RAT PLACENTATION: AN EXPERIMENTAL MODEL FOR INVESTIGATING THE HEMOCHORIAL MATERNAL-FETAL INTERFACE

Abstract

The rat possesses hemochorial placentation with deep intrauterine trophoblast cell invasion and trophoblast-directed uterine spiral artery remodeling; features shared with human placentation. Recognition of these similarities spurred the establishment of in vitro and in vivo research methods using the rat as an animal model to address mechanistic questions regarding development of the hemochorial placenta. The purpose of this review is to provide the requisite background to help move the rat to the forefront in placentation research.

INTRODUCTION

Hemochorial placentation is a strategy involving modification of the maternal-fetal interface for the purpose of facilitating nutrient and waste exchange and development of healthy offspring. The hemochorial placenta possesses two elemental functions: i) ensuring the delivery of maternal nutrients to the placenta; ii) transferring nutrients from the placenta to the developing fetus. Accomplishment of each task requires execution of vital ancillary functions (e.g. immunoregulatory, endocrine, etc). A key feature of hemochorial placentation is the extensive vascular remodeling of maternal uterine spiral arteries, which facilitates nutrient flow and gas exchange to sustain the growing fetus [1–5]. This uteroplacental specialization is used by several species, including higher primates, rodents, lagomorphs, and others [6, 7].

Building a hemochorial placenta is a dynamic process involving the expansion of stem cell populations capable of differentiating into distinct trophoblast cell lineages. Mature trophoblast cells exhibit migratory and invasive properties and possess the capacity to recognize, modify, and emulate the behaviors of cells within their host environment. These developmental fates emerge under the direction of an intrinsic genetic program, the modulatory effects of the maternal environment, and the vitality of the fetus. Plasticity is the operative process in constructing a robust hemochorial placenta. A successful maternal-fetal interface is one that adapts most effectively to the stresses and challenges of pregnancy. Organization of the placentation site is sometimes flawed. Errors within regulatory pathways controlling placental development or maladaptive responses can negatively impact the health of the mother, the progression of fetal development, and can also have lasting effects on postnatal fitness. The intriguing biology and human health relevance make understanding the regulation of hemochorial placentation a compelling scientific pursuit.

Modern biomedical research has benefitted from the use of accessible model organisms (especially rodents) to study fundamental physiological and pathological processes. The benefits of such a research strategy for gaining insights into hemochorial placentation is not universally agreed upon. There are concerns about ‘evolutionary divergence’. Some differences in structure and function of rodent versus human hemochorial placentas and the biology of rodent versus human pregnancies exist [Table 1; 8]. However, there are also compelling similarities among species utilizing hemochorial placentation. Positive selection has resulted in the conservation of genes regulating mammalian placentation [9–13]. The mouse has been a valuable animal model for studying many aspects of placentation [13–15]; however, in contrast to the human, intrauterine trophoblast invasion is shallow and placentation superficial [16–18]. Organization of rat and human placentation sites exhibit striking similarities (Fig. 1; Table 1), especially regarding trophoblast-directed remodeling of the uterine spiral arteries [2, 17, 19–22]. Both species exhibit deep trophoblast invasion [19]. The rat has many of the advantages of the mouse, including the capacity for genetic manipulation [23]. Other rodents possess deep trophoblast invasion (e.g. guinea pig and hamster) [19]; however, experimental tools are not readily available for mechanistic analyses of placentation in these species. Nonhuman primates are advantageous in secondary evaluation of pathways established in other models but have practical limitations in most primary analyses.

Table 1.

Comparison of rat and human pregnancy and placentation

Parameter	Rat	Human
1. Uterine and vascular structure	Duplex uterus – bidirectional blood supply from aortic and internal iliac arteries; externally located arcuate and radial arteries	Simplex uterus-bidirectional blood supply from aortic and internal iliac arteries; internally located arcuate and radial arteries
2. Maternal recognition of pregnancy	Mating-activated luteotropin secretion (pituitary prolactin)/placental-derived luteotropin (secondary)	Placental-derived luteotropin (chorionic gonadotropin)
3. Embryo implantation:	Eccentric (secondarily interstitial); abembryonic	Interstitial; embryonic
i) Uterine modifications	Decidualization (hormonal dependence + implantation stimulus)	Decidualization (hormonal dependence)
ii) Immune cell trafficking	Abundance of NK cells associated with uterine spiral arteries	Abundance of NK cells associated with uterine spiral arteries
4. Placental structure:	Hemochorial (discoid-chorioallantoic)	Hemochorial (discoid-chorioallantoic)
i) Placental-fetal interface	Labyrinth zone (hemotrichorial – syncytial barrier)	Villous trophoblast (hemomonochorial – syncytial barrier)
ii) Uterine-placental interface	Junctional zone and invasive trophoblast cells	Extravillous trophoblast cells
iii) Trophoblast invasion	Deep intrauterine – endovascular and interstitial	Deep intrauterine – endovascular and interstitial
iv) Spiral artery remodeling	Disappearance of smooth muscle, basement membrane restructuring, trophoblast cell replacement of endothelium and acquisition of a pseudo-endothelial phenotype	Disappearance of smooth muscle, basement membrane restructuring, trophoblast cell replacement of endothelium and acquisition of a pseudo-endothelial phenotype
5. Hormonal maintenance of pregnancy: (source of sex steroid hormones)	Corpus luteum (primary)/placenta (secondary – progesterone and androgens)	Corpus luteum (1st trimester); placenta (remainder of pregnancy – progesterone and estrogens)
6. Gestation length	Short (~21 days)	Long (~9 months)
7. Number of offspring at birth	Polytocous	Monotocous/ditocous

Fig. 1. Hemochorial placentation.

Schematic diagram showing homologous structures within human and rat hemochorial placentation sites.

The purpose of this review is to extol the merits of the rat as a model system for studying physiologically relevant mechanisms controlling hemochorial placentation. We describe aspects of the organization and physiology of the rat placenta and the unique features and tools available to generate mechanistic insights into the regulation of hemochorial placentation. The emphasis of the effort is on the invasive trophoblast cell lineage and uterine spiral artery remodeling.

I. DEVELOPMENT OF THE RAT HEMOCHORIAL PLACENTA

The parenchymal cell of the hemochorial placenta is the trophoblast cell. Trophoblast cells have a variety of phenotypes, which are generated from a multi-lineage differentiation pathway [24, 25]. Cellular specializations develop that facilitate trophoblast cell interactions with two vascular beds. Trophoblast cells associated with the maternal vasculature specialize in facilitating nutrient flow to the placenta (rat: junctional zone; human: extravillous trophoblast), whereas trophoblast cells developing in proximity to the fetal vasculature promote nutrient transfer to the fetus (rat: labyrinth zone; human: villous trophoblast).

The rat hemochorial placenta develops from stem cells arising from extraembryonic ectoderm, a derivative of the outer cellular layer of the blastocyst termed trophectoderm. Extraembryonic ectoderm differentiates into chorionic ectoderm and the ectoplacental cone, which subsequently contribute to the labyrinth zone and junctional zone, respectively [25, 26]. The labyrinth zone arises from the interaction of allantoic mesoderm with chorionic ectoderm, yielding trophoblast cell syncytialization and establishment of the barrier for maternal-fetal exchange [27]. The junctional zone borders the mesometrial uterine decidua. This region is also referred to as the ‘trophospongiosum’, ‘spongiotrophoblast’, and ‘spongy region’. Four trophoblast cell lineages differentiate from trophoblast progenitor cells within the junctional zone: i) trophoblast giant cells, ii) spongiotrophoblast cells, iii) glycogen cells, and iv) invasive trophoblast cells (Fig. 2). Trophoblast giant cells are the first trophoblast cell lineage to develop. They are the main endocrine cell of the placenta and possess some invasive abilities. Additionally there are subpopulations of trophoblast giant cells with specific intraplacental locations and functions [28]. Spongiotrophoblast cells are the main constituents of the junctional zone and also contributors to the endocrine function of the placenta. Glycogen cells first appear during midgestation, notably accumulate glycogen, and are probably progenitors for at least a subset of invasive trophoblast cells. Invasive trophoblast cells first appear at midgestation and consist of trophoblast cells that exit the junctional zone and enter the mesometrial uterine compartment, where they penetrate and surround the uterine spiral arteries. Endovascular invasive trophoblast cells replace the endothelium, while interstitial invasive trophoblast cells are situated between the vasculature [17, 20–22]. Invasive trophoblast cells are proposed to play key roles in uterine spiral artery remodeling [5, 29].

Fig. 2. Trophoblast stem cells and their differentiated lineages.

In the rat trophoblast stem cells can be directed toward distinct differentiated trophoblast cell lineages: trophoblast giant cells, spongiotrophoblast cells, glycogen cells, invasive trophoblast cells, and syncytial trophoblast.

Uterine spiral artery remodeling

Placentation sites are fed by uterine spiral arteries. These blood vessels undergo pregnancy-specific changes that facilitate delivery of nutrients. Specific structural modifications create conduits for high volume transfer free of maternal regulatory interference. This includes restructuring all components of the artery, including endothelium, basement membrane, and smooth muscle. Precise orchestration of this process is critical for ensuring appropriate nutrient delivery and preventing inappropriate exposure to deleterious reactive oxygen species [30]. Failures in uterine spiral artery remodeling are linked to pregnancy-associated diseases such as preeclampsia, intrauterine growth restriction, and premature pregnancy termination [1–5]. Mechanisms controlling uterine vascular remodeling are poorly understood. Putative regulators include natural killer (NK) cells and the specialized invasive/extravillous trophoblast cells [1, 5, 31, 32]. In the rat, NK cells and invasive trophoblast cells direct two distinct waves of uterine spiral artery remodeling (Fig. 3).

Fig. 3. Natural killer cell, uterine spiral artery, and trophoblast cell dynamics within rat placentation sites throughout gestation.

The schematic diagram highlights the two waves of uterine spiral artery remodeling executed through the actions of natural killer cells (first wave) and invasive trophoblast cells (second wave).

NK cells at the placentation site

Uterine adaptations to pregnancy include regulated intrauterine immune cell trafficking [31–35]. Following implantation, most maternal leukocytes are excluded from the implantation site except for uterine NK cells. NK cells are conspicuous cellular constituents of uteroplacental compartments in primates and rodents [31, 32, 35–37].

After implantation, in the mouse and rat, NK cells increase in number within the uterine mesometrial compartment [17, 32, 38–40]. This is the site where maternal blood vessels enter the uterus and is the region overlying the developing chorioallantoic placenta. By midgestation, NK cells migrate away from the placenta and establish more prominent relationships with the uterine spiral arteries. The mesometrial vasculature is situated between the mesometrial decidua and the mesometrial surface of the uterus. This morphologically distinct region is referred to as the ‘metrial gland’, ‘mesometrial triangle’, ‘mesometrial lymphoid aggregate of pregnancy’, or the ‘decidualized mesometrial triangle’ [41–45]. As gestation progresses NK cells disappear from the uterine mesometrial compartment. In the mouse, the disappearance of NK cells is caused by both necrotic and apoptotic mechanisms, and is possibly mediated by a Fas ligand-Fas-dependent signaling pathway [46, 47].

NK cells are linked to uterine spiral artery development. NK cell deficiency is associated with profound alterations in the uterine mesometrial vasculature [32, 48]. Genetic deficiency of uterine NK cells in the mouse leads to a lack of remodeling of uterine spiral arteries resulting in hypertrophied vascular media, swollen endothelial cells, and narrow vessel lumens [32, 49, 50]. NK cell depletion in the rat delays uterine spiral artery development resulting in a trophoblast cell-dependent compensatory response characterized by enhanced trophoblast-directed uterine spiral artery remodeling [48]. Experimentation with the mouse and rat as well as human tissues indicate that NK cell effects on the uterine vasculature may be achieved through the production of interferon γ (IFNγ) [50], nitric oxide [51], and/or an assortment of angiogenic growth factors, including vascular endothelial growth factors (VEGFs) [52–57].

Invasive trophoblast cell lineages

Specializations develop that facilitate trophoblast cell interactions with the maternal uterine vascular bed. Trophoblast cells connected to the maternal vasculature specialize in facilitating nutrient flow to the placenta. Intrauterine trophoblast cell invasion is a prominent feature of rat placentation. Rat invasive trophoblast cells have been characterized by their epithelial nature (expression of cytokeratin), polyploidy, accumulation of glycogen, expression of a unique subset of prolactin family cytokines, and their location at the uterine site of placental attachment [17, 20, 21, 58–61]. The first trophoblast cells penetrating into the uterine mesometrial compartment take an endovascular route and the depth of their invasion is generally limited to the mesometrial decidua compartment. Rat endovascular trophoblast cell invasion is a progressive process commencing at midgestation and migrating in a countercurrent fashion up the vessels. Endovascular trophoblast cells replace endothelial cells and become embedded in fibrin [19, 20]. Regulatory mechanisms controlling the propulsion of endovascular trophoblast cells along the inner lining of the uterine spiral arteries, their embedding in fibrin, and their replacement of the endothelium have not been elucidated. A second wave of trophoblast cell invasion begins after gestation day 13.5 and includes both interstitial and endovascular trophoblast cells [17, 20, 21]. Interstitial trophoblast cell migration extends throughout the metrial gland [17, 21]. Their expansion is linked to migration out of the junctional zone and is not associated with proliferation once they arrive in the metrial gland [21, 62]. Invasive trophoblast cells have been isolated from the metrial gland and analyzed by flow cytometry and determined to be primarily diploid [63]. Polyploid trophoblast giant cells arising from the ectoplacental cone do exhibit some limited penetration into the deciduum [21]. Interestingly, a recent report indicates that the depth of trophoblast cell invasion at the mouse placentation site is influenced by parity and extended in multiparous mice [66]. The impact of parity on placentation in the rat has not been reported.

Following parturition, evacuation of intrauterine trophoblast cells is associated with a return of uterine vasomotor control and restructuring of the uterus for the next pregnancy [65, 66]. Efficient postpartum demise of invasive trophoblast cells is critical to the health of the mother and success of subsequent pregnancies. Invasive trophoblast cells are initially retained in the rat uterus following parturition. Removal of intrauterine invasive trophoblast cells is complete within a few days following parturition, associated with an influx of macrophages into the uterine mesometrial compartment, and the return of smooth muscle cells to the uterine mesometrial vasculature [66]. The trigger for demise of retained intrauterine trophoblast cells is, at least in part, associated with their separation from the chorioallantoic placenta at parturition [66]. This observation implicates the placenta as a source of trophic factors, which sustain intrauterine invasive trophoblast cells.

NK cell – invasive trophoblast cell interactions

Although NK cells direct the initial wave of uterine spiral artery remodeling, their actions on the uterine mesometrial compartment are not required for the invasive trophoblast-directed second wave of uterine spiral artery remodeling [48]. This latter process is actually accelerated and more robust in the absence of NK cells [17, 48]. We hypothesize that by promoting spiral artery development, oxygen delivery, and establishing the hemochorial placenta, NK cells effectively delay endovascular trophoblast invasion. Delaying trophoblast entry into spiral arteries and the concomitant enhanced flow of maternal resources to the placenta minimizes the demands on the mother. Rat NK cells also appear to modulate the pseudo-endothelial phenotype of endovascular trophoblast cells [48]. NK cells from other species have been proposed to have direct interactions with trophoblast cells [67].

II. THE RAT AS AN EXPERIMENTAL SYSTEM FOR INVESTIGATING HEMOCHORIAL PLACENTATION

Experimental rat models have been established to investigate the impact of genetics and environmental challenges on placental development.

Genetics of placentation

The genetics of the rat offers benefits for investigating the biology of hemochorial placentation. There are striking strain differences in the organization of the rat placentation site. The Brown Norway (BN) inbred rat strain exhibits a subfertility phenotype and possesses growth-restricted placentation sites [22]. Litter size in the BN rat is limited to 3–5 pups versus 10–15 pups in more fertile inbred and outbred rat strains [22, 68–70]. Subfertility in the BN rat is associated with ovarian and uterine dysfunction [71]. BN rats ovulate fewer eggs and their ovaries show disruptions in their organization and in their abilities to produce steroid hormones. These ovarian anomalies contribute to uterine dysfunction. The BN rat uterus is less responsive to progesterone, which leads to an attenuated decidual reaction and a less than optimal maternal milieu for placental/fetal development [71]. Collectively, these maternal factors and potential intrinsic differences in BN rat trophoblast development contribute to growth-restricted placentas. The most prominent deficits in the BN rat placentation site are in the development of the junctional zone and in the depth of intrauterine invasion of trophoblast cells [22]. The junctional zone is thin and trophoblast invasion is shallow (Fig. 4). This relationship is probably causal and reflects the origin of invasive trophoblast within the junctional zone. The BN rat placental growth deficits are more striking in placentas associated with male fetuses than for placentas from female fetuses [72]. Whether this observation also extends to the unique features of BN rat junctional zone development and intrauterine trophoblast invasion remains to be determined. Some insights about the structural features of BN rat placentation sites have been obtained from gene expression profiling [72]. In comparison to late gestation Sprague Dawley rat placentas, BN rat placentas express higher levels of transcripts encoding proteins contributing to the renin-angiotensin system, and lesser amounts of transcripts encoding proteins regulating angiogenesis. These features may be responsible for the deficits observed in BN rat placentation or alternatively they may represent adaptive responses required for a successful BN rat pregnancy. Late gestation BN rat placentation sites are also unusual in that they possess abundant numbers of NK cells in close proximity to the mesometrial uterine vasculature [22]. The retention of NK cells may signify a BN rat adaptation to poor trophoblast invasion and a requirement to ensure sufficient uterine vascular delivery of nutrients to the placenta and fetus.

Fig. 4. Genetics of rat placentation.

Rat strains exhibit striking differences in the depth and extent of intrauterine trophoblast invasion. Trophoblast cell invasion can be tracked by fluorescence in transgenic placentas constitutively expressing enhanced green fluorescence protein in a wild type uterus. Panel A, Holtzman Sprague-Dawley rat control outbred placentation site exhibiting extensive trophoblast invasion (Adapted from Ref. 81). Panel B, Brown Norway rat placentation site showing limited intrauterine trophoblast invasion.

The quantifiable trait differences in placentation can be exploited to discover regulatory genes controlling placentation. Genetic control of physiological processes can be investigated in recombinant and in chromosome-substituted inbred rat strains [73, 74]. Both strategies have utilized the BN rat representing a normotensive rat and a variety of hypertensive inbred rat strains. Quantitative trait loci have been identified for litter size as well as placental and fetal weights using BN rat and Spontaneous Hypertensive rat recombinant inbred strains [69, 70]. A litter size locus was mapped to rat Chromosome 8 [69], whereas loci affecting placental and fetal weights were mapped to Chromosomes 15 and 1, respectively [70]). Chromosome-substituted rat strains have been generated with BN rat chromosomes introgressed into Dahl Salt Sensitive (DSS) or Fawn Hooded Hypertensive inbred rat strains [75, 76]. The strains have been used extensively to elucidate the genetic control of cardiovascular function [75, 76] and are also proving to be valuable for the characterization of the genetics of reproduction. These analyses are greatly facilitated because the BN rat was used for the first rat genome-sequencing project [77]. BN-DSS rat chromosome-substituted strains were used to identify specific chromosomes linked to placentation [78]. Chromosomes 14 and 17 were identified as possessing regulatory information controlling a quantitative trait associated with rat placentation. Once a quantitative trait is located to a specific chromosome then congenic rats can be generated to further define the locus. There are strategies for ascribing a gene or regulatory sequence to a trait, including comparative genomics, deep sequencing, and the use of bacterial artificial chromosome transgenics for phenotypic rescue [74].

Environmental manipulation of rat placentation

The organization of the rat placenta is sensitive to the maternal milieu. Several experimental manipulations have been performed that demonstrate the ability of the rat placentation site to adapt to an environmental challenge. These manipulations include alterations in maternal nutrition, oxygen delivery, and the induction of various disease states. In the next few paragraphs we provide some examples of environmental manipulations that impact rat placentation.

a) Nutritional modulation of the placentation site

Maternal nutrient availability affects the structure and development of the rat placenta. Maternal under nutrition impacts the entire placenta resulting in increased cell death in both the junctional and labyrinth compartments of the rat placenta [79]. Protein restriction leads to an expansion of the junctional zone, a decrease in the size of the labyrinth zone, and an increase in the surface area comprising the trophoblast-fetal interface [80]. In contrast, feeding pregnant rats a diet that promotes obesity is associated with decreased placental size, including reductions in both junctional and labyrinth zone volumes [81]. Similarly a high fat diet also decreases the size of the junctional zone but not the placental labyrinth zone [82]. The size of the junctional zone may be positively linked to circulating insulin-like growth factor (IGF) 2 levels [81]. In summary, it appears that within certain limits the developing placenta can respond to nutrient availability by adjusting allocation of trophoblast cells to the junctional zone compartment (uterine-trophoblast interface).

b) Hypoxia activation of the invasive trophoblast lineage

Oxygen is an essential cellular nutrient and a fundamental regulator of hemochorial placentation. During early pregnancy oxygen tensions tend to be low and then increase following the establishment of the hemochorial placenta [30, 83–86]. Insights about the role of oxygen supply as an intrinsic regulator of placentation have been derived from mutagenesis of genes in the mouse genome controlling cellular responses to oxygen deprivation [87]. The rat has proven to be an effective model for elucidating oxygen-dependent mechanisms controlling placentation. In vivo and in vitro experiments have demonstrated that development of the invasive/extravillous trophoblast lineage is activated by hypoxia [48, 64]. Low oxygen tension at the maternal-fetal interface alters cellular allocation, creating a preference for trophoblast lineages associated with junctional zone development, including the invasive trophoblast lineage. A critical period of sensitivity to hypoxia has been identified that coincides temporally with early stages of the placentation process [64]. In vivo responses to low oxygen are conserved in rodent and primate placentation [48, 88, 89].

c) Placentation responses to disease

Hypertension

Numerous rat models have been established to investigate pregnancy-associated hypertension and preeclampsia [90]. However, few of these experimental systems have examined the impact of the manipulation on placentation. The emphasis with these models has been examining perturbations in blood pressure regulation and kidney pathologies. An exception is the ‘renin-angiotensinogen transgenic rat model’. In this model system, dams possessing the human angiotensinogen transgene exhibit proteinuria, increased vascular resistance, and hypertension when mated to males expressing the human renin transgene [91, 92]. Endovascular invasive trophoblast cells invade more deeply into the uterine spiral arteries in this transgenic rat model [93, 94]. The increased vascular resistance observed in the renin-angiotensinogen rat transgenic model is linked to the presence of agonistic autoantibodies to the angiotensin II type 1 receptor and to increased uterine angiotensin II concentrations [95–97]. Maternal angiotensin II may also promote trophoblast invasion [98]. The vascular pathologies observed in this rat model may result in a hypoxia-driven activation of invasive trophoblast and thus an intact adaptive response, which is not observed in other pregnancy hypertension conditions such as human preeclampsia [1, 4, 5].

Diabetes

Glucose intolerance can be induced in the rat by treatment with various agents that destroy pancreatic islet cells (source of insulin) leading to a diabetic state [99]. In the pregnant rat, diabetes is associated with an increase in the glycogen cell contribution of the junctional zone and sustained trophoblast cell proliferation resulting in placentomegaly [100–103]. Maternal hyperinsulinemia decreases the number and depth of endovascular invasive trophoblast cells [104].

Maternal ethanol intake

Excess maternal ethanol consumption leads to a recognizable disease state referred as fetal alcohol syndrome [105]. This condition is associated with intrauterine growth restriction and postnatal deficits in brain function. Maternal ethanol intake also impacts rat placentation, including impairment of trophoblast-directed remodeling of uterine spiral arteries [106].

Inflammation

Inflammatory conditions are associated with a number of disease states, and are associated with the development of preeclampsia, intrauterine growth restriction, and spontaneous pregnancy termination. Experimentally, inflammation can be induced by systemic exposure to lipopolysaccharide (LPS). In rats, LPS treatment results in deficits in uteroplacental perfusion and trophoblast-directed uterine spiral artery remodeling [107]. This condition can be blocked by treatment with anti-inflammatory cytokines, such as interleukin 10, or by interfering with the action of tumor necrosis factor alpha.

Stress

Glucocorticoids are part of a response to physiological stressors that is associated with a number of disease states. Elevation of systemic glucocorticoids concentrations during pregnancy in the rat results in placental and fetal growth restriction [108–110]. Increases in apoptosis are observed in trophoblast cells throughout the placenta. These actions of excess glucocorticoids on placentation may be mediated through disruption of VEGF, IGF, peroxisome proliferator-activated receptor gamma, and/or Wnt signaling [110–113].

Rat experimental system overview

Insights gained from investigating the impact of genetics and environment on trophoblast cell invasion and spiral artery remodeling in the rat should have considerable relevance to other species, including the human. Human placental development and function are profoundly affected by the maternal environment [114]. An experimental animal model, such as the rat, offers the opportunity to investigate the progression of the diseased state and mechanisms underlying placental dysfunction. However, an important consideration in interpreting the impact of an experimental perturbation on pregnancy in the rat versus a disease state in the human is that the rat is a polytocous species. Consequently, the rat has an advantage in ensuring at least some offspring survive an environmental challenge or a disease state. They accomplish this fete by limiting the number of embryos implanting and by redirecting resources to fewer placentation sites, which are not routine options available to the human.

III. IN VITRO STRATEGIES FOR INVESTIGATING THE RAT TROPHOBLAST LINEAGE

Methodologies have been developed for in vitro investigation of rat trophoblast cells. These include the establishment of trophoblast stem (TS) cell lines, which have been shown to reflect the behavior of trophoblast cells developing in situ.

Rcho-1 TS cells

Almost three decades ago, Dr. Shinichi Teshima and colleagues at the National Cancer Institute (Tokyo, Japan) induced a transplantable rat choriocarcinoma with extraordinary attributes [115]. The tumor contained cells resembling trophoblast giant cells and they produced placental lactogens [115, 116]. Two research groups established cell lines from the rat choriocarcinoma termed RCHO [117] and Rcho-1 TS cells [118]. These cells are easy to expand and when factors that promote proliferation are removed the cells differentiate into mature trophoblast lineages [118]. Differentiation is directed primarily toward trophoblast giant cells; however, based on gene expression profiles it is evident that other differentiated trophoblast cell lineages are also present. Features distinguishing Rcho-1 TS cells from other established TS cells populations; include their independence of fibroblast growth factor-4 (FGF4) supplementation and their aneuploidy. These transformed rat TS cell lines have been used to investigate various aspects of trophoblast cell biology [119]. Most recently, the Rcho-1 TS cells have been utilized to elucidate the involvement of a phosphatidylinositol 3-kinase/AKT/FOS like antigen 1 signaling pathway controlling the invasive trophoblast phenotype [120–122].

Rat TS cells

A strategy for the ex vivo propagation of TS cells from mouse blastocysts was reported in 1998 [123]. The technique utilized FGF4, heparin, and mouse embryonic feeder layers as agents facilitating the establishment of the mouse TS cell cultures. FGF4-dependent rat TS cell lines have recently been described [124]. The cells were derived from rat blastocysts under conditions similar to those used for the establishment of mouse TS cells. Rat TS cells share many properties with mouse TS cells, including their ability to self renew and to differentiate into trophoblast cell lineages [124]. Similar to mouse TS cells and Rcho-1 TS cells, rat TS cells preferentially differentiate into trophoblast giant cells. They produce steroid hormones and a spectrum of peptide hormones, including members of the prolactin family. However, they also exhibit some differences [124]. Unlike mouse TS cells, rat TS cells do not express the pluripotency gene, Sox2 [125], in their stem state [124] and they do not readily contribute to the formation of a placenta following reintroduction into a blastocyst and transfer to a pseudopregnant rat [124]. Rat TS cells are also distinguished by their expression of Ascl2 [124], a gene implicated in regulating mouse placental morphogenesis [126]. The implication from these observations is that rat TS cells may have advanced beyond the developmental state of mouse TS cells [124]; potentially analogous to the relationship between mouse embryonic stem cells and mouse epiblast stem cells [127]. In vitro behavior of rat TS cells mimics trophoblast cell in vivo adaptive responses during rat placental morphogenesis. For example, low oxygen acutely induces rat TS cells to differentiate along an invasive lineage consistent with in vivo trophoblast cell responses to maternal hypoxia [48, 64].

TR-TBT cells

Nakashima and colleagues established immortalized rat labyrinth trophoblast cells (TR-TBT cells) from gestation day 18 pregnant transgenic rats possessing a temperature-sensitive SV40 large T antigen [128, 129]. The cells proliferate at 33°C. Growth is slowed at 37°C and differentiated functions are evident, such as expression of nutrient and waste transporters. TR-TBT cells have proven useful for studying drug metabolism and transport [128, 129].

Other rat trophoblast-related cell models

There is an assortment of additional cell models that have been developed from rat embryos and placentas that share features with trophoblast cells [130–134]. Some of these cell lines were byproducts of efforts to establish rat embryonic stem cells [132–134]. The mixed developmental phenotypes of the cultures and their limited capacity for differentiation toward trophoblast cells have restricted their usefulness for placental research. Pluripotent rat embryonic stem cells have also been established and could serve as a model for investigating early signals the derivation of the trophoblast cell lineage [135, 136]. Methodologies for establishing primary rat trophoblast cultures from the junctional and labyrinth zones have also been established, and have contributed to our understanding of placental biology [137, 138]. Finally, a precision-cut slice explant culture method has recently been introduced for ex vivo analysis of rat placental tissue [139].

IV. STRATEGIES FOR IN VIVO INVESTIGATION OF RAT PLACENTATION

Experimental approaches have been adapted for monitoring rat trophoblast cells in situ and for isolating and manipulating rat trophoblast cells situated at the rat placentation site.

Phenotypic analysis of the rat placenta

Protocols have been described for mating and gestational staging of rat pregnancy, dissection of placentation sites, and phenotypic characterization of trophoblast cell types [140, 141]. It is most meaningful to examine individual placentation sites in the context of the intact uterus and their associated fetuses. Unique gene and protein expression patterns have been used to distinguish specific rat trophoblast cell lineages developing in situ [140]. Imaging techniques have also been established for quantification of placental development, including the depth and extent of intrauterine trophoblast cell invasion [21, 22, 64].

Tracking invasive trophoblast cells

A couple of useful transgenic rat models are available for in situ monitoring of the invasive trophoblast cell lineage [142–144]. In each case, the transgene consists of a ‘constitutive’ promoter (ROSA 26 or chicken β actin, chβA) driving the expression of a reporter (human placental alkaline phosphatase, hAP or enhanced green fluorescent protein, EGFP). Intrauterine invasive trophoblast cells can be easily visualized on the wild-type uterine background by histochemical detection of heat stable alkaline phosphatase [145] or by fluorescence [66; Fig. 4]. ROSA 26-hAP invasive trophoblast can be quantified in the uterine mesometrial compartment by alkaline phosphatase enzymatic activity measurements [145], while chβA-EGFP positive invasive trophoblast can be recovered, analyzed, and sorted by flow cytometry [64]. There is an interesting caveat. The so-called ‘constitutive’ promoters are not equally expressed in all differentiated trophoblast lineages [64, 145, 146]; thus necessitating parallel experiments monitoring cell-specific reporter activity using histological techniques.

Manipulation of the trophoblast lineage

The true value of an animal model is its use for addressing in vivo mechanistic questions. There are several strategies for manipulating the rat genome [23, 147, 148]. A variety of approaches have been used to generate ‘gain-of-function’ mutations in the rat, including pronuclear injection and viral delivery [149–153]. There is also a growing list of techniques for producing ‘loss-of-function’ mutations, including the use of zinc finger nuclease and TALEN mutagenesis [154, 155, 156], chemical and transposon-based mutagenesis [157–159], and gene targeting using rat embryonic stem cells [160–162]. At this juncture, zinc finger mutagenesis is the most efficient of the techniques for producing knock-out rats [23]. Jacob and colleagues have recently generated ~100 strains of rats with specific mutations in genes implicated in the regulation of the cardiovascular system (http://rgd.mcw.edu/wg/physgenknockouts). Some of these mutant rat strains may be of interest to investigators studying placentation. The limitation with this strategy is that the generated mutations cannot be conditionally regulated in specific cell types. Gene targeting with rat embryonic stem cells represents a potential strategy for generating rats possessing cell-type specific mutations [162]. The latter will also require co-development of suitable mutant rats with trophoblast cell-specific manipulations capable of driving excision of relevant DNA sequences. Such tools are not currently available.

Another strategy of considerable promise is trophectoderm-specific lentiviral gene delivery, which was first developed in the mouse [163, 164]. In this technique, blastocysts are infected with lentiviral particles containing a gene construct of interest and then transferred into the uteri of pseudopregnant females. Gene delivery and activation is restricted to trophectoderm and its derivatives. Placentas can be harvested at various times during gestation and analyzed. The strategy has been successfully adapted for the rat and allows for the generation of both ‘gain-of-function’ and ‘loss-of-function’ [146; Fig. 5]. This technique has been used to investigate regulatory pathways controlling the invasive rat trophoblast cell lineage [122].

Fig. 5. Genetic manipulation through trophoblast-specific lentiviral delivery.

Panel A) Schematic diagram of the procedure. Lentiviral gene constructs are delivered to zona pellucida free blastocysts, which are then transferred to pseudopregnant recipients, and placentation sites analyzed during various stages of gestation. Panel B) Rat blastocysts examined under phase or fluorescence microscopy following incubation with lentiviral constructs expressing EGFP under the control of the phosphoglycerate kinase promoter. Panel C) Trophoblast-specific lentiviral gene delivery assessed at gestation d13.5. Transduced blastocysts were transferred into uteri of pseudopregnant rats and harvested at gestation d13.5 and examined under bright field or fluorescence microscopy. (Adapted from Ref. 135).

V. CONCLUDING COMMENTS

The use of animal model systems provides an essential tool for dissecting molecular mechanisms controlling cellular development. The premise of employing any animal model system is that if the process being studied is fundamental it will likely demonstrate conservation across species. Historically, the rat has been a valuable model for studying most aspects of reproduction and in many areas it still remains the preferred model system. The rat has been the preferred animal model for physiologists because of its size, which makes it more amenable to experimental manipulation, resulting in considerable understanding of rat physiology and its relationship to the human [165]. This led to the rat being the dominant pre-clinical model system used by the pharmaceutical and agro-chemical industries. Although there are some differences in the organization of the rat versus the human maternal-fetal interface, as indicated above similarities in the lineages of cells comprising the placentation site and their function exist. Paramount among the similarities between these two species is the process of trophoblast cell invasion and uterine spiral artery remodeling. If we understand biological processes at the maternal-fetal interface in species that can be experimentally manipulated, such as the rat, then we can more intelligently study the development of the human maternal-fetal interface and identify pivotal junctures of cellular control, facilitating diagnosis and therapeutic intervention. In some instances, cross species similarities may prevail, while in other cases the differences may be most compelling. Nonetheless, our appreciation for the biology of pregnancy and hemochorial placentation increases.