Myriam Hemberger

Trophoblast functions, angiogenesis and remodeling of the maternal vasculature in the placenta.

One of the most important local adaptations to pregnancy is the change in maternal blood flow to the implantation site. In rodents and primates, new blood vessels form through angiogenesis, dilate and then become modified such that the blood enters into trophoblast cell-lined sinuses (hemochorial). Evidence from gene knockout mice suggests that factors from the placenta regulate the uterine vasculature. Consistent with this, trophoblast giant cells produce a number of angiogenic and vasoactive substances that may mediate these effects. Teratocarcinomas containing large numbers of trophoblast giant cells (derived from Parp1 gene-deficient ES cells) show similar hemochorial host blood flow, implying that the effects are not specific to the uterine vascular bed. As in primates, murine trophoblast cells also invade into the uterine arteries of the mother. However, in normal pregnancy, dilation of the uterine arteries may be largely mediated by the effect of uterine natural killer cells.

Genes governing placental development

The placenta is essential for fetal growth because it promotes the delivery of nutrients and oxygen from the maternal circulation. In mice, many gene mutations disrupt formation of the placenta, with specific effects at different times and on different components. Studies of these mutations are beginning to provide insights into both the molecular pathways required for formation of different placental substructures and the nature of intercellular interactions, between trophoblast, mesenchymal and vascular components, that regulate placental development. Conserved gene expression patterns in humans should enable the elucidation of the molecular basis of human placental dysfunction.

Differential expression of angiogenic and vasodilatory factors by invasive trophoblast giant cells depending on depth of invasion

The uterine bed undergoes remarkable changes during pregnancy, including proliferation and decidualization of the uterine stroma and remodeling and angiogenesis of the maternal vasculature. Fetal‐derived trophoblast giant cells invade into the uterus where they gain access to the maternal blood circulation to ensure sufficient nutrient supply of the embryo. In serial sections through early‐ to mid‐gestation conceptuses, we have determined the exact distance of trophoblast invasion and the expression of angiogenic, vasodilatory, and anticoagulative factors that are likely to influence remodeling and redirection of the maternal circulatory system. Trophoblast derivatives were detected at a distance as far as ∼300 μm from the placental border, where they are allocated exclusively along the mid‐line of the decidua. The farthest invading cells characteristically expressed proliferin and proliferin‐related protein, hormones that affect endothelial cell migration and vascularization. Occasionally, these cells replaced the normal vascular endothelium and acquired a “pseudo‐endothelial” shape. Complete vascular disintegration was observed 50–80 μm outside of the placental border where maternal blood was entirely lined by a trophoblast giant cell‐derived network of blood sinuses. This transition in blood space lining correlated with trophoblast expression of various vasodilatory and anticoagulative factors that are likely to promote blood flow toward the placenta. Analysis of teratocarcinoma‐like tumors demonstrated that trophoblast giant cell‐induced promotion and redirection of blood flow is not restricted to the uterine environment. These results show that trophoblast giant cells have the intrinsic capacity to attract and increase blood flow and to gradually displace the vascular endothelium resulting in the formation of canals entirely lined by trophoblast cells. Developmental Dynamics 227:185–191, 2003.

Epigenetic dynamics of the Kcnq1 imprinted domain in the early embryo

The mouse Kcnq1 imprinted domain is located on distal chromosome 7 and contains several imprinted genes that are paternally repressed. Repression of these genes is regulated by a non-coding antisense transcript, Kcnq1ot1, which is paternally expressed. Maternal repression of Kcnq1ot1 is controlled by DNA methylation originating in the oocyte. Some genes in the region are imprinted only in the placenta, whereas others are imprinted in both extra-embryonic and embryonic lineages. Here, we show that Kcnq1ot1 is paternally expressed in preimplantation embryos from the two-cell stage, and that ubiquitously imprinted genes proximal to Kcnq1ot1 are already repressed in blastocysts, ES cells and TS cells. Repressive histone marks such as H3K27me3 are present on the paternal allele of these genes in both ES and TS cells. Placentally imprinted genes that are distal to Kcnq1ot1, by contrast, are not imprinted in blastocysts, ES or TS cells. In these genes, paternal silencing and differential histone marks arise during differentiation of the trophoblast lineage between E4.5 and E7.5. Our findings show that the dynamics during preimplantation development of gene inactivation and acquisition of repressive histone marks in ubiquitously imprinted genes of the Kcnq1 domain are very similar to those of imprinted X inactivation. By contrast, genes that are only imprinted in the placenta, while regulated by the same non-coding RNA transcript Kcnq1ot1, undergo epigenetic inactivation during differentiation of the trophoblast lineage. Our findings establish a model for how epigenetic gene silencing by non-coding RNA may depend on distance from the non-coding RNA and on lineage and differentiation specific factors.

Parp1-deficiency induces differentiation of ES cells into trophoblast derivatives

Embryonic stem (ES) cells deficient in the enzyme poly(ADP-ribose) polymerase (Parp1) develop into teratocarcinoma-like tumors when injected subcutaneously into nude mice that contain cells with giant cell-like morphology. We show here that these cells express genes characteristic of trophoblast giant cells and thus belong to the trophectoderm lineage. In addition, Parp1(-/-) tumors contained other trophoblast subtypes as revealed by expression of spongiotrophoblast-specific marker genes. The extent of giant cell differentiation was enhanced, however, as compared with spongiotrophoblast. A similar shift toward trophoblast giant cell differentiation was observed in cultures of Parp1-deficient ES cells and in placentae of Parp1(-/-) embryos. Analysis of other cell lineage markers demonstrated that Parp1 acts exclusively in trophoblast to suppress differentiation. Surprisingly, trophoblast derivatives were also detected in wildtype tumors and cultured ES cells, albeit at significantly lower frequency. These data show that wildtype ES cells contain a small population of cells with trophectoderm potential and that absence of Parp1 renders ES cells more susceptible to adopting a trophoblast phenotype.

The importance of cysteine cathepsin proteases for placental development

The typically lysosomal family of cysteine cathepsin proteases has been implicated in the development of the placenta in particular, from studies in the mouse. Here, we analysed overall expression, regulation and presence of transcript isoforms of cysteine cathepsins during human extra-embryonic development. All 11 family members are expressed in human placental tissues, and many are differentially regulated during gestation. Several cysteine cathepsins exhibit deregulated expression levels in placentas from pregnancies complicated by pre-eclampsia. The localization of cathepsin B predominantly in placental and decidual macrophages suggests a role in the physiological functions of these cells in mediating villous angiogenesis and decidual apoptosis. Cathepsin L levels are highest in a subpopulation of invasive cytotrophoblasts. Reflecting the expression pattern of two murine cathepsins, these data give insights into the evolutionary conservation of cathepsin function that is not necessarily exhibited by gene pairs defined by highest sequence similarity. Furthermore, cathepsin L protein localization in uterine epithelial cells demonstrates the in vivo occurrence of intranuclear cathepsin L isoforms. The zonally restricted expression of cathepsin in the syncytiotrophoblast may be important for the metabolic breakdown of maternal nutrients. Overall, the distribution and abnormal expression levels in pre-eclamptic placentas indicate that cysteine cathepsins may play important roles during normal placentation and in the etiology of pre-eclampsia.

H19 and Igf2 are expressed and differentially imprinted in neuroectoderm-derived cells in the mouse brain

Abstractu2002Igf2 and H19 are reciprocally imprinted genes that are closely linked and coexpressed in tissues of mesodermal and endodermal origin. Here we report that coexpression of these genes is also found in specific fetal tissues of neuroectodermal origin, that is in the ventral midline region of both the hindbrain and spinal cord. For cells of neuroectodermal origin, complete absence of Igf2 and H19 transcription was previously described. Analysis of allele-specific expression of both Igf2 and H19 in the ventral midline region of the hindbrain shows that H19 is expressed monoallelically, with the paternal allele being silent, whereas Igf2 is expressed biallelically. Furthermore, we observed a strong influence of the parental species background, in that the Mus musculus allele was always expressed at higher levels than the M. spretus allele. This was observed when the M. spretus allele was contributed by the mother or by the father. An analysis of Igf2 methylation by bisulphite genomic sequencing provided no clear answer as to whether Igf2 expression and methylation are linked in a tissue of neuroectodermal origin. Taken together, our results provide novel information on H19 and Igf2 expression and imprinting patterns in the fetal mouse brain. In addition, they indicate that some aspects of Igf2 regulation in cells of neuroectodermal origin do not follow the pattern that exists in mesoderm- and endoderm-derived tissues. Apart from the ventral midline region, H19 and Igf2 were found to be coexpressed in the ectodermally derived Rathke’s pouch and in some circumventricular organs of the brain, such as the organum vasculosum of the lamina terminalis (OVLT) and the pineal gland.

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 heterochro- matin 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.

Proliferation and growth factor expression in abnormally enlarged placentas of mouse interspecific hybrids

It has been shown previously that abnormal placental growth occurs in crosses and backcrosses between different mouse (Mus) species. In such crosses, late gestation placentas may weigh between 13 and 848 mg compared with a mean placental weight of approximately 100 mg in late gestation M. musculus intraspecific crosses. A locus on the X‐chromosome was shown to segregate with placental dysplasia. Thus in the (M. musculus × M. spretus)F1 × M. musculus backcross, placental hyperplasia cosegregates with a M. spretus derived X‐chromosome. Here we have investigated whether increased cell proliferation and aberrant expression of two genes that are involved in placental growth control, Igf2 and Esx1, may cause, or contribute to placental hyperplasia. Increased bromodeoxyuridine labeling of nuclei, reflecting enhanced proliferation, was indeed observed in hyperplastic placentas when compared with normal littermate placentas. Also, increased expression of Igf2 was seen in giant cells and spongiotrophoblast. However, when M. musculus × M. spretus F1 females were backcrossed with males that were heterozygous for a targeted mutation of the Igf2 gene, placentas that carried a M. spretus derived X‐chromosome and were negative for a functional Igf2 allele exhibited an intermediate placental phenotype. Furthermore, in early developmental stages of placental hyperplasia, we observed a decreased expression of the X‐chromosomal Esx1 gene. This finding suggests that abnormal expression of both Igf2 and Esx1 contributes to abnormal placental development in mouse interspecific hybrids. However, Esx1 is not regulated by IGF2.

Divergent genetic and epigenetic post‐zygotic isolation mechanisms in Mus and Peromyscus

Interspecific hybridization in the rodent genera Peromyscus and Mus results in abnormal placentation. In the Peromyscus interspecies hybrids, abnormal allelic interaction between an X‐linked locus and the imprinted paternally expressed Peg3 locus was shown to cause the placental defects. In addition, loss‐of‐imprinting (LOI) of Peg3 was positively correlated with increased placental size. As in extreme cases this placental dysplasia constitutes a post‐zygotic barrier against interspecies hybridization, this finding was the first direct proof that imprinted genes may be important in speciation and thus in evolution. In the Mus interspecies hybrids, a strong role of an X‐linked locus in placental dysplasia has also been detected. However, here we show by backcross and allele specific expression analyses that neither LOI of Peg3 nor abnormal interactions between Peg3 and an X‐linked locus are involved in generating placental dysplasia in Mus hybrids, although the placental phenotypes observed in the two genera seem to be identical. In contrast to this, another dysgenesis effect common to Peromyscus and Mus hybrids, altered foetal growth, is caused at least in part by the same X‐chromosomal regions in both genera. These findings first underline the strong involvement of the X‐chromosome in the genetics of speciation. Secondly, they indicate that disruption of epigenetic states, such as LOI, at specific loci may be involved in hybrid dysgenesis effects in one group, but not in another. Thus, we conclude that even in closely related groups divergent molecular mechanisms may be involved in the production of phenotypically similar post‐zygotic barriers against hybridization.