Birth of parthenogenetic mice that can develop to adulthood

Only mammals have relinquished parthenogenesis, a means of producing descendants solely from maternal germ cells. Mouse parthenogenetic embryos die by day 10 of gestation. Bi-parental reproduction is necessary because of parent-specific epigenetic modification of the genome during gametogenesis. This leads to unequal expression of imprinted genes from the maternal and paternal alleles. However, there is no direct evidence that genomic imprinting is the only barrier to parthenogenetic development. Here we show the development of a viable parthenogenetic mouse individual from a reconstructed oocyte containing two haploid sets of maternal genome, derived from non-growing and fully grown oocytes. This development was made possible by the appropriate expression of the Igf2 and H19 genes with other imprinted genes, using mutant mice with a 13-kilobase deletion in the H19 gene as non-growing oocytes donors. This full-term development is associated with a marked reduction in aberrantly expressed genes. The parthenote developed to adulthood with the ability to reproduce offspring. These results suggest that paternal imprinting prevents parthenogenesis, ensuring that the paternal contribution is obligatory for the descendant.     

Embryological and molecular investigations of parental imprinting on mouse chromosome 7

Mouse embryos with duplications of whole maternal (parthenogenetic and gynogenetic) or paternal (androgenetic) genomes show reciprocal phenotypes and do not develop to term. Genetic complementation has identified the distal region of chromosome 7 (Chr 7) as one of the regions for which both a maternal and paternal chromosome copy are essential for normal development, presumably because of the presence of imprinted genes whose expression is dependent on their parental origin. Embryos with the maternal duplication and paternal deficiency of distal Chr 7 are growth retarded and die around day 16 of gestation; the reciprocal paternal duplication embryos die at an unidentified earlier stage. We report here the incorporation of cells with the paternal duplication into chimaeras, resulting in a striking growth enhancement of the embryos. One gene located on mouse distal Chr 7 (ref. 5) is the insulin-like growth factor 2 (Igf2) gene, an embryonic mitogen. In embryos with the maternal duplication of distal Chr 7, the two maternal alleles of the Igf2 gene are repressed. The presence of two paternal alleles of this gene in many cells is probably responsible for the growth enhancement observed in chimaeras. We propose that there are other imprinted genes in this Chr 7 region. We also compare the imprinting of this subgenomic region with phenotypes resulting from the duplication of the whole parental genome in parthenogenones and androgenones.     

Seed development: Early paternal gene activity in Arabidopsis.

Both parental genomes are expressed during embryogenesis, although the time of activation of the paternally inherited genes varies between organisms. Results reported by Vielle-Calzada et al. have suggested that delayed activation of the paternal genome seems to be the rule in plant development. We find, however, that during early embryogenesis in Arabidopsis, paternal genes are expressed and are sufficient for normal development. Our findings indicate that there is no overall maternal control of early embryogenesis, and that the contribution of the parental alleles needs to be assessed for each gene individually.     

Methylation of a CTCF-dependent boundary controls imprinted expression of the Igf2 gene

The expression of the insulin-like growth factor 2 (Igf2) and H19 genes is imprinted. Although these neighbouring genes share an enhancer, H19 is expressed only from the maternal allele, and Igf2 only from the paternally inherited allele. A region of paternal-specific methylation upstream of H19 appears to be the site of an epigenetic mark that is required for the imprinting of these genes. A deletion within this region results in loss of imprinting of both H19 and Igf2 (ref. 5). Here we show that this methylated region contains an element that blocks enhancer activity. The activity of this element is dependent upon the vertebrate enhancer-blocking protein CTCF. Methylation of CpGs within the CTCF-binding sites eliminates binding of CTCF in vitro, and deletion of these sites results in loss of enhancer-blocking activity in vivo, thereby allowing gene expression. This CTCF-dependent enhancer-blocking element acts as an insulator. We suggest that it controls imprinting of Igf2. The activity of this insulator is restricted to the maternal allele by specific DNA methylation of the paternal allele. Our results reveal that DNA methylation can control gene expression by modulating enhancer access to the gene promoter through regulation of an enhancer boundary.     

Parental imprinting of the mouse H19 gene.

THE mouse H19 gene encodes one of the most abundant RNAs in the developing mouse embryo. It is expressed at the blastocyst stage of development, and accumulates to high levels in tissues of endodermal and mesodermal origin (H. Kim, unpublished result). After birth the gene is expressed in all tissues except skeletal muscle. It lacks a common open reading frame in the 2.5-kilobase RNA, but has considerable nucleotide sequence similarity between the genes of rodents and humans. Expression of the gene in transgenic mice results in late prenatal lethality, suggesting that the dosage of its gene product is strictly controlled. The H19 gene maps to the distal segment of mouse chromosome 7, in a region that is parentally imprinted, a process by which genes are differentially expressed on the maternal and paternal chromosomes. We have now used an RNase protection assay that can distinguish between H19 alleles in four subspecies of Mus, to demonstrate that the H19 gene is parentally imprinted, with the active copy derived from the mother. This assay will be of general use in assaying allele-specific gene expression.     

Essential role for de novo DNA methyltransferase Dnmt3a in paternal and maternal imprinting

Imprinted genes are epigenetically marked during gametogenesis so that they are exclusively expressed from either the paternal or the maternal allele in offspring. Imprinting prevents parthenogenesis in mammals and is often disrupted in congenital malformation syndromes, tumours and cloned animals. Although de novo DNA methyltransferases of the Dnmt3 family are implicated in maternal imprinting, the lethality of Dnmt3a and Dnmt3b knockout mice has precluded further studies. We here report the disruption of Dnmt3a and Dnmt3b in germ cells, with their preservation in somatic cells, by conditional knockout technology. Offspring from Dnmt3a conditional mutant females die in utero and lack methylation and allele-specific expression at all maternally imprinted loci examined. Dnmt3a conditional mutant males show impaired spermatogenesis and lack methylation at two of three paternally imprinted loci examined in spermatogonia. By contrast, Dnmt3b conditional mutants and their offspring show no apparent phenotype. The phenotype of Dnmt3a conditional mutants is indistinguishable from that of Dnmt3L knockout mice, except for the discrepancy in methylation at one locus. These results indicate that both Dnmt3a and Dnmt3L are required for methylation of most imprinted loci in germ cells, but also suggest the involvement of other factors.     

Delayed activation of the paternal genome during seed development

Little is known about the timing of the maternal-to-zygotic transition during seed development in flowering plants. Because plant embryos can develop from somatic cells or microspores, maternal contributions are not considered to be crucial in early embryogensis. Early-acting embryo-lethal mutants in Arabidopsis, including emb30/gnom which affects the first zygotic division, have fuelled the perception that both maternal and paternal genomes are active immediately after fertilization. Here we show that none of the paternally inherited alleles of 20 loci that we tested is expressed during early seed development in Arabidopsis. For genes that are expressed at later stages, the paternally inherited allele becomes active three to four days after fertilization. The genes that we tested are involved in various processes and distributed throughout the genome, indicating that most, if not all, of the paternal genome may be initially silenced. Our findings are corroborated by genetic studies showing that emb30/gnom has a maternal-effect phenotype that is paternally rescuable in addition to its zygotic lethality. Thus, contrary to previous interpretations, early embryo and endosperm development are mainly under maternal control.     

Genomic imprinting determines methylation of parental alleles in transgenic mice

Mouse embryogenesis relies on the presence of both the maternal and the paternal genome for development to term. It has been proposed that specific modifications are imprinted onto the chromosomes during gametogenesis; these modifications are stably propagated, and their expression results in distinct and complementary contributions of the two parental genomes to the development of the embryo and the extraembryonic membranes. Genetic data further suggest that a substantial proportion of the genome could be subject to chromosomal imprinting, the molecular nature of which is unknown. We used random DNA insertions in transgenic mice to probe the genome for modified regions. The DNA methylation patterns of transgenic alleles were compared after transmission from mother or father in seven mouse strains carrying autosomal insertions of the same transgenic marker. One of these loci showed a clear difference in DNA methylation specific for its parental origin, with the paternally inherited copy being relatively undermethylated. This difference was observed in embryos on day 10 of gestation, but not in their extraembryonic membranes. Moreover, the methylation pattern was faithfully reversed upon each germline transmission to the opposite sex. Our findings provide evidence for heritable molecular differences between maternally and paternally derived alleles on mouse chromosomes.     

Uniparental paternal disomy in a genetic cancer-predisposing syndrome.

The 11p15.5 region of human chromosome 11 seems to contain a locus or loci involved in congenital overgrowth anomalies as well as in the genesis of many tumours associated with the Beckwith-Wiedemann syndrome (BWS). Given the unusual differential parental allele involvement in the different aetiological forms of BWS and the loss of maternal alleles in associated tumours, we have now used 11p15.5 markers to determine the parental origin of chromosome 11 in eight sporadic cases of BWS. Probands in three informative families had uniparental paternal disomy for region 11p15.5. Further, an overall greatly increased frequency of homozygosity for several 11p15.5 markers in 21 sporadic BWS patients suggests that isodisomy probably accounts for an even higher proportion of BWS sporadic cases. This demonstrates that uniparental paternal disomy can be associated with a genetic cancer-predisposing syndrome.     

Complex social behaviour derived from maternal reproductive traits

A fundamental goal of sociobiology is to explain how complex social behaviour evolves, especially in social insects, the exemplars of social living. Although still the subject of much controversy, recent theoretical explanations have focused on the evolutionary origins of worker behaviour (assistance from daughters that remain in the nest and help their mother to reproduce) through expression of maternal care behaviour towards siblings. A key prediction of this evolutionary model is that traits involved in maternal care have been co-opted through heterochronous expression of maternal genes to result in sib-care, the hallmark of highly evolved social life in insects. A coupling of maternal behaviour to reproductive status evolved in solitary insects, and was a ready substrate for the evolution of worker-containing societies. Here we show that division of foraging labour among worker honey bees (Apis mellifera) is linked to the reproductive status of facultatively sterile females. We thereby identify the evolutionary origin of a widely expressed social-insect behavioural syndrome, and provide a direct demonstration of how variation in maternal reproductive traits gives rise to complex social behaviour in non-reproductive helpers.     

A global disorder of imprinting in the human female germ line

Imprinted genes are expressed differently depending on whether they are carried by a chromosome of maternal or paternal origin. Correct imprinting is established by germline-specific modifications; failure of this process underlies several inherited human syndromes. All these imprinting control defects are cis-acting, disrupting establishment or maintenance of allele-specific epigenetic modifications across one contiguous segment of the genome. In contrast, we report here an inherited global imprinting defect. This recessive maternal-effect mutation disrupts the specification of imprints at multiple, non-contiguous loci, with the result that genes normally carrying a maternal methylation imprint assume a paternal epigenetic pattern on the maternal allele. The resulting conception is phenotypically indistinguishable from an androgenetic complete hydatidiform mole, in which abnormal extra-embryonic tissue proliferates while development of the embryo is absent or nearly so. This disorder offers a genetic route to the identification of trans-acting oocyte factors that mediate maternal imprint establishment.     

CTCF mediates methylation-sensitive enhancer-blocking activity at the H19/Igf2 locus

The Insulin-like growth factor 2 (Igf2) and H19 genes are imprinted, resulting in silencing of the maternal and paternal alleles, respectively. This event is dependent upon an imprinted-control region two kilobases upstream of H19 (refs 1, 2). On the paternal chromosome this element is methylated and required for the silencing of H19 (refs 2-4). On the maternal chromosome the region is unmethylated and required for silencing of the Igf2 gene 90 kilobases upstream. We have proposed that the unmethylated imprinted-control region acts as a chromatin boundary that blocks the interaction of Igf2 with enhancers that lie 3' of H19 (refs 5, 6). This enhancer-blocking activity would then be lost when the region was methylated, thereby allowing expression of Igf2 paternally. Here we show, using transgenic mice and tissue culture, that the unmethylated imprinted-control regions from mouse and human H19 exhibit enhancer-blocking activity. Furthermore, we show that CTCF, a zinc finger protein implicated in vertebrate boundary function, binds to several sites in the unmethylated imprinted-control region that are essential for enhancer blocking. Consistent with our model, CTCF binding is abolished by DNA methylation. This is the first example, to our knowledge, of a regulated chromatin boundary in vertebrates.     

Maternal Rnf12/RLIM is required for imprinted X-chromosome inactivation in mice

Two forms of X-chromosome inactivation (XCI) ensure the selective silencing of female sex chromosomes during mouse embryogenesis. Imprinted XCI begins with the detection of Xist RNA expression on the paternal X?chromosome (Xp) at about the four-cell stage of embryonic development. In the embryonic tissues of the inner cell mass, a random form of XCI occurs in blastocysts that inactivates either Xp or the maternal X?chromosome (Xm). Both forms of XCI require the non-coding Xist RNA that coats the inactive X?chromosome from which it is expressed. Xist has crucial functions in the silencing of X-linked genes, including Rnf12 (refs 3, 4) encoding the ubiquitin ligase RLIM (RING finger LIM-domain-interacting protein). Here we show, by targeting a conditional knockout of Rnf12 to oocytes where RLIM accumulates to high levels, that the maternal transmission of the mutant X?chromosome (Δm) leads to lethality in female embryos as a result of defective imprinted XCI. We provide evidence that in Δm female embryos the initial formation of Xist clouds and Xp silencing are inhibited. In contrast, embryonic stem cells lacking RLIM are able to form Xist clouds and silence at least some X-linked genes during random XCI. These results assign crucial functions to the maternal deposit of Rnf12/RLIM for the initiation of imprinted XCI.     

The mouse insulin-like growth factor type-2 receptor is imprinted and closely linked to the Tme locus.

T-associated maternal effect (Tme) is the only known maternal-effect mutation in the mouse. The defect is nuclear-encoded and embryos that inherit a deletion of the Tme locus from their mother die at day 15 of gestation. There are many genomically imprinted regions known in the mouse genome but so far no imprinted genes have been cloned. The Tme locus is absent in two chromosome-17 deletion mutants, Thp and the tLub2, and its position has been localized using these deletions to a 1-cM region. We report here that the genes for insulin-like growth factor type-2 receptor (Igf2r) and mitochondrial superoxide dismutase-2 (Sod-2) are absent from both deletions. Probes for these genes and for plasminogen (Plg) and T-complex peptide 1 (Tcp-1) were used in pulsed-field gel mapping to show that Tme must lie within a region of 800-1,100 kb. We also demonstrate that embryos express Igf2r only from the maternal chromosome, and that Tcp-1, Plg and Sod-2 are expressed from both chromosomes. Therefore Igf2r is imprinted and closely linked or identical to Tme.     

The genetic basis of family conflict resolution in mice

Asymmetries in the costs and benefits of parental investment for mothers, fathers and offspring result in family conflict over the production and provisioning of young. In species where females provide most resources before and after birth, the resolution of this conflict may be influenced by genes expressed in mothers and by maternally and paternally inherited genes expressed in offspring. Here we disentangle these effects by means of reciprocal mating and cross-fostering of litters between two strains of mice that differ with respect to the typical resolution of family conflict. We find that differences in litter size between these two strains are determined by paternal genotype, whereas differences in provisioning are under maternal control, showing that there is antagonistic coadaptation of maternal and paternal effects on distinct life-history traits. Maternal provisioning is also influenced by the type of foster offspring. Contradictory to theoretical expectations, however, we find no evidence for a negative correlation across strains between maternal provisioning and offspring demand. Instead, we show that there is positive coadaptation such that offspring obtain more resources from foster mothers of the same strain as their natural mother, irrespective of their father's strain.