Lateral Cell Surface

From as early as 1900, it was documented that sea urchin embryos can be dissociated into single cells after a simple treatment with Ca2+-free seawater (Fig. 8A; Herbst 1900). Later, based on Herbst's observation, Giudice (1962) observed that dissociated cells were able to spontaneously reaggregate and differentiate into structures closely resembling normal larvae if Ca2+ and Mg2+ were restored in the seawater (Fig. 8B). The possibility of obtaining normal bipinnaria larvae from dissociated cells has been also demonstrated in the starfish Asterina pectinifera (Dan-Sohkawa et al. 1986).

Cell-Cell Adhesion and Communication: The Discovery of Toposome

The ease of dissociating and reaggregating cells of the sea urchin embryo offered an exceptional model for studies on molecules involved in cell-cell adhesion and for the development of methods for their isolation. In the late

+ BuOH extract 1

^ o o © o

1

• «<3

v © © ^

I

T

1

«ft

Fig. 8A-D. Schematic drawing illustrating dissociation and reaggregation experiments leading to identification of the active component mediating cell adhesion in the sea urchin embryo. BuOH n-Butanol; PM plasma membrane

1970s, it was first shown that Fab fragments (Fabs) from antibodies to plasma membranes purified from blastula embryos were able to prevent reaggregation of dissociated cells (Fig. 8C). The inhibition was reversed if soluble proteins extracted with n-butanol from purified membranes were added (Fig. 8C). In addition, these n-butanol extracts were able to strongly stimulate the rate of reaggregation of dissociated cells, suggesting the presence of some aggregating-factor(s) (Fig. 8B; Noll et al. 1979). In agreement with this hypothesis, it was observed that the exposure of dissociated cells to n-butanol completely removed the protein(s) responsible for reaggregation, since cells were not able to reaggregate even in the presence of Ca2+ and Mg2+ (Fig. 8D). However, the treatment with n-butanol did not affect the viability of the cells, since both reaggregation and embryonic development were completely restored by readdition of extracted proteins to the butanol-treated cells (Fig. 8D; Noll et al. 1979). As a consequence, an easy and low-cost procedure, i.e. the non-cytolytic treatment of live dissociated cells with diluted n-butanol, allowed the preparation of large quantities of crude extracts, from which the isolation, purification and characterisation of the active component could be attempted. In fact, the biochemical identity and biological activity of a large and oligomeric glycoprotein complex, called toposome, was later achieved (Noll et al. 1985; Matranga et al. 1986). Toposome is a 22S complex consisting of six 160-kDa subunits that are processed proteolytically as development proceeds. Cuts are revealed only after analysis by SDS-PAGE, since nicks introduced by specific enzymes, probably a cathepsin B-like protease (Yokota and Kato 1988), do not cause fragmentation of the native protein, thus securing the embryonic integrity. The need for toposome processing has been explained by postulating that a limited number of subunits could be generated by this strategy. Their differential association could then give rise to various molecular populations, each specifying a positional code guiding the cell in the embryo; from this comes the name toposome (Noll et al. 1985). Supporting this interesting hypothesis, toposome-specific monoclonal antibodies have been shown to stain cell surface structures in a pattern consistent with a positional code. The biological activity of the whole toposome complex, or parts of it, in mediating cell adhesion of dissociated cells has been tested and it was found that the oligomeric integrity of toposome is essential for its function (Matranga et al. 1986; Scaturro et al. 1998). The ultrastructural localisation of toposome has been investigated by electron microscopy of immunogold-labeled eggs and hatched blastulae (Gratwohl et al. 1991). Toposomes were seen on the surface of the egg, as well as stored in yolk granules and in the electron-dense lamellar compartment of the cortical granules. In the hatched blastula, toposomes modified by limited proteolysis in the yolk granules have been found associated with the plasma membranes, while unmodified 160-kDa toposomes, originating from the cortical granules, have been found on the outside of the hyaline layer. The latter distribution suggests that these toposomes function by attaching the apical lamina to the surface of the microvilli and thereby to the cytoskeleton of the growing embryo. Therefore, the authors proposed a different function for the two differently localised populations of toposomes (Gratwohl et al. 1991).

An important aspect of dissociation and reaggregation experiments is the effect on the transduction of signals from the exterior of the cell to the nucleus, ultimately involving DNA synthesis. An intriguing paradox is that in the sea urchin embryos, a "no-contact inhibition" was found. In fact, dissociated cells stop DNA synthesis until cell contacts are re-established, i.e. in the formation of the reaggregates (Sconzo et al. 1970; De Petrocellis and Vittorelli 1975). Later, we found that toposomes were responsible for the required signal transduction since reaggregation-inhibiting Fab restored DNA synthesis in dissociated cells (Vittorelli et al. 1980). Apparently, the binding of Fabs to the contact sites mimics cell-cell adhesion and thus stimulates DNA synthesis, in the same way that binding of Fabs to the receptors of epidermal growth factor or insulin mimics the action of those hormones (Kahn et al. 1978; Schreiber et al. 1981).

Further analysis on toposome molecules led to the characterisation of its precursor from sea urchin coelomic fluids of both male and female adults (Cervello and Matranga 1989). The authors produced, for the first time, evidence that the so-called vitellogenin (Vg), found in the coelomic fluid of both male and female sea urchin adults, and its intermediate form, the so-called major yolk protein (MYP) present in granules of unfertilised eggs, are both unprocessed precursor forms of toposome. Both proteins promote cell reag-gregation of dissociated blastula cells, suggesting that processing is not required for the cell-adhesion function, but rather directs their localisation during development (Cervello and Matranga 1989). Historical work on the 22S particle assigned to the protein a nutritional role due to its accidental occurrence in granules of the egg and to the need for the non-feeding embryo to develop soon. Similarly, the so-called vitellogenin, although found in both male and female sea urchins, has been recognised as the bona fide precursor to yolk protein, asking for genetic and functional analogies with the vertebrate homologue. It was then important to find the coding sequence for the protein. Previous efforts to gain decisive evidence of a yolk-related nutritional role by cloning the gene for the 22S glycoprotein particle from S. purpuratus failed because they resulted in the isolation of only a short cDNA and genomic DNA fragments (Shyu et al. 1986,1987).

Recently, full-length cDNAs from four different sea urchin species have been reported. They include the Hawaiian Tripneustus gratilla toposome mRNA complete coding sequence (cds), submitted back in 2001 and released in 2003 (accession no.AY026514) and the Mediterranean P.lividus cds submitted in 2003 and released in 2004 (accession no. AY274929). Strikingly, it has been discovered that the protein, also ironless, is a member of the transferrin family (H. Noll et al., pers. comm., work in prep.), in agreement with sequence data from reports on Pseudocentrotus depressus (Unuma et al. 2001), S. purpuratus (Brooks and Wessel 2002) and H. pulcherrimus (Yokota et al. 2003) cDNAs. Other ironless members of the transferrin family continue to be discovered, two of which are also membrane-associated (Morabito and Moczyd-lowski 1994; McNagny et al. 1996). Their functions, however, remain unknown.

Other Cell-Cell Adhesion Molecules

Among cell-cell adhesion molecules, cadherins have been fully characterised in vertebrate organisms (see review by Koch et al. 2004). These proteins are transmembrane glycoproteins that mediate homophilic calcium-dependent cell-cell adhesion in a number of cellular junctions, and their function has been shown to be critical both in normal development and in the development of the invasive and metastatic phenotype (see reviews by Wheelock and Johnsony 2003a; Hazan et al. 2004). Several pathways are activated by cad-herin-mediated cell-cell interactions and numerous studies are in progress to elucidate the complex relationships among them (Wheelock and Johnsony 2003b). Although recent work on sequence analysis has shed new light on the molecular basis of cadherin adhesion, understanding the specificity of these interactions remains a major challenge (Patel et al. 2003). The first immuno-logical evidence for the presence of a cell adhesion protein similar to the mouse E-cadherin in the sea urchin embryo has been shown in P. lividus (Ghersi and Vittorelli 1990; Ghersi et al. 1993). Furthermore, the use of poly-clonal antibodies raised against a cloned sea urchin cadherin, which recognises at least three major polypeptides, and the cloning of a novel sea-urchin-specific cadherin molecule support the hypothesis for the presence of several cadherins in this system (Miller and McClay 1997).

Cell Junctions

The presence of cell junctions in the sea urchin embryo has been described since the 1960s. The blastular wall has the structure of a simple epithelium, similar to that of vertebrates. At their apical surfaces, the cells are joined by typical junctional complexes, including zonulae adherens or belt desmo-somes, septate junctions and spot desmosomes, while hemidesmosome-like structures appear localised at their basal surface (Fig. 6; Wolpert and Mercer 1963; Spiegel and Howard 1983). Septate junctions have also been observed at the four-blastomere stage (Chang and Afzelius 1973), although they constitute a continuous layer only later in development (Gilula 1973).At least three types of desmosome, two types of septate junction and a tricellular junction have been described (Spiegel and Howard 1983). Studies on the osmotic and structural properties of the blastular wall date back to 1940, when a decrease in the permeability of the blastula wall to small molecules such as sucrose (Moore 1940) was observed, coinciding with the formation of junctional structures and with an apparent increase in the adhesion between cells.

0 0

Post a comment