Parental genomic imprinting of the human IGF2 gene

p57K1P2 is expressed in Wilms' tumor with LOH of 11p15.5

p57KIP2 is a cyclin-dependent kinase inhibitor that maps to human chromosome band 11p15.5, placing it in a genomically imprinted region that has been implicated in the etiology of Wilms' tumor and in the Beckwith-Wiedemann syndrome. Recent analysis of p57KIP2 expression in the mouse has determined that this gene is exclusively expressed from the maternal allele. It has been suggested that p57KIP2 is the WT2 tumor suppressor gene in the 11p15.5 region. We have used reverse transcriptase PCR to determine whether loss of p57KIP2 expression occurs in Wilms' tumor samples that have undergone maternal loss of heterozygosity of 11p15.5. p57KIP2 mRNA was amplified in both the Wilms' tumor tissue and in normal kidney tissue of all five patients analyzed. Semi-quantitative PCR analyses demonstrated that the relative level of p57KIP2 expression in tumor tissue is not markedly different from that in normal kidney. Our data indicate that if the p57KIP2 gene is imprinted in humans and expressed exclusively from the maternal allele, reactivation of the paternal allele has occurred in all five Wilms' tumor samples analyzed in this study. Sequence analysis of the p57KIP2 Cdk inhibitory domain in genomic DNA from primary and secondary tumors from two patients showed only a single base change in one secondary WT, resulting in a predicted methionine to isoleucine substitution at amino acid position 70. These studies suggest that p57KIP2 may not be the WT2 gene.

Genomic imprinting in disorders of growth

This review has briefly considered some of the vast amount of information that has been gathered on genomic imprinting and its role in PWS, AS, BWS and Russell-Silver syndrome. The pace of investigation into the phenomenon of imprinting will undoubtedly continue, because our understanding remains far from complete. Newer approaches to identifying imprinted genes based on their expression rather than their location are likely to uncover currently unknown genes. We can also look forward to more insight into the fascinating complexities of the imprinting process.

Promoter-specific modulation of insulin-like growth factor II genomic imprinting by inhibitors of DNA methylation

The insulin-like growth factor II (IGF-II) gene is maternally imprinted in most normal tissues with only the paternal allele being transcribed. In several human tumors, however, IGF-II is expressed from both parental alleles. To explore the underlying mechanism of IGF-II imprinting, we have examined the effect of DNA demethylation in cultured human and mouse astrocyte cells. An increased expression of IGF-II was observed when these cells were treated with the DNA demethylating agents, 5-azacytidine or 2-deoxy-5-azacytidine. Allelic analysis indicated that, following DNA demethylation, the increment in IGF-II mRNA was primarily derived from the normally suppressed maternal allele. Examination of promoter usage revealed that only the most proximal promoter (mP3 in mouse and hP4 in human) responded to DNA demethylating agents, whereas the expression of IGF-II from the other promoters remained unchanged. The enhanced expression of IGF-II from these promoters suggests the presence of a methylation-response element in or near mP3 and hP4. This study indicates that DNA demethylating agents increase IGF-II expression primarily by stimulating the normally imprinted allele through the activation of the most proximal IGF-II promoter.

Imprinted genes and regulation of gene expression by epigenetic inheritance

Six new imprinted genes have recently been identified by association with established imprinted regions, in systematic screens or by serendipity. This brings the total to seventeen imprinted genes, which display a wide variety of functions. Some imprinted genes have been shown to be both physically and mechanistically linked within domains that are under the control of an imprinting centre. Others may apparently undergo imprinting independently. Methylation is clearly required for maintenance of mono-allelic expression while chromatin structure and non-coding RNAs may also play a role.

Alterations in the promoter-specific imprinting of the insulin-like growth factor-II gene in Wilms' tumor

The human IGF2 gene, which encodes a mitogenic peptide required for normal fetal development, is overexpressed in many types of tumors. IGF2 is transcribed from four promoters (P1-P4), and in most tissues, the gene is imprinted. In this study, we have analyzed IGF2 promoter usage and determined the allelic expression from each promoter in 19 ApaI- and 22 AluI-heterozygous Wilms' tumors. Loss of IGF2 imprinting (LOI) was observed in 8 ApaI-informative tumors. In these tumors, each parental allele was expressed in equal abundance, indicating that there was complete relaxation of IGF2 imprinting. In each LOI tumor, expression from promoter P1 as well as from the normally imprinted promoters P2-P4 was biallelic. In the 11 ApaI-informative tumors which maintained IGF2 imprinting (maintenance of imprinting), transcription from promoters P2-P4 was always monoallelic, while transcripts from P1 were derived from either one or both alleles. The lack of consistency of IGF2 imprinting of promoter P1 in maintenance of imprinting tumors was also observed in normal fetal tissues of 6 12 weeks gestation, suggesting a similarity in IGF2 regulation between Wilms' tumors and embryonic tissue development. These data suggest that the increased expression of IGF2 in Wilms' tumor may be caused either by biallelic gene expression in LOI tumors from promoters P2-P4 and/or by a reversion to an earlier stage of development which is characterized by increased synthesis of this fetal growth factor.

Lack of imprinting of the human dopamine D4 receptor (DRD4) gene

The term genomic imprinting has been used to refer to the differential expression of genetic material depending on whether it has come from the male or female parent. In humans, the chromosomal region 11p15.5 has been shown to contain 2 imprinted genes (H19 and IGF2). The gene for the dopamine D4 receptor (DRD4), which is of great interest for research into neuropsychiatric disorders and psychopharmacology, is also located in this area. In the present study, we have examined the imprinting status of the DRD4 gene in brain tissue of an epileptic patient who was heterozygous for a 12 bp repeat polymorphism in exon 1 of the DRD4 gene. We show that both alleles are expressed in equivalent amounts. We therefore conclude that the DRD4 gene is not imprinted in the human brain.

Stage-specific and cell type-specific aspects of genomic imprinting effects in mammals

Recent studies have revealed that maternal and paternal alleles of some imprinted genes are differentially expressed from the earliest time of expression, with virtually no expression from one of the two alleles, while for other imprinted genes the normally silent allele can be transcribed during early development. In addition, a number of imprinted genes manifest their imprints only in select tissues. These observations indicate that the marks that denote parental chromosome origin need not directly determine allele expression, but rather bias later epigenetic modifications toward a particular allele. Thus, factors expressed at specific stages or in specific cell types are required to silence one parental allele or another. Stage-dependent and tissue-specific epigenetic modifications include the progressive establishment of the mature adult parental allele-specific DNA methylation patterns. These changes resemble and may share a common mechanistic basis with other epigenetic modifications that occur during development. Understanding the mechanisms by which these post-fertilization epigenetic modifications are mediated and regulated will be essential for understanding how genomic imprinting leads to differences in parental allele expression.

Good prognosis of cellular mesoblastic nephroma with hyperdiploidy and relaxation of imprinting of the maternal IGF2 gene

A large congenital mesoblastic nephroma (CMN) of combined classical and and cellular histological structure was removed from a 1-month-old female infant. The tumor extended extrarenally and may have been incompletely excised. Tumor tissue showed a mosaic hyperdiploidy with 54 chromosomes in the hyperdiploid line. No other antitumor therapy was given and there has been no recurrence after 4 years. Genomic imprinting normally prevents transcription of the maternal gene for insulin-like growth factor 2 (IGF2). Relaxation of IGF2 imprinting leading to abnormal transcription of the maternal gene is found in a majority of Wilms' tumors and in other malignant neoplasms. The biallelic transcription of IGF2 demonstrated in the CMN from this case is consistent with abnormal transcription of the maternal allele. Relaxation of imprinting of the maternal IGF2 gene or abnormal expression of the gene through other mechanisms may have a role in the genesis of CMN or the cellular subtype.

Monoallelic expression of IGF2 at the human fetal/maternal boundary

IGF2 is expressed in both placental and decidual tissues, enabling an analysis of the parental imprinting over the fetomaternal boundary. Evidence is provided that IGF2 is monoallelically expressed in both placenta and pregnant, as well as nonpregnant, endometrium. These observations suggest that the maternally derived IGF2 allele is inactivated during germline transmission. Comparison of promoter usage in decidua and placental samples shows that the P3 promoter appears to regulated independently of the others. These observations are discussed with respect to current models of IGF2 imprinting and the hypothesized conflict of parental reproductive interests which bears on the phenomenon of parental imprinting.

Parental imprinting effect at the INS-IGF2 diabetes susceptibility locus

Although association of insulin-dependent diabetes mellitus with a haplotype at a locus encompassing the genes for insulin and the insulin-like growth factor II has been well established, two major studies disagree as to whether linkage to this locus is confined to paternally inherited alleles, or is present in alleles transmitted from either parental sex. Towards resolving this discrepancy, we examined parent-of-origin specific association rather than linkage, using the haplotype relative risk method in a mixed Caucasian population. We find that the haplotype relative risk (HRR) conferred by paternal chromosomes was much higher (5.1, p < 0.01) than the corresponding maternal value (2.3, p = 0.07), which narrowly failed to reach statistical significance. Thus, although we cannot exclude an effect of the maternal allele, such an effect appears to be considerably weaker. We review evidence that parental imprinting is genotype-dependent, which may explain the different degrees to which the paternal effect is seen in different populations.

Epigenetic changes encompassing the IGF2/H19 locus associated with relaxation of IGF2 imprinting and silencing of H19 in Wilms tumor

In most tissues IGF2 is expressed from the paternal allele while H19 is expressed from the maternal allele. We have previously shown that in some Wilms tumors the maternal IGF2 imprint is relaxed such that the gene is expressed biallelically. We have now investigated this subset of tumors further and found that biallelic expression of IGF2 was associated with undetectable or very low levels of H19 expression. The relaxation of IGF2 imprinting in Wilms tumors also involved a concomitant reversal in the patterns of DNA methylation of the maternally inherited IGF2 and H19 alleles. Furthermore, the only specific methylation changes that occurred in tumors with relaxation of IGF2 imprinting were solely restricted to the maternal IGF2 and H19 alleles. These data suggest that there has been an acquisition of a paternal epigenotype in these tumors as the result of a pathologic disruption in the normal imprinting of the IGF2 and H19 genes.

Susceptibility to human type 1 diabetes at IDDM2 is determined by tandem repeat variation at the insulin gene minisatellite locus

The IDDM2 locus encoding susceptibility to type 1 diabetes was mapped previously to a 4.1-kb region spanning the insulin gene and a minisatellite or variable number of tandem repeats (VNTR) locus on human chromosome 11p15.5. By 'cross-match' haplotype analysis and linkage disequilibrium mapping, we have mapped the mutation IDDM2 to within the VNTR itself. Other polymorphisms were systematically excluded as primary disease determinants. Transmission of IDDM2 may be influenced by parent-of-origin phenomena. Although we show that the insulin gene is expressed biallelically in the adult pancreas, we present preliminary evidence that the level of transcription in vivo is correlated with allelic variation within the VNTR. Allelic variation at VNTRs may play an important general role in human disease.

Promoter-specific IGF2 imprinting status and its plasticity during human liver development

IGF2 has been shown to be expressed preferentially from the paternally derived allele, although the maternal allele can be found active during both prenatal and postnatal development as well as in neoplastic tumours in humans. We addressed here whether or not the biallelic expression patterns that can be seen during postnatal human liver development reflected a coordinated change in the activities of the four promoters of human IGF2. We show here that the P2, P3 and P4 promoters, but not the P1 promoter, display monoallelic activity in embryonic, neonatal and younger infant liver specimens. The P2, P3 and P4 promoters can, however, be found active either monoallelically or biallelically or even monoallelically on opposite parental alleles in older infant and adult liver specimens. In contrast, H19, which is closely linked to IGF2, is monoallelically expressed in all postnatal liver samples analysed. We conclude that the functional imprinting status of IGF2 during postnatal liver development appears to be promoter/enhancer-specific and either partly or completely independent of H19.

Imprinting of IGF2, insulin-dependent diabetes, immune function, and apoptosis: a hypothesis

Parental genomic imprinting is the phenomenon in which the behavior of a gene is modified, depending on the sex of the transmitting parent [Peterson and Sapienza (1993): Annu Rev Genet 27:7-31]. Recent observations have revealed that the inheritance patterns, age-of-onset, severity, and etiology of certain human diseases can be explained by aberrations in the establishment or the maintenance of the imprint. Examples include the Prader-Willi, Angelman, and Beckwith-Wiedemann syndromes [Nicholls (1994): Am J Hum Genet 54:733-740], malignancy [Sapienza (1990): Biochim Biophys Acta 1072:51-61; Feinberg (1993): Nat Genet 4:110-113], and insulin-dependent diabetes mellitus (IDDM) [Julier et al. (1994) Nature 354:155-159; Bennett et al. (1995) Nat Genet 9:284-292]. We review the evidence that implicates an imprinted gene in the INS-IGF2 region of chromosome 11p15 in the etiology of IDDM (referred to as the IDDM2 locus) and show that in human fetal pancreas, INS is not imprinted, thus providing an argument against INS as the candidate gene. We also examine imprinting effects on the expression of IGF2 in components of the human immune system believed to be important in IDDM and show imprinted expression in fetal thymus as early as 15 weeks gestation. We demonstrate further that in the circulating mononuclear cells of two individuals, lectin-stimulated IGF2 transcription was biallelic, indicating relaxation of imprinting, whereas in one individual, transcription was monoallelic. Finally, we review the current available data supporting a role for insulin-like growth factor-II (IGF-II) in the immune system and, more specifically, discuss the evidence supporting a role for the IGFs in the prevention of apoptosis. These data have led us to formulate a novel hypothesis that could mechanistically explain the involvement of the IDDM2 locus in the pathogenesis of IDDM.

Uniparental disomy and genomic imprinting as causes of human genetic disease

The existence of parent-of-origin differences in the expression of some genes, a process known as genomic imprinting, has been recognized and documented over the past several years. This epigenetic marking process results in the differential expression of normal genes depending upon whether they were inherited from the mother or the father. A number of human disorders have been identified as resulting from alterations in genomic imprinting. One process which can unmask genomic imprinting is uniparental disomy, in which both members of a chromosome pair are contributed by one sex parent. When uniparental disomy is present, genetic abnormality can result either from homozygosity of a single mutant allele which is present in two doses, or from the presence of two copies of an imprinted unexpressed gene or genes, rather than the usual one expressed and one unexpressed. Examples of human genetic disorders that are the consequence of genomic imprinting, and a discussion of current knowledge about the mechanisms of imprinting and the causes of uniparental disomy, are reviewed.

Mechanistic and developmental aspects of genetic imprinting in mammals

Genetic imprinting in mammals allows the recognition and differential expression of maternal and paternal alleles of certain genes. Recent results from a number of laboratories indicate that, at least for some genes, gametic imprints, which must exist in order to mark chromosomes or genes as having been transmitted via sperm or ovum, are not by themselves sufficient to determine allele expression. Other postfertilization events are required, and these events are subject to both tissue-specific and developmental stage-specific regulation. Changes in imprinted gene methylation during preimplantation and fetal life indicate that the establishment of additional allele-specific modifications is likely to contribute to imprinted regulation. Disruptions in imprinting processes, loss of imprints, and loss of nonimprinted alleles through uniparental disomy are likely to contribute to a variety of developmental abnormalities and pathological conditions in both mice and humans.

Promoter-specific imprinting of the human insulin-like growth factor-II gene

Genomic imprinting is a mechanism whereby only one of the two parental alleles is expressed. Loss or relaxation of genomic imprinting has been proposed as an epigenetic mechanism for oncogenesis in a variety of human tumours. Although the mechanism of imprinting is unknown, differential CpG methylation of the parental alleles has been implicated. The human insulin-like growth factor-II (IGF2) gene, which is transcribed from four promoters, P1-P4 (ref. 13), is imprinted in fetal liver but biallelic expression occurs in adult liver. Like most tissues, fetal liver uses primarily promoters P3 and P4 (ref. 17). Adult liver, however, transcribes IGF2 from promoter P1, and it has been suggested that the recruitment of P1 may be responsible for the absence of imprinting in human liver, and in choroid plexus and leptomeninges. We report here that in liver and chondrocytes, IGF2 transcripts from promoter P1 are always derived from both parental alleles, whereas transcripts from promoters P2, P3 and P4 are always from one parental allele. These findings demonstrate that imprinting and a lack of imprinting can both occur within a single gene in a single tissue, suggesting that regional imprinting factors may be important.

Localization of Beckwith-Wiedemann and rhabdoid tumor chromosome rearrangements to a defined interval in chromosome band 11p15.5

Chromosome rearrangements have provided useful landmarks to identify disease loci and have served as starting points for positional cloning strategies for candidate genes. We have used fluorescence in situ hybridization (FISH) and pulsed-field gel electrophoresis (PFGE) to map three Beckwith-Wiedemann syndrome (BWS) breakpoints and a rhabdoid tumor breakpoint more precisely. These breakpoints mapped to the interval between D11S679 and the insulin-like growth factor 2 (IGF2) gene on 11p15.5. A cosmid (c15-2) was identified that mapped centromeric to the BWS t(11;16) and the rhabdoid tumor-associated t(11;22), telomeric to the BWS t(11;22), and was found to span the BWS-associated inv(11) breakpoint. Pulsed-field gel analysis placed all four breakpoints into a 250-675 kb interval distal to D11S679 and at least 270 kb centromeric to the IGF2 and H19 loci. These data locate all three BWS rearrangements and the rhabdoid tumor t(11;22) breakpoint in the same region of 11p15.5, suggesting that they may be affecting the same locus or closely linked loci. Cosmid c15-2 provides a well-defined starting point in the search for candidate disease genes.

Parental origin of transcription from the human GNAS1 gene

Variation in the phenotypic expression of Albright's hereditary osteodystrophy (AHO) determined by the parent of transmission, suggests that the human Gs alpha gene (GNAS1), in which mutations occur in AHO, may be under imprinted control. GNAS1 is also known to map to a chromosomal region (20q13.11) showing syntenic homology with the imprinted mouse region 2E1-2H3. To establish if GNAS1 is indeed imprinted, we have examined the parental origin of GNAS1 transcription in human fetal tissues. Of 75 fetuses genotyped, at gestational ages ranging from 6 to 13 weeks, 13 heterozygous for a FokI polymorphism in exon 5 of GNAS1 were identified whose mothers were homozygous for one or other allele. RNA from up to 10 different tissues from each fetus was analysed by RT-PCR. In all cases expression from both parental alleles was shown by FokI digestion of RT-PCR products and quantification of the resulting fragments. No tissue specific pattern of expression was discerned in these experiments. If genomic imprinting regulates the expression of the human GNAS1 gene, our data suggest that the effect must either be subtle and quantitative, or be confined to a small subset of specialised hormone responsive cells within the target tissues.

Human maternal uniparental disomy for chromosome 16 and fetal development

Two severely growth-retarded fetuses found to have maternal uniparental disomy (UPD) for chromosome 16 and trisomy 16 placental mosaicism both had an unfavourable outcome. Antenatally, the first case was complicated by an unexplained raised maternal serum alpha-fetoprotein concentration, preterm premature rupture of the membranes, and growth retardation detectable at 21 weeks' gestation, whilst the other had an unexplained raised maternal serum human chorionic gonadotrophin level, a two-vessel cord on ultrasound, and cessation of growth at 25 weeks. At post-mortem, both babies had an imperforate anus. Fetal maternal UPD may explain the poor outcome that occurs in some cases of confined placental mosaicism for chromosome 16 and is also associated with specific fetal abnormalities.

Loss of imprinting of IGF2 is linked to reduced expression and abnormal methylation of H19 in Wilms' tumour

The insulin-like growth factor-II (IGF2) and H19 genes are imprinted in mouse and human, with expression of the paternal IGF2 and maternal H19 alleles. IGF2 undergoes loss of imprinting (LOI) in most Wilms' tumours (WT). We now show that: (i) LOI of IGF2 is associated with a 80-fold down regulation of H19 expression; (ii) these changes are associated with alterations in parental-origin-specific, tissue-independent sites of DNA methylation in the H19 promoter; and (iii) loss of heterozygosity is also associated with loss of H19 expression. Thus, imprinting of a large domain of the maternal chromosome results in a reversal to a paternal epigenotype. These data also suggest an epigenetic mechanism for inactivation of H19 as a tumour suppressor gene.

Parental imprinting of autosomal mammalian genes

The molecular mechanisms introducing epigenetic modifications that lead to differential silencing of some autosomal alleles depending on their parental legacy are still largely unknown, but recent results from studies of endogenously imprinted genes and particular transgenes make DNA methylation a strong candidate. At the same time, these results have raised new questions about the details of the imprinting process.

Spatially restricted imprinting of mouse chromosome 7

Three of the four known imprinted genes (Igf-2, H19, and Snrpn) map to mouse chromosome 7. We used mRNA phenotyping to examine the tissue-specific transcription of Igf-1r, H-ras-1, and Gabrb3, which map to chromosome 7 between Snrpn and the Igf-2/H19 domain, and Myod-1, which maps proximal to Snrpn. We found that all of these genes were expressed by both parental alleles in tissues from day 1 neonates. The fact that imprinted genes can flank or map closely to genes that escape such epigenetic modification suggests that autosomal imprinting is not manifested globally along imprinted chromosomes but rather is spatially restricted, perhaps even defined by specific DNA consensus sequences or an "imprint box" associated with imprintable genes.

Mosaic and polymorphic imprinting of the WT1 gene in humans

We have examined the imprinting of the Wilms' tumour suppressor gene (WT1) in human tissues. We confirm that WT1 is biallelically expressed in the kidney, however, in five of nine preterm placentae WT1 was expressed largely or exclusively from the maternal allele. Monoallelic expression of WT1 was also found in two fetal brains. These data demonstrate that WT1 can undergo tissue specific imprinting. Furthermore, because monoallelic expression of WT1 was not found in all placentae examined, WT1 imprinting may be genetically polymorphic within the human population.

Maternal imprinting of human SNRPN, a gene deleted in Prader-Willi syndrome

Prader-Willi syndrome (PWS), a human neuroendocrine disorder, is associated with deficiencies of paternal chromosome 15q12. Small nuclear ribonucleoprotein polypeptide N (SNRPN) is the first expressed gene identified in the PWS critically deleted region. Following our demonstration that the murine homologue of SNRPN is imprinted, we have characterized a sequence polymorphism within expressed portions of human SNRPN and show that human SNRPN is monoallelically expressed in fetal brain and heart and in adult brain. Analysis of maternal DNA and SNRPN cDNA confirmed that the maternal allele of SNRPN is not expressed in fetal brain and heart. Maternal imprinting of SNRPN supports the hypothesis that paternal absence of SNRPN is responsible for the PWS phenotype.

Overlapping patterns of IGF2 and H19 expression during human development: biallelic IGF2 expression correlates with a lack of H19 expression

The spatial patterns of IGF2 and H19 gene expression are strikingly similar during parts of human embryonic/fetal and early postnatal development. Notable exceptions were found with the ciliary anlage of the embryonic retina and the choroid plexus/leptomeninges, where transcripts from the IGF2 but not the H19 locus could be detected. Moreover, in contrast to the other tissue samples examined, the choroid plexus/leptomeninges expressed both parental IGF2 alleles. Whilst RNase protection analysis revealed a weak activity of the P1 promoter in the choroid plexus/leptomeninges, the P2, P3 and P4 promoters were all active wherever IGF2 was expressed. We discuss these observations with respect to a hypothesized coordinated control of the reciprocally imprinted and closely linked IGF2 and H19 loci.

Mechanisms of genomic imprinting in mammals

This chapter can be summarized by the following main points: Genomic imprinting results in the functional nonequivalence of the maternal and paternal genomes, thereby preventing the development of viable parthenogenotes and androgenotes in eutherian mammals. Imprinting may have arisen as a result of the specialized evolutionary requirements of the parental genomes or may have been an obligatory step in the development of placentation. A substantial proportion of transgenes and a smaller number of endogenous genes demonstrate imprinted pattern of expression in mice and humans. An analysis of DNA methylation in somatic tissues and germ cells during embryonic and postnatal development reveals dynamic changes, particularly during gametogenesis and early embryogenesis. The nature and timing of these changes suggest that DNA methylation may be involved in genomic imprinting. Imprinted genes display complex methylation patterns. Many aspects of these patterns are consistent with a role for methylation in the imprinted phenotype, although it is currently unclear whether methylation functions in the establishment of imprinting or plays a secondary role in the maintenance of the imprinted pattern of expression. Studies underway to identify new imprinted genes may help elucidate both the function and mechanism of genomic imprinting.

Insulin-like growth factor mediated stromal-epithelial interactions in human breast cancer

The prominent 'desmoplastic' or stromal reaction seen in many invasive breast carcinomas lead to early speculation that stromal cells play a role in breast cancer pathogenesis. Experimental evidence now supports this hypothesis and interactions between stromal cells and epithelial cells appear to be important for both normal mammary development and neoplasia. The identification of genes that are selectively expressed in the stroma of malignant breast lesions has recently provided new insights into the molecular basis of stromal-epithelial interactions. Stromally expressed genes include growth factors, proteases and extracellular matrix proteins, all biological activities with potential roles in malignant progression. Investigations discussed here concern the nature of the paracrine signals provided by malignant epithelial cells that activate changes in stromal gene expression, the effect that the stromally derived factors have on the behavior of malignant epithelial cells and the identification of novel factors and receptors in either stroma or epithelia that contribute to their mutual interactions. These questions will be addressed in the context of this laboratory's studies on insulin-like growth factors, as these molecules show marked differences in stromal expression between benign and malignant breast tissue and thus provide a useful paradigm for investigations into the paracrine environment of an evolving breast tumor.

Gene mapping of the mammalian genome: the CEPH and Genethon initiative

Over the past four years, the CEPH (Jean Dausset Foundation) has expanded its linkage mapping effort to include physical mapping and, in 1990, co-founded the Genethon to ensure that a combined physical and genetic map of the entire human genome would be achieved. The Genethon has applied methods developed at CEPH on an industrial scale to accomplish the colossal task of constructing an integrated map. It is the role of such an integrated map to accelerate the search for the genes responsible for inherited diseases, and the results of the past 12 months encourage our optimism that this goal will be realized rapidly. These discoveries are providing not only an approach to the diagnosis of genetically based disease but also some of the first breakthroughs in the area of gene therapy.

Constitutional relaxation of insulin-like growth factor II gene imprinting associated with Wilms' tumour and gigantism

We have examined the imprinting of the insulin-like growth factor II gene (IGF2) in ten normal kidney samples from children with renal embryonal neoplasms. In kidney samples from nine children with normal growth profiles, IGF2 mRNA was transcribed monoallelically, consistent with normal imprinting of the gene. But in one child who had generalized somatic overgrowth, IGF2 was transcribed from both alleles in her kidney, peripheral blood leukocytes and Wilms' tumour. These findings suggest that a defect in genomic imprinting can occur constitutionally, leading to growth abnormalities and predisposition to Wilms' tumour.

Disruption of insulin-like growth factor 2 imprinting in Beckwith-Wiedemann syndrome

To study insulin-like growth factor 2 (IGF2) imprinting in BWS (Beckwith-Wiedemann syndrome, an overgrowth syndrome associated with Wilms and other embryonal tumours), we examined allele-specific expression using an Apal polymorphism in the 3' untranslated region of IGF2. Four of six BWS fibroblast strains demonstrated biallelic expression, as did the tongue tissue from one of these patients. Paternal heterodisomy was excluded for all BWS patients with biallelic expression, suggesting strongly that the BWS phenotype in some patients involves disruption of IGF2 imprinting. Constitutional loss of IGF2 imprinting in a subgroup of our BWS patients, and recent reports of loss of imprinting in sporadic Wilms tumour, further strengthens the view that IGF2 overexpression plays an important role in somatic overgrowth and the development of embryonal tumours.

The insulin-like growth factor type-2 receptor gene is imprinted in the mouse but not in humans

In mouse, four genes have been found to undergo genomic imprinting resulting in differential expression of maternally and paternally inherited alleles. To determine whether the cognate genes are also subject to imprinting in humans, we have studied allele-specific expression patterns of insulin-like growth factor 2, IGF2-receptor and H19 in human fetal and adult tissues. In keeping with previous findings in mice, our results indicate that in human fetal tissues the paternal H19 alleles is inactive. IGF2 is monoallelically expressed in various tissues but surprisingly not in adult human liver. The human IGF2R gene, another classic example of imprinting in mice, was found to be expressed from both alleles. We provide the first direct evidence for differential imprinting in the human and murine genome.