Successful pregnancy requires a healthy placenta. The placenta supports fetal growth and development by mediating nutrient and gas exchange between the mother and fetus, as well as inducing maternal-fetal immuno-tolerance. The major cellular components of placenta, which mediate this cross talk, are different trophoblasts, which are subdivided by location and function. For example, syncytiotrophoblasts reside within the placental villi and are responsible for hormone secretion; whereas extravillous trophoblasts are located within the decidua and play a role in remodeling maternal spiral artery. The proper functions of all trophoblasts subtypes are critical for a successful pregnancy.
Trophectoderm (TE) differentiation and embryo implantation
Trophoblast lineage initiates at the pre-implantation embryo stage, morula to blastocyst. The trophectoderm of blastocyst gives rise to all subtypes of trophoblasts in placenta. Upon embryo implantation, trophoblasts derived from trophoectoderm attach to the uterine epithelia and induces endometrial stroma cell decidualization. Proper function of trophectoderm is required for the successful implantation and subsequent embryo development as trophectoderm dysfunction is associated with failure of implantation failure and/or aberrant placental development. The morphological integrity of day-5 trophectoderm of human blastocysts has been shown to significantly affect rates of embryo implantation failure and 1st trimester miscarriage in human Assisted Reproductive Technology, confirming the critical role of trophectoderm during the peri-implantation stage .
Studying the molecular mechanism underlying human trophectoderm cells function in vivo during the first trimester of human pregnancy is critical for improving success rates of Assisted Reproductive Technology. However, research is hampered by ethical and practical issues, thereby limiting the availability of human blastocysts and early stage placenta (<=6 week). Therefore, the establishments of a human trophoblast stem cell line or trophoblast progenitor cell line, an in vitro model of trophoectoderm, would be valuable tool in this research field. Similar trophoblast stem cells of murine origin are well established, and have shown the multipotency to differentiate into terminal trophoblasts both in vivo and in vitro. Since human trophoblast stem cells have not yet been established. we and others have generated trophoblast progenitor cells using human embryonic stem cells  and induced pluripotent stem cells . Alternatively, trophoblast progenitor cells have been generated from first trimester placenta  or by ectopic expression of trophoblast specific transcription factors within fibroblasts . These human trophoblast progenitor cell models retain the capacity to differentiate and represent a useful model for the study of human early placentation events.
Placenta trophoblast function and pregnancy diseases
In the placenta, cytotrophoblasts represent the progenitor cells which give rise to both extravillous trophoblasts and syncytiotrophoblasts. Syncytiotrophoblasts secrete placental hormones to maintain pregnancy, whereas extravillous trophoblasts invade the decidua and remodel the maternal spiral arteries to increase fetal blood supply. Because of their critical functions, defective trophoblast differentiation results in diseases of pregnancy, e.g., implantation failure, preeclampsia, and intrauterine growth restriction [1,6]. For example, syncytio- and extravillous trophoblast dysfunction is directly associated with the shallow placentation seen in the preeclamptic placenta [6-8]. Therefore, understanding the molecular events, especially transcription factors that regulate trophoblast differentiation and placentation, will be beneficial in diagnosing, preventing, and/or treating placenta-related diseases of pregnancy.
The biochemistry of GATA transcription factors
The GATA transcription factors family is composed of 6 members and belongs to the zinc finger family [9,10]. With their highly conserved dual zinc finger domains, GATAs regulate gene transcription via binding to a consensus DNA binding motif [5'-(A/T) GATA (A/G)-3'] within a target genes regulatory region. GATA transcription factors play critical roles during embryonic development, and are expressed in a tissue-specific manner. GATA-1~3 is expressed within the hematopoietic system and ectoderm whereas GATA-4~6 is expressed within mesoderm and endoderm [9-11]. For instance, GATA-1 knockout mouse die due to anemia [12,13] and GATA-2 knockout mouse die before embryonic day-11.5 due to hematopoietic defect [14,15], confirming GATA-1 and -2 critical roles in hematopoietic lineage; GATA-3 knockout mouse die before embryonic day-12.5 due to both hematopoietic and neuron defects [16,17], suggesting GATA-3 plays a critical role in hematopoietic lineage and ectoderm; GATA-4 knockout mouse die at embryonic day-8~9 of cardiovascular defects, suggesting GATA-4 plays the role in mesoderm [18,19]; GATA-5 knockout female mouse exhibit abnormalities of the genitourinary tract, suggesting GATA-5 plays a role in the endoderm ; GATA-6 knockout mouse die at embryonic day-5.5 due to defects in extraembryonic endoderm development, suggesting GATA-6 is required for the formation of visceral endoderm [21, 22]. Hence, knockout mouse models suggest that all GATA transcription factors, except GATA-5, are critical in both the 3-germ layer and extraembryonic endoderm during embryonic development. The placenta phenotype of GATA knockout mice has not been documented as fetal death occurs prior to placenta formation/maturation. However, in 4 of these 6 GATA knockout mouse models, embryonic lethality occurs during the development phase of trophoblast differentiation, placenta development stages (Implantation window (embryonic day-4.5 to -5.5), and placenta formation/maturation (embryonic day-7.5 to -12.5) as shown in (Figure 1). These data strongly suggest potential roles for GATAs in trophoblast differentiation and placenta development. Hence, in this review, we summarize the studies to date of GATA transcription factors with respect to trophoblast lineage differentiation and related diseases of pregnancy.
Life span of GATA transcription factors knockout model.
Figure 1 Life span of GATA transcription factors knockout model.
Expression profile of GATA transcription factors in placenta
GATA-3 is highly expressed in mouse and human trophoectoderm at the blastocyst stage [23-25]. In mouse, GATA-3 is detected in the ectoplacental cone by in situ hybridization [24,26]. GATA-3 and GATA-2 are activated during human embryonic stem cell differentiation into trophoblast lineage cells [3,5, 27], and within pig trophoblast lineage cells [27,28]. GATA-6 is recognized as an extraembryonic endoderm marker [29-32], which is also expressed in bovine blastocyst trophoectoderm outgrowths [33-36]. In addition, GATA-1 has been reported in ovine trophoblasts  whereas GATA-4 and GATA-5 expression is absent in trophoblasts.
Function of GATA transcription factors in trophoblast and the dys-regulated GATA transcription factors in placenta- related diseases
Compared to other GATA factors, functions of GATA-2 and -3 in trophoblasts are more intensely studied. GATA-2 and -3 are critical for trophoblast establishment, maintenance and differentiation. Overexpression of GATA-3 in mouse embryonic stem cells results in mouse embryonic stem cells differentiation into trophoblasts . Further, knockdown of GATA-3 inhibits the morula to blastocyst transformation in mice, strongly suggesting that GATA-3 is required for trophoectoderm maintenance . Furthermore, decreased expression of GATA-3 in mouse blastocysts compromises embryo hatching and implantation, indicating GATA-3 also plays an important role in the differentiation and biofunction of trophectoderm . Lastly, GATA-2 and GATA-3 are required for mouse trophoblast giant cell differentiation and function (hormone secretion: proliferin and placental lactogen I), suggesting GATA-2 and GATA-3 are also important for trophoblast giant cells formation and biofunction [40-44]. In human trophoblasts, GATA-2 and GATA-3 are significantly increased during placental trophoblast differentiation and required for regulation of syncytin gene expression . The results demonstrate the dual functions of GATA-3 in the trophoblast stem cells and terminally differentiated trophoblasts.
Data governing the relationship between GATA factors and placenta-related diseases is still rather limited. GATA-1 expression is significantly increased in the peripheral blood of women with preeclampsia . Dys-regulation of GATA-3 has been linked with compromised uterine decidualization in the placenta of women with intrauterine growth restriction (IUGR) and preeclampsia .
Molecular regulations of GATA transcription factors in trophoblast lineage
GATA family factors play important roles in trophoblast differentiation and placenta-related diseases. Hence understanding their molecular regulation will extend our knowledge of their function with respect to placentation and related diseases of pregnancy. GATA transcription factor expressions are broadly regulated by transcription factors, DNA methylation, histone modification and microRNAs during embryonic development.
Transcription factors: Along with CDX2 (caudal type homeobox 2), TEAD4 (TEA domain family member 4), EOMES (eomesodermin), GATA-2 and GATA-3 are key transcription factors of mouse trophoblast stem cells . These key genes activate each other and form a regulatory circuit. TEAD4 and EOMES bind the promoter of GATA-3 to activate its transcription in trophoectoderm of blastocyst [24,48]. EOMES and Tcfap2c (transcription factor AP-2 gamma (activating enhancer binding protein 2 gamma)) positively increase GATA-3 in mouse trophoblast stem cells . On the other hand, GATA-3 occupies CDX2 intron 1 in mouse blastocysts and trophoblast stem cells . Whereas in the inner cells mass of blastocyst, the origin of embryonic stem cells, GATA-3 and GATA-2 are repressed by POU5F1 (POU class 5 homeobox 1) binding to their promoters [33,49]. Furthermore, adding retinoic acid in culture media of human embryonic stem cells induces GATA-3 expression, along with trophoblast lineage differentiation .
Histone modification: Most recently, it has been reported that GATA- 3 is regulated by histon modification in mouse trophoblast lineage . The Histone modification status of GATA promoters can promote or inhibit their gene transcription. Embryonic ectoderm development (EED), a Polycomb group (PcG) protein and positive regulator of methylation of H3K27 (histon 3 Acetyl K27), oppositely regulates GATA-2 and -3 gene expression within the inner cell mass and trophectoderm of blastocysts. High levels of EED recruit H3K27Me3 to occupy GATA-3 promoter and thus repress GATA3 expression within inner cell mass; whereas, trophectoderm has low expression of EED and H3K27Me3 and high expression of GATA-3. Further, ectopic expression of EED in mouse trophoblast stem cells has been shown to increase H3K27Me3 levels at the GATA-3 promoter and repress GATA-3 expression. Furthermore, knockdown of EED expression in mouse trophoblast stem cells activates GATA-3 and decreases the H3K27Me3 of the GATA-3 promoter . On the other hand, lysine-specific demethylase 6B (KDM6B), a histone demethylase and negative regulator of H3K27Me3, has a reverse expression pattern within the inner cell mass and trophectoderm of the blastocyst. Elevated KDM6B in trophectoderm prevents the binding of PcG in trophectoderm and promotes the GATA-3 gene expression. There could be another possibility that H3K27Ac, a transcription factor permissive to histone modification, prevents PcG binding at the GATA-3 promoter in the trophoblast lineage . Together, these data suggest that histone modification regulates GATA-3 expression in mouse trophoblast stem cells and trophoectoderm by binding its promoter.
MicroRNA regulation: MicroRNAs mediated GATA-3 regulation has been well studied in immune systems and cancer. Though the mechanism of microRNA regulation of GATA-3 within trophoblasts remains unknown, there is evidence showing that GATA-3 may be similarly regulated by microRNAs in the placenta. It has been shown that MiR-29b significantly highly expresses in preeclamptic placenta. Further, MiR-29 has been found to increase apoptosis, decrease invasion and angiogenesis in human trophoblasts , which is consistent with the observation that GATA-3 promotes miR-29b expression in breast cancer cells to suppress metastasis (invasion) . It has also been shown that miR-155 contributes to preeclampsia by decreasing trophoblast proliferation/migration , which is supported by studies showing miR-155-induced decreases in GATA-3 expression within T cells to inhibit cell differentiation . Further, decreased miR-21 has been documented in the placenta of infants with lowest birth weights , which agrees with the findings that upregulating miR-21 increases GATA-3 expression in T cells . Together, these data suggest that multiple microRNAs mediated GATA-3 regulation could be critical in the process of trophoblast differentiation and the pathogenesis of placenta-related diseases.
DNA methylation: GATA-3 is known to be regulated by DNA methylation during human embryonic stem cells differentiation ; however, the methylation status of GATA-3 during trophoblast differentiation is still not clear.
Conclusions and future Perspective
GATA transcription factors play important roles during trophoblast lineage and placenta formation. However, there are several open questions that need to be addressed: 1. what are the in vivo roles of GATA transcription factors in placenta? 2. What are their roles in human placenta-related diseases of pregnancy? 3. What epigenetic regulations are in play as GATA transcription factors regulating trophoblast lineage, especially, DNA methylation and microRNAs? To answer these questions: A placental conditional knockout mouse model represents a powerful tool to discover the roles of GATA transcription factors in placenta in vivo. Human trophoblast progenitor cells could be a useful tool to study the epigenetic regulation of GATA transcription factors in human placenta development.
The authors would like to thank Ms. Susan Ferguson and Dr. Trixie Smith for critically reading the manuscript and providing constructive criticism.