Cell fate determination

The fate map of the sea urchin embryo was originally created by observing each of the cell layers and what its descendants became. More recent investigations have refined these maps by following the fates of individual cells injected with fluorescent dyes such as diI (see Chapter 1). These studies have shown that by the 60-cell stage, most of the embryonic cell fates are specified, but that the cells are not irreversibly committed. In other words, particular blastomeres consistently produce the same cell types in each embryo, but these cells remain pluripotent and can give rise to other cell types if experimentally placed in a different part of the embryo.

A fate map of the 60-cell sea urchin embryo is shown in Figure 8.12 (Logan and McClay 1999; Wray 1999). The animal half of the embryo consistently gives rise to the ectoderm—the larval skin and its neurons. The veg₁ layer produces cells that can enter into either the ectodermal or endodermal organs. The veg₂ layer gives rise to cells that can populate three different structures—the endoderm, the coelom (body wall), and secondary mesenchyme (pigment cells, immunocytes, and muscle cells). The first tier of micromeres produces the primary mesenchyme cells that form the larval skeleton, while the second tier of micromeres contributes cells to the coelom (Logan and McClay 1997, 1999).

Figure 8.12

Fate map and cell lineage of the sea urchin Strongylocentrotus purpuratus. (A) The 60-cell embryo is shown, with the left side facing the viewer. Blastomere fates are segregated along the animal-vegetal axis of the egg. (B) Cell lineage map of the embryo. (more...)

Although the early blastomeres have consistent fates in the larva, most of these fates are achieved by conditional specification. The only cells whose fates are determined autonomously are the skeletogenic micromeres. If these micromeres are isolated from the embryo and placed in test tubes, they will still form skeletal spicules. Moreover, if these micromeres are transplanted into the animal region of the blastula, not only will their descendants form skeletal spicules, but the transplanted micromeres will alter the fates of nearby cells by inducing a secondary site for gastrulation. Cells that would normally have produced ectodermal skin cells will be respecified as endoderm and will produce a secondary gut (Figure 8.13; Hörstadius 1973; Ransick and Davidson 1993). The micromeres appear to produce a signal that tells the cells adjacent to them to become endoderm and induces them to invaginate into the embryo. Their ability to reorganize the embryonic cells is so pronounced that if the isolated micromeres are recombined with an isolated animal cap (the top two animal tiers), the animal cap cells will generate endoderm, and a more or less normal larva will develop (Figure 8.14; Hörstadius 1939).

Figure 8.13

Ability of the micromeres to induce a secondary axis in sea urchin embryos. (A) Micromeres are transplanted from the vegetal pole of a 16-cell embryo into the animal pole of a host 16-cell embryo. (B) The transplanted micromeres invaginate into the blastocoel (more...)

Figure 8.14

Ability of the micromeres to induce presumptive ectodermal cells to acquire other fates. (A) Normal development of the 64-cell sea urchin embryo, showing the fates of the different layers. (B) An isolated animal hemisphere becomes a ciliated ball of ectodermal (more...)

In a normal embryo, the veg₂ cells become specified by the micromeres, and they, in turn, help specify the veg₁ layer. Without the veg₂ layer, the veg₁ cells are able to produce endoderm, but the endoderm is not specified as foregut, midgut, or hindgut. Thus, there appears to be a cascade wherein the vegetal pole micromeres induce the cells above them to become the veg₂ cells, and the veg₂ cells induce the cells above them to assume the veg₁ fates.Thus, the micromeres undergo autonomous specification to become skeletogenic mesenchyme, and these micromeres produce the initial signals that specify the other tiers of cells.

The identities of the signaling molecules involved in this process are just now becoming known (Sherwood and McClay 1997). The molecule responsible for specifying the micromeres (and their ability to induce the neighboring cells) appears to be β-catenin. As we saw in Chapter 6, β-catenin is a transcription factor that is often activated by the Wnt pathway, and several pieces of evidence suggest it for this role. First, during normal sea urchin development, β-catenin accumulates in the nuclei of those cells fated to become endoderm and mesoderm (Figure 8.15A). This accumulation is autonomous and can occur even if the micromere precursors are separated from the rest of the embryo. Second, this accumulation appears to be responsible for specifying the vegetal half of the embryo. Treating sea urchin embryos with lithium chloride causes the accumulation of β-catenin in all their cells, and this treatment transforms the presumptive ectoderm into endoderm. Conversely, experimental procedures that inhibit β-catenin accumulation in the vegetal cell nuclei prevent the formation of endoderm and mesoderm (Figure 8.15B,C; Logan et al. 1998; Wikramanayake et al. 1998). It is possible that the levels of nuclear β-catenin accumulation help to determine the mesodermal and endodermal fates of the vegetal cells. Third, β-catenin is essential for giving the micromeres their inductive ability.^* Experiments described above demonstrated that the micromeres were able to induce a second embryonic axis when transplanted to the animal hemisphere. However, micromeres from embryos in which β-catenin was prevented from entering the nucleus were unable to induce the animal hemisphere cells to form endoderm, and a second axis was not formed (Logan et al. 1998).

Figure 8.15

The role of β-catenin in specifying the vegetal cells of the sea urchin embryo. β-catenin is stained by a fluorescently labeled antibody. (A) During normal development, β-catenin accumulates predominantly in the micromeres and (more...)

It appears that the specification of cell fates in sea urchins involves a cascade initiated by the micromeres. It has been hypothesized (Davidson 1989) that the signal from the micromeres induces a post-translational modification in some factor in the veg₂ layer. Similarly, the signal from the veg₂ layer probably modifies some protein (possibly a transcription factor) in the veg₁ layer. In this way, the different tiers are assigned different fates.

Axis specification

In the sea urchin blastula, the cell fates line up along the animal-vegetal axis established in the egg cytoplasm prior to fertilization. The animal-vegetal axis also appears to structure the future anterior-posterior axis, with the vegetal region sequestering those maternal components necessary for posterior development (Boveri 1901; Maruyama et al. 1985).

In most sea urchins, the dorsal-ventral and left-right axes are specified after fertilization, but the manner of their specification is not well understood. Since the first cleavage plane can be either parallel, perpendicular, or oblique with respect to the eventual dorsal-ventral axis, it is probable that the dorsal-ventral axis is not specified until the 8-cell stage, when there are cell boundaries that correspond to these positions (Kominami 1983; Henry et al. 1992). Interestingly, in those sea urchins that bypass the larval stage to develop directly into juveniles, the dorsal-ventral axis is specified maternally in the egg cytoplasm (Henry and Raff 1990).

Function of primary mesenchyme cells

Shortly after the blastula hatches from its fertilization envelope, the vegetal side of the spherical blastula begins to thicken and flatten (Figure 8.17, 9 hours). At the center of this flat vegetal plate, a cluster of small cells begins to change. These cells begin extending and contracting long, thin (30 × 5 μm) processes called filopodia from their inner surfaces. The cells then dissociate from the epithelial monolayer and ingress into the blastocoel (Figure 8.17, 9–10 hours). These cells, derived from the micromeres, are called the primary mesenchyme. They will form the larval skeleton, so they are sometimes called the skeletogenic mesenchyme. At first the cells appear to move randomly along the inner blastocoel surface, actively making and breaking filopodial connections to the wall of the blastocoel. Eventually, however, they become localized within the prospective ventrolateral region of the blastocoel. Here they fuse into syncytial cables, which will form the axis of the calcium carbonate spicules of the larval skeleton (Figure 8.18).

Figure 8.18

Formation of syncytial cables by mesenchyme cells of the sea urchin. (A) Primary mesenchyme cells in the early gastrula align and fuse to lay down the matrix of the calcium carbonate spicule (arrows). (B) Scanning electron micrograph (SEM) of spicules (more...)

Importance of extracellular lamina inside the blastocoel

The ingression of the micromere descendants into the blastocoel is caused by these cells losing their affinity for their neighbors and for the hyaline membrane and acquiring a strong affinity for a group of proteins that line the blastocoel. This model was first proposed by Gustafson and Wolpert (1967) and was confirmed in 1985, when Rachel Fink and David McClay measured the strengths of sea urchin blastomere adhesion to the hyaline layer, to the basal lamina lining the blastocoel, and to other blastomeres. Originally, all the cells of the blastula are connected on their outer surface to the hyaline layer and on their inner surface to a basal lamina secreted by the cells. On their lateral sides, each cell has another cell for a neighbor. Fink and McClay found that the prospective ectoderm and endoderm cells (descendants of the mesomeres and macromeres, respectively) bind tightly to one another and to the hyaline layer, but adhere only loosely to the basal lamina (Table 8.2). The micromeres originally display a similar pattern of binding. However, the micromere pattern changes at gastrulation. Whereas the other cells retain their tight binding to the hyaline layer and to their neighbors, the primary mesenchyme precursors lose their affinity for these structures (which drops to about 2% of its original value) while their affinity for components of the basal lamina and extracellular matrix (such as fibronectin) increases a hundredfold. This change in affinity causes the micromeres to release their attachments to the external hyaline layer and their neighboring cells and, drawn in by the basal lamina, migrate up into the blastocoel (Figures 8.18, 8.19). These changes in affinity have been correlated with changes in cell surface molecules that occur during this time (Wessel and McClay 1985).

Table 8.2

Affinities of mesenchymal and nonmesenchymal cells to cellular and extracellular components^a.

Figure 8.19

Ingression of primary mesenchyme cells. (A-E) Interpretative diagrams depicting changes in adhesive interactions in the presumptive primary mesenchyme cells (color). These cells lose their affinities for hyaline and for their neighboring blastomeres while (more...)

As shown in Figure 8.18, there is a heavy concentration of extracellular material around the ingressing primary mesenchyme cells (Galileo and Morrill 1985; Cherr et al. 1992). Once inside the blastocoel, the primary mesenchyme cells appear to migrate along the extracellular matrix of the blastocoel wall, extending their filopodia in front of them (Galileo and Morrill 1985; Karp and Solursh 1985). Several proteins (including fibronectin and a particular sulfated glycoprotein) are necessary to initiate and maintain this migration (Wessel et al. 1984; Sugiyama 1972; Lane and Solursh 1991; Berg et al. 1996).

But these guidance cues cannot be sufficient, since the migrating cells “know” when to stop their movement and form spicules near the equator of the blastocoel. The primary mesenchyme cells arrange themselves in a ring at a specific position along the animal-vegetal axis. At two sites near the future ventral side of the larva, many of these primary mesenchyme cells cluster together, fuse with each other, and initiate spicule formation (Figure 8.18A; Hodor and Ettensohn 1998). If a labeled micromere from another embryo is injected into the blastocoel of a gastrulating sea urchin embryo, it migrates to the correct location and contributes to the formation of the embryonic spicules (Figure 8.20; Ettensohn 1990). It is thought that this positional information is provided by the prospective ectodermal cells and their basal laminae (von Übisch 1939; Harkey and Whiteley 1980). Only the primary mesenchyme cells (and not other cell types or latex beads) are capable of responding to these patterning cues (Ettensohn and McClay 1986). Miller and colleagues (1995) have reported the existence of extremely fine (0.3-μm diameter) filopodia on the skeleton-forming mesenchyme. These filopodia are not thought to function in locomotion; rather, they appear to explore and sense the blastocoel wall and may be responsible for picking up dorsal-ventral and animal-vegetal patterning cues from the ectoderm (Figure 8.20; Malinda et al. 1995).

Figure 8.20

Placement of the primary mesenchyme cells. (A) Nomarski videomicrograph showing a long, thin filopodium from a primary mesenchyme cell extending to the ectodermal wall of the gastrula, as well as a shorter filopodium extending inward from the ectoderm. (more...)