Random X Inactivation and Extensive Mosaicism in Human Placenta Revealed by Analysis of Allele-Specific Gene Expression along the X Chromosome

Imprinted inactivation of the paternal X chromosome in marsupials is the primordial mechanism of dosage compensation for X-linked genes between females and males in Therians. In Eutherian mammals, X chromosome inactivation (XCI) evolved into a random process in cells from the embryo proper, where either the maternal or paternal X can be inactivated. However, species like mouse and bovine maintained imprinted XCI exclusively in extraembryonic tissues. The existence of imprinted XCI in humans remains controversial, with studies based on the analyses of only one or two X-linked genes in different extraembryonic tissues. Here we readdress this issue in human term placenta by performing a robust analysis of allele-specific expression of 22 X-linked genes, including XIST, using 27 SNPs in transcribed regions. We show that XCI is random in human placenta, and that this organ is arranged in relatively large patches of cells with either maternal or paternal inactive X. In addition, this analysis indicated heterogeneous maintenance of gene silencing along the inactive X, which combined with the extensive mosaicism found in placenta, can explain the lack of agreement among previous studies. Our results illustrate the differences of XCI mechanism between humans and mice, and highlight the importance of addressing the issue of imprinted XCI in other species in order to understand the evolution of dosage compensation in placental mammals.     

CpG methylation of an X-linked transgene is determined by somatic events postfertilization and not germline imprinting

The process of X-inactivation in mammals requires at least two events, the initiation of inactivation and the maintenance of the inactive state. One possible mechanism of control is by methylation of DNA at CpG dinucleotides to maintain the inactive state. Furthermore, the paternal X-chromosome is frequently inactivated in the extraembryonic membranes. The relationship between the parental origin of the chromosome, nonrandom inactivation and DNA methylation is not clear. In this paper, we report on the CpG methylation of an X-linked transgene, CAT-32. The levels of methylation in embryonic, extraembryonic and germline cells indicates that the modifications of the transgene are broadly similar to those reported for endogenous X-linked genes. Interestingly, the methylation of CAT-32 transgene in extraembryonic tissues displays patterns that could be linked to the germline origin of each allele. Hence, the maternally derived copy of CAT-32 was relatively undermethylated when compared to the paternal one. The changes in DNA methylation were attributed to de novo methylation occurring after fertilization, most probably during differentiation of extraembryonic tissues. In order to determine whether or not the patterns of DNA methylation reflected the germline origin of the X-chromosome, we constructed triploid embryos specifically to introduce two maternal X-chromosomes in the same embryo. In some of these triploid conceptuses, methylation patterns characteristic of the paternally derived transgene were observed. This observation indicates that the methylation patterns are not necessarily dependent on the parental origin of the X-chromosome, but could be changed by somatic events after fertilization. One of the more likely mechanisms is methylation of the transgene following inactivation of the X-chromosome in extraembryonic tissues.     

The role of the X chromosome in mammalian extra embryonic development

Accumulating evidence points to the importance of the X chromosome for trophoblast development. In rodents, the extraembryonic cell lineage differs from somatic tissues in that X chromosome inactivation is imprinted, preferentially silencing the paternal X chromosome. As a consequence, trophoblast development is extremely susceptible to deviations from normal X inactivation and is impaired in situations of increased and reduced X-linked gene dosage. Mouse mutants have also shown that maintenance of X chromosome silencing in extraembryonic tissues requires a special set of heterochromatin proteins. Moreover, the X chromosome has been implicated in causing several malformations of the placenta. The observed importance of the X chromosome for placental development can be explained by the presence of many trophoblast-expressed genes, especially in the proximal and central regions. Given that the placenta represents a postzygotic barrier to reproduction, evolutionary constraints may be responsible for the presence of placental genes on the X chromosome that are often co-expressed in brain and testis.     

Genomic imprinting in the placenta

Genomic imprinting is an epigenetic mechanism that is important for the development and function of the extra-embryonic tissues in the mouse. Remarkably all the autosomal genes which were found to be imprinted in the trophoblast (placenta) only are active on the maternal and repressed on the paternal allele. It was shown for several of these genes that their paternal silencing is not dependent on DNA methylation, at least not in its somatic maintenance. Rather, recent studies in the mouse suggest that placenta-specific imprinting involves repressive histone modifications and non-coding RNAs. This mechanism of autosomal imprinting is similar to imprinted X chromosome inactivation in the placenta. Although the underlying reasons remain to be explored, this suggests that imprinting in the placenta and imprinted X inactivation are evolutionarily related.     

Sex chromosome elimination,X chromosome inactivation and reactivation in the southern brown bandicoot Isoodon obesulus (Marsupialia: Peramelidae)

Cytogenetic studies have shown that bandicoots (family Peramelidae) eliminate one X chromosome in females and the Y chromosome in males from some somatic tissues at different stages during development. The discovery of a polymorphism for X-linked phosphoglycerate kinase (PGK-1) in a population of Isoodon obesulus from Mount Gambier, South Australia, has allowed us to answer a number of long standing questions relating to the parental source of the eliminated X chromosome, X chromosome inactivation and reactivation in somatic and germ cells of female bandicoots. We have found no evidence of paternal PGK-1 allele expression in a wide range of somatic tissues and cell types from known female heterozygotes. We conclude that paternal X chromosome inactivation occurs in bandicoots as in other marsupial groups and that it is the paternally derived X chromosome that is eliminated from some cell types of females. The absence of PGK-1 paternal activity in somatic cells allowed us to examine the state of X chromosome activity in germ cells. Electrophoresis of germ cells from different aged pouch young heterozygotes showed only maternal allele expression in oogonia whereas an additional paternally derived band was observed in pre-dictyate oocytes. We conclude that reactivation of the inactive X chromosome occurs around the onset of meiosis in female bandicoots. As in other mammals, late replication is a common feature of the Y chromosome in male and the inactive X chromosome in female bandicoots. The basis of sex chromosome loss is still not known; however later timing of DNA synthesis is involved. Our finding that the paternally derived X chromosome is eliminated in females suggests that late DNA replication may provide the imprint for paternal X inactivation and the elimination of sex chromosomes in bandicoots.     

Parental source of chromosome imprinting and its relevance for X chromosome inactivation

In imprinting, homologous chromosomes behave differently during development according to their parental origin. Typically, paternally derived chromosomes are preferentially inactivated or eliminated. Examples of such phenomena include inactivation of the mammalian X chromosome, inactivation or elimination of one haploid chromosome set in male coccids, and elimination of paternal X chromosomes in the fly Sciara. It has generally been thought that the paternal chromosomes bear an imprint leading to their inactivation or elimination. However, alteration of the parental origin of chromosomes, as in the study of parthenogenotes in mammals and coccids, shows that passage of chromosomes through a male germ cell or fertilization is not essential for inactivation or elimination. It appears that neither chromosome set is programmed to resist or undergo inactivation. Instead the two sets differ in relative sensitivity, and the question is whether the maternal set have an imprint for resistance, or the paternal set one for susceptibility. Very early in development of mammals both X chromosomes are active. This makes it simpler to envisage the maternal X bearing an imprint for resistance to inactivation, which persists through the early developmental period. Similar considerations also apply in coccids and Sciara. Thus, imprinting should be regarded as a phenomenon conferred on the maternal chromosomes in the oocyte. This permits simpler models for the mechanism of X-inactivation, and weakens the case for evolution of X-inactivation from an earlier form of inactivation during male gametogenesis. One may speculate whether imprinting affects timing of gene action in development.     

Transformation of Hprt gene with sperm DNA

Inactivation of the X chromosome during mammalian spermatogenesis has been postulated to occur by the same mechanism that controls female somatic X chromosome inactivation. We have used DNA-mediated transformation of HPRT- cells to test this idea, because it has been shown previously that inactive X chromosome DNA from somatic cells will not transform HPRT- cells. Isolated DNA from the mature sperm of five mammals (human, mouse, horse, bull, rabbit) were all capable of HPRT transformation, and transformants were confirmed electrophoretically. Measures were taken to ensure that the transformation frequencies observed could not be due to somatic contamination. The positive HPRT transformation result indicates that mature sperm X chromosomal DNA is not modified in the same manner as that of female inactive X chromosomal DNA. Since there is evidence for methylation of the somatic inactive X chromosome, it is possible that methylation, at least for the genes studied, is not involved in sperm X chromosome inactivation.     

Differential X reactivation in human placental cells: implications for reversal of X inactivation

X inactivation--the mammalian method of X chromosome dosage compensation--is extremely stable in human somatic cells; only fetal germ cells have a developmental program to reverse the process. The human placenta, at term, differs from other somatic tissues, since it has the ability to reverse the X-inactivation program. To determine whether reversal can be induced at other stages of placental development, we examined earlier placental specimens using a cell-hybridization assay. We found that global X reactivation is also inducible in villi cells from first-trimester spontaneous abortions but not from first-trimester elective terminations. These differences in inducibility are not associated with detectable variation in histone H4 acetylation, DNA methylation, or XIST expression--hallmarks of the inactivation process--so other factors must have a role. One notable feature is that the permissive cells, unlike nonpermissive ones, have ceased to proliferate in vivo and are either beginning or in the process of programmed cell death. Cessation of mitotic proliferation also characterizes oocytes at the stage at which they undergo X reactivation. We suggest that, along with undermethylation, the apoptotic changes accompanying cessation of cell proliferation contribute to the reversal of inactivation, not only in placental cells, but also in oocytes entering meiosis.     

Parental source of chromosome imprinting and its relevance for X chromosome inactivation

Abstract. In imprinting, homologous chromosomes behave differently during development according to their parental origin. Typically, paternally derived chromosomes are preferentially inactivated or eliminated. Examples of such phenomena include inactivation of the mammalian X chromosome, inactivation or elimination of one haploid chromosome set in male coccids, and elimination of paternal X chromosomes in the fly Sciara . It has generally been thought that the paternal chromosomes bear an imprint leading to their inactivation or elimination. However, alteration of the parental origin of chromosomes, as in the study of parthenogenotes in mammals and coccids, shows that passage of chromosomes through a male germ cell or fertilization is not essential for inactivation or elimination. It appears that neither chromosome set is programmed to resist or undergo inactivation. Instead the two sets differ in relative sensitivity, and the question is whether the maternal set have an imprint for resistance, or the paternal set one for susceptibility. Very early in development of mammals both X chromosomes are active. This makes it simpler to envisage the maternal X bearing an imprint for resistance to inactivation, which persists through the early developmental period. Similar considerations also apply in coccids and Sciara . Thus, imprinting should be regarded as a phenomenon conferred on the maternal chromosomes in the oocyte. This permits simpler models for the mechanism of X-inactivation, and weakens the case for evolution of X-inactivation from an earlier form of inactivation during male gametogenesis. One may speculate whether imprinting affects timing of gene action in development.     

Paternal RLIM/Rnf12 is a survival factor for milk-producing alveolar cells

In female mouse embryos, somatic cells undergo a random form of X chromosome inactivation (XCI), whereas extraembryonic trophoblast cells in the placenta undergo imprinted XCI, silencing exclusively the paternal X chromosome. Initiation of imprinted XCI requires a functional maternal allele of the X-linked gene Rnf12, which encodes the ubiquitin ligase Rnf12/RLIM. We find that knockout (KO) of Rnf12 in female mammary glands inhibits alveolar differentiation and milk production upon pregnancy, with alveolar cells that lack RLIM undergoing apoptosis as they begin to differentiate. Genetic analyses demonstrate that these functions are mediated primarily by the paternal Rnf12 allele due to nonrandom maternal XCI in mammary epithelial cells. These results identify paternal Rnf12/RLIM as a critical survival factor for milk-producing alveolar cells and, together with population models, reveal implications of transgenerational epigenetic inheritance.     

Variability of X chromosome inactivation: effect on levels of TIMP1 RNA and role of DNA methylation

X chromosome inactivation results in dosage equivalency for X-linked gene expression between males and females. However, some X-linked genes show variable X inactivation, being expressed from the inactive X in some females but subject to inactivation in other women. The human tissue inhibitor of metalloproteinases-1 ( TIMP1) gene falls into this category. As TIMP1 and its target metalloproteinases are involved in many biological processes, women with elevated TIMP1 expression may exhibit different disease susceptibilities. To address the potential impact of variable X inactivation, we analyzed TIMP1 expression levels by using an RNase protection assay. The substantial variation of TIMP1 expression observed in cells with monoallelic TIMP1 expression precluded analysis of the contribution of the inactive X to total TIMP1 RNA levels in females, so we examined expression in rodent/human somatic cell hybrids. TIMP1 expression levels varied more widely in hybrids retaining an inactive X than in those with an active X chromosome, suggesting variable retention of the epigenetic silencing mechanisms associated with X inactivation. Therefore, we investigated the contribution of methylation at the promoter to expression level variation and found that methylation of the TIMP1 promoter correlated with instability and low level expression, whereas stable TIMP1expression from the inactive X equivalent to that seen from the active X chromosome was observed when the promoter was unmethylated. Since all female cell lines examined showed methylation of the TIMP1 promoter, the contribution of expression from the inactive X appears minimal. However, as women age, they may accumulate cells stably expressing TIMP1 from the inactive X, with a resulting increase of TIMP1, which may explain some sex differences in various late-onset disorders.     

DNA methylation analysis of ade novo balanced X;13 translocation in a girl with abnormal phenotype: evidence for functional duplication of the whole short arm of the X chromosome

We report on a 13-month-old girl showing dysmorphic features and a delay in psychomotor development. She was diagnosed with a balancedde novo translocation 46, X, t(X;13)(p11. 2;p13) and non-random inactivation of the X chromosome. FISH analysis, employing the X chromosome centromere andXIST-region-specific probes, showed that theXIST locus was not involved in the translocation. Selective inactivation of paternal X, which was involved in translocation, was revealed by the HUMARA assay. The pattern of methylation of 5 genes located within Xp, which are normally silenced on an inactive X chromosome, corresponded to an active (unmethylated) X chromosome. These results revealed that in our proband the X chromosome involved in translocation (Xt) was preferentially inactivated. However, genes located on the translocated Xp did not includeXIST. This resulted in functional Xp disomy, which most probably accounts for the abnormal phenotype in our patient.     

Preferential X inactivation in human placenta membranes: Is the paternal X inactive in early embryonic development of female mammals?

Summary In placenta membranes of newborn girls carrying electrophoretically distinguishable G6PD alleles, the maternally derived isozyme is expressed preferentially. This phenomenon cannot be explained by allelic differences in enzyme activity or by somatic selection directed against cells with particular G6PD phenotypes. Instead, it may be that in this tissue X inactivation is nonrandom. Preferential expression of the maternal X chromosome, as has been shown in marsupials and in extraembryonic membranes of rodents and now in man, may reflect the state of activity of the X chromosomes in the early stages of female embryonic development.H.-H. R. is supported by the Deutsche Forschungsgemeinschaft     

Nrk, an X-linked protein kinase in the germinal center kinase family, is required for placental development and fetoplacental induction of labor

The complete mechanism of labor induction in eutherian mammals remains unclear. Although important roles for the fetus and placenta in triggering labor have been proposed, no gene has been shown to be required in the fetus/placenta for labor induction. Here we show that Nrk, an X-linked gene encoding a Ser/Thr kinase of the germinal center kinase family, is essential in the fetus/placenta for labor in mice. Nrk was specifically expressed in the spongiotrophoblast layer, a fetus-derived region of the placenta, and Nrk disruption caused dysregulated overgrowth of the layer. Due to preferential inactivation of the paternally derived X chromosome in placenta, Nrk heterozygous mutant placentas exhibited a similar defect to that in Nrk-null tissues when the wild-type allele was paternally derived. However, the phenotype was weaker than in Nrk-null placentas due to leaky Nrk expression from the inactivated X chromosome. Crossing of Nrk-null females to wild-type and Nrk-null males, as well as uterine transfer of Nrk-null fetuses to wild-type females, revealed that pregnant mice exhibit a severe defect in delivery when all fetuses/placentas are Nrk-null. In addition, Nrk was not expressed in female reproductive tissues such as the uterus and ovary, as well as the fetal amnion and yolk sac, in pregnant mice. Progesterone and estrogen levels in the maternal circulation and placenta, which control the timing of labor, were unaffected upon Nrk disruption. We thus provide evidence for a novel labor-inducing fetoplacental signal that depends on the X chromosome and possibly arises from the placenta.     

Genetic and developmental analysis of X-inactivation in interspecific hybrid mice suggests a role for the Y chromosome in placental dysplasia

It has been shown previously that abnormal placental growth, i.e., hyper- and hypoplasia, occurs in crosses and backcrosses between different mouse (Mus) species. A locus that contributes to this abnormal development has been mapped to the X chromosome. Unexpectedly, an influence of fetal sex on placental development has been observed, in that placentas attached to male fetuses tended to exhibit a more pronounced phenotype than placentas attached to females. Here, we have analyzed this sex dependence in more detail. Our results show that differences between male and female placental weights are characteristic of interspecific matings and are not observed in intraspecific Mus musculus matings. The effect is retained in congenic lines that contain differing lengths of M. spretus-derived X chromosome. Expression of the X-linked gene Pgk1 from the maternal allele only and lack of overall activity of two paternally inherited X-linked transgenes indicate that reactivation or lack of inactivation of the paternal X chromosome in trophoblasts of interspecific hybrids is not a frequent occurrence. Thus, the difference between male and female placentas seems not to be caused by faulty preferential X-inactivation. Therefore, these data suggest that the sex difference of placental weights in interspecific hybrids is caused by interactions with the Y chromosome.