Human Molecular Genetics

Human Molecular Genetics Advance Access originally published online on December 17, 2003

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 13/3/353    most recent ddh028v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (9) Request Permissions Google Scholar Articles by Vernucci, M. Articles by Riccio, A. Search for Related Content PubMed PubMed Citation Articles by Vernucci, M. Articles by Riccio, A. Social Bookmarking          What's this?

Human Molecular Genetics, 2004, Vol. 13, No. 3 353-361
DOI: 10.1093/hmg/ddh028

Developmentally regulated functions of the H19 differentially methylated domain

Maria Vernucci¹, Flavia Cerrato¹, Paolo V. Pedone¹, Luisa Dandolo², Carmelo B. Bruni^3,4 and Andrea Riccio^1,*

¹Dipartimento di Scienze Ambientali, Seconda Università di Napoli, Caserta, Italy, ²GDPM Department, Institut Cochin, Paris, France, ³Dipartimento di Biologia e Patologia Cellulare e Molecolare ‘L. Califano’, Università di Napoli ‘Federico II’, Napoli, Italy and ⁴Istituto di Endocrinologia ed Oncologia Sperimentale ‘G. Salvatore’, CNR, Napoli, Italy

Received October 20, 2003; Accepted December 1, 2003

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Igf2 and H19 are physically linked imprinted genes. In embryonic liver, their reciprocal expression (paternal for Igf2 and maternal for H19) is controlled by a paternally methylated region (H19 DMD) located 5' of H19. This region contains a methylation-sensitive insulator that prevents the Igf2 promoters being activated by downstream enhancers on the maternal chromosome. In adult liver, Igf2 is normally not expressed but is reactivated upon tumour formation. By analysing three deletions of the H19 locus, we investigated the mechanism regulating the imprinted expression of the Igf2 gene in the course of liver tumourigenesis. We observed that the role of the H19 DMD in the control of Igf2 expression changes during tumourigenesis. The H19 DMD is required on the paternal chromosome for Igf2 activation in the early stages while its maternal allele is necessary for maintaining Igf2 imprinting only in the late stages. A positive regulatory function of the paternal H19 DMD is also evident in normal neonatal liver, but its relevance for Igf2 expression becomes higher in the second post-natal week. Our results support a model in which both methylated and non-methylated parental copies of the H19 DMD have active roles in the regulation of Igf2 expression in the liver and these activities are under developmental control.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
A subset of mammalian genes is controlled by genomic imprinting. These genes have their alleles expressed or repressed depending on their parental origin from mother or father. Correct imprinting is important in normal development and growth and disordered imprinting has been implicated in the pathogenesis of genetic diseases and cancer (1–3). Imprinted genes are epigenetically modified during gametogenesis so that their expression becomes dependent on the gametic origin (4,5). DNA methylation is required to maintain genomic imprinting and many imprinted genes have sequences which are differentially methylated between the parental alleles (DMR or DMD). These often correspond to regulatory elements controlling the expression of one or more imprinted genes. Many different mechanisms are involved in the process of translating the imprinting marks into monoallelic expression (4–6). While much information on imprinting mechanisms is available in normal organisms, the way imprinted expression works and is deregulated in pathological conditions and particularly in cancer is less understood (3).

The insulin-like growth factor 2 gene (Igf2) codes for a mitogenic peptide with important roles in the control of embryo and placenta growth and implicated in the pathogenesis of human cancer and the overgrowth and cancer-associated Beckwith–Wiedemann Syndrome (1–3). Igf2 forms with H19 a couple of physically and functionally linked imprinted genes located in an evolutionary conserved chromosome region (human 11p15.5 and mouse distal 7) (4,7). The Igf2 and H19 genes are reciprocally imprinted. However, although Igf2 is expressed only by its paternally inherited allele and H19 only by the maternal allele, these two genes are coordinately transcribed during embryonic development. This is due to tissue-specific enhancers located downstream of the H19 gene, which are used by both Igf2 and H19 (8–10). A paternally methylated DMR (H19 DMD) is present 4 kb upstream of H19 (11). The first evidence that this region was required for Igf2 imprinting was obtained from mice carrying a deletion (13) including the H19 gene and 10 kb of its 5'-flanking sequence (12). This mutation resulted in biallelic expression of the Igf2 gene and overgrowth when maternally transmitted (12). A similar although less extensive reactivation of the maternal Igf2 allele was observed in the mice following maternal transmission of a 1.6 kb deletion (DMD) that removed most of the H19 DMD but left the H19 gene intact (13). In contrast, removal of the H19 transcriptional unit had little or no effect on Igf2 imprinting (14,15). More recently, the H19 DMD was shown to harbour a chromatin insulator (16–19). This element binds the protein CTCF and its binding is prevented by CpG methylation (16,17). A targeted mutation of the CTCF sites demonstrated that these are needed for the enhancer-blocking activity and for preventing the maternal DMD from becoming methylated post-zygotically (20). Thus, the reciprocal imprinting of the Igf2 and H19 genes is explained by the gamete of origin-specific methylation of the H19 DMD. According to this model, the unmethylated DMD present on the maternal chromosome limits the action of the enhancers to the H19 promoter, while methylation of the DMD inactivates the insulator on the paternal homologue, thereby allowing activation of the Igf2 promoter by the enhancers.

The expression of the Igf2 and H19 genes is shut off after birth in the majority of tissues. We have previously shown that Igf2 and H19 were coordinately reactivated in adult liver upon induction of tumourigenesis and that their imprinting was maintained (21). The H19 endodermal enhancer was necessary for Igf2 activation in the liver tumours and the deletion of the paternal enhancer led to marked delay in tumour development (22). By analysing three different deletions of the H19 locus (Fig. 1), in this paper we investigate the mechanisms regulating the imprinted expression of the Igf2 gene in the course of liver tumourigenesis. We found that the H19 DMD has different functions in the control of Igf2 expression: (1) its paternal allele is required for Igf2 activation in the early stages of tumourigenesis; and (2) its maternal allele is needed for maintaining the imprinting in the late-occurring tumours. We also observed that the role of the H19 DMD changes during normal liver development. The maternal H19 DMD is required for maintaining the imprinting of Igf2 in the embryo and newborn, while the paternal H19 DMD is needed for Igf2 expression mainly after the first post-natal week.

View larger version (15K): [in this window] [in a new window]   Figure 1. The wild-type Igf2–H19 locus and the 13, 3 and DMD deletions. The diagram shows the relative positions of the Igf2 and H19 genes (open rectangles) and of the DMD (solid rectangle) and endodermal enhancer (EE, ovals) elements and extension of the deletions.

 

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Deletion of H19 and 10 kb of 5' flanking region results in lack of activation of Igf2 during early liver tumourigenesis
It was previously shown that a deletion (H1913) covering the H19 gene and 10 kb of its 5'-flanking sequence and including the H19 DMD (Fig. 1) resulted in a strong reactivation of the maternal Igf2 allele in the embryos following maternal transmission of this mutation (12). No phenotype was evident upon paternal transmission. Since we had previously observed that Igf2 was reactivated in adult liver during tumourigenesis and that its imprinting was maintained (21), we wanted to analyse the effect of the H1913 mutation during tumourigenesis. The experimental liver tumourigenesis system previously used by our group (21) was chosen for this purpose. This system consists of a transgenic mouse line carrying the SV40 Tag oncogene under the control of a liver-specific promoter (23). The males of this transgenic line have a low but constitutive expression of the viral oncogene, which after formation of hyperplastic foci and neoplastic nodules, eventually leads to the development of multiple hepatocellular carcinomas by 4–5 months of age. The females do not express the SV40 Tag and do not develop tumours, unless they are treated with the bacterial lipopolysaccharide (LPS), which induces the transgene promoter. Activation of the viral oncogene triggers the liver carcinogenesis process. This allows the analysis of tumourigenesis from its very early stages. Before breeding, the mice of the H1913 and SV40 Tag lines were maintained on different genetic backgrounds (C57BL/6 and BALB/c, respectively; see Materials and Methods), so that maternal and paternal Igf2 alleles could be distinguished on the basis of polymorphisms in the offspring. Reciprocal crosses were set up between heterozygous H1913 mice and homozygous SV40-Tag transgenic mice. The offspring carried the SV40 Tag in hemizygosis and either the wild-type H19 locus in homozygosis (wt) or H1913 in heterozygosis (either paternally inherited, pat 13 or maternally inherited, mat 13). In these mice, the relative contribution of the maternal and paternal alleles to Igf2 expression was determined by RT–PCR amplification (Fig. 2A and C) and the overall expression of the Igf2 gene was measured by RNAse protection assay (Fig. 2B and D). Igf2 imprinting and overall expression were analysed at post-natal day 1 and in 2-month-old females after induction of tumourigenesis. In the newborns, we observed exclusive expression from the paternal Igf2 allele in the wt mice (Fig. 2A, lanes 2 and 6), as expected from the imprinting of this gene. The allele-specific expression was lost and both parental alleles were efficiently expressed in the mice following maternal transmission of the 13 mutation (mat 13, Fig. 2A, lane 8), but imprinting was normal in the mice inheriting 13 from the father (pat 13 mice, Fig. 2A, lane 4), consistent with previously published data (12). The overall level of Igf2 mRNA was normal in the pat 13 newborn mice (Fig. 2B) and approximately doubled in the mat 13 mice (Fig. 2B). Igf2 imprinting and expression was then determined in the adult mice (Fig. 2C and D). In the 2-month-old females (either wt or 13), Igf2 (as well as H19) was not expressed (Fig. 2D, first lane and data not shown). We have previously observed that Igf2 is activated early during liver tumourigenesis (22), so the effect of 13 on Igf2 imprinting and expression was first analysed in total liver, 3 days after the induction of the SV40 Tag oncogene. The results showed that the Igf2 gene was activated more than 100-fold in the LPS-treated wt mice and that only the paternal allele was expressed (Fig. 2D and C, lanes 1 and 5). Surprisingly, however, the effect of the H1913 mutation was different from that seen in the newborns. The allele-specific RT–PCR showed that the maternal Igf2 allele was not derepressed in the adult mat 13 mice (mat/pat ratio, 0.05; see Fig. 2C, lane 7) as it was in the mat 13 newborns (mat/pat ratio, 0.72; see Fig. 2A, lane 8). Likewise, the paternal Igf2 allele was not activated in the adult pat 13 mice (Fig. 2D) as it was in the pat 13 newborns (Fig. 2B). In other wt and H1913 female mice, oncogene activation was perpetuated for up to 90 days by longer LPS treatment. We have previously observed that, under such treatment, Igf2 activation reaches very high levels in the liver of wt mice, although neoplastic nodules develop only a few months later (21). Even after this longer treatment, the expression of the paternal Igf2 allele in the pat 13 mice and the expression of the maternal Igf2 allele in the mat 13 mice remained very poor (data not shown). This clearly indicates that the region deleted in H1913 is required in cis for activation of the paternal Igf2 allele during the early stages of liver tumourigenesis. The lack of derepression of the maternal Igf2 allele upon maternal transmission of the H1913 deletion is consistent with a positive function of this region for Igf2 expression.

View larger version (43K): [in this window] [in a new window]   Figure 2. Effect of the H1913 deletion on imprinting and overall expression of the Igf2 gene in neonatal liver and early stages of liver tumourigenesis. (A and B) Analysis of newborns. RNAs extracted from the liver of 1-day-old mice transgenic for the SV40 Tag oncogene following paternal (pat 13) or maternal (mat 13) transmission of the H1913 deletion and corresponding wt were analysed by allele-specific RT–PCR with Igf2-specific primers (A) or RNase protection with Igf2-, H19- and Gapdh-specific probes (B). (C and D) Analysis of the early stages of liver tumourigenesis. RNAs extracted from 2-month-old female mice with the same genotypes of (A, B) either before or after induction of the transgenic oncogene by bacterial LPS (3 day-treatment) were analysed as in (A, B). Igf2 alleles were RT-PCR amplified with primers flanking the Igf2 (CA)n repeat and the PCR products separated by electrophoresis on denaturing polyacrylamide–urea gels (21). The a and b alleles differed by 8–12 bp in length. The relative expression of the maternal and paternal Igf2 alleles (mat/pat) was calculated from the intensities of the RT-PCR bands. The mice derived from reciprocal crosses between the H1913 (C57BL/6) and the SV40 Tag (BALB/c) lines. The wt animals therefore expressed either the Igf2a (BALB/c) or b (C57BL/6) alleles, depending on parental inheritance. To test for contamination by genomic DNA, RNA samples were run in duplicate with (+RT) or without (–RT) addition of reverse transcriptase. The DNA derived from a CRP-Tag H19 wt mouse was analysed as control (A, lane 1). The shorter PCR product (b allele) was amplified to some extent more efficiently than the longer one (a allele), but the relative intensities were reproducible within the same experimental conditions. The Igf2 and H19 RNA levels measured by RNase protection are expressed in arbitrary units (wt=100) and reported in tables. The exposure time of the gels analysing the adult mice was 5-fold longer than that of the newborns. The samples shown in this figure are representative of a series of at least three experiments with very similar results. Note that the maternal Igf2 allele is not derepressed in the mat 13 mice and the paternal Igf2 allele is not activated in the pat 13 mice upon induction of tumourigenesis in adult liver.

 
We have previously demonstrated that the hemizygous SV40 Tag male mice (that constitutively express the transgenic oncogene) spontaneously develop liver tumours with an increasingly higher frequency between the ages of 120 and 141 days and that the lack of Igf2 activation due to the deletion of the H19 endodermal enhancer causes a significant delay in tumour formation (22). In order to test the relevance of the H1913 deletion for Igf2 activation during tumourigenesis, we compared the development of liver tumours in the pat 13 and wt male mice. The wt mice showed neoplastic nodules since the age of 120 days and reached the frequency of 75–80% animals with tumours by the age of 141–148 days (14/17 and 12/16, respectively). In contrast, none out of six pat 13 mice had developed any tumour by 141 days and none out of three by 148 days (Table 1). Thus, paternal transmission of the H1913 deletion causes a delay in the process of liver tumourigenesis of more than 3 weeks, indicating that the impairment in Igf2 activation is very relevant in these mice.

View this table: [in this window] [in a new window]   Table 1. Occurrence of liver tumours in the pat H1913 and corresponding wild-type mice

 
After 5 months of age, almost all the SV40 Tag male mice with the wt H19 locus developed liver tumours which soon became large neoplastic masses and caused death of the animals. The pat 13 mice lived longer (>6 months) than the wt, probably because of the delayed tumour development. During the sixth month of life, liver tumours began to develop also in the pat 13 mice. By 180 days of age, multiple tumours were found in the majority of these animals. We dissected the neoplasms and analysed the expression of the Igf2 gene. About two-thirds of the tumours arisen in the Pat 13 mice did express Igf2 at levels which were more than 100-fold higher than those present in the peri-neoplastic tissue (P) and approached those detected in the wt mice (Fig. 3B and data not shown). In the pat 13 as in the wt tumours, Igf2 imprinting was maintained (Fig. 3A and data not shown). We then analysed the neoplasms developed in the mat 13 mice. Exclusive or predominant expression of the paternal Igf2 allele was detected in one-third of the tumours arisen in these mice and in the peri-neoplastic tissue (P, Fig. 3C). However, the other two-thirds of the tumours expressed the maternal Igf2 allele to a level much closer to that of the paternal allele (Fig. 3C). All mat 13 tumours showed elevated Igf2 expression and no correlation was found between Igf2 mRNA levels and Igf2 imprinting (Fig. 3D). So, Igf2 was highly expressed in many liver tumours developed in the pat 13 mice and the maternal Igf2 allele was efficiently derepressed in many tumours arisen in the mat 13 mice. Thus the region deleted in H1913, which is required for activation of the paternal Igf2 allele in early tumourigenesis, is not necessary in the late stages. Consistent with this observation, maternal transmission of H1913 results in derepression of the maternal Igf2 allele in the liver tumours but not in early liver tumourigenesis.

View larger version (49K): [in this window] [in a new window]   Figure 3. Effect of H1913 on Igf2 expression and imprinting in late stages of liver tumourigenesis. (A and B) Analysis of liver tumours arisen in mice following paternal transmission of the H19 13 deletion. RNAs extracted from the peri-neoplastic liver tissue (P) or from the liver tumours (T1–T8) developed in two 180-day-old CRP-Tag Pat 13 male mice were analysed by allele-specific RT–PCR with Igf2-specific primers (A) or RNase protection with Igf2- and Gapdh-specific probes (B) as in Figure 2. (C and D) Analysis of liver tumours arisen in mice following maternal transmission of the H1913 deletion. RNAs extracted from the peri-neoplastic liver tissue (P) or from three independent liver tumours (T1–T3) arisen in a 120-day-old CRP-Tag mat 13 male mouse were analysed by allele-specific RT–PCR with Igf2-specific (C) or RNase protection with Igf2- and Gapdh-specific probes (D) as in Figure 2. The tumour samples shown in this figure are representative of a large series of liver neoplasms analysed that gave very similar results. Note that the paternal Igf2 allele is actively expressed in many liver tumours developed in the pat 13 mice and the maternal Igf2 allele is derepressed in the liver tumours developed in the mat 13 mice.

 
Igf2 activation is not affected by deletion of the H19 gene
The region missing in the H1913 mice including the H19 gene and 10 kb of 5' region is required for Igf2 activation during early liver tumourigenesis. In order to better define the DNA region needed for Igf2 expression, we took advantage of two mouse lines carrying smaller deletions of the H19 locus shorter than H1913. The first of these lines (3), generated by Ripoche et al. (14), harbours a deletion covering the entire transcription unit of the H19 gene and 50 bp of its promoter (Fig. 1). As with the H1913 line, the heterozygous H193 mice were reciprocally crossed with the homozygous SV40 Tag transgenic mice and imprinting and overall expression of the Igf2 gene were analysed in their offspring. These mice carried the SV40 Tag oncogene in hemizygosis and either the wild-type H19 locus in homozygosis or H193 in heterozygosis (either paternally inherited, pat 3 or maternally inherited, mat 3). Igf2 imprinting and overall expression were analysed in the newborn and in the adult mice early after the induction of tumourigenesis (3 day-LPS treatment). The results obtained showed that the H193 mutation (either paternally or maternally inherited) did not affect Igf2 imprinting and expression in the liver of the newborns, consistent with previous data (14 and data not shown). Similarly, no difference was observed in Igf2 imprinting and overall Igf2 expression between the wt and either the pat 3 or mat 3 adult mice after LPS-induction of the transgenic oncogene (data not shown). This clearly demonstrates that the region deleted in H193 is dispensable for Igf2 activation in the liver, both in normal development and tumourigenesis.

The H19 DMD is required for Igf2 activation during early liver tumourigenesis
The observations that the maternal Igf2 allele is not derepressed in the mat 13 mice and the paternal Igf2 allele is not activated in the pat 13 mice during early liver tumourigenesis but Igf2 activation is normal in the pat 3 mice, indicate that the cis-acting sequences required for Igf2 activation resides in the 10 kb region upstream of H19. We then analysed the H19DMD line harbouring a 1.6 kb deletion of the H19 DMD (13) (Fig. 1). H19DMD causes reactivation of the maternal Igf2 allele associated with a reduction of H19 expression in the embryos following maternal transmission of the mutation and reactivation of the paternal H19 allele associated with a reduction of Igf2 expression in the embryos following its paternal transmission (13). Also with the H19DMD mice, we followed the scheme of reciprocal crosses with the SV40 Tag mice described above. In this case, the expression and imprinting of both the Igf2 and H19 genes were analysed in the liver of the offspring (Fig. 4). Consistent with the results of Thorvaldsen et al. (13), we observed derepression of the maternal Igf2 allele in the mat DMD newborns (mat/pat ratio, 0.51; Fig. 4A, lane 8) and derepression of the paternal H19 allele in the pat DMD mice (pat/mat ratio, 0.41; Fig. 4A, lane 13). Also consistent with what was reported by these authors, the RNase protection assay showed a 2–3-fold reduction in the overall Igf2 mRNA level of the pat DMD newborn mice and a 2-fold reduction in the H19 RNA level of the mat DMD newborns, when compared with the wt animals (Fig. 4B). We then analysed the effect of DMD in the adult mice, after induction of the transgenic oncogene by LPS (Fig. 4C and D). Differently from what was observed in the mat and pat DMD newborns (Fig. 4A, lanes 8 and 13), the maternal Igf2 allele was not derepressed in the adult mat DMD mice (mat/pat ratio, 0.03; Fig. 4C, lane 7) and the paternal H19 allele was poorly activated in the adult pat DMD mice (pat/mat ratio, 0.12; see Fig. 4C, lane 11). Also, the overall activation of the Igf2 gene was very inefficient (11-fold less than wt) in the adult pat DMD mice (Fig. 4D) and the overall H19 activation significantly impaired (5-fold weaker than wt) in the adult mat DMD mice (Fig. 4D), when compared with the pat and mat DMD newborns (Fig. 4B). These results clearly demonstrate that the H19 DMD is required for the activation of the Igf2 (and H19) gene during early tumourigenesis. Interestingly, the H19DMD deletion also affects Igf2 (and H19) expression in neonatal liver, but this effect becomes dramatic upon induction of tumourigenesis in adult liver. Moreover, in the LPS-treated adult pat DMD mice, the lack of activation of the paternal Igf2 allele did not correspond to a stronger reactivation of the paternal H19 allele and, in the LPS-treated adult mat DMD mice, the weaker activation of the maternal H19 allele was not compensated by a more intense derepression of the maternal Igf2 allele. Rather, the DMD deletion affected the activation of both the Igf2 and H19 genes on the mutant chromosome, indicating that the effect of the inhibition in the adult liver was not mediated by promoter competition (in the absence of the insulator) but likely caused by a positive regulatory function of the DMD.

View larger version (43K): [in this window] [in a new window]   Figure 4. Effect of the H19DMD deletion on imprinting and overall expression of the Igf2 and H19 genes in neonatal liver and early stages of liver tumourigenesis. (A and B) Analysis of newborns. RNAs extracted from the liver of 1-day-old mice transgenic for the SV40 Tag oncogene following paternal (pat DMD) or maternal (mat DMD) transmission of the H19DMD deletion and corresponding wt were analysed by allele-specific RT–PCR with Igf2-specific or H19-specific primers (A) or RNase protection with Igf2-, H19- and Gapdh-specific probes (B). (C and D) Analysis of early stages of liver tumourigenesis. RNAs extracted from 2-month-old female mice with the same genotypes as (A and B) after induction of the transgenic oncogene by bacterial LPS (3, day treatment) were analysed as in (A, B). RT-PCR of Igf2 and RNase protection were as in Figure 2. H19 alleles were amplified by using primers flanking a MspI RFLP and the PCR products separated by electrophoresis on non-denaturing polyacrylamide gels after digestion with MspI (34). As described in Figure 2 for the Igf2 gene, the wt mice expressed either the Igf2a (BALB/c) or b (C57BL/6) alleles and either the H19a (C57BL/6) or b (BALB/c) allele, depending on parental inheritance. Note that upon induction of tumourigenesis the maternal Igf2 allele is not derepressed and the maternal H19 allele is only weakly activated in the adult mat DMD mice and the activation of the paternal Igf2 allele is very inefficient and that of the paternal H19 allele severely impaired in the adult pat DMD mice.

 
Accelerated post-natal silencing of the Igf2 gene in the mice lacking the H19 DMD
The results reported in the previous section demonstrates that the H19 DMD has an important positive regulatory function needed for activation of the Igf2 gene during the early phases of liver tumourigenesis occurring in adult mice. This activity is absent or less relevant in the liver tumours and in neonatal liver. We therefore wanted to investigate the effect of the DMD deletion in normal mouse liver during the first 3 weeks of life. In the rodents, the expression of the Igf2 gene is progressively extinguished during this period, in all tissues but choroid plexus and leptomeninges (24,25). The mat DMD mice were analysed first (Fig. 5A and B). These mice express Igf2 biallelically at birth (13). We observed that from post-natal day 1 to day 14 Igf2 expression derived from the mutant DMD chromosome was down-regulated more rapidly than that from the wild-type allele (Fig. 5A). At day 1, the expression from the wild-type allele exceeded that of the mutant allele 2-fold, while at day 14 the difference was 12-fold. This effect was liver-specific, since the Igf2 alleles from the mutant DMD and wild-type chromosomes had equivalent expression at post-natal day 14 in skeletal muscle, spleen, gut and brain (Fig. 5B). We then analysed the effect of the paternal transmission of DMD. The results showed that in the liver the post-natal decline of the Igf2 mRNA level in the pat DMD mice was more precocious and rapid than in the wt mice, in which Igf2 expression only decreases after day 7 (Fig. 5C). However, Igf2 expression profiles were similar for wt and pat DMD mice in the gut, skeletal muscle and heart between day 1 and day 21 after birth (Fig. 5D). These results clearly indicate that the H19 DMD has two distinct functions in the control of Igf2 expression: a positive role which is exerted by the paternal allele and is particularly relevant in the post-natal liver and a negative role which is exerted on the maternal chromosome and is needed for imprinting control in embryonic–neonatal liver and in non-hepatic tissues.

View larger version (40K): [in this window] [in a new window]   Figure 5. Effect of the H19DMD deletion on post-natal silencing of the Igf2 gene. (A and B) Analysis of the mice following maternal (mat DMD) transmission of the H19DMD deletion. RNAs extracted from the liver of 1- to 14-day-old mat DMD mice (A) or from skeletal muscle, spleen, gut and brain of 14-day-old mat DMD mice (B) were analysed by allele-specific RT–PCR with Igf2-specific primers as in Figure 2A. (C and D) Analysis of the mice following paternal (pat DMD) transmission of the H19DMD deletion. RNAs extracted from the liver (C) or from skeletal muscle, gut and heart of 1- to 21-day-old pat DMD mice or corresponding wild-type (D) were analysed by RNase protection with Igf2- and Gapdh-specific probes as in Figure 2B. Levels of Igf2 mRNA relative to day 1 are reported as histograms. Note the accelerated post-natal silencing of the maternal Igf2 allele in the mat DMD mice and the accelereted post-natal silencing of the paternal Igf2 allele in the pat DMD mice.

 

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
During embryogenesis, the H19 DMD controls the reciprocal imprinting of the Igf2 and H19 genes by means of its methylation-sensitive insulator activity (13,16–20). In this paper we show that the H19 DMD plays a different role in the control of Igf2 expression in adult liver. In the early phases of liver tumourigenesis and in normal liver since the second post-natal week, the H19 DMD is strongly required for activation of the Igf2 gene. This is demonstrated by the poor activation of the paternal Igf2 allele in the mice following paternal transmission of the H19 DMD deletion and by the lack of derepression of the maternal Igf2 allele in the mice following maternal transmission of the H19 DMD. However, in the liver tumours the paternal Igf2 allele is activated and the maternal Igf2 allele is derepressed in the absence of the H19 DMD in cis, indicating that the positive function of this regulatory region is dispensable in late liver tumourigenesis and that the H19 DMD is required on the maternal chromosome for imprinting control at this stage.

In neonatal liver, the expression of the paternal Igf2 allele is unaffected by the H1913 deletion in cis, while it is reduced 3-fold by the H19DMD deletion (12,13,26). We observed that the activation of the paternal Igf2 allele was strongly impaired (>10-fold) in the adult liver of both pat 13 and pat DMD mice, upon induction of tumourigenesis. The severity of Igf2 inhibition was confirmed by the observation that tumour formation was significantly delayed in the liver of the pat 13 mice. In addition, a progressively increasing defect in the expression of the Igf2 gene from the paternal DMD chromosome was detected between day 1 and day 14 after birth, a time window during which liver maturation is completed. No such effect was observed in tissues other than the liver, indicating that the positive regulatory function of the H19 DMD in the control of Igf2 expression is liver-specific and developmentally regulated. It has been proposed that the reason DMD affects paternal Igf2 expression in neonatal liver and H1913 does not is the presence of the H19 gene in DMD (13). At this developmental stage, this gene is derepressed and may compete with Igf2 on the paternal chromosome of the pat DMD mice, once the insulator has been removed. In the adult liver, however, we observed that Igf2 activation occurring during early tumourigenesis is affected by the lack of the H19 DMD even in the absence of the H19 gene (13). Moreover, the lack of Igf2 activation is not associated with the derepression of the paternal H19 allele (DMD). This clearly shows that the paternal H19 DMD has a direct positive control function on Igf2 expression in the liver.

The positive function of the H19 DMD in the liver is particularly relevant post-natally. After birth, Igf2 is down-regulated in the majority of rodent tissues. However, the kinetic of this decline shows tissue-specific differences, with the liver being the tissue in which Igf2 expression persists longer (24,25). Interestingly, an element controlling negatively the post-natal expression of Igf2 in the mesodermal tissues has been identified near the Igf2 gene (25). It has been proposed that the silencing of the Igf2 gene may be linked to major events of tissue maturation occurring over the perinatal and neonatal period (24). It is possible that the fine regulation of the levels of the Igf2 peptide required in the various tissues at this developmental stage is exerted through the activities of several tissue-specific cis-acting elements.

Maternal transmission of the H1913 and H19DMD deletions results in derepression of the maternal Igf2 allele in newborn mice (12,13). Strikingly, we observed no activation of the maternal Igf2 allele in the adult mat 13 and mat DMD mice during the early phases of liver tumourigenesis and in the mat DMD mice in normal liver at post-natal day 14. Since the effect of the deletions on the maternal Igf2 allele paralleled that on the paternal allele, it is likely that the lack of activation of the maternal Igf2 allele is due to the absence of the positive function of the H19 DMD on the maternal chromosome. Thorvaldsen et al. (13,26) reported that the presence of DMD in cis decreases the expression of the maternal H19 allele in newborn mice in a tissue-specific manner. We observed that maternal transmission of DMD strongly affects H19 expression in early liver tumourigenesis, indicating that the positive function of the H19 DMD on H19 expression is particularly relevant in the adult liver. This is consistent with the observation that the paternal H19 allele is not derepressed in the pat DMD mice during early liver tumourigenesis.

The data available do not define how the DMD exerts its positive function on Igf2 and H19 expression. The H19 DMD is methylated on the paternal chromosome and unmethylated on the maternal one throughout mouse development (11). We have not observed any evidence of demethylation in this region upon liver tumourigenesis (data not shown). Therefore, our results indicate that the deletion of the methylated copy of the DMD prevents the activation of the paternal Igf2 allele while the deletion of the non-methylated DMD strongly affects the activation of the maternal H19 allele in early liver tumorigenesis. Both activities could be mediated by the same cis-acting element and the interaction with the cognate trans-acting factor not influenced by DNA methylation. On the wt maternal chromosome, the insulator function of the DMD would limit the effect of this positive element to the H19 promoter. Alternatively, DNA methylation may be required for Igf2 activation on the paternal chromosome. In this context, it is worth mentioning that two additional DMRs (DMR1 and DMR2) methylated on the paternal chromosome are present in the Igf2 gene (27) and that one of these (DMR2) is a positive element, whose activity is increased by DNA methylation (28). The H19 DMD lacks liver-specific enhancer activity, if transfected into human hepatoma cells (16,29). It is possible that an enhancer function is evident only in normal liver or that the H19 DMD is required for the higher-order chromatin structure of the Igf2-H19 locus and this activity may be difficult to detect in transfection studies.

Contrary to the early stages of tumourigenesis, the paternal Igf2 allele was strongly activated in many liver tumours that arose in the pat 13 mice and the maternal Igf2 allele was efficiently derepressed in the majority of the neoplasms developed in the mat 13 mice. This indicates that the positive function of the H19 DMD is not absolutely required for Igf2 activation in late stages of liver tumourigenesis. However, the unmethylated maternal copy of the H19 DMD is needed for maintaining Igf2 imprinting at this stage. Therefore, the H19 DMD changes its function during liver tumourigenesis and plays similar roles in normal neonatal liver and liver tumours. Perhaps, in mitotically active cells, such as those of the embryonic–neonatal liver or liver tumours, the endodermal enhancer can activate the Igf2 promoter more efficiently, but the auxiliary function of the H19 DMD might be needed in a quiescent tissue such as the adult liver. Alternatively, the trans-acting factors interacting with the endodermal enhancer may be more abundant or active in the embryonic–neonatal liver and liver tumours than in the normal adult liver. The genetic heterogeneity (high in the SV40 Tag-dependent tumours) and/or the different grade of differentiation can explain the absence of Igf2 activation in a minority of liver tumours. An important consideration on the function of Igf2 in tumourigenesis also derives from these results. Although Igf2 activation was reduced only in early liver tumourigenesis, tumour development was significantly delayed in the pat 13 mice. This indicates that, at least in this experimental model, Igf2 expression is required for tumour formation since the very early pre-neoplastic stages of tumourigenesis.

In summary, our results support a model in which two different functions of the H19 DMD are involved in the control of Igf2 expression in the liver and at least one of these activities is under developmental control (Fig. 6). On the maternal chromosome, the unmethylated DMD works as enhancer-blocking barrier and insulates the Igf2 promoter from the downstream enhancers. On the paternal chromosome, the methylated DMD has a positive regulatory function. This activity is already present in neonatal liver, but becomes absolutely required together with that of the endodermal enhancer in later development and conditions in which Igf2 is reactivated in adult animals, such as during tumourigenesis. The positive function of the H19 DMD is liver-specific and becomes dispensable in the late stages of liver tumourigenesis. Our results and data reported by other groups (26) indicate that the maternal H19 DMD also has an activator function for the H19 gene. In addition, a silencer activity toward the H19 promoter has been attributed to the methylated paternal DMD (30). This makes the H19 DMD a multi-functional and developmentally controlled regulatory region. Further experimental work is needed to identify the elements and the protein factors implicated in each specific activity.

View larger version (21K): [in this window] [in a new window]   Figure 6. Model for developmental and tissue-specific control of the H19 DMD function. The H19 and Igf2 genes and the DMD and endodermal enhancers are shown as in Figure 1. The allele-specific DNA methylation of H19 and of the DMD is indicated as CH3. The positive regulatory actions of the DMD and endodermal enhancers on the expression of Igf2 and H19 are indicated by horizontal arrows. No data are available on the role of the maternal DMD in the control of H19 expression in late liver tumourigenesis (B, +*). Note that the positive effect of the paternal H19 DMD on Igf2 expression is strong (thick arrow) in normal liver at post-natal day 14 and during early liver tumourigenesis (A), weak (thin arrow) in normal embryonic-neonatal liver and late liver tumourigenesis (B) and absent (no arrow) in tissues other than the liver (C). In addition, the positive effect of the maternal DMD on H19 expression is strong (thick arrow) in normal post-natal day 14 liver and in early liver tumourigenesis (A), weak (thin arrow) in normal embryonic-neonatal liver (B) and variable (broken arrows) (26) in tissues other than the liver (C).

 

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Materials
Oligonucleotides were purchased from PRIMM (Milano, Italy). ³²P- and ³⁵S-labelled isotopes were from Amersham (UK). Restriction enzymes were from New England Biolabs. All other materials were of the best grade commercially available.

Animals
The SV40-Tag 60-3, H1913, H193 and H19DMD mouse transgenic lines have been previously described (12–14,23). In order to distinguish paternal from maternal alleles at the Igf2 and H19 loci, animals of the CRP-Tag 60-3 line were maintained in a BALB/c background and the H19 deletion lines in a C57BL/6 background. Reciprocal crosses were then obtained between homozygote CRP-Tag 60-3 mice and animals heterozygote for each of the three H19 deletion lines. Igf2 and H19 expression was analysed in their progeny. The mice of each litter were typed for the presence of the H1913, H193 or H19DMD as described (12–14). The 2-month-old females were treated with intraperitoneal injections of E. coli lipopolysaccharide (Sigma) for 3–90 days, as previously described (23). The males were not treated and sacrificed between 120 and 180 days of age. Samples of pre-neoplastic liver and liver tumours were isolated in these animals and used for histological examination and nucleic acid analysis.

Isolation of DNA and RNA
Genomic DNA was prepared from tissues by standard proteinase K digestion and phenol–chloroform extraction methods. Total RNA was isolated using the single-step acid guanidinium thiocyanate–phenol–chloroform extraction method (31).

H19 and Igf2 allelotype analysis
The MspI RFLP of the H19 gene and the VNTR of the Igf2 gene were analysed by PCR as described by Casola et al. (21) and Pedone et al. (32).

RNase protection
RNase protection assays were carried out with the previously described Igf2-, H19- and Gapdh-specific probes (21,22). The Igf2 probe derived from the 3' untraslated sequence and detected transcript originated from all Igf2 promoters. The reaction conditions were those reported by Ungaro et al. (33). RNA levels were quantified by scanning the gels using a Molecular Dynamics PhosphorImager and ImageQuaNT software.

	ACKNOWLEDGEMENTS

 
We wish to thank S.M. Tilghman for the H19 13 mouse line, M.S. Bartolomei for the H19 DMD line and G. Ciliberto for the SV40-Tag line, and M. Constancia and A. Murrell for critical reading of the manuscript. We also acknowledge Giovanni and Salvatore Sequino for excellent technical assistance in the animal house. This work was supported by grants from Associazione Italiana Ricerca sul Cancro and MURST 60% (to A.R.).

	FOOTNOTES

 
^* To whom correspondence should be addressed. Tel: +39 823274599; Fax: +39 823274605; Email: andrea.riccio{at}unina2.it

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 

1. Tycko, B. and Morison, I.M. (2002) Physiological functions of imprinted genes. J. Cell Physiol., 192, 245–258.[CrossRef][Web of Science][Medline]

2. Maher, E.R. and Reik, W. (2000) Beckwith–Wiedemann syndrome: imprinting in clusters revisited. J. Clin. Invest., 105, 247–252.[Web of Science][Medline]

3. Feinberg, A.P., Cui, H. and Ohlsson, R. (2002) DNA methylation and genomic imprinting: insights from cancer into epigenetic mechanisms. Semin. Cancer Biol., 12, 389–398.[CrossRef][Web of Science][Medline]

4. Reik, W. and Walter, J. (2001) Genomic imprinting: parental influence on the genome. Nat. Rev. Genet., 2, 21–32.[Web of Science][Medline]

5. Ferguson-Smith, A.C. and Surani, M.A. (2001) Imprinting and the epigenetic asymmetry between parental genomes. Science, 293, 1086–1089.[Abstract/Free Full Text]

6. Rand, E. and Cedar, H. (2003) Regulation of imprinting: A multi-tiered process. J. Cell. Biochem. 88, 400–407.[CrossRef][Web of Science][Medline]

7. Brannan, C.I. and Bartolomei, M.S. (1999) Mechanisms of genomic imprinting. Curr. Opin. Genet. Dev., 9, 164–170.[CrossRef][Web of Science][Medline]

8. 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.[Abstract/Free Full Text]

9. Kaffer, C.R., Grinberg, A. and Pfeifer, K. (2001) Regulatory mechanisms at the mouse Igf2/H19 locus. Mol. Cell. Biol., 21, 8189–8196.[Abstract/Free Full Text]

10. Davies, K., Bowden, L., Smith, P., Dean, W., Hill, D., Furuumi, H., Sasaki, H., Cattanach, B. and Reik, W. (2002) Disruption of mesodermal enhancers for Igf2 in the minute mutant. Development, 129, 1657–1668.[Abstract/Free Full Text]

11. Tremblay, K.D., Duran, K.L. and Bartolomei, M.S. (1997) A 5' 2-kilobase-pair region of the imprinted mouse H19 gene exhibits exclusive paternal methylation throughout development. Mol. Cell. Biol., 17, 4322–4329.[Abstract]

12. 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.[CrossRef][Medline]

13. 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.[Abstract/Free Full Text]

14. 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. Genes Dev., 11, 1596–1604.[Abstract/Free Full Text]

15. 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.[Abstract/Free Full Text]

16. Hark, A.T., Schoenherr, C., 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.[CrossRef][Medline]

17. Bell, A.C. and Felsenfeld, G. (2000) Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene. Nature, 405, 482–485.[CrossRef][Medline]

18. Kanduri, C., Pant, V., Loukinov, D., Pugacheva, E., Qi, C.F., Wolffe, A., Ohlsson, R. and Lobanenkov, V.V. (2000) Functional association of CTCF with the insulator upstream of the H19 gene is parent of origin-specific and methylation-sensitive. Curr. Biol., 10, 853–856.[CrossRef][Web of Science][Medline]

19. Szabo, P., Tang, S.H., 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.[CrossRef][Medline]

20. Schoennher, C.J., Levorse, J.M. and Tilghman, S.M. (2003). CTCF maintains differential methylation at the Igf2/H19 locus. Nat. Genet., 33, 66–69.[CrossRef][Web of Science][Medline]

21. Casola, S., Ungaro, P., Pedone, P.V., Lazzaro, D., Fattori, E., Ciliberto, G., Zarrilli, R., Bruni, C.B. and Riccio, A. (1995). Loss of heterozygosity of imprinted genes in SV40 t/T antigen-induced hepatocellular carcinomas. Oncogene, 11, 711–721.[Web of Science][Medline]

22. Vernucci, M., Cerrato, F., Besnard, N., Casola, S., Pedone, P.V., Bruni, C.B. and Riccio, A. (2000) The H19 endodermal enhancer is required for IGF2 activation and tumour formation in experimental liver carcinogenesis. Oncogene, 19, 6376–6385.[CrossRef][Web of Science][Medline]

23. Ruther, U., Woodroofe, C., Fattori, E. and Ciliberto, G. (1993) Inducible formation of liver tumors in transgenic mice. Oncogene, 8, 87–93.[Web of Science][Medline]

24. Brown, A.L., Graham, D.E., Nissley, S.P., Hill, D.J., Strain, A.J. and Rechler, M.M. (1986) Developmental regulation of insulin-like growth factor II mRNA in different rat tissues. J. Biol. Chem., 261, 13144–13150.[Abstract/Free Full Text]

25. Constancia, M., Dean, W., Lopes, S., Moore, T., Kelsey, G. and Reik, W. (2000) Deletion of a silencer element in lgf2 results in loss of imprinting independent of H19. Nat. Genet., 26, 203–206.[CrossRef][Web of Science][Medline]

26. Thorvaldsen, J.L., Mann, M.R., Nwoko, O., Duran, K.L. and Bartolomei, M.S. (2002) Analysis of sequence upstream of the endogenous H19 gene reveals elements both essential and dispensable for imprinting. Mol. Cell. Biol., 22, 2450–2462.[Abstract/Free Full Text]

27. Lopes, S., Lewis, A., Hajkova, P., Dean, W., Oswald, J., Forne', T., Murrell, A., Constancia, M., Bartolomei, M., Walter, J. and Reik, W. (2003). Epigenetic modifications in an imprinting cluster are controlled by a hierarchy of DMRs suggesting long-range chromatin interactions. Hum. Mol. Genet., 12, 295–305.[Abstract/Free Full Text]

28. Murrell, A., Heeson, S., Bowden, L., Constancia, M., Dean, W., Kelsey, G. and Reik, W. (2001) An intragenic methylated region in the imprinted Igf2 gene augments transcription. EMBO Rep., 2, 1101–1106.[CrossRef][Web of Science][Medline]

29. Yoo-Warren, H., Pachnis, V., Ingram, R.S. and Tilghman, S.M. (1988) Two regulatory domains flank the mouse H19 gene. Mol. Cell. Biol., 8, 4707–4715.[Abstract/Free Full Text]

30. Drewell, R.A., Brenton, J.D., Ainscough, J.F., Barton, S.C., Hilton, K.J., Arney, K.L., Dandolo, L. and Surani, M.A. (2000) Deletion of a silencer element disrupts H19 imprinting independently of a DNA methylation epigenetic switch. Development, 127, 3419–3428.[Abstract]

31. Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Anal. Biochem., 162, 156–159.[Web of Science][Medline]

32. Pedone, P.V., Pikaart, M.J., Cerrato, F., Vernucci, M., Ungaro, P., Bruni, C.B. and Riccio, A. (1999) Role of histone acetylation and DNA methylation in the maintenance of the imprinted expression of the H19 and Igf2 genes. FEBS Lett., 458, 45–50.[CrossRef][Web of Science][Medline]

33. Ungaro, P., Casola, S., Vernucci, M., Pedone, P.V., Bruni, C.B. and Riccio, A. (1997). Relaxation of insulin-like growth factor-2 imprinting in rat cultured cells. Mol. Cell. Endocrinol., 135, 153–163.[CrossRef][Web of Science][Medline]

34. Pachnis, V., Belayew, A. and Tilghman, S.M. (1984) Locus unlinked to alpha-fetoprotein under the control of the murine raf and Rif genes. Proc. Natl Acad. Sci. USA, 81, 5523–5527.[Abstract/Free Full Text]

CiteULike    Connotea    Del.icio.us    What's this?

This article has been cited by other articles:

				T. Yoshimizu, A. Miroglio, M.-A. Ripoche, A. Gabory, M. Vernucci, A. Riccio, S. Colnot, C. Godard, B. Terris, H. Jammes, et al. The H19 locus acts in vivo as a tumor suppressor PNAS, August 26, 2008; 105(34): 12417 - 12422. [Abstract] [Full Text] [PDF]

				W. P. Tsang, T. W.L. Wong, A. H.H. Cheung, C. N.N. Co, and T. T. Kwok Induction of drug resistance and transformation in human cancer cells by the noncoding RNA CUDR RNA, June 1, 2007; 13(6): 890 - 898. [Abstract] [Full Text] [PDF]

				A. Sparago, S. Russo, F. Cerrato, S. Ferraiuolo, P. Castorina, A. Selicorni, C. Schwienbacher, M. Negrini, G. B. Ferrero, M. C. Silengo, et al. Mechanisms causing imprinting defects in familial Beckwith-Wiedemann syndrome with Wilms' tumour Hum. Mol. Genet., February 1, 2007; 16(3): 254 - 264. [Abstract] [Full Text] [PDF]

				N. Engel, J. L. Thorvaldsen, and M. S. Bartolomei CTCF binding sites promote transcription initiation and prevent DNA methylation on the maternal allele at the imprinted H19/Igf2 locus Hum. Mol. Genet., October 1, 2006; 15(19): 2945 - 2954. [Abstract] [Full Text] [PDF]

				J. L. Thorvaldsen, A. M. Fedoriw, S. Nguyen, and M. S. Bartolomei Developmental Profile of H19 Differentially Methylated Domain (DMD) Deletion Alleles Reveals Multiple Roles of the DMD in Regulating Allelic Expression and DNA Methylation at the Imprinted H19/Igf2 Locus Mol. Cell. Biol., February 15, 2006; 26(4): 1245 - 1258. [Abstract] [Full Text] [PDF]

				Y Fellig, I Ariel, P Ohana, P Schachter, I Sinelnikov, T Birman, S Ayesh, T Schneider, N de Groot, A Czerniak, et al. H19 expression in hepatic metastases from a range of human carcinomas J. Clin. Pathol., October 1, 2005; 58(10): 1064 - 1068. [Abstract] [Full Text] [PDF]

This Article Abstract FREE Full Text (PDF) All Versions of this Article: 13/3/353    most recent ddh028v1 Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to citation manager Search for citing articles in: ISI Web of Science (9) Request Permissions Google Scholar Articles by Vernucci, M. Articles by Riccio, A. Search for Related Content PubMed PubMed Citation Articles by Vernucci, M. Articles by Riccio, A. Social Bookmarking          What's this?

Online ISSN 1460-2083 - Print ISSN 0964-6906

Copyright © 2009 Oxford University Press

Oxford Journals Oxford University Press

• Site Map
• Privacy Policy
• Frequently Asked Questions

Other Oxford University Press sites: Oxford University Press Oxford Journals China Oxford Journals Japan American National Biography Booksellers' Information Service Children's Fiction and Poetry Children's Reference Corporate & Special Sales Dictionaries Dictionary of National Biography Digital Reference English Language Teaching Higher Education Textbooks Humanities International Education Unit Law Medicine Music Online Products Oxford English Dictionary Reference Rights and Permissions Science School Books Social Sciences Very Short Introductions World's Classics