Hemochorial placentation: development, function, and adaptations

Abstract

Placentation is a reproductive adaptation that permits fetal growth and development within the protected confines of the female reproductive tract. Through this important role, the placenta also determines postnatal health and susceptibility to disease. The hemochorial placenta is a prominent feature in primate and rodent development. This manuscript provides an overview of the basics of hemochorial placental development and function, provides perspectives on major discoveries that have shaped placental research, and thoughts on strategies for future investigation.

The review presents basic concepts of hemochorial placentation, discusses significant contributions to the field, and highlights experimental approaches for future investigation.

Introduction

Everyone had a placenta, but few of us know much about its contributions to the most critical time of our existence, and at a broader level appreciate its importance in the survival of our species. The placenta is much more than reproduction. It sets the stage for all aspects of embryonic and fetal development and largely determines postnatal health. The placenta has a rich history of scientific discovery, some designed and some through serendipity. Investigation of the placenta is somewhat unique and has had its challenges. Evolutionary divergence and ethical considerations have presented experimental obstacles for placental research. Additionally, nomenclature differences across species have also proven to be a hindrance. In this review, we provide a perspective on the placenta and placental research, as well as offer some thoughts for future investigation. The focus will be on hemochorial placentation, which reflects the authors’ familiarity. We approach the task from a trophoblast-centric perspective that is biased by our past research efforts. Initially, we will provide a general framework for understanding placenta development and function and then proceed to highlight discoveries that have shaped placenta research and our understanding of hemochorial placentation. The reader is directed to a series of excellent recent reviews, which expertly discuss a wide range of topics related to placental development and function [1–7].

What is a placenta?

A placenta is a reproductive adaptation that enabled viviparity to become a successful reproductive strategy [8]. The placenta is comprised of specialized epithelial cell types, collectively referred to as trophoblast cells, situated among mesenchymal cells and vasculature, at the maternal-fetal interface. Trophoblast cells are specialized cell types capable of accessing and modifying maternal structures and controlling the bidirectional flow of nutrients and wastes [9]. Effectively, the placenta and its cellular constituents regulate the milieu in which the fetus develops. Among species, different types of placentas exist and exhibit a range of shapes, sizes, cellular constituents, and levels of integration within the female reproductive tract [6]. Humans, the rat and mouse, and many nonhuman primates possess a hemochorial placenta, characterized by a limited cellular barrier for maternal-fetal delivery of nutrients and, in some instances, extensive intrauterine trophoblast cell invasion and uterine arterial remodeling.

Trophoblast cell lineage

Embryonic development is initiated at fertilization by the blending of sperm and egg haploid genomes. The initial cell differentiation event during embryonic development is the establishment of the trophoblast cell lineage [10–12]. As an embryo increases in cell number, cell–cell interactions, and microenvironments change. Cells on the surface of the embryo are directed to differentiate into the trophoblast lineage. These early trophoblast cells, referred to as trophectoderm, encase a cluster of cells called the inner cell mass (ICM). Trophectoderm cells further differentiate into parenchymal cell lineages comprising the placenta, whereas cells from the ICM have the capacity to develop into the embryo and other extraembryonic tissues (allantois, amnion, yolk sac). ICM–trophoblast cell signaling is vital to the growth and development of the trophoblast lineage [13, 14]. Development of specialized trophoblast cell lineages is engineered through interactions with maternal and fetal structures.

Specialized trophoblast cell lineages

Not all trophoblast cells are alike. As stated above, trophoblast stem (TS) and progenitor cell populations give rise to specialized trophoblast cell lineages (Table 1). In the mouse and the rat, TS and progenitor cells differentiate into trophoblast giant cells, spongiotrophoblast cells, glycogen trophoblast cells, invasive trophoblast cells, and syncytiotrophoblast [15–17]. Trophoblast giant cells and syncytiotrophoblast arise via endoreduplication and cell fusion, respectively, which are processes that overtly alter nucleocytoplasmic organization. The human trophoblast lineage also includes TS and progenitor cell populations, which are generically classified within the moniker of cytotrophoblast [2, 18]. This classification is broad and includes both mitotic and postmitotic cell types. Invasive trophoblast cell lineages (extravillous trophoblast) and syncytiotrophoblast also exist in the human placentation site [19, 20]. Collectively, these specialized trophoblast cell types can be defined by their epithelial nature, their location within the placentation site, and their functionality. The precise hierarchy of trophoblast cell lineage development has not been determined in rodents or the human.

Table 1.

Trophoblast cell types.

Cell type	Speciesa	Placentation site location	Principal function(s)
Trophectoderm	Mouse, rat, human	Outer cellular layer of blastocyst	Embryo implantation; inner cell mass/epiblast protection; source of trophoblast stem and progenitor cells
Trophoblast stem progenitor cells	Mouse, rat	Prominently located in extraembryonic ectoderm; ectoplacental cone	Seed differentiated trophoblast lineages
Cytotrophoblast	Human	Beneath syncytiotrophoblast layer (villous compartment); Base of extravillous trophoblast column (extravillous compartment)	Trophoblast stem and progenitor cell population; seed differentiated trophoblast lineages; also, postmitotic component of villous compartment
Primary syncytium	Human	Multinucleated trophoblast at the embryo implantation site of early pregnancy	Invasion/facilitating embryo interstitial implantation into the uterine decidua
Trophoblast giant cell	Mouse, rat	Junctional zone: uterine-placental interface (primary), labyrinth zone (secondary)	Endocrine activity: steroid and peptide hormonogenesis
Glycogen trophoblast cell	Mouse, rat	Junctional zone (transient)	Energy reserve
Spongiotrophoblast cell	Mouse, rat	Junctional zone	Endocrine activity: peptide hormonogenesis
Endovascular invasive trophoblast cell	Mouse, rat, human	Intrauterine: intravascular; extravillous compartment (human)	Uterine spiral artery restructuring
Interstitial invasive trophoblast cell	Mouse, rat, human	Intrauterine: extravascular; extravillous compartment (human)	Placental anchor; uterine stroma remodeling
Syncytiotrophoblast	Mouse, rat, human	Labyrinth zone (mouse, rat); villous compartment (human)	Barrier/transport (mouse, rat, human); endocrine activity: steroid and peptide hormonogenesis (human)

^aTrophoblast cell types found in mouse, rat, and human.

How are trophoblast cells organized into a placentation site?

Embryo implantation is different across species, which affects organization of trophoblast cells and placentation [12]. In the mouse and rat, folds develop within the uterine lumen to create a chamber where mural (abembryonic) trophectoderm attaches to the antimesometrial uterine epithelium. Trophectoderm attachment initiates differentiation of uterine stromal cells into decidual cells, followed by mesometrial uterine epithelial cell death and expansion of the primordial placentation site into the mesometrial decidua. Alternatively, interstitial embryo implantation occurs in the human when polar trophectoderm cells attach to uterine epithelium facilitating complete penetration of the embryo into the underlying decidualized uterine stroma. In the human, this embedding process is facilitated by differentiation of trophectoderm into a primary syncytium [12, 21]. Formation of a primary syncytium associated with embryo implantation is also observed in the guinea pig but not the mouse or rat [22, 23]. As hemochorial placentation progresses, two organizing centers emerge and include the (i) maternal interface and (ii) fetal interface (Figure 1). These organizing centers are determined by the depth extraembryonic mesenchyme and associated vasculature penetrate into the trophoblast compartment. The region encompassing trophoblast and extraembryonic mesenchyme interaction delineates the labyrinth zone of the mouse and rat placenta and the villous compartment of the human placenta [16, 21, 24]. Labyrinth and villous compartments are further demarcated by the presence of syncytiotrophoblast layers. The villous compartment of the human placenta is subdivided into functional units referred to as cotyledons, which are each bathed by maternal blood. Trophoblast cells extending beyond the trophoblast-extraembryonic mesenchyme admixture are located at the maternal boundary. These trophoblast cells are arranged into defined structures in both rodents and the human, and are referred to as the junctional zone and extravillous trophoblast column, respectively [15, 16, 18, 25]. As gestation advances, invasive trophoblast cells arise from the junctional zone in the mouse and rat, and from the extravillous trophoblast column in the human, and migrate into the uterine parenchyma [25–28]. Two types of invasive trophoblast cells have been identified and include (i) interstitial and (ii) endovascular. Interstitial invasive trophoblast cells coalesce between the uterine vasculature, whereas endovascular invasive trophoblast infiltrate uterine blood vessels, especially arteries/arterioles, and supplant the vascular endothelium. During invasive trophoblast cell differentiation, shifts occur in integrin expression altering interactions with surrounding extracellular matrices [29]. As endovascular invasive trophoblast cells differentiate, they acquire an endothelial cell-like phenotype [30–32]. The rat and human exhibit deep intrauterine trophoblast cell invasion, whereas dissemination of trophoblast cells into the mouse uterine parenchyma is nominal [16, 33, 34]. Cellular constituents of the maternal uterine interface, including decidual cells, endometrial glands, and immune/inflammatory cell populations, modulate trophoblast cell behavior (see below).

Figure 1.

Schematic representations of hemochorial placentation sites showing homologous structures within human, rat, and mouse. Note human and rat exhibit deep intrauterine trophoblast invasion, whereas trophoblast invasion in the mouse is shallow. (adapted from Ref. 241)

Placental function

Trophoblast cells support development through facilitating delivery of maternal resources to the fetus and protecting the fetus from adverse exposures emanating from the mother. Acquisition of maternal nutrients is facilitated through several trophoblast-guided processes, including (i) direct modification of arterial vasculature; (ii) production of hormones, which affect a range of maternal targets that act to maintain pregnancy and mobilize nutrient reserves for fetal development; and (iii) expression of nutrient transporters. Invasive trophoblast cell populations are principal directors of uterine arterial vascular remodeling [27, 28]. Several trophoblast cell populations contribute to the endocrine activities of the placenta; however, chief among them are trophoblast giant cells and spongiotrophoblast in the mouse and rat, and syncytiotrophoblast in the human placenta [35–37]. Syncytiotrophoblast is paramount in facilitating nutrient delivery to the fetus [38–40]. The protective function of the placenta is multifaceted and includes the ability to abate harmful actions of the maternal immune system [41] and through the exclusion and biomodification of xenobiotics [4, 42, 43]. Although all trophoblast cell populations interact at some level with the maternal immune system, those embedded within the uterine parenchyma experience the greatest immune challenge. Syncytiotrophoblast express a robust assortment of efflux transporters and xenobiotic metabolizing enzymes, which act to limit delivery of injurious compounds to the fetus [42, 43]. The barrier function of the placenta is also of critical importance in regulating the vertical transmission of pathogens [44].

Experimental tools for placental investigation

In the preceding paragraphs, we have described several trophoblast cell lineages, morphogenetic processes, and placental functions. Each is critical to the success of pregnancy. Mechanistic insights about placentation have been achieved through key experimental advancements (Figure 2).

Figure 2.

Timeline of selected technical advancements in placental research. In vitro and in vivo approaches that have positively impacted placental research over the past 50 years are presented.

Learning from placental pathology and pregnancy diseases

Historically, the tools of anatomy and cell biology have had a major impact on placental research, especially studies of human placenta organization as well as inferences about placenta function. Attributes of specific trophoblast cell lineages have been defined and can be monitored within tissues using techniques such as immunocytochemistry, in situ hybridization, or laser capture microdissection. The arrangement of specific trophoblast cell lineages within the placentation site is precisely determined. Aberrations in the organization and behaviors of specific trophoblast cell populations are correlated with pregnancy disease states [27, 28]. This basic premise of linking placental structure and/or function with disease has been examined at many levels with a variety of approaches, including transcriptomics, epigenomics, proteomics, metabolomics, etc. Genome-wide association studies have also been implemented to identify candidate genes regulating placentation in disease states [45].

Assessment of analyses based on placental pathology and pregnancy diseases

The effectiveness of this approach on human placental investigation is based on the availability of a diversity of tissue specimens. Placental samples spanning gestation from both normal and pregnancy disease states are useful resources. However, access to tissue specimens can be dependent upon geographic location and the socio-political climate. The establishment of placental tissue biobanks (e.g. Research Centre for Women's and Infant's Health BioBank, Sinai Health System, Toronto, Ontario, Canada; National Centers for Translational Research in Reproduction and Infertility Human Placental Tissue Bank, University of California, San Francisco, CA, etc.) has helped facilitate placental research. The strength of a human tissue specimen-based research strategy is in establishing correlations that can be further tested with experimental model systems. However, it is important to appreciate that analyses of human placental specimens do not permit the separation of cause versus effect. A 9-month-old (term) placenta can be carefully interrogated, but it represents a compounding history of successful and failed adaptations. Early gestation samples can be informative; however, they are generally devoid of clinical outcomes. Chorionic villus specimens are an intriguing exception [46], with the caveat that their acquisition presaged a potential disease state. Noninvasive retrieval of sloughed trophoblast cells accumulating at the cervix is a relatively new sampling method that provides sufficient material for molecular analyses, including prenatal diagnostics, and opportunities for establishing correlations with pregnancy outcomes [47, 48]. The utility of the trophoblast cells residing in the cervix as harbingers of in situ placenta function is yet to be determined.

Pregnancy disease can arise from multiple etiologies and can be difficult to compartmentalize into homogenous classifications. Biochemical and molecular analyses of the placentation site are advancing beyond assessment of heterogeneous tissue specimens, with current approaches enabling interrogation of specific cell populations and single cells [49–51]. Tissue architecture impacts cell behavior. Thus, in situ validations will be critical in evaluating isolated cell populations and single cell data.

In vitro approaches

Placental research expanded exponentially as regulatory processes governing trophoblast cell development and function were investigated in a culture dish. Several landmark developments that spurred these efforts are highlighted in the following paragraphs.

Transformed/immortalized trophoblast cell lines

First and foremost, the establishment of cell lines from trophoblast cancers, referred to as choriocarcinomas, represented a pivotal development in placental research. These efforts, first reported approximately 50 years ago, yielded choriocarcinoma cell lines including BeWo (ATCC CCL98), JEG-3 (ATCC HTB-36), and JAR (ATCC HTB-144) [52–55]. Importantly, these cell lines continue to be used in placental research. As methodologies for immortalizing and adapting cells to culture emerged, additional cell lines with trophoblast properties were established, e.g., HTR-8/SVneo (ATCC CRL-3271) [56], SGHPL-4 [57], SWAN-71 (Sw.71, Applied Biological Materials Inc., Richmond, BC, Canada) [58], and others [59–61]. Among these various cell models, BeWo cells are unique in that they can be induced to differentiate into syncytiotrophoblast via elevation of intracellular cyclic adenosine monophosphate [62]. BeWo cells have proven particularly useful for elucidating molecular mechanisms controlling syncytialization [63, 64]. Much has been learned with these cell models; however, several challenges remain due to the nature of the procedures used to establish the cell lines. First, it is a challenge to ascribe a specific trophoblast lineage identity to a cell line, and it is difficult to understand the impact of the transformation/immortalization process on the elaboration of an exact trophoblast cell phenotype. The latter is an especially problematic issue when features of trophoblast cell and transformed/immortalized cell behaviors overlap, e.g., proliferation, invasive properties, immune cell evasion, etc.

Primary trophoblast cell and explant cultures

An assortment of primary cultures derived from isolated trophoblast cells and placental explants has provided a means for investigating trophoblast cell biology [65–70]. Strauss and colleagues established an efficient and reproducible technique for isolating human cytotrophoblast [65]. The procedure, or modified versions of the procedure, has been effectively used by many laboratories to investigate fundamental properties of human trophoblast cells. Primary trophoblast culture systems resolve issues of cellular identity, but present challenges due to biological variation associated with their origin, and their limited life span in culture, and their potential heterogeneity. In addition to primary culture systems, in vitro perfusion of isolated human placental cotyledons has proven to be an effective method for examining bidirectional transport and barrier function of the human placenta [71].

Rodent TS cells

In 1983, Teshima and colleagues published a report describing the generation of a transplantable rat cell population with trophoblast properties, including the presence of large cells that resemble trophoblast giant cells [72]. The cells originated from the mid-gestation rat placenta following fetus removal and peritoneal exposure. The resulting tissue mass, was referred to as a choriocarcinoma, and was used to establish cell lines referred to as RCHO [73] and Rcho-1 [74]. The cells proved remarkable in that they could be maintained in a stem/proliferative state or induced to differentiate by mitogen removal [74]. Differentiation was signified by the development of trophoblast giant cells that exhibited functional attributes comparable to trophoblast giant cells of the placenta [74, 75]. Although Rcho-1 cells have been utilized to effectively investigate the regulation of trophoblast cell differentiation [76–80], they also exhibit some limitations which include aneuploidy (pseudo-tetraploid) and autonomy from developmental growth regulators [81].

Authentic TS cells were first derived from mouse blastocysts and extraembryonic ectoderm [82] and subsequently from single blastomeres [83]. Ex vivo proliferation of TS cells is dependent on fibroblast growth factor (FGF) and transforming growth factor beta (TGFB) family regulatory cues, the same signals that control TS, and progenitor cell expansion in the early embryo [82, 84–86]. Mouse TS cells are diploid and can be indefinitely propagated in vitro [82]. When mitogenic factors are removed, TS cells undergo differentiation and when TS cells are reintroduced into an early embryo, they can contribute to all differentiated trophoblast lineages [82]. Two groups independently established defined culture conditions for maintaining mouse TS cells [86, 87]. The essential components defined by Kubaczka and coworkers include FGF4, heparin, and TGFB, with TS cells plated on an extracellular Matrigel matrix [86]. Alternatively, Ohinata and Tsukiyama identified FGF2, activin A, and inhibitors of Rho-associated protein kinase and WNT signaling as key components for TS cell culture [87]. TS cells have also been derived from rat blastocysts [88] and from blastocysts of mouse and rat mutant models [89–91]. Collectively, mouse and rat TS cells have proven to be extraordinary in vitro model systems for investigating regulatory pathways controlling trophoblast cell stem and differentiation states.

Human embryonic stem cell-derived trophoblast

Unlike the rat and mouse, direct expansion of TS cells from human embryos has proven more difficult. Human embryonic stem (ES) cells are readily propagated from human embryos [92], and have become a model system for investigating development of the human trophoblast cell lineage [93]. A serendipitous observation, first reported in 2002 by Thomson and colleagues, showed that exposure of human ES cells to bone morphogenetic protein-4 (BMP4) resulted in the development of the trophoblast cell lineage [94]. Trophoblast derivation from human ES cells has been optimized by the addition of FGF and activin/nodal signaling inhibitors. These human ES-derived trophoblast cells are proposed to represent a window into early events of human trophoblast lineage development [95]. The physiological relevance of BMP-directed differentiation of human ES cells to trophoblast has been debated [61, 93, 96]. Capture of human TS cells from BMP4-treated human ES cell cultures has been elusive.

Human TS cells

At the end of 2017, a report appeared that described a stem cell population established from first trimester human trophoblast tissue or human blastocysts with all the known and expected features of human TS cells [97]. To identify potential signaling pathways necessary to support a stem and proliferative state, Okae and coworkers performed transcriptome profiling of first trimester cytotrophoblast. Their investigation implicated epidermal growth factor (EGF) and WNT signals as candidate regulators of human TS cells. EGF and WNT pathway activation combined with inhibition of epithelial cell differentiation pathways (TGFB, histone deacetylase, and Rho-associated protein kinase inhibitors) resulted in culture conditions conducive to human TS cell expansion. The stem cell population could be induced to differentiate into both syncytiotrophoblast and extravillous trophoblast through modulation of culture conditions. Similarities and differences were noted between rodent and human TS cells. Importantly, a core group of transcription factors associated with the TS cell stem state, including GATA binding protein 3 (GATA3), TEA domain transcription factor 4 (TEAD4), and transcription factor AP-2 gamma (TFAP2C), are shared between mouse and human [61, 97].

“Extended/expanded” potential stem cell populations

Two recent reports describe exciting new tools for investigating trophoblast development in both the human and mouse [98, 99]. Each report used a different approach to generate pluripotent stem cell populations with “extended/expanded” properties. In the first approach, a chemical library screen was used to generate key components of a cocktail (dimethindene maleate, and minocycline hydrochloride) that could be added to leukemia inhibitory factor (LIF) and a glycogen synthase kinase-3 inhibitor to establish human ES cells with an ICM-like, “naïve” state, from human ES cells with an epiblast-like, “primed” state. Subsequently, an equivalent cell population was also established from the mouse [98]. The second approach used a combination of small molecule signaling pathway inhibitors (mitogen-activated protein kinases, SRC proto-oncogene, non-receptor tyrosine kinase, WNT/HIPPO/poly-ADP-ribosylation), known to interfere with blastomere differentiation and LIF to capture a stem cell population equivalent to a blastomere from a 4- or 8-cell mouse embryo [99]. When reintroduced into early embryos, both “extended/expanded” stem cell populations formed chimeras and contributed to both ICM-derived and extraembryonic cell lineages. These results differ dramatically from chimeras generated using mouse ES cells or TS cells, which contribute only to epiblast or trophoblast lineages, respectively [14]. Thus, “extended/expanded” pluripotent stem cells exhibit properties expected for a multipotent blastomere of an early embryo prior to trophoblast lineage determination.

An earlier report, from Fisher and colleagues described the ex vivo expansion of stem cells from 8-cell human embryos [100]. These cells have the capacity for self-renewal, development of ICM-derived germ layers, and differentiation towards both villous and extravillous human trophoblast cell lineages. Interestingly, the human trophoblast progenitor cells prominently express caudal type homeobox 2 (CDX2) [100], a key regulator of mouse trophoblast lineage development [5] that is not a feature characteristic of human TS cells [97]. Collectively, based on these initial reports, it appears that “extended/expanded” potential stem cells may represent new tools for investigation of early decision-making within the trophoblast lineage.

Induced mouse TS cells

Mouse TS cells have also been engineered through introduction of a core group of transcription factors (Eomesodermin, EOMES; GATA3, TFAP2C) into fibroblasts [101, 102]. This significant accomplishment built on earlier discoveries, in which genes controlling trophoblast lineage development were identified using mouse mutagenesis and mouse TS cells [5]. Induced-TS cells behave like authentic TS cells isolated from mouse embryos. These two cell populations possess similar transcriptomic and epigenomic profiles. Furthermore, they can differentiate into specialized trophoblast lineages and when introduced into blastocysts can contribute to the formation of the placenta. Capturing mouse-induced TS cells was facilitated by well-established culture conditions for mouse TS cell propagation.

Assembly of embryo-like structures with mouse ES and TS cells

Mouse ES cells and TS cells can be assembled into three-dimensional extracellular matrix scaffolds to recapitulate several significant embryonic and extraembryonic developmental events [103]. Such an in vitro model system is remarkable, and can be exploited to investigate molecular mechanisms controlling embryonic-extraembryonic cell cross-talk during embryogenesis.

Overview and assessment of in vitro approaches

In vitro experimentation is a powerful tool to investigate molecular mechanisms controlling trophoblast cell biology. An extensive evaluation of the merits and limitations of various human trophoblast cell models is provided by Moffett and coworkers [61]. The culture and manipulation of rodent and, very recently, human TS cells represent major methodological advancements that permit the discovery of regulatory events that control the TS cell stem state as well as trophoblast differentiation. Differences in rodent and human TS cells exist and delayed progress in the ex vivo expansion of human TS cells. Differentiation along the trophoblast giant cell lineage has been the focus of much effort with rodent TS cells. Effective strategies for directing TS cells to other differentiated trophoblast cell lineages are emerging [104]. Several approaches have surfaced for investigating human trophoblast cell differentiation, including the recent establishment of human TS cells [97]. The discovery of culture conditions for sustaining human TS cells will facilitate the identification of core conserved signaling pathways defining the trophoblast lineage. Human TS cells may supplant other in vitro culture systems and become the in vitro model of choice for investigating human trophoblast cell development. Culture conditions have also been established for monkey, vole, rabbit, cow, and pig TS or TS-like cells [105-109]. Finally, it is important to recognize that in vitro analyses illuminate the “potential” of a cell in the context of the culture environment, which may or may not reflect physiology. Hypotheses generated using in vitro models are best tested in vivo. Knowledge of comparative placentation is a prerequisite for identifying the appropriate animal model to match the regulatory process under investigation.

In vivo approaches

Conservation of genes regulating mammalian placentation is evident [2, 110-113]. Approximately, 80% of genes regulating placentation are conserved in their expression and biological activities between mouse and human [112]. Thus, the mouse has been a valuable animal model for studying many aspects of placentation [2, 114]. Organization of rat and human placentation sites shows conservation regarding deep trophoblast invasion and trophoblast-directed remodeling of uterine spiral arteries [16, 33, 34] (Figure 1). Similarities between guinea pig and human placentation are evident, making the guinea pig another intriguing model for investigating the hemochorial placenta [115, 116]. However, the requisite experimental tools (TS cell culture models, antibodies, in vivo genetic manipulation, etc.) for implementing the guinea pig as a model system for placentation research are not yet available. Nonhuman primate models are advantageous in secondary evaluation of regulatory pathways established in other models but have practical limitations (costs, compliance issues, etc.) in most primary analyses. However, in some instances nonhuman primates may be the only relevant alternative for in vivo analysis of primate-specific aspects of placentation.

Targeted mutagenesis in the mouse

The advent of gene targeting in mouse ES cells and subsequent production of mutant mouse models provided useful discovery tools for identifying and dissecting pathways controlling trophoblast lineage development and placentation [114]. Identification of some genes critical to placental development was based on a foundation of earlier experimentation, while identification of others was a consequence of serendipity. Global in vivo gene disruption experiments identified numerous key regulators of the trophoblast lineage and placental development including the following selected genes: Achaete-scute family bHLH transcription factor 2 (Ascl2), Cdx2, Estrogen related receptor, beta (Esrrb), Gata3, Heart and neural crest derivatives expressed 1 (Hand1), ETS proto-oncogene 2 (Ets2), Peroxisome proliferator activated receptor gamma (Pparg), Glial cells missing homolog 1 (Gcm1), Eomes, Tfap2c, E74 like ETS transcription factor 5 (Elf5), and Tead4 [117-132]. These phenotypic outcomes in the mouse directly led to evaluations of functional conservation in human placentation [61, 91, 133-145]. Refinements to mouse mutagenesis techniques resulted in trophoblast lineage-specific targeting using Cre/Lox strategies. Regulatory sequences associated with the human Cytochrome P450 family 19 subfamily A member 1 (Cyp19a1) gene, several mouse genes (Trophoblast specific protein alpha, Tpbpa; Prolactin family 3, subfamily d, member 1, Prl3d1; Prolactin family 2, subfamily c, member 2, Prl2c2; PR/SET domain 1, Prdm1; Keratin 5, Krt5; Prolactin family 3, subfamily b, member 1, Prl3b1; Gcm1; Cathepsin Q, Ctsq), and a chimeric Tpbpa/Adenosine deaminase (Ada) enhancer have all been effectively utilized to produce gene disruptions in specific lineages of trophoblast cells [146-155] (Table 2). Strengths and limitations of Cre mouse lines for placenta-specific gene disruptions have been discussed [156].

Table 2.

Regulatory sequences demonstrated to drive Cre expression in mouse trophoblast cell lineages.

Regulatory sequencea	Cellular target(s)	References
Cyp19a1	All trophoblast lineages first detectable in extraembryonic ectoderm and ectoplacental cone structures	146
Chimeric Tpbpa/Ada enhancer	All trophoblast lineages	151
Krt5	All trophoblast lineages first detectable in the ectoplacental cone; visceral yolk sac	152
Tpbpa	Trophoblast progenitors in the ectoplacental cone that give rise to subsets of trophoblast giant cells, spongiotrophoblast, glycogen trophoblast cells	148
Prdm1	Trophoblast progenitors in the junctional zone, spongiotrophoblast, subsets of trophoblast giant cells, and glycogen trophoblast cells	149
Prl3d1	Trophoblast giant cells	150
Prl2c2	Trophoblast giant cells	150
Prl3b1	Placenta (specific cell types were not reported; would expect trophoblast giant cell expression)	153
Ctsq	Trophoblast giant cells of the labyrinth zone and lining channels in the junctional zone	155
Gcm1	Syncytiotrophoblast layer-II	147, 154

^aAll regulatory sequences are from the mouse except for human Cyp19a1.

Targeted mutagenesis in other species

Over the past few years, an assortment of genome editing strategies (zinc finger nucleases, transcription activator-like effector nucleases, Crispr/Cas9) has been implemented in additional species to generate global gene disruptions [157, 158]. These technical breakthroughs have expanded the utility of other species that may be more appropriate for investigating specific events in the hemochorial placentation process. Genome editing has led to a renaissance in using the rat as an experimental model for investigating hemochorial placental development [16, 91], and will continue to create additional opportunities for in vivo mechanistic approaches in a wider range of species.

Lentiviral-mediated trophectoderm gene delivery

A clever strategy to genetically manipulate trophectoderm on the surface of the mouse blastocyst was independently reported by three groups [159-161]. The approach takes advantage of the tight surface epithelium of the blastocyst for selective and efficient gene delivery to trophectoderm without ICM access as well as the capacity of trophectoderm to develop into all trophoblast cell lineages. In brief, the zona pellucida encasing the embryo is removed, the denuded blastocyst is incubated with lentiviral constructs and transferred into appropriately timed pseudopregnant mice, followed by subsequent analysis of placentation sites at various stages of gestation (Figure 3). Importantly, both gain-of-function and loss-of-function gene manipulations can be performed [162]. This technique has been adapted for manipulating trophectoderm of embryos from other species, including the rat, rabbit, pig, and sheep [163-166]. The merit of the lentiviral trophectoderm delivery approach is the ability to selectively modify the trophoblast lineage.

Figure 3.

Lentiviral-mediated trophectoderm gene delivery. (a) Procedure for lentiviral-mediated gene delivery. (b–d) Ubiquitin C promoter driven enhanced green fluorescent protein (EGFP) trophectoderm lentiviral delivery, transfer to pseudopregnant rats, and assessment at gestation day 13.5. Fetal–placental tissues were examined under bright field (b), fluorescence microscopy (c), or a merge of panels b and c (d). Note the placental-specific expression of EGFP. Scale bar = 2 mm (adapted from Ref. 163).

Overview and assessment of in vivo approaches

Understanding molecular mechanisms in a physiological context is essential. However, the efficacy of the approach is not simply in using an animal model but in using a relevant animal model. Furthermore, the usefulness of the approach depends on what you are trying to model. Modeling specific physiological processes, or events such as trophoblast cell intrauterine invasion, uterine spiral artery remodeling, or transplacental movement of solutes, can be achieved using an appropriate species. In contrast, modeling the full clinical spectrum of a human pregnancy/placentation disease state or syndrome is inherently more problematic. There are fundamental differences in litter-bearing species versus singleton or twin pregnancies. A negatively impacted pregnancy in a litter-bearing species is associated with activation of mechanisms that cull progeny during pregnancy. The mistake is trying to use one species to model all aspects of placentation and related diseases. Knowledge of hemochorial placentation in a range of species is critical for matching the appropriate animal model to the biology that is being investigated.

Future discovery of conserved regulatory processes controlling hemochorial placental development

Placental research is at an exciting stage. In vitro and in vivo model systems have been established that permit careful dissection of regulatory processes controlling placental development. An effective strategy for future placental research will be the use of a range of TS cell culture systems, appropriately validated animal models that exhibit the relevant trait and that can be genetically manipulated, and the continued interrogation of archived normal and diseased human placental specimens. Combining these experimental tools to address specific biological questions will spur scientific advancement.

Placental biology: selected discoveries

In this section, we highlight discoveries that have impacted our understanding of hemochorial placentation.

Placental structure and pregnancy disease

Relationships between placentation and diseases of pregnancy have been established. This is best illustrated through a landmark discovery of the connection between extravillous trophoblast cells and uterine spiral arteries with preeclampsia [166]. Shallow trophoblast invasion and failures in the modification of the uterine vasculature were identified as pathological features of preeclampsia and subsequently other pregnancy diseases including early pregnancy loss, intrauterine growth restriction, pre-term birth, and placenta abruption [27, 28, 167–169]. Aberrations in trophoblast cell differentiation are apparent in diseases of placentation. Specifically, in preeclampsia, trophoblast cells at the uterine interface fail to acquire an appropriate integrin expression pattern and endothelial cell-like phenotype [170, 171]. Molecular mechanisms underlying the development of invasive trophoblast cell lineages and uterine spiral artery remodeling are poorly understood, as is their efficient removal at the end of pregnancy [172].

Uterine environment—directing placentation

In cancer biology, reference is made to “seed and soil” when describing the ability of tumor cells (seed) to successfully colonize target organs (soil) [173]. Trophoblast cell development within the uterus can be viewed in an analogous manner (trophoblast cell: seed; uterus: soil). During the establishment of pregnancy, the uterus undergoes pivotal changes including the differentiation of uterine stromal cells to multifunctional decidual cells, modulation of endometrial gland function, alterations of the uterine vasculature, and a pronounced redistribution of immune cells within the uterine endometrium [41, 174-177]. Disruptions in uterine development compromise placentation [178-182]. Impairments in uterine stromal cell decidualization may contribute to diseases of placentation, including preeclampsia [183-185] and aging-associated placental insufficiency and pregnancy failure [186].

Identifying the function of various immune cells (natural killer cells, macrophages, dendritic cells, and T-regulatory cells, etc.) within the uterus has been the focus of substantial investigation [41, 176]. As an example, NK cells accumulate in significant numbers at the placentation site and are viewed as both maternal protectors and collaborators in promoting placentation [175, 187, 188]. NK cells carry out several functions including the ability to modify uterine arterial vasculature to facilitate the delivery of nutrients to the placenta [175, 189-191], promote placental development [192], and restrain trophoblast invasion. NK cells effectively regulate the gestational timing of deep placentation [33, 191, 193]. The role of uterine NK cells in restraining trophoblast invasion is not fully accepted. Instead, it has been proffered from in vitro experimentation that uterine NK cells promote trophoblast cell invasiveness [194]; however, this in vitro observation contradicts in vivo findings described above [33, 191, 193]. In human pregnancies, extravillous trophoblast-NK cell signaling is influenced by histocompatibility antigen-killer immunoglobulin-like receptor polymorphic profiles, which influence the depth and extent of intrauterine trophoblast invasion and uterine spiral artery remodeling [195]. Failed pregnancies trigger NK cell killing activity, which contributes to ridding the uterus of compromised extraembryonic and embryonic tissues [188, 196, 197]. Although the NK cell actions listed above are compelling, it is also evident that NK cells are not required for progression of pregnancy, at least when NK cell-deficient mice or rats are maintained in the controlled conditions of a modern research vivarium [191, 193]. It remains possible that resident uterine NK cells, in addition to other immune cells, possess an expanded spectrum of biological roles and an intrinsic purpose in pregnancies that occur in feral environments [198].

Placental endocrine signaling—fodder for evolutionary thought

The placenta is a rich source of hormones, growth factors, and cytokines. Some of the most innovative experimentation prior to the genome era is represented in the identification and characterization of hormones produced by trophoblast cells. These hormones are placental-specific mimics of hormones produced by a wide range of endocrine organs and notably include chorionic gonadotropin (CG), placental lactogen (PL), interferon (IFN), as well as the enzymatic machinery for the biosynthesis of steroid hormones [36, 37, 199-201]. These discoveries were consequential, and in the case of CG became the central component of treatments for infertility and early pregnancy tests [202]. As individual hormone signaling systems were characterized and genomes were sequenced, it became evident that the endocrinology of the placenta exhibits considerable species diversity. This is best demonstrated with a few examples.

CG is a hormone characteristic of primate and equine placentas, but is not present in placentas of other species, including the mouse and rat [203]. PL, which is also referred to as chorionic somatomammotropin, represents an expansion of the growth hormone family in the human [204]. In the rat and mouse, PL was derived from duplications of an ancestral prolactin gene [36]. Furthermore, the expanded prolactin family of the mouse and rat consists of approximately two dozen related proteins representing a diverse array of functions, most of which are not yet fully understood [205-207]. Placentas of some species do not produce hormones with prolactin or growth hormone-related actions [36, 205]. Trophoblast IFNs were discovered through efforts to identify factors controlling the establishment of pregnancy in sheep, a species not exhibiting a hemochorial placenta but instead the less invasive synepitheliochorial placenta [201]. Specifically, these studies identified a protein secreted by trophoblast cells of elongating embryos capable of facilitating the maintenance of corpus luteum function during early pregnancy. This protein was determined to be a new member of the Type I IFN family and was termed IFN tau [208-212]. Other IFNs have been identified in placentas of nonruminant species, but not IFN tau [200]. Finally, the enzymes responsible for the biosynthesis of androgens and estrogens, key hormones regulating reproductive function, exhibit species-specific placental expression profiles. Specifically, 17 alpha hydroxylase (CYP17A1), an enzyme essential for androgen production, is prominently expressed in mouse and rat placentas, but less so in the human placenta [35, 213-216]. Conversely, aromatase (CYP19A1), the rate-limiting enzyme in estrogen biosynthesis, is prominently expressed in human placenta [217], but not in placentas of the mouse and rat [218]. In addition to the hormones listed here, several other examples could be presented to underscore the lack of placenta endocrine system conservation.

Even though most of the tasks required to establish and maintain pregnancy are shared among species, the hormones regulating the physiological adaptations to pregnancy are not necessarily conserved across species. Historically, the extensive divergence in placental endocrinology has been one of the best arguments for not using animal models to investigate the human placenta. Origins of placental hormone diversity arose from fundamental adaptive drivers of evolution such as (i) gene amplification; (ii) selection-induced subfunctionalization and neofunctionalization; (iii) birth of new genes and regulatory sequences. Gene amplification increases protein production and is a known adaptive response to environmental stressors [219]. Once the stressor is removed, genes within the amplified locus can mutate and expand functionality. New genes and regulatory sequences can also arise from co-optation of transposable elements.

Co-opting transposable elements and placental development

As genomes were sequenced, it became apparent that the contribution of DNA designated for coding proteins was a small portion of the total genome, and that the accumulation of repetitive transposable elements in the genome was substantial [220]. Transposable elements arose from ancient viral infections and have served as substrates for the evolution of species, providing contributions to the birth of new genes and regulatory elements controlling gene expression [221-224]. The placenta has provided several examples for co-opting transposable elements. The syncytin genes are one of the best examples of the endogenization of viral genes. Heidmann and colleagues showed that retroviral genes encoding envelope proteins have independently been used in multiple mammalian species to facilitate the process of cell fusion in the development of syncytiotrophoblast [225, 226]. Syncytin gene disruption interferes with the formation of syncytial trophoblast layers in the mouse placenta and compromises pregnancy [227]. In sheep, endogenous Jaagsiekte retroviral genes regulate synepitheliochorial placental development [228]. Furthermore, gene regulatory elements associated with the trophoblast lineage have been generated from transposable elements [229]. The increased involvement of transposable elements in placental biology may be linked to a global hypomethylation of DNA in the placenta [230-234].

Placental plasticity

Ten years ago, we published a report that shifted our thinking regarding the process of placentation [235]. Until that time, we viewed placentation as a programmed outcome of gene regulatory pathways controlling the differentiation of trophoblast cell lineages within the uterine milieu. We did not appreciate how placental development could be shaped by the environment. Prior to the report, we had rediscovered deep placentation in the rat [33] and were motivated to investigate its regulation. Reports suggested that low oxygen hinders the in vitro invasive properties of trophoblast cells [236, 237]. We seized on these findings and sought to use oxygen tension as an in vivo experimental tool with the intent to disrupt intrauterine trophoblast invasion and compromise pregnancy. To our surprise, the low oxygen tensions selected (10%–11% oxygen) evoked a remarkable adaptation. Hypoxia (e.g., low oxygen) exposure led to changes at the placentation site, including expansion of the uterine mesometrial vasculature, alterations to placental compartment sizes, activation of invasive endovascular trophoblast cells, uterine spiral artery remodeling, and, amazingly, maintenance of pregnancy [235] (Figure 4). A critical 24-h period of sensitivity to oxygen tension could be mapped between gestation days 8.5 and 9.5. This adaptive response is highly reproducible in the rat [90, 190] and elements of the adaptive response are also observed in the mouse [238]. Much of what was learned can be summarized in three main lessons: (i) use caution when extrapolating in vitro findings to in vivo animal models; (ii) the intrinsic value for in vivo experimentation is the ability to test mechanisms controlling placentation; and (iii) although hypoxia can be a consequence of a failed placenta and compromised pregnancy, it is also an important signal guiding placental development. Our observations are illustrative of an intrinsic placental/placentation plasticity. Placentas are engineered to adapt to the maternal environment, which has led us to conclude that a signature feature of a healthy placenta is its plasticity [25].

Figure 4.

Placentation site plasticity: responses to hypoxia exposure during rat pregnancy. (a) Chicken beta actin promoter-enhanced green fluorescent protein (chβA-EGFP) transgenic model for monitoring intrauterine trophoblast cell invasion. (b) Gestation day 13.5 placentation site from a rat exposed to ambient oxygen. (c) Gestation day 13.5 placentation site from a rat exposed to hypoxia (∼11% oxygen) from gestation day 6.5–13.5. Maternal hypoxia activated endovascular trophoblast invasion. Scale bar = 0.5 mm (adapted from Ref. 235).

We have not been the only laboratory to observe an effect of an environmental challenge on the placenta. The organization of placentation sites is affected by an assortment of environmental exposures, including maternal diet, chemical exposures, disease states, infectious agents, stressors, etc. [16, 239, 240]. Although hypoxia-inducible factor has been identified as a regulatory mechanism mediating placental adaptations to low oxygen tensions [90, 190, 241-243], regulatory pathways controlling placental adaptations to other environmental challenges are yet to be identified. We infer that disease results when adaptive pathways impacting placentation are overwhelmed or dysregulated.

Sexual dimorphisms exist in placental responses to environmental challenges [244-247]. In general, early stages of female placenta development show heightened sensitivity to disruption, while male placentas are more vulnerable as gestation advances [247]. Sexual dimorphisms may be due to differences in sex chromosome composition, sex-specific modulation of the epigenome, or the ability to relax allele-specific gene repression when exposed to stressors [240, 248, 249]. Genes exhibiting allele-specific expression (e.g., imprinted genes) are fundamentally involved in the regulation of placentation [250]. Epigenetic modifications dictate allele-specific expression and are mediators of environmental impacts on the genome [240, 251].

It appears that the plasticity of the placenta may be enhanced in comparison to other tissues. Mechanisms that impact the epigenome and genome structure are likely at the core of cellular plasticity. The global DNA hypomethylation characteristic of trophoblast cells is not common, and among other cell types has only been observed in cancer cells [230-234]. Epigenetic modifications can relax DNA repression, thereby enabling structural changes, including copy number variation, activation of transposable elements, access to typically unavailable regions of the genome, and increased responses to environmental stressors [233, 245]. Importantly, these structural changes are the same adaptive mechanisms utilized in the evolution of species. Unlike other cell types, trophoblast cells can tolerate profound changes in the organization of their genomes and continue to function normally [252-254]. Interestingly, DNA structural variations are less abundant in diseased versus normal placentas [255]. If plasticity is a positive attribute of the placenta, and is enabled, at least in part, by the epigenetic landscape, then understanding dynamics of epigenome modification during trophoblast cell development and following environmental exposures should be fruitful research endeavors.

Final Thoughts

In the preceding paragraphs, we described tactics for investigating placentation and highlighted selected observations defining the placentation process. We will close by commenting on some potentially provocative directions of placental research.

We have attempted to convince the reader that meaningful tests of hypotheses directed at understanding hemochorial placentation require in vivo animal experimentation. However, different from other organs, the placenta exhibits considerable species diversity, which has led some to a Homo sapian-centric approach to placenta research [256]. No single species represents the perfect animal model for investigating the human placenta, but species exist that have elements of placentation shared with humans. Consequently, rather than be restrictive and relegated to in vitro human trophoblast cell models and establishing correlations of putative regulators impacted by pathology, we need to embrace species diversity. Experimental tools are in hand that permit expanding the effective repertoire of animal models for placenta research.

The placenta is charged with a monumental task: “provide a controlled environment for fetal development that is optimal for postnatal growth, maturation, and survival.” The origins of adult health and disease can be traced to the placenta [4]. Moreover, postnatal success can be predicted by simple measurements of the size and shape of the placenta at birth [257, 258]. The implications are profound. How do disruptions in placental form and function affect the trajectory of postnatal development? Can we identify sensitive events in placentation affecting postnatal development and in a preventative mode introduce placental modifications via stem cell or pharmaceutic treatments with the goal of optimizing adult health? Such an approach will have merits but circumventing an injurious maternal environment may be a limiting factor in the prospects of a successful pregnancy.

Preterm birth is a serious health problem and financial burden [259]. The impact of a shortened gestation that obviates the protective and nutritive functions of the placenta in conjunction with premature fetal exposure to extrauterine life can have devastating life-long consequences. However, prolonging exposure to an unhealthy in utero environment also has costs. Flake and colleagues recently provided a strategy for extrauterine life maintained by an artificial placenta [260]. Fetal sheep were removed during the last month of gestation by caesarian delivery and placed into a plastic bag instrumented to provide a secure and supportive environment for the promotion of growth and maturation. At term, newborn sheep extricated from the artificial womb were of appropriate size and in good health. Optimization of conditions for extrauterine fetal development is required and will benefit from knowledge of placentation and placental function. Extension of the technology to other species, including the human, will present challenges, including ethical issues, which are commonplace for placenta research. To the lay person, extrauterine fetal development may seem to be science fiction, but to the placental researcher it represents a new opportunity.