SEARCH

SEARCH BY CITATION

Keywords:

  • Cdx2;
  • Hippo signal;
  • Tead;
  • Tead4;
  • trophectoderm;
  • Yap

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

During preimplantation mouse development, embryos establish two distinct cell lineages by the time of blastocyst formation: trophectoderm (TE) and inner cell mass (ICM). To explain the mechanism of this cell fate specification, two classical models, namely the inside–outside model and polarity model have been proposed based on experimental manipulation studies on embryos. This review summarizes recent findings on the molecular mechanisms of fate specification, and discusses how these findings fit into the classical models. TE development is regulated by a transcription factor cascade, the core transcription factors of which are Tead4 and Cdx2. The transcriptional activity of Tead4 is regulated by the position-dependent Hippo signaling pathway, thus supporting the inside–outside model. In contrast, several findings support the polarity model; some other findings suggest different mechanisms. We also discuss how the two classical models could be further developed in the light of recent molecular findings.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

During most stages of mouse development, the embryo develops by attaching to the maternal uterine epithelium. Thus, one of the characteristic features of mouse (or mammalian) development is the generation of two groups of cells with distinct fates: embryonic and extraembryonic tissues. Therefore, the major role of preimplantation development is to establish extraembryonic tissues that are required for implanting the embryo into the uterus. During preimplantation development, the mouse embryo develops into a cyst-like structure called the blastocyst (Fig. 1). The blastocyst consists of two types of cells: trophectoderm (TE), an epithelial sheet surrounding the fluid filled cavity, blastocoels (also called blastocyst cavity), and inner cell mass (ICM), a group of cells attached to the inside of the trophoectoderm. The TE cells are involved in implantation by forming extraembryonic tissues including the placenta. Therefore, the specification of TE is the first cell specification event in mouse embryos. Although this first cell fate specification process seems to be simple, its underlying mechanisms still remain elusive. Historically, several models have been proposed based on the results of experimental manipulations. Recent identification of the transcription factors and signaling pathways involved in this process provides clues for understanding the molecular mechanisms. In this review, I summarize the current understanding of how the fate of TE is determined by focusing on its underlying molecular mechanisms and also discuss how these mechanisms fit into the classical models.

image

Figure 1.  Outline of preimplantation mouse development. At the 8-cell stage, embryos undergo compaction, and the blastomeres become flattened and polarized. After the 8-cell stage, the inner cells are formed, and inner cells form inner cell mass and the outer cells form trophectoderm. ab, abembryonic pole; em, embryonic pole.

Download figure to PowerPoint

Outline of preimplantation mouse development

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

The mouse ovulates and the eggs are fertilized around midnight (day 0). After fertilization, the female set of chromosomes resume meiosis and the second polar body is extruded. The zygotes undergo cleavage and become a two-cell embryo on the second day (1.5 days post coitus [dpc]). By repeated cleavage, cell numbers increase to eight on the third day (2.5 dpc) (Fig. 1). Up to the 8-cell stage, the boundaries of blastomeres are clear and each blastomere is morphologically identifiable. At the 8-cell stage, cell-to-cell adhesion mediated by the homophilic cell adhesion molecule E-cadherin increases (Hyafil et al. 1980; Shirayoshi et al. 1983) and cell boundaries become less evident. This process is called compaction. With compaction, each blastomere becomes flattened and polarized by acquiring apicobasal cell polarity. Up to this stage, all blastomeres are topologically equivalent; they have a free apical surface, which makes contact with the external environment. In the following cleavages, the blastomeres adopt either of the following two topological locations: one is an outside position, which maintains contact with the external environment, and the other is the inside position, which is completely surrounded by other cells. The outside cells remain polarized, while the inside cells become apolar (Fig. 2, upper panel). A cell division generating two outside cells from one outside cell is called a conservative division, while a division generating one outside and one inside cell from one outside cell is called a differentiative division (Fig. 2). Generation of outside cells from the inside cells has not been reported. A majority of the inside cells are formed during the cell divisions from 8 to 16 cells (first wave) and 16 to 32 cells (second wave), and the numbers of cells undergoing differentiative division in each wave are variable (Graham & Deussen 1978; Johnson & Ziomek 1981; Pedersen et al. 1986; Fleming 1987). The inside cells seem to be also formed in the subsequent divisions, but detailed analysis has not been performed. Small cavities are formed between the cells and these cavities are combined into a large cavity called the blastocoel by around the 32-cell stage (Motosugi et al. 2005). Embryos with blastocoels are called blastocysts. In the blastocysts, the outside cells constitute the trophectoderm (TE), epithelial cells that form extraembryonic tissues including the extraembryonic ectoderm, and trophoblasts producing the ectoplacental cone and placenta. The inside cells constitute the inner cell mass (ICM). In the blastocysts, the ICM is attached to one end of the TE. The TE covering the ICM is called polar TE, while the one covering the blastocoel is called mural TE. The ICM side of the blastocyst is the embryonic side and the opposite is the abembryonic side. Thus, the blastocyst has an embryonic–abembryonic (Em–Ab) axis (Fig. 1). The ICM further differentiates into an epiblast and primitive endoderm by E4.5. The former maintains pluripotency and forms the embryo proper, while the latter forms extraembryonic tissues that play important roles in patterning embryos.

image

Figure 2.  Historical models for cell fate specification in preimplantation embryos. The polarity model proposes the importance of polarization of blastomeres prior to cell division and differential inheritance of polarity. Polarized cells are indicated with the presence of apical domain (red bar). The inside–outside model proposes that differentiation into inner cell mass (ICM) requires internal position surrounded by other cells.

Download figure to PowerPoint

Preimplantation mouse embryos have strong regulative ability

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

One of the most characteristic features of preimplantation mouse development is that the embryos have a strong regulative ability, which is clearly different from other oviparous lower vertebrates like frogs and fishes. In case of frogs and fishes, the determinant(s) for the future dorsal side is localized to the vegetal pole of the unfertilized eggs (Schier & Talbot 2005; White & Heasman 2008). Upon fertilization, the determinant migrates to one side and this side becomes dorsal. The anteroposterior axis is formed along the animal–vegetal axis of the egg. Thus, the future body axes are already predictable in the zygote, and the removal or ectopic addition of dorsal determinants result in ventralized or dorsalized (or axis-duplicated) embryos. In contrast, mouse embryos appear to lack such localized determinants. Removal of the animal or vegetal pole of the zygotes has no deleterious effect on development (Zernicka-Goetz 1998). Embryos give rise to a normal mouse even if one blastomere of the 2-, 4-, or 8-cell stage embryos was killed, and isolated blastomeres of 4- or 8-cell stage embryos showed totipotency when aggregated with host embryos (Tarkowski 1959a,b; Tarkowski & Wroblewska 1967; Rossant 1976; Kelly 1977). Chimeric mice generated by aggregating two 8-cell stage embryos, or by introducing embryonal carcinoma cells or embryonic stem cells into embryos all developed into normal mice (Garner & Mclaren 1974; Bradley et al. 1984). Therefore, no essential embryonic axis is set up before the blastocyst stage.

Over the past decade, several groups have reported a correlation between the animal–vegetal axis of the zygote (or the polarity of the two-cell-stage embryos) and Em–Ab axis (Gardner 1997; Piotrowska et al. 2001; Plusa et al. 2005b). Others reported the absence of a correlation (Alarcon & Marikawa 2005; Motosugi et al. 2005; Kurotaki et al. 2007), and thus, the existence of a correlation became contentious. As described above, if any intrinsic bias is present in unmanipulated embryos, such bias is overcome in manipulated embryos. This fact clearly indicates that fundamentally important mechanisms exist for cell fate specification other than intrinsic bias, and such mechanisms should support the strong regulative ability of preimplantation embryos.

Two models for explaining the process of cell fate specification of the TE and ICM

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

Historically, three models have been proposed as mechanisms for the segregation of cell fates based on embryo manipulation experiments. The first is the mosaic (segregation) model in which cell fates are specified by the selective segregation of determinants (Dalcq 1957) (see also a review by [Johnson 2009]). Although similar ideas have been reiterated at intervals, the absence of critical determinants in mouse zygotes has been experimentally demonstrated as already discussed above. Thus, this model is no longer supported. In contrast, the other two models are still valid and provide conceptual frameworks for interpreting molecular mechanisms.

The first is the inside–outside (positional) model proposed by Tarkowski (Tarkowski & Wroblewska 1967) (Fig. 2). In this model, cell fate is determined by the position of cells within the embryos. Tarkowski and Wroblewska analyzed the developmental potential of every single-cell blastomere isolated from the 4- or 8-cell stage embryos. Most of the 1/8 embryos developed into trophoblasts suggesting that differentiation into ICM is not regulated by the inheritance of determinants (thus the mosaic model is not supported), and there is an alternative route of differentiation (Tarkowski & Wroblewska 1967). They also proposed a hypothesis that differentiation into ICM requires an intercellular (or inside) environment. This inside–outside model was directly tested (Hillman et al. 1972; Kelly 1977; Suwinska et al. 2008), and forced repositioning of labeled blastomeres of up to 32-cell stage embryos to the inside or outside positions in the aggregates resulted in preferential differentiation into ICM or TE, respectively. Thus, the outside and inside cells adopt position-dependent fates TE and ICM, respectively.

The other model is the polarity (polarization) model originally proposed by Johnson and Ziomek (Johnson & Ziomek 1981) (Fig. 2). In this model, cell polarity controls cell fates. Each 1/8 blastomere of the 8-cell stage embryos is polarized along the apicobasal axis, and this polarity is conserved during its division to 2/16 cells. When the cell division plane is parallel to this axis, two polarized cells or outside cells are formed (conservative division). In contrast, when the division plane is perpendicular to the apicobasal axis, one polar (outside) and one apolar (inside) cells are formed (differentiative division). Because the differentiative division forms two populations that differ both in their positions and properties from the time of their formation, Johnson proposed that the polarization of the 8-cell stage blastomeres is the critical event in the initiation of lineage divergence. The inside cells derived from the bases of polarized cells form the ICM, and the outside cells inheriting the apical surface form the TE. Therefore, the segregation of TE and ICM fates occurs by a process of differential inheritance. This model is supported by a range of observations and is further developed (Johnson & Mcconnell 2004; Yamanaka et al. 2006; Johnson 2009), the details of which are discussed later.

Transcriptional regulators involved in cell fate specification

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

Recent studies identified the transcription factors involved in cell fate specification of ICM and TE. Numerous transcription factors have been required to maintain pluripotency of the ICM, epiblast, and/or embryonic stem (ES) cells. Among them, the key transcription factors for ICM development are the POU family of transcription factor Oct3 (also known as Oct4, encoded by Pou5f1), Sox family of transcription factor Sox2, and a homeodomain protein Nanog. The embryos that have null mutations in any one of these three genes form blastocysts, but the peri-implantation stage embryos lack an epiblast (Nichols et al. 1998; Avilion et al. 2003; Chambers et al. 2003; Mitsui et al. 2003). In the in vitro culture of blastocysts, the ICM of the null mutant blastocysts fails to grow out. ICMs lacking Oct3/4, Sox2, or Nanog differentiate into trophoblasts indicating their involvement in cell fate specification (Nichols et al. 1998; Avilion et al. 2003; Chambers et al. 2003; Mitsui et al. 2003). The pluripotent state of E3.5 ICM is transitional and becomes the ground state in E4.5 naïve epiblasts. Nanog is required to establish a ground state from the transitional state (Silva et al. 2009).

In the TE fate specification, a caudal-like homeodomain transcription factor Cdx2 is the key factor. Cdx2 expression starts from the compacted 8-cell stage, and the expression pattern appears to be stochastic up to the early blastocyst stage. Then, the expression becomes restricted to the outside cells or TE in the blastocysts (Dietrich & Hiiragi 2007; Ralston & Rossant 2008). Cdx2 null mutants develop into early blastocysts, but the blastocoel eventually collapses (Strumpf et al. 2005). The expression of TE-specific genes is significantly decreased, and all cells are positive for Oct3/4 and Nanog. The Cdx2-null embryos fail to grow out into trophoblasts in an in vitro culture (Strumpf et al. 2005). A T-box transcription factor Eomesodermin (Eomes) is another factor required for TE development. The Eomes-null mutants develop into normal-looking blastocysts but fail to grow out when cultured in vitro, indicating that Eomes is required for the differentiation of fate-specified TE into differentiated trophoblasts (Russ et al. 2000). The Ets-transcription factor Elf5 is also involved in TE development. It acts downstream of the initial cell fate specification, and reinforces the expression of Cdx2 and Eomes (Ng et al. 2008). Thus, Cdx2, Eomes, and Elf5 establish a positive feedback loop, which stabilizes the TE lineage (Fig. 3). In ES cells, its expression is suppressed by DNA methylation, and loss of DNA methylation in ES cells induces differentiation into the TE lineage (Ng et al. 2008). A zinc-finger transcription factor Gata3 is another candidate factor involved in TE development. It is strongly expressed in trophoblast stem (TS) cells, and its expression in embryos starts from the 8-cell stage onward and is restricted to the TE in blastocysts (Home et al. 2009). It regulates Cdx2 in TS cells and blastocysts (Home et al. 2009). Although Gata3-knockdown embryos show decreased Cdx2 expression and developmental rate into blastocysts (Home et al. 2009), Gata3-null mutant embryos develop up to the postimplantation stage, probably because of the redundancy of Gata2 (Ma et al. 1997). Thus, the exact role of Gata3 in TE fate specification is currently unknown.

image

Figure 3.  Transcriptional network regulating trophectoderm and inner cell mass (ICM) fates. Tead4 is the most upstream transcription factor regulating trophectoderm (TE) lineage development, and regulates multiple TE regulators in parallel. Mutual suppression between Cdx2 and Oct3/Nanog and feedback loop regulation by Elf5 stabilize TE and ICM fates.

Download figure to PowerPoint

The network of transcription factors governing the ICM-TE fate specification has been elegantly analyzed in ES cells by Niwa and his colleagues (Fig. 3). Mimicking the ICM of Oct3/4 mutants, inactivation of Oct3/4 in ES cells transforms ES cells into trophoblasts (Niwa et al. 2000). Furthermore, overexpression of Cdx2 is sufficient to promote trophoblast differentiation of ES cells (Niwa et al. 2005). Eomes is regulated by Cdx2 and promotes trophoblast differentiation (Niwa et al. 2005). ES cell studies revealed that Oct3/4 and Cdx2 are the key players for networking the segregation of ES-TE fates. At the cleavage stage, Oct3/4 and Cdx2 expression is weak and overlaps, and expression level increase gradually and segregates in the blastocysts (Niwa et al. 2005; Dietrich & Hiiragi 2007). Oct3/4 and Cdx2 enhance their own expression, while mutually suppressing the expression of one another (Niwa et al. 2005). This transcriptional network likely plays a role in segregating their expression domains. Similar mutual suppression is also observed between Nanog and Cdx2 (Chen et al. 2009). Considering that sorting of the expression domains of Cdx2 to TE, and Oct3/4 and Nanog to ICM occurs at different times (early and late blastocyst stages, respectively) (Dietrich & Hiiragi 2007), the transcriptional network observed in ES cells likely plays a major role in the stabilization and/or enhancement of established cell fates rather than initial segregation of cell fates.

The TEA domain transcription factor Tead4 has been recently identified as an essential factor for TE development acting upstream of Cdx2 (Yagi et al. 2007; Nishioka et al. 2008) (Fig. 3). The Tead4 mutants do not express TE/trophoblast-specific genes, fail to form blastocoels, and all cells follow the ICM fate. In Tead4-null mutant embryos, Cdx2 expression is almost completely lost (only faint expression is transiently observed in some embryos between the 8- and 16-cell stages). The abnormalities of Tead4-null embryos are more severe than those of Cdx2-null mutants (Nishioka et al. 2008). In ES cells, expression of activator-modified Tead4 promotes trophoblast differentiation by regulating multiple TE/trophoblast genes even in Cdx2-deficient ES cells (Nishioka et al. 2009). Therefore, Tead4 is the most upstream regulator of TE development, and Tead4 promotes TE/trophoblast development through both Cdx2-dependent and Cdx2-independent pathways.

Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

In preimplantation embryos, Tead4 expression is not lineage-restricted, although its expression is upregulated in the trophoblast lineage after implantation (Yagi et al. 2007; Nishioka et al. 2008). Tead4 is expressed in both ICM and TE, but it is required only for the TE lineage and is dispensable for the ICM lineage (Yagi et al. 2007; Nishioka et al. 2008). Thus the question is why widely expressed Tead4 regulates TE development in the outside cells. Activator-modified Tead4 is sufficient to promote trophoblast differentiation of ES cells and induce Cdx2 in embryos (Nishioka et al. 2009) suggesting that Tead4 instructively regulates TE development and its activity should be controlled within an embryo.

Tead family of transcription factors are characterized as downstream mediators of the Hippo signaling pathway in both mammals (Ota & Sasaki 2008; Zhao et al. 2008b) and flies (Goulev et al. 2008; Wu et al. 2008; Zhang et al. 2008). The Hippo signaling pathway was originally identified as a tumor suppressor signaling pathway in Drosophila (see reviews and references therein, [Harvey & Tapon 2007; Pan 2007; Saucedo & Edgar 2007]). Cells proliferate actively at a low cell density and stop proliferation at a high cell density in mammalian culture cells. This phenomenon, known as contact inhibition of proliferation, is regulated by the Hippo signaling pathway (Zhao et al. 2007; Ota & Sasaki 2008). Thus, cell–cell contact information activates Hippo signaling leading to phosphorylation of a Tead coactivator protein, Yap, and inhibits its nuclear localization (Zhao et al. 2007).

In preimplantation embryos, Yap RNA is widely expressed, but Yap proteins show differential subcellular localization within an embryo. Yap is present in the nuclei of the outside cells, while it is excluded from the nuclei in the inside cells (Nishioka et al. 2009). Subcellular localization of Yap is regulated by differential activation of Hippo signaling within embryos. Activation of the Hippo signaling pathway, monitored by phosphorylation of Yap, is strong in the inside cells and weak in the outside cells (Nishioka et al. 2009). When Hippo signaling is suppressed by overexpression of the dominant-negative form of Lats2, a protein kinase and a Hippo pathway component, or in the Lats1−/−;Lats2−/− double mutant embryos, the inside cells exhibit nuclear accumulation of Yap and strong Cdx2 expression (Nishioka et al. 2009). Thus, Hippo signaling regulates Tead4 activity by modulating the subcellular localization of Yap.

Activation of Hippo signaling depends on cell–cell contacts. When E-cadherin-mediated cell adhesion is inhibited by treating embryos with an anti-E-cadherin antibody, ECCD-1, Hippo signaling is reduced, and inside cells show nuclear Yap and express Cdx2 (Nishioka et al. 2009). Hippo signaling activity also depends on cell position. When cells are forced to adopt an inside position by aggregating three 8-cell stage embryos into one large chimera, only the outside cells show nuclear Yap and express strong Cdx2 and the inside cells do not show nuclear Yap or Cdx2 expression (Nishioka et al. 2009). When an ICM is isolated from an early blastocyst by immunologically removing the outside (TE) cells (immunosurgery), the TE cells are regenerated from the outside cells of the isolated ICM and form a miniature blastocyst (Spindle 1978; Rossant & Lis 1979). In this process, the regenerating outside cells exhibit nuclear Yap and Cdx2 expression (Nishioka et al. 2009). Taken together, in the inside cells, cell–cell contact activates Hippo signaling and suppresses nuclear accumulation of Yap. Therefore, Tead4 is inactive and these cells adopt the ICM fate. In contrast, in the outside cells, weak Hippo signaling allows nuclear accumulation of Yap, which activates Tead4. Active Tead4 induces Cdx2 as well as other TE regulators and promotes TE development (Fig. 4). The exact reason for weak Hippo signaling in the outside cells is currently unknown, but one likely reason is the differential degree of cell–cell contacts: inside cells are completely surrounded by other cells, while outside cells have a free apical surface. Such cell–cell contact ratio-dependent regulation of Yap distribution is also observed in two-dimensional colonies (Zhao et al. 2007) and three-dimensional aggregates of cultured cells (Nishioka et al. 2009). Thus, differential Hippo signaling along the inside–outside axis translates the positional information into TE-ICM cell fates. This model is consistent with the inside–outside model.

image

Figure 4.  Model of cell fate specification by Hippo signaling. In the inside cells, cell–cell contacts/adhesion activates Hippo signaling, which phosphorylates and inhibits nuclear localization of Yap. Therefore, Tead4 is inactive. In the outside cells, weak Hippo signaling allows nuclear accumulation of Yap, which activates Tead4. Active Yap induces Cdx2 and other trophectoderm (TE) regulators and promotes TE differentiation.

Download figure to PowerPoint

Polarization is a key step for cell fate specification: Support for the polarity model

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

Although the Hippo signaling pathway supports the operation of the inside–outside model, it does not exclude the possible involvement of the polarity model. In fact, it is difficult to separate cell position and cell polarity because cell polarity always correlates with cell position in both normal and manipulated embryos.

In the polarity model, each blastomere establishes an apicobasal polarity at the 8-cell stage, and inheritance of apical and basal regions leads to differential cell fates. The first important component of the polarity model is that cell polarity is stable during mitosis. The second is that cell fate is regulated by the inheritance of polarity (Yamanaka et al. 2006). These assumptions are experimentally supported. Blastomeres acquire the apicobasal polarity at the late 8-cell stage accompanied by compaction mediated by E-cadherin. Polarization is observed with the organization of plasma membrane, cytoplasmic components, and cytoskeletons. The cortical apical membrane has dense microvilli (Handyside 1980; Reeve & Ziomek 1981), and apical cytoplasm is enriched in endosomes and clathrin-coated vesicles, while the nuclei are localized more basally (Reeve 1981; Reeve & Kelly 1983; Fleming & Pickering 1985; Maro et al. 1985). The microtubule and actin filaments are also enriched on the apical side (Johnson & Maro 1984; Fleming et al. 1986; Houliston et al. 1987). Among them, the polar apical membrane is most stable, and the cells inheriting the apical membrane acquire cell polarity. During mitosis of dissociated blastomeres of 8-cell stage embryos (1/8-cells), the polarization of the cytoplasm and cytoskeletons (Maro et al. 1985), as well as most junctional contact between cells are lost (Goodall & Maro 1986). However, polarized organization of the apical cortical domain is maintained (Johnson et al. 1988). Such 1/8-cells undergo differentiative or conservative division and produce polar–apolar or polar–polar couplets, but no apolar–apolar couplets (Johnson & Ziomek 1981).

Cdx2 mRNA was recently shown to be enriched in the apical region of 8- to 16-cell stage embryos (Jedrusik et al. 2008). This may cause differential distribution of Cdx2 mRNA between the daughters. Because Cdx2 overexpression enhances cell polarization, such differential distribution of Cdx2 mRNA has been proposed as a mechanism of cell fate specification. However, Cdx2 is not the first factor regulating TE fate and acting downstream of Tead4 and Yap (Yagi et al. 2007; Nishioka et al. 2008, 2009). Furthermore, other studies have shown that cell polarization is independent of zygotic Cdx2, and Cdx2 acts downstream of first lineage specification, because the outside cells of Tead4 mutant embryos establish apicobasal polarity without Cdx2 expression (Nishioka et al. 2008), and Cdx2 mutant cells do not preferentially contribute to ICM in chimeric blastocysts (Ralston & Rossant 2008). Therefore, it is likely that such mechanisms contribute to the stabilization or enhancement of differences between inside and outside cells generated by other mechanisms rather than initiating the differences.

The presence of cell polarity divides the plasma membrane into subdomains and restricts the distribution of membrane-related proteins. The apicobasal polarity of epithelial cells is regulated by interactions among several polarity protein complexes including PAR-aPKC, Crumbs, and Scribble (see reviews by Suzuki & Ohno [2006]; Assemat et al. [2008]). In preimplatation embryos, PAR-aPKC is the only system where the roles have been examined so far. An apical protein Par-6 (Pard6b) becomes enriched in the apical membrane after compaction and its localization is maintained in the outside cells up to blastocysts (Vinot et al. 2005). In the inside cells, Par-6 is distributed throughout the plasma membrane. A similar distribution is also observed for Par-3 (Pard3) (Plusa et al. 2005a). aPKC proteins (PKClambda/Prkci and PKCzeta/Prkcz) are gradually enriched in the apical membrane of the outside cells and are distributed throughout the membrane in the inside cells (Pauken & Capco 2000; Plusa et al. 2005a; Vinot et al. 2005). In contrast, Par-1 (EMK1/Mark2) is localized to the cell contact area (basolateral membrane) in compacted 8-cell stage embryos, but is localized in the nuclei from the 16-cell stage onwards (Vinot et al. 2005). Tight junction protein ZO-1 is localized to the apicolateral edges of the outer cells. Adherens junction proteins E-cadherin and β-catenin are localized to the basolateral membrane of the outside cells and throughout the inside cells. Disruption of the Par-aPKC system by overexpression of either dominant negative aPKC (PKClambda) or knockdown of Par-3 (Pard3) by injecting siRNA in one blastomere of the 4-cell stage embryos causes injected cells to preferentially occupy the ICM position (Plusa et al. 2005a). Although the differentiation status of injected cells is not known, the results are consistent with the hypothesis that the presence of polarity promotes the adoption of the TE fate.

Other mechanisms affecting cell fate specification

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

Differences in developmental potential at the 4-cell stage

Although preimplantation embryos have high regulative ability, some blastomeres appear to exhibit different developmental potentials in certain conditions. The cleavage of two blastomeres takes place sequentially in the second round of cleavage, and the cleavage planes are either meridional (M) or equatorial (E). Based on the orientation and order of cleavage, embryos are classified into four categories: ME, EM, MM, and EE. Among them, in case of ME embryos, the embryonic–abembryonic axis of the blastocysts seems to correlate with that of 2-cell stage embryos (Piotrowska-Nitsche & Zernicka-Goetz 2005). The vegetal blastomeres of the ME embryos express a high level of Cdx2 (Jedrusik et al. 2008), and chimeric embryos obtained solely from such blastomeres have significantly lower developmental potential compared with similar chimeras obtained from other blastomeres (Piotrowska-Nitsche et al. 2005). Biochemically, these blastomeres have a lower level of histone H3 methylation at Arg26 (H3R26me), H3R2me, and H3R17me, and a lower level of the histone methyltransferase CARM1, which is involved in the regulation of ICM-specific transcription factors, Oct4, Nanog, and Sox2 in ES cells (Wu et al. 2009), than those in other blastomeres (Torres-Padilla et al. 2007). All data indicate differences in the developmental potential of blastomeres in 4-cell stage embryos in specific developmental conditions. This seems to fit into the mosaic (or segregation) model. However, such differences are not observed with embryos adopting the other three cleavage patterns. Further studies are needed to elucidate the mechanisms by which differences are generated and the biological significance of the observed differences.

Ras/MAPK signaling

Besides Hippo signaling, the involvement of Ras/mitogen-activated protein kinase (MAPK) signaling has been reported in TE regulation. However, there are discrepancies among three reports. One group reported that conditional activation of the activated Ras allele (Hras1Q61L) in ES cells induces their TE differentiation and trophoblast stem (TS) cell derivation, suggesting that Ras-activated signaling is involved in TE development (Lu et al. 2008). One of the downstream effectors of Ras/MAPK signaling, Erk2, showed polarized distribution at the apical membrane in the 8-cell stage embryos, but in later stage embryos, Erk2 is uniformly distributed (Lu et al. 2008). Treatment of embryos with an Erk inhibitor, PD98059, from 8-cell stage onward inhibits expression of Cdx2 and cavitation, but inhibitor treatment from the morula stage has no effect (Lu et al. 2008). Another group observed no abnormality with the same inhibitor treatment even at high concentrations (Maekawa et al. 2005). The third report also observed no defect in TE fate specification in Erk inhibitor-treated embryos, but instead observed defects in the formation of hypoblast/primitive endoderm from ICM (Nichols et al. 2009). The inhibitor-treated embryos have a decreased number of TE cells (Nichols et al. 2009). This phenotype is consistent with the stimulation of trophoblast proliferation by Fgf4 (Tanaka et al. 1998; Nichols et al. 2009), and Erk is a downstream component of Fgf signaling. Therefore, the significance of this signaling pathway for TE fate specification is currently unknown.

Perspective

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References

The identification of Hippo signaling components as key regulators for position-dependent gene expression provides us with an important clue to understand molecular mechanisms for position-dependent cell fate specification. Although involvement of the Hippo signaling components Lats1/2, Yap, and Tead4, and regulation of Hippo signaling by cell–cell contact has been demonstrated (Nishioka et al. 2009), the exact mechanism linking cell position with Hippo signaling still remains unknown. The most upstream component of the Drosophila Hippo signaling pathway is the large cadherin-like molecule Fat (Bryant et al. 1988), and its activity is regulated by interaction with its ligand Dachsous, another adhesion molecule with large cadherin repeats (Matakatsu & Blair 2006). The mammalian ortholog of fly Fat is Fat4, but Fat4 mutant mice do not exhibit defects related to abnormal Hippo signaling (Saburi et al. 2008). Mice mutants for Dachsous1/2 have not been reported. Thus, their involvement in preimplantation development is currently unknown. The involvement of more downstream components of the Hippo pathway, FERM domain proteins Merlin (Merlin) and Frmd6 (Expanded), protein kinase Mst1/2 (Hippo), and adaptor protein WW45 (Salvador) in the regulation of cell proliferation in mammals (cells and mouse mutants) has been demonstrated (see following reviews and references therein, Edgar [2006]; Pan [2007]; Zhao et al. [2008a]), but their roles in preimplantation development are also unknown. Whether differential Fat-Dachsous signaling causes differential Hippo signaling along the inside–outside axis is an important question to be addressed in the future.

In spite of the importance of cell polarity for cell fate specification, its underlying mechanism remains elusive. An important question is whether the regulation of cell fates by cell polarity also involves Hippo signaling. Several circumstantial observations suggest connections between polarity and Hippo signaling. Nuclear localization of Yap first becomes apparent at the 8-cell stage (Nishioka et al. 2009), concomitant with the acquisition of polarity. In Drosophila, Hippo signaling regulates the apical domain size independent of cell proliferation (Genevet et al. 2009; Hamaratoglu et al. 2009), suggesting the connection between cell polarity and Hippo signaling. If cell polarity regulates Hippo signaling, then it should involve the segregation of signaling components, possibly membrane localized proteins, into subdomains as observed with polarity and cell adhesion-related proteins. Such segregation could block activation of the signaling pathway.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Outline of preimplantation mouse development
  5. Preimplantation mouse embryos have strong regulative ability
  6. Two models for explaining the process of cell fate specification of the TE and ICM
  7. Transcriptional regulators involved in cell fate specification
  8. Tead4 activity is regulated by cell–cell contacts through Hippo signaling pathway components: Support for the inside–outside model
  9. Polarization is a key step for cell fate specification: Support for the polarity model
  10. Other mechanisms affecting cell fate specification
  11. Perspective
  12. References