Biochemical Society Transactions

Biochemical Society Annual Symposium No. 80

Epigenetic regulation of placental endocrine lineages and complications of pregnancy

Rosalind M. John


A defining feature of mammals is the development in utero of the fetus supported by the constant flow of nutrients from the mother obtained via a specialized organ: the placenta. The placenta is also a major endocrine organ that synthesizes vast quantities of hormones and cytokines to instruct both maternal and fetal physiology. Nearly 20 years ago, David Haig and colleagues proposed that placental hormones were likely targets of the epigenetic process of genomic imprinting in response to the genetic conflicts imposed by in utero development [Haig (1993) Q. Rev. Biol. 68, 495–532]. There are two simple mechanisms through which genomic imprinting could regulate placental hormones. First, imprints could directly switch on or off alleles of specific genes. Secondly, imprinted genes could alter the expression of placental hormones by regulating the development of placental endocrine lineages. In mice, the placental hormones are synthesized in the trophoblast giant cells and spongiotrophoblast cells of the mature placenta. In the present article, I review the functional role of imprinted genes in regulating these endocrine lineages, which lends support to Haig's original hypothesis. I also discuss how imprinting defects in the placenta may adversely affect the health of the fetus and its mother during pregnancy and beyond.

  • fetal development
  • imprinted gene
  • placenta
  • placental endocrine signalling
  • pregnancy


Two extraordinary characteristics set apart mammals from other classes of vertebrate. One is the high degree of maternal resources provided during development of the young and the other is the epigenetic process of genomic imprinting. Although neither of these characteristics is fully exclusive to mammals, their coexistence is unique. Genomic imprinting involves the epigenetic modification of discrete DNA regions in the male and female germlines that results in the preferential expression of genes from the maternal or paternal allele in the somatic cell [1] (Figure 1). Two of the earliest imprinted genes identified in mice, Igf2 (insulin-like growth factor 2) and Igf2r (insulin-like growth factor 2 receptor), were found to function oppositely to regulate embryonic growth and to be expressed from opposing parental alleles [26] (Figure 1). In response to these and other studies, the parental conflict hypothesis was proposed which suggested that imprinting occurred in response to the increasing conflict between males and females brought about by vivaparity and the asymmetry in parental contribution [7]. Fundamentally, because females provide more resources to her offspring than the male, there is opportunity for males to exploit this relationship. Since pregnancy involves usurping the female physiology to the benefit of the growing fetus, it has been suggested that the processes driving the maternal adaptations required to support pregnancy might also be subject to genomic imprinting [8]. However, because both imprinting and vivaparity arose millions of years ago and because these two characteristics are so intimately linked in extant mammals, much debate still exists and it may well be that imprinting evolved in response to a number of different selective pressures.

Figure 1 Genomic imprinting

(A) In mammals, DNA methylation marks (black lollipops) are established within discrete DNA regions by either the maternal (as shown) or the paternal germline. After fertilization, the epigenetic machinery of the somatic cells builds on these marks to discriminate between the maternal and paternal alleles and, through a complex process of differential DNA methylation and histone modification, nearby genes acquire a heritably silenced state on one of the parental alleles. (B) Targeted mutagenesis in the mouse has demonstrated that genes silenced by the maternal genome are often growth promoters, whereas genes silenced by the paternal genome are those that constrain growth. Grb10, growth factor receptor-bound protein 10; KO, knockout.

Mammals and the placenta

There are three groups of extant mammals, i.e. monotremes, marsupials and eutherians, which exhibit increasing investment towards viviparity and internal development of the embryo, and the subsequent nursing of their young via specialized milk-secreting glands [9]. Extra-embryonic membranes (amnion, allantois, chorion and yolk sac) perform essential functions, such as protection, nutrient transfer, gas exchange and waste removal in all amniotes (mammals, reptiles and birds), but in mammals, and some lizards and snakes, these membranes facilitate ongoing feto–maternal exchange (reviewed recently in [10]). Although early development and growth is dependent on these membranes, some marsupials and all eutherian mammals switch to a more sophisticated and specialized extra-embryonic structure, the chorioallantoic placenta, later in gestation. The chorioallantoic placenta is the main site of feto–maternal exchange until term and has been proposed that this more substantial structure allowed for longer gestation periods resulting in the live birth of larger, better-adapted offspring [11]. However, this longer gestation also entails a greater dependency on maternal resources and consequently more extensive adaptation of the mother.

The mouse placenta

Genomic imprinting has been explored extensively in mouse models aided by our ability to genetically manipulate gene expression. For the purpose of the present review, it is helpful to briefly describe what is known about the development of the trophoblast and the chorioallantoic placenta in the mouse, which has been described more extensively elsewhere [1222]. The first differentiation event after fertilization involves the allocation of the totipotent cells of the newly fertilized zygote to the pluripotent epiblast or to the trophectoderm. This occurs during the early eight-cell morula stage and it is this trophectoderm that contributes most to the mature placenta. A second lineage, the primitive endoderm, emerges from the inner cell mass at the 32–64-cell stage and it is this lineage which will give rise to the endodermal component of the visceral and parietal yolk sacs. As development continues, the trophectoderm expands to encompass the blastocyst, and acquires two spatially determined fates. Cells of the polar trophectoderm, which overlie the inner cell mass, retain a stem cell character whereas those within the mural trophectoderm, which overlies the blastocoel cavity, terminally differentiate at implantation and endoreduplicate to produce primary TGCs (trophoblast giant cells). After implantation, the mural trophectoderm continues to proliferate to give rise to the diploid extra-embryonic ectoderm which contributes to the ectoplacental cone region and, along with the extra-embryonic mesoderm, forms the chorion. It is within the ectoplacental cone and the chorion that the progenitors reside which ultimately give rise to eight discernable trophoblast-derived cell types of the mature placenta (Figure 2).

Figure 2 Allocation of cells to mature mouse placenta

ICM, inner cell mass. Adapted from Developmental Biology, 284(1), David G. Simmons, James C. Cross, Determinants of trophoblast lineage and cell subtype specification in the mouse placenta, 12–24, © 2005, with permission from Elsevier. Additional data taken from [68].

The mature placenta is organized into the histologically distinct labyrinth zone, junctional zone, giant cell layer and maternal decidua (Figure 3). The labyrinth zone is located nearest the embryo and is the region where nutrient transport takes places. Fetal blood vessels invade the base of the placenta after chorioallantoic fusion takes place and are surrounded by a trilaminar layer of trophoblast-derived cells which are in direct contact with maternal blood. Adjacent to the labyrinth is the junctional zone composed of two trophoblast-derived cell types, the spongiotrophoblast and the glycogen cells. As suggested by the names, this region has both a spongy appearance and clearly discernable glycogen stores that accumulate later in gestation. The spongiotrophoblast forms the bulk of the junctional zone [16]. The placenta achieves its maximum weight and is structurally developed by E (embryonic day) 16.5, but whereas the labyrinth keeps on expanding after E16.5, the junctional zone reduces in volume, which may reflect the lack of proliferation of spongiotrophoblast cells after E16.5 alongside the migration of glycogen cells into the maternal decidua [16]. A single layer of secondary parietal TGCs lines the junctional zone forming a barrier between this zone and the maternal decidua. Giant cells are so named because they display a high degree of polyploidy generated by a process called endoreduplication thought to enable large-scale protein synthesis. Consequently, they have giant nuclei. In addition to the parietal TGCs, three other distinct subtypes of trophoblast giant cell have been identified on the basis of their anatomical location and gene signature [17]. The spiral artery TGCs invade the maternal decidua and are involved in modulating the maternal vasculature feeding into the placenta [20], the canal TGCs line the maternal blood supply as it enters the labyrinth, and the sinusoidal TGCs line the small maternal sinusoids within the labyrinth where nutrient transport takes place. These three subtypes are involved in remodelling maternal vasculature and are also in intimate contact with the maternal blood supply throughout the later stages of pregnancy.

Figure 3 The mature mouse placenta

The chorioallantoic placenta is composed of eight discernable trophoblast lineages plus fetal blood vessels. The spongiotrophoblast, the four TGC lineages and the glycogen cell lineage express various combinations of placental prolactins.

Placental hormones and maternal adaptations to pregnancy

Female mammals undergo substantial changes in order to sustain this internal development of the embryos and also for lactation. To support the energetic demands of the fetus both during gestation and immediately afterwards, metabolic changes in the mother occur, including increased food consumption, insulin secretion and body fat, followed by increased peripheral insulin resistance later in gestation. To facilitate the maternal–fetal exchanges, changes in cardiovascular system occur including increased blood volume, red blood cell number and cardiac output. To prevent rejection of the semi-allogenic fetus, the maternal immune system becomes less responsive. Additionally, pregnancy induces bouts of neurogenesis in the female brain. Such changes occur in a controlled sequence, and the tight regulation of the signals inducing these changes is critically important to both the health of the fetus and of the mother. In rodents, ruminants and primates, and probably all mammalian species, the placenta plays a role in inducing and maintaining these changes through the action of secreted hormones related to the pituitary hormones prolactin and growth hormone. The ancestral genes have undergone duplication to give rise to large families of prolactin-related proteins including the placental lactogens (rodents) or duplications of the related growth hormone gene (primates). In mice, the expansion of the placental prolactin family has been remarkable, with at least 23 independent genes encoding placental prolactin homologues [23]. In mice early in pregnancy, prolactin expression from the pituitary induced by the process of mating maintains the corpus luteum, and thus progesterone secretion. A switch occurs halfway through pregnancy, when the mature chorioallantoic placenta forms and placental Prl (prolactin) 3d1, also known as Pl1, and then placental Prl3b1, also known as Pl2, perform this maintenance function. There is a large body of literature that indicates that these lactogenic hormones play additional roles in rodent pregnancy, including the stimulation of insulin-producing β-cell proliferation [24,25], the stimulation of maternal neurogenesis [26] and the control of the transition from a proliferative to a secretory mammary gland [27,28].

In mice, the ectoplacental cone and the adjacent parietal TGCs are the main sites of placental prolactin expression early in pregnancy, whereas the key endocrine lineages of the mature mouse placenta are the TGCs and the spongiotrophoblast [23]. These are also the main sites of expression of the pregnancy-specific glycoproteins, a family of highly similar secreted proteins that contribute to the protection of the semi-allotypic fetus from the maternal immune system and to remodel placental and maternal vasculature [29,30]. There is a distinct pattern of spatial and temporal expression of the placental prolactins, and not all family members act through the prolactin receptor (Figure 4). Thus alterations in both the transcriptional competence of these lineages and the relative contributions of these lineages to the placenta will affect placental hormone production with different outcomes.

Figure 4 Lineage specificity of the placental lactogen gene family and the action of imprinted genes

Representation of the expression specificity of individual placental prolactins as determined by in situ hybridization [23]. Each of the six lineages is represented as in Figures 2 and 3. Genes in red are maternally expressed, and genes in blue are paternally expressed. In cases where the phenotype has been described after targeted deletion, these genes are only presumed to act reciprocally, i.e. loss of function of Peg10 results in the absence of spongiotrophoblast, but this does not necessarily demonstrate a role for Peg10 in promoting the formation of this lineage, which would require an overexpression study.

Genomic imprinting and the placental endocrine lineages

Studies on a number of imprinted genes suggest a selective pressure on the paternal genome to silence genes which limit embryonic and/or placental growth which is offset by the silencing of growth-promoting genes by the maternal genome. Since embryonic growth requires maternal resources, this predicts the existence of imprinted genes that function beyond the immediate feto–placental unit to regulate the distribution of maternal resources. There are two simple mechanisms through which genomic imprinting could regulate maternal resources. First, imprints could directly switch on or off key genes involved in diverting maternal resources to the fetus. An example of this is found in the New World mouse Peromyscus [31]. In this study, one of five placental lactogens identified was expressed only from the paternal allele. To date, there is no evidence for imprinted expression of placental lactogens or the prolactin receptor in other mammals, but only mice and humans have been extensively surveyed. An alternative mechanism through which imprinting could influence placental hormones is by regulating the cell types that express these hormones. The first suggestion that this might be the case came from studies on the transcription factor Ascl2 (achaete–scute complex homologue 2) and an increasing number of imprinted genes are now known to converge on the endocrine lineages of the mouse placenta (Table 1).

View this table:
Table 1 Summary of imprinting phenotypes affecting the mouse endocrine lineages

For references, see the text.

Ascl2 (also known as Mash2)

Ascl2 is primarily expressed in the diploid trophoblast cells of the placenta from the maternal allele [32,33]. Loss of maternal allele expression of Ascl2 (Ascl2−/+) results in a complete absence of the spongiotrophoblast lineage, a reduced labyrinthine trophoblast, which may be secondary to the loss of the spongiotrophoblast [34], and an expanded parietal TGCs layer. Although chorioallantoic fusion takes place, the embryos fail by E10.5. However, in a transgenic model expressing Ascl2 at 50% of the wild-type level, embryos are viable to term, albeit slightly growth-restricted, with a mature placenta almost completely lacking spongiotrophoblast, a complete lack of glycogen cells and an expansion of parietal TGCs [35]. Ascl2 may act as a simple cellular switch between the spongiotrophoblast/glycogen cell lineages and the TGC lineages. However, recent work on a model in which Ascl2 expression is elevated in the placenta and in which there is a reduction of both the spongiotrophoblast and the giant trophoblast giant cell lineages suggests that Ascl2 is more likely to play a role in maintaining progenitors in a specific proliferative state rather than in specifying cell fate decisions (S.J. Tunster and R.M. John, unpublished work).

Phlda2 (pleckstrin homology-like domain, family A, member 2) (also known as Ipl)

Phlda2 is a maternally expressed gene that physically and mechanistically maps to the same imprinted domain on mouse distal chromosome 7 as does Ascl2 and is expressed in the ectoplacental cone, the yolk sac and the syncytiotrophoblast (layers I and II) [3638]. Phlda2 encodes a PH (pleckstrin homology) domain-only protein which may moderate signalling by competing with the PH domains of other proteins preventing their recruitment to the plasma membrane [39]. Phlda2 deficiency results in a specific increase in the junctional zone alongside increased staining for glycogen [40]. Two-fold elevated expression of Phlda2, induced by a Phlda2 BAC (bacterial artificial chromosome) transgene, drives a reciprocal reduction of the junctional zone and a reduction in glycogen staining [41]. Although alterations in placental glycogen suggested a role for Phlda2 in regulating the expansion of this population, marker analysis uncovered a substantial loss of the spongiotrophoblast population in response to the double dose of Phlda2. Further analysis on both the loss-of-function and gain-in-expression models suggests that altering the dosage of Phlda2 does not substantially affect other placental lineages (S.J. Tunster and R.M. John, unpublished work). Thus Phlda2 acts specifically and negatively to regulate the spongiotrophoblast lineage.

Peg10 (paternally expressed gene 10)

Peg10 is a paternally expressed placenta-specific gene derived from a Ty3/gypsy LTR (long terminal repeat) retrotransposon not present in non-mammalian vertebrates [42]. Like Ascl2−/+ embryos, Peg10+/− embryos undergo chorioallantoic fusion but die in utero by E10.5. The developing placenta completely lacks spongiotrophoblast cells, but unlike loss of the active Ascl2 allele, loss-of-function of Peg10 does not result in an expansion of the giant cell population.

Peg3 (paternally expressed gene 3)

Loss of the paternal allele of Peg3, which encodes a large zinc-finger protein with 11 widely spaced C2H2 (Cys2-His2) motifs, results in decreased embryonic and placental weight [43]. Although the placental phenotype has not been characterized specifically with respect to the placental lineages, altered expression of five Prl genes (Prl8a8, Prl3c1, Prl7b1, Prl7a2 and Prl8a6) and the spongiotrophoblast/glycogen/giant cell marker Tpbpa (trophoblast-specific protein β) in the Peg3+/− placenta at E12.5 [44] suggests a specific impact on endocrine cells. Altered expression of Prl8a8 and Prl3c1, which are specific to the spongiotrophoblast, indicate that this lineage is affected by loss of function of Peg3. Altered expression of Prl7b1, which is not expressed in the spongiotrophoblast, suggests that another endocrine lineage is also affected (Figure 4).

Other classically imprinted genes

Imprinted genes that either play a general role in regulating placental growth or ones that alter the relative proportions of the different cell types may also affect endocrine production. Whereas Igf2r−/+ placenta are proportionately increased in size [45], this translates as a 35% increase in the endocrine lineages and thus potentially a 35% increase in the expression of placental hormones. Overexpression of the paternally expressed Igf2 gene induced by loss of imprinting results in a larger labyrinth, a 2-fold increase in the glycogen cell lineage and a 2-fold increase in TGCs with no effect on the spongiotrophoblast [46]. Paternal loss of expression, either globally or just of the placenta-specific transcript, results in a reduction in both the labyrinth and junctional zones with a 40–80% loss of glycogen cells and a reduction in the number of TGCs [47]. Loss of the maternal allele of Cdkn1c (cyclin-dependent kinase inhibitor 1c), a cell-cycle inhibitor, results in placentomegaly, but histological and gene marker analysis indicates a specific reduction in the spongiotrophoblast lineage and the sinusoidal TGCs of the labyrinth with lower expression of prolactin genes [48] (Figure 3). Thus altering the dose of these genes changes the relative proportions of the different cell types affecting the amount of hormones secreted into the maternal blood.

Non-classically imprinted genes

Studies in the mouse have identified four genes that directly limit the endocrine lineages of the placenta but are not classically imprinted. These genes are expressed preferentially from the maternally inherited allele by virtue of their location on the X chromosome, which is paternally silenced in the mouse placenta [49].

Esx1 (extra-embryonic, spermatogenesis, homeobox 1)

Esx1 encodes a paired-like homeobox gene expressed in chorionic ectoderm, visceral yolk sac endoderm and the labyrinthine trophoblast. Esx1−/+ placentas are 151% of the weight of wild-type placentas at E16.5 with significantly more junctional zone and with Tpbpa-expressing cells invading the labyrinth layer [50]. Although an increase in the number of glycogen cells was reported in this study, specific lineage markers were not analysed and there may be alterations to the spongiotrophoblast lineage. Parietal TGC differentiation is not altered, but the other three subtypes were not investigated in detail. Esx1 deficiency is also linked to late embryonic growth retardation.

Cited1 {CBP [CREB (cAMP-response-element-binding protein)-binding protein]/p300-interacting transactivator with glutamate/aspartate-rich C-terminal domain 1}

Cited1 is expressed in all trophoblast-derived tissues of the placenta. Cited1−/+ placentas weigh the same as wild-type placentas, but possess a much larger junctional zone and a concomitant decrease in the volume occupied by the labyrinth [51]. Increased expression Prl2c2 (Plf) determined by in situ hybridization suggests that an increased spongiotrophoblast lineage accounts for the larger junctional zone and there may also be increases in some or all of the TGC lineages.

Nrk (Nik-related kinase)

Nrk encodes a serine/threonine kinase expressed in the junctional zone [52]. Nrk−/+ placentas are 240% of the weight of wild-type placentas at E18.5 with a much larger junctional zone encroaching on the labyrinth. Since only the Tpbpa marker was used to characterize the junctional zone defect, the increased volume could occur as a consequence of an expansion of either the spongiotrophoblast or the glycogen cell lineage, or perhaps both.

Plac1 (placenta-specific protein 1)

Plac1 encodes a putative cell-surface protein expressed in all trophoblastic cells [53]. Plac1−/+ placentas are 200% of the weight of wild-type placentas at E16.5. The junctional zone is greatly expanded. Staining for Prl8a8 confirmed the identity of the expanded region as spongiotrophoblast which, as seen in other models with expanded junctional zones, has encroached into the labyrinth. Increased TGCs numbers were evident in the labyrinth. Plac1 deficiency is also linked to late embryonic growth retardation.

Genomic imprinting and pregnancy adaptations

Although there is compelling evidence that imprinted genes regulates placental endocrine lineages, little work has been performed on mouse dams to lend experimental support for the role for imprinted genes in modulating pregnancy adaptations. Wild-type dams carrying Peg3+/− litters fail to gain the same weight as those carrying wild-type litters [43], which might suggest altered placental signalling. Wild-type dams carrying litters of mixed wild-type and Nrk−/+ pups experience delayed labour, which again may indicate an abnormal maternal state. Heterozygous (Cdkn1c+/−; paternal inheritance of targeted allele thus wild-type for Cdkn1c expression) dams carrying Cdkn1c-deficient pups go into premature labour and show increased blood pressure, proteinuria and glomerular lesions [5456]. Heterozygous (H19+/Δ13; with paternal inheritance of a targeted deletion of the imprinting centre for Igf2, thus wild-type Igf2 levels) dams carrying litters of wild-type and H19Δ13/+ pups (higher Igf2 levels) show altered glucose homoeostasis [57]. Wild-type dams carrying litters of wild-type and Igf2P0+/− fetuses (lacking the placenta-specific Igf2 transcript expressed exclusively in the labyrinthine trophoblast which accounts for 10% of placental Igf2 expression) do not show alterations in glucose management [58]. However, dams carrying fully Igf2P0+/− litters display significantly increased plasma insulin, α-amino nitrogen and corticosterone late in pregnancy [59]. Since, in the case of the Igf2P0, the dams are wild-type and the mutation is restricted to the placenta, alterations in the placenta are likely to drive the changes in the maternal state.

Altered endocrine signalling and pregnancy complications in humans

Beckwith–Weidemann syndrome (MIM #130650), Silver–Russell syndrome (MIM #180860), Prader–Willi syndrome (MIM #176270) and Angelman syndrome (MIM #105830) are four rare childhood developmental disorders which arise in response to the aberrant expression of imprinted genes in humans. However, imprinted genes may play an important role in driving some more frequent complications of pregnancy. IUGR (intrauterine growth restriction), which generally results in low birthweight, is commonly associated with abnormal placental function. Expression of PHLDA2, which we know from animal studies regulates placental development and birthweight [40,41,60], has been reported to be elevated in human IUGR babies in several studies. Initially, McMinn et al. [61] reported increased placental PHLDA2 expression in nine out of 38 IUGR placentas (approximately 25%). Two more recent studies reported similar findings [62,63]. Even within the normal birthweight range, placental PHLDA2 expression inversely correlates with fetal growth or birthweight [64,65]. IUGR may be explained by the failure of the placenta to support the nutrient demands of the rapidly growing fetus. However, if PHLDA2 regulates endocrine lineages in the human placenta, low birthweight may occur as a consequence of abnormal maternal adaptations to pregnancy. Thus aberrant expression of PHLDA2 in the placenta may be linked not only to low birthweight in humans, but also to additional complications of pregnancy such as gestational diabetes, something we can explore using our animal models. Alterations in the expression of imprinted genes, including PHLDA2, have been reported in cases of miscarriage and stillbirth [66] which could be explained by the failure of the placenta to effectively remodel maternal vasculature or by the failure of the mother to undergo systemic adaptations to pregnancy. Some mothers carrying babies with an inactivated maternal copy of the CDKN1C gene develop HELLP (haemolysis/elevated liver enzymes/low platelet count) syndrome, a very severe form of pre-eclampsia [67], which again could be explained by altered placental signalling. Thus there is increasing evidence that this small, peculiar family of genes has great relevance to both our understanding of and treatment of pregnancy complications.


Although there is much direct experimental evidence that links imprinted genes to the endocrine lineages of the mouse placenta, much work is required to determine how this translates into alterations in maternal physiology, metabolism and behaviour. However, the emergence of imprinting models in which specific lineages are altered to either boost or reduce placental signalling provide powerful tools for future studies. The huge variety in the structures of the placenta in various mammalian species, alongside substantial differences in the nature and action of placental compared with maternal hormones during pregnancy, may mean that not all studies on the mouse will be relevant across all mammals. However, it seems likely that there will be parallels that may be useful both in the identification and in the management of ‘at-risk’ human pregnancies.


R.M.J.'s research is funded by the Biotechnology and Biological Sciences Research Council [grant number BB/J015].


  • Biochemical Society Annual Symposium No. 80: Biochemical Society Annual Symposium No. 80 held at University of Leeds, U.K., 11–13 December 2012. Organized and Edited by Paul Hurd (Queen Mary, University of London, U.K.), Adele Murrell (Cancer Research UK) and Ian Wood (Leeds, U.K.).

Abbreviations: Ascl2, achaete–scute complex homologue 2; Cdkn1c, cyclin-dependent kinase inhibitor 1c; Cited1, {CBP [CREB (cAMP-response-element-binding protein)-binding protein]/p300-interacting transactivator with glutamate/aspartate-rich C-terminal domain 1}; E, embryonic day; Esx1, extra-embryonic, spermatogenesis, homeobox 1; Igf2, insulin-like growth factor 2; Igf2r, insulin-like growth factor 2 receptor; IUGR, intrauterine growth restriction; Nrk, Nik-related kinase; Peg3, paternally expressed gene 3; Peg10, paternally expressed gene 10; PH, pleckstrin homology; Phlda2, pleckstrin homology-like domain, family A, member 2; Plac1, placenta-specific protein 1; Prl, prolactin; TGC, trophoblast giant cell; Tpbpa, trophoblast-specific protein β


View Abstract