hypomethylation. Pooled 8 to 32-stage cells were predigested with methylation-sensitive HpaII or insensitive MspI, and subjected to sequential PCR amplification with primers of II, III, IV and V in the first PCR and then with internal primers of IIa, IIIb, IVa and V in the nested PCR. The nested PCR of HpaII-digested DNA should yield a 320 bp and 399 bp product from the target and control regions, respectively, when the HpaII sequences are methylated in the regions. However, no products should be amplified in PCR of HpaII-digested DNA as in that of MspI-digested DNA if the methylation were not preserved in these cells. As shown in Figure 3 , the 320 bp product was amplified in the PCR of HpaII-digests as equally as in that of undigested DNA, whereas significant amplification of the 399 bp product was not obtained either in the PCR of HpaII- or MspI-digested DNA. Thus, the methylation at HpaII sites in this upstream region seems to be preserved during pre-implantation. Although we could not analyze the methylation of more upstream HpaII sequences in the paternally methylated region due to difficulties resulting from the limited amount of DNA and the repeated structure as mentioned below, it would be reasonable that methylation of these might be preserved as that in the region analyzed. The technique used here does not distinguish between maternal and paternal alleles because of necessitated use of pooled materials, nor does it tell how many fractions of the embryos are methylated.
As the far 5' sequences upstream to the 3' PstI site (-830 bp from the transcription initiation site) of the human H19 gene were not available, we determined the upstream sequences (-830 bp to -3447 bp). Sequence analysis of 2617 bp revealed a high GC content (59%) over the entire region and a relatively high frequency of CpG dinucleotides. The ratios of CpG versus GpC and observed versus expected CpG dinucleotides were 0.61 and 0.49, respectively. This feature of high frequencies of G+C and CpG was most prominent in the most downstream repeat unit (-1994 to -2397; described below). This region contained 60% G+C residues and CpG versus GpC frequencies were 0.86. These values are comparable to those of 226 bp just upstream to the transcription start (65% GC and 0.63 CpG/GpC ratio, ref. 20 ) which shows 76% homology with the mouse H19 gene (21 ). Thus, the differentially methylated region shares features of CpG islands. Another remarkable feature was a huge direct repeat structure. The entire stretch of the paternally methylated region consisted of 3.5 repeats of a 400 bp nucleotides sequence. Homology of each 400 bp repeat almost reached 90% to each other except the most 5' sequences of 200 bp which was devoid of the 5' half of the repeat unit and showed 71% homology to other counterparts. Sequence comparisons with the GenBank database revealed no significant sequence homology with any known gene. However, it seems noteworthy that there was no organizational or sequence similarity between the differentially methylated regions of the mouse and human. The G-rich sequence of a 460 bp region (-1285 to -1745 bp relative to the start of the mouse H19 transcription, ref. 12 ), which was identified to have a paternal-specific methylation imprint, was not at all found in the upstream region of the human H19 gene. The human large repeated sequence spanning from -1994 to -3403 bp relative to the H19 transcription initiation site was not represented in the mouse H19 upstream region of 5212 nucleotides sequence (K.D.Tremblay et al., GenBank accession no. U19619). The only similarity was the relative locations of their unique sequences, although the human sequence lay somewhat more upstream (Fig. 4 ).
The H19 gene is imprinted both in human and mouse (9 ,10 ), and differential methylation of the gene has been detected specifically on the inactive paternal allele in human and mouse tissues as well (15 -17 ). However, unlike the gametic imprint of mouse H19 (12 ), paternal methylation in the human had not met the criteria for such an imprint, due to observations of hypomethylation in the promoter and 3' regions in the human placenta at an early gestational stage (11 ). In the current study, we have extended our analysis further 5' relative to the transcription start of H19. The methods employed in our study, Southern blot hybridization and PCR-based methylation assays, have several limitations as indicated above. Under these limitations we have found a region which fulfilled criteria to be a gametic distinguishing characteristic: (i) paternal specific methylation in somatic cells; (ii) gametic methylation in sperm; and (iii) preservation during pre-implantation, as far as possible. While the study does not exclude the possibility that a gametic methylation imprint would be found in other regions of CpG dinucleotides which are not represented in sequences recognized by methylation sensitive enzymes, it suggests a possible region worthy of testing as cis-elements controlling the human H19 imprinting.
This region extending from -2.0 to -3.3 kb relative to the transcription start is methylated in sperm and on the paternal allele in somatic cells. It showed a 50% methylation in early stage placentae as well as in late stage placentae and other tissues. It is likely that methylation in this region is preserved during pre-implantation when generalized demethylation occurs. Thus, methylation in this region has a potential as a human paternal-specific methylation imprint of H19.
Sequence analysis of this region revealed features of a CpG island, including a high proportion of G plus C residues and a high frequency of CpG dinucleotides, and additionally revealed a long stretch of repeated sequences. Observation of repeated sequences, including direct repeats, at other imprinted genes has led to speculation for possible involvement in the imprinting control mechanism (22 ,23 ). In this case, the differentially methylated region was confined to the repeated sequences and did not extend downstream. Therefore, it is unlikely that methylation itself inhibits binding of basal transcription factors to the promoter. This seems to be exemplified by the fact that H19 is expressed in the hydatidiform mole, in which most upstream regions except the promoter is heavily methylated, as well as in the placenta (our own observations and ref. 24 ). Instead, binding of methyl-CpG binding proteins to the region might deprive structural flexibility to inhibit the formation of a secondary or tertiary structure required for active transcription. Alternatively, some specific sequences for trans-acting positive regulators might be embedded in the region, though we were unable to detect any consensus sequence for DNA binding proteins.
We previously reported that H19 is expressed from both alleles in human placentae before 10-week gestation (11 ). Although we searched for DNA methylation patterns correlated to this change of expression pattern, such a pattern has not been seen in upstream regions studied thus far. Methylation analyses showed a 50% methylation in the repeat region, a mild methylation (less than 50%) in the downstream region and almost no methylation in the promoter region in all the human placentae examined regardless of H19 biallelic or monoallelic expression. These results indicate that the change of H19 expression pattern is attributed to methylation changes in other regions or other, as yet unknown, modifiers of H19 expression. In addition to the 3' enhancers, the presence of additional regulatory elements was suggested since a significant decline in H19 RNA was not observed in placenta as well as in muscle in enhancer deletion mice (25 ). Although such elements seem to act in a parent-of-origin dependent manner in mice (26 ), it is possible that they are active on both alleles in humans, because upstream sequences are quite different between the two species.
It seems somewhat puzzling that no significant sequence homology of the human H19 upstream region to the mouse corresponding region was found. Although the physiological function of H19 is not known, it must have essential roles in normal development of embryos because more than 70% sequence homology of the H19 gene and the promoter region is maintained between human and mouse. On the other hand, its imprinted nature and methylation patterns are conserved between two species. Given these facts, we expected homology or conservation of key sequences within the upstream region; however, this was not observed. It is interesting to note that while we show that the human upstream region has a 400 bp sequence reiterated several times, the upstream mouse region has a much shorter 8 or 9 bp sequence repeated a number of times. Whether there is anything functionally similar about these sequences in the human and mouse is unclear. Likewise, what factors or sequences functionally mediate the difference in expression between human and mouse in placenta is also unclear. While experimental evidence for a role of the human upstream repeat sequences is still required, it is tempting to speculate that sequences responsible for controlling the pattern of imprinted expression could be quite different for the same gene between species as well as between imprinted genes within the same species.
Genomic DNA was prepared from placentae, trophoblasts of the complete hydatidiform moles and a fetal liver (15-week gestation) as described (27 ). Parental DNA was extracted from peripheral blood leukocytes, and in some cases decidua DNA was substituted for maternal leukocytes DNA.
Genomic DNA was predigested with a combination of PstI and PvuII, or PstI and BglII, then with HpaII (10 units per [mu]g DNA) and subjected to electrophoresis on 1 or 1.6% agarose gels. Hybridization blots were finally washed with 0.1* SSC/0.1% SDS at 60oC for 10 min. The blots were stripped and re-used for hybridization with an internal control probe when the control signal overlapped to any signal derived from HpaII partial digests. Signal intensities were measured using a BAS1000 (Fujix). The probe 2 was prepared from a plasmid subclone carrying a 3.3 kb SfiI/EcoRI fragment of the human H19 upstream region. The probe 1 and an internal control probe were made by nested PCR. For the probe 1, genomic DNA (0.5 [mu]g) was amplified with primers H19PU1 and H19PD1 for 35 cycles at 94/60/72oC for 100/70/120 s (DNA Thermal Cycler PJ 2000, Perkin Elmer), and a 1.8 kb PCR product was purified from a gel to be used for nested PCR driven with primers H19PU2 and H19PD2 (94/60/72oC for 100/60/100 s). Sequence data for the primers were from the GenBank database (M32053). For the control probe DNA, primers, WTR1/WTR2 for first PCR and WTR1a/WTR2b for nested PCR, were designed based on the GenBank database (X69950). Cycle conditions were at 94/56/72oC for 100/60/120 s (first PCR) and at 94/60/72oC for 100/60/120 s (nested PCR). The 1.8 kb nested PCR product was digested with PstI and a 358 bp PstI fragment was extracted from a gel. Primer sequences as follows: H19PU1, 5'-TCACCAAAGGCCAAGGTGGT; H19PD1, 5'-CAGCCTTGCTGCGCAATGT; H19PU2, 5'-AGGCCAAGGTGGTGACCGA; H19- PD2, 5'-TCCTTTGGCATCCGGAGACA; WTR1, 5'-TCAGGACTGGAAGCAACTCT; WTR2, 5'-GTCTCCGTACGACCCCAAC; WTR1a, 5'-ACGTATCTGCTCTGGCCACG; WTR2b, 5'-AAGCCAGCGCTGGGGAACTG.
Placenta DNA heterozygous for the AvaI or the HhaI RFLP was digested with HpaII or HpaII plus HhaI at 37oC for 7 h in a condition of 0.1 [mu]g DNA/[mu]l and 10 U/[mu]g DNA. To determine allele-specificity of methylation by means of the AvaI RFLP, 1.5 [mu]g of restriction digested and 0.75 [mu]g of undigested DNA were amplified in a 100 [mu]l reaction volume by PCR with primers I and III for 32 cycles at 94/59/72oC for 40/30/110 s (Gene Amp PCR System 9600, Perkin Elmer). The 1.5 kb PCR product was digested with BglI and a 405 bp fragment was purified from a gel prior to AvaI digestion. For the HhaI RFLP, 0.8 [mu]g or 0.4 [mu]g of restriction digested or undigested DNA was amplified with primers II and III for 30 cycles at 94/59/72oC for 105/60/100 s, and then subjected to HhaI digestion. The AvaI and HhaI digests were analyzed on 6% polyacrylamide gels and visualized by ethidium bromide staining. PCR primers for detection of the AvaI and HhaI RFLPs: I, 5'-GAGCCTGCCAAGCAGAGCG; II, 5'-CAATGAGGTGTCCCAGTTCCA; III, 5'-CACATAAGTAGGCGTGACTTGA.
Fertilized human embryos were obtained following in vitro fertilization. All embryos were surplus to the patients' needs and were donated with informed consent. DNA was prepared from a pool of 192-272 cells at 8 to 32-cell stages according to Tremblay et al. (12 ).The extracted DNA was suspended in 21 [mu]l of H2O and seven [mu]l was subjected to restriction digestion with HpaII or MspI. The restriction digests were once extracted with phenol/chloroform and ethanol-precipitated prior to PCR amplification. Each sample was amplified in a 25 [mu]l reaction containing 200 [mu]M each dNTP/1.5 mM MgCl2/ 10% glycerol/0.75 U of AmpliTaq DNA polymerase (Perkin Elmer)/0.5 [mu]M of each primer (II, III, IV or V) by 27 cycles at 94/58/72oC for 105/60/100 s. The products were 10-fold diluted and two [mu]l was subjected to a nested PCR by 20 cycles with primers, IIa, IIIb, IVa and V in a 50 [mu]l reaction containing the same composition as the first PCR. The final products were separated on a 4% polyacrylamide gel and visualized by ethidium bromide staining. Primer sequences were as follows: IIa, 5'-CCTAGTGTGAAACCCTTCTCG; IIIb, 5'-TGTGGATAATGCCCGACCTGA; IV, 5'-CCAGGAACGTGAGGTCTGAG; IVa, 5'-CCCTCTGTGCCATCCGAGTC; V, 5'-CACC- CTGCTCCTCGGTCCTA.
A human genomic library, EMBL3 SP6/T7 (Clontech), was screened for the H19 upstream region with the probe 1. A 3.3 kb SfiI (in the right arm of vector)-EcoRI fragment was purified from a phage clone carrying the most far reaching upstream region and subcloned into pUC19. Several subclones were further prepared from it. Sequencing was performed on these subclones using the Auto Read and the Cycle Sequencing kits (Pharmacia). The GenBank accession number is U50731.
We thank K. Yun, J. Gunn and A. E. Reeve for critical reading. The present study was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan and by a Grant-in-Aid for Basic and Clinical Research of Biological Basis of Growth and Development and their Disturbance from the Ministry of Health and Welfare of Japan.
1 Editorial. (1992) Signs of an imprinted domain. Nature Genet., 2, 1-2.
2 Zemel,S., Bartolomei,M.S. and Tilghman,S.M. (1992) Physical linkage of two mammalian imprinted genes, H19 and insulin-like growth factor 2. Nature Genet., 2, 61-65. MEDLINE Abstract
3 Barlow,D.P., Stöger,R., Herrmann,B.G., Saito,K. and Schweifer,N. (1991) The mouse insulin-like growth factor type-2 receptor is imprinted and closely linked to the Tme locus. Nature,349, 84-87.MEDLINE Abstract
4 Ogawa,O., McNoe,L.A., Eccles,R., Morrison,I.M. and Reeve,A.E. (1993) Human insulin-like growth factor type I and type II receptors are not imprinted. Hum. Mol. Genet., 2, 2163-2165.MEDLINE Abstract
5 Xu,Y., Goodyer,C.G., Deal,C. and Polychronakos,C. (1993) Functional polymorphism in the parental imprinting of the human IGF2R gene. Biochem. Biophys. Res. Commun., 197, 747-754.MEDLINE Abstract
6 DeChiara,T.M., Robertson,E.J. and Efstratiadis,A. (1991) Parental imprinting of the mouse insulin-like growth factor II gene. Cell,64, 849-859.MEDLINE Abstract
7 Ohlsson,R., Nyström,A., Pfeifer-Ohlsson,S., Töhönen,V., Hedborg,F., Schofield,P., Flam,F. and Ekström,T.J. (1993) IGF2 is parentally imprinted during human embryogenesis and in the Beckwith-Wiedemann syndrome. Nature Genet., 4, 94-97.MEDLINE Abstract
8 Giannoukakis,N., Deal,C., Paquette,J., Goodyer,C.G. and Polychronakos,C. (1993) Parental genomic imprinting of the human IGF2 gene. Nature Genet., 4, 98-101.MEDLINE Abstract
9 Bartolomei,M.S., Zemel,S. and Tilghman,S.M. (1991) Parental imprinting of the mouse H19 gene. Nature,351, 153-155.MEDLINE Abstract
10 Zhang,Y. and Tycko,B. (1992) Monoallelic expression of the human H19 gene. Nature Genet., 1, 40-44.MEDLINE Abstract
11 Jinno,Y., Ikeda,Y., Yun,K., Maw,M., Masuzaki,H., Fukuda,H., Inuzuka,K., Fujishita,A., Ohtani,Y., Okimoto,T., Ishimaru,T. and Niikawa,N. (1995) Establishment of functional imprinting of the H19 gene in human developing placentae. Nature Genet., 10, 318-324.MEDLINE Abstract
12 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. Nature Genet., 9, 407-413.MEDLINE Abstract
13 Razin,A. and Cedar,H. (1994) DNA methylation and genomic imprinting. Cell, 77, 473-476.MEDLINE Abstract
14 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.MEDLINE Abstract
15 Zhang,Y., Shields,T., Crenshaw,T., Hao,Y., Moulton,T. and Tycko,B. (1993) Imprinting of human H19: allele-specific CpG methylation, loss of the active allele in Wilms tumor, and potential for somatic allele switching. Am. J. Hum. Genet., 53, 113-124.MEDLINE Abstract
16 Steenman,M.J.C., Rainier,S., Dobry,C.J., Grundy,P., Horon,I.L. and Feinberg,A.P. (1994) Loss of imprinting of IGF2 is linked to reduced expression and abnormal methylation of H19 in Wilms' tumour. Nature Genet., 7, 433-439.
17 Moulton,T., Crenshaw,T., Hao,Y., Moosikasuwan,J., Lin,N., Dembitzer,F., Hensle,T., Weiss,L., McMorrow,L., Loew,T., Kraus,W, Gerald,W. and Tycko,B. (1994) Epigenetic lesions at the H19 locus in Wilms' tumour patients. Nature Genet., 7, 440-447.MEDLINE Abstract
18 Miyatake,S., Ikeda,Y., Jinno,Y. and Niikawa,N. (1996) Jpn. J. Hum. Genet. In Press.MEDLINE Abstract
19 Bird,A.P. (1986) CpG-rich islands and the function of DNA methylation. Nature, 321, 209-213.MEDLINE Abstract
20 Brannan,C.I., Dees,E.C., Ingram,R.S. and Tilghman,S.M. (1990) The product of the H19 gene may function as an RNA. Mol. Cell. Biol., 10, 28-36.MEDLINE Abstract
21 Pachnis,V., Brannan,C.I. and Tilghman,S.M. (1988) The structure and expression of a novel gene activated in early mouse embryogenesis. EMBO J., 7, 673-681MEDLINE Abstract
22 Smrzka,O.W., Faé,I., Stöger,R., Kurzbauer,R., Fischer,G.F., Henn,T., Weith,A. and Barlow,D.P. (1995) Conservation of a maternal-specific methylation signal at the human IGF2R locus. Hum. Mol. Genet., 4, 1945-1952.MEDLINE Abstract
23 Neumann,B., Kubicka,P. and Barlow,D.P. (1995) Characteristics of imprinted genes. Nature Genet., 9, 12-13.MEDLINE Abstract
24 Mutter,G.L., Stewart,C.L., Chaponot,M.L. and Pomponio,R.J. (1993) Oppositely imprinted genes H19 and insulin-like growth factor 2 are coexpressed in human androgenetic trophoblast. Am. J. Hum. Genet., 53, 1096-1102.MEDLINE Abstract
25 Leighton,P.A., Saam,J.R., Ingram,R.S., Stewart,C.L. and Tilghman,S.M. (1995) An enhancer deletion affects both H19 and Igf2 expression. Genes Dev., 9, 2079-2089. MEDLINE Abstract
26 Leighton,P.A., Ingram,R.S., Eggenschwiler,J., Efstratiadis,A. and Tilghman,S.M. (1995) Disruption of imprinting caused by deletion of the H19 gene region in mice. Nature,375, 34-39.MEDLINE Abstract
27 Jinno,Y., Yun,K., Nishiwaki,K., Kubota,T., Ogawa,O., Reeve,A.E. and Niikawa,N. (1994) Mosaic and polymorphic imprinting of the WT1 gene in humans. Nature Genet., 6, 305-309.MEDLINE Abstract
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:26:30 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
