Studies on the gastrulation of amphibian embryos : pseudopodia in the gastrula of Bufo bufo japonicus and their significance to gastrulation

When morphogenetic movements, such as gastrulation, occur in embryos, many factors, such as migration, adhesion, contraction and expansion of embryonic cells, are thought to play important roles. In the case of echinoderm gastrulation, cells of the invaginating archenteron protrude filopodia toward the blastocoelic wall and these filopodia play a main role in the invagination (Gustafson & Kinnander, 1956; Dan & Okazaki, 1956; Gustafson & Wolpert, 1967). During the epibolic movement in teleost embryos the deep cells form pseudopodia and migrate actively (Lenz & Trinkaus, 1967; Trinkaus & Lenz, 1967; Wourms, 1972; Trinkaus, 1973). Invaginating mesodermal cells of chick embryos, in the primitive streak and between the epiblast and hypoblast, are fibroblast-like and form filopodia (Trelstad, Hay & Revel, 1967; Granholm & Baker, 1970).

In the case of amphibian gastrulation, the cells lining the top of the invaginating archenteron resemble a bottle in shape. These cells have been termed ‘bottle cells’ or ‘flask cells’ and it was proposed that they play a main role in the invagination of the archenteron (Holtfreter, 1943a, b; Baker, 1965). Epibolic movement of the ectodermal layer was also thought to be involved in the gastrulation process (Holtfreter, 1944).

It has been shown that some cells dissociated from the gastrula form pseudopodia in vitro. It has been suggested that this pseudopodial activity may be a significant factor in the process of gastrulation (Holtfreter, 1943b; Sirakami, 1959, 1963).

It is not yet clear, however, whether cells in the intact embryo also form such pseudopodia and play a role in gastrulation. In this paper the existence of pseudopodia in the gastrula of Bufo bufo japonicus is shown and the significance of this for gastrulation is discussed.

Fertilized eggs of Bufo bufo japonicus (Bufo vulgaris) were obtained from the natural mating of adults collected at Iwakura, Kyoto. The stages of the embryos were determined according to Baba & Okada (1932). Embryos of stage 10–14 (late blastula to late gastrula) were dejellied with fine forceps and were fixed for 2 days at 3–5 °C in a buffered (0·05 M phosphate buffer, pH 7·2) 2·5 % glutaraldehyde solution containing paraformaldehyde (2 %) and picric acid (0·1 %) (Ito & Karnovsky, 1968). They were then rinsed in the same buffer that contained sucrose (4 %), and were post-fixed for 2 h in a buffered (pH 7·2–7·4) 1 % osmium tetroxide solution that contained sucrose (2·25 %). The samples were dehydrated in a graded series of ethanol and transferred through propyleneoxide, propylene-oxide Epon mixtures and into Epon. Finally the samples were embedded in Epon containers. To facilitate orientation and to allow better fixation, the embryos were cut bilaterally before the post-fixation step. One μm sections were mounted on glass slides that had been previously coated with 0·5% collodion (Aoki & Gutievvez, 1967) and were stained with 1 % toluidine blue (in 0·1 M phosphate buffer, pH 6·9) for 20 min at 90 °C. They were examined with a bright-field microscope.

Blastomeres are closely packed and polygonal in shape (Fig. 1). Pseudopodia were not observed in any region at this stage.

The first sign of invagination of the archenteron, accompanied by the appearance of bottle cells, is noticed in the area of the dorsal lip (Fig. 2A). In the marginal zone of the dorsal part, the cells of the presumptive pharyngeal endoderm and the mesoderm are the first to form pseudopodia (Fig. 2A). The pseudopodia are mostly lobopodia (Fig. 2B). Pseudopodia are also observed in the ventral part, though the bottle cells are not yet observed in the area of the presumptive ventral lip (Fig. 2C).

The archenteron is now formed, the top of which is lined by the bottle cells (Fig. 3A). Many pseudopodia are observed among the cells of the presumptive pharyngeal endoderm and mesoderm (Fig. 3B). The pseudopodia are conspicuous in the vicinity of the inner surface of the ectodermal layer (though mesodermal cells may be present in this layer at this stage) (Fig. 3B, C). The cells lining the inner surface of the ectodermal layer are tightly packed and the invaginating cells (presumptive pharyngeal endoderm and mesoderm) are more loosely packed.

The archenteron elongates parallel to the inner surface of the ectodermal layer (Fig. 4A). Invagination of the endodermal and mesodermal cells is progressing in the dorsal part (Fig. 4A) and in the ventral part as well (Fig. 4C). Some cells of the endodermal mass become loosely packed and large intercellular spaces are present (Fig. 4A) but formation of pseudopodia was not observed in this region. Besides the invaginating cells, some cells of the inner surface of the ectodermal layer are observed to form pseudopodia (Fig. 4B).

Formation and distribution of the pseudopodia described above are schematically summarized in Fig. 5.

The sinking-in of the bottle cells into the endodermal mass and epibolic movement of the ectoderm are processes that have been proposed as the major factors that cause gastrulation (Holtfreter, 1943b, 1944; Baker, 1965). The sinking-in activity of the isolated bottle cells was clearly shown by stepwise observation of their migration into an endodermal mass (Holtfreter, 1944). In the gastrula of Bufo bufo the bottle cells are present among the endodermal and mesodermal cells, and these cells move toward the animal pole during gastrulation.

Mesodermal cells seem to originate from the internal marginal zone of the blastula as in other anurans such as Xenopus laevis (Nieuwkoop & Florschütz, 1950). These cells become displaced toward the blastoporal lip, then turn the direction of their displacement toward the animal pole by rolling around the internally situated lip, and come to underlie the ectodermal layer. It is not clear how such a displacement of the mesodermal cells is regulated.

The cells of the presumptive pharyngeal endoderm and mesoderm move along the inner surface of the ectodermal layer toward the animal pole. These cells are loosely packed and observed to form pseudopodia coinciding with their invagination. The pseudopodia are conspicuous in the vicinity of the ectodermal layer. Some cells lining the inner surface of the ectodermal layer also form pseudopodia. The formation of the pseudopodia and the shape and arrangement of the cells seem to suggest that the invagination of the presumptive pharyngeal endoderm and mesoderm is brought about by the active migration of the individual cells along the inner surface of the ectodermal layer.

Movement of embryonic cells during gastrulation is dependent on the pseudopodia in other species such as in echinoderms and fishes (Gustafson & Kinnander, 1956; Dan & Okazaki, 1956; Gustafson & Wolpert, 1967; Trinkaus, 1973; Wourms, 1972). The migrating cells in these cases move along the inner surface of the cell layer that envelops the embryo. This is also suggested for Bufo bufo as described above. The pseudopodia observed in the gastrula of Bufo bufo are morphologically similar to those observed in teleost embryos (Lenz & Trinkaus, 1967; Trinkaus & Lenz, 1967; Trinkaus, 1973; Wourms, 1972) and could also be responsible for the movement of the cells.

Dissociated cells from the gastrula of amphibians were observed to form pseudopodia (Holtfreter, 1944, 1946, 1947; Sirakami, 1959, 1963). This seems to have a correlation with the pseudopodia observed in the gastrula of Bufo bufo.

By cell electrophoresis, an increase in the negative charge on the surface of cells from the dorsal lip region of Rana pipiens embryos, was shown to occur at the beginning of gastrulation (B. E. Schaeffer, Schaeffer & Brick, 1973). This change of the surface charge might be related to pseudopodial activity, because the increase of pH that increases the negative charge on the cell surface (H. E. Schaeffer, Schaeffer & Brick, 1973), activates the formation of the pseudopodia in the dissociated cells from the amphibian embryos (Holtfreter, 1943 b, 1944, 1946) and accelerates movement of the cells from the chick embryo in vitro (Taylor, 1962).

The whole process of gastrulation must be looked on as the integration of many factors. The sinking-in activity of the bottle cells and constriction of the fibrillar layer, observed in the apical end of the bottle cells (Baker, 1965; Perry & Waddington, 1966), seem to bring about the initial formation of the blastoporal groove. Spreading of the ectodermal layer then engulfs the endodermal mass. In anuran amphibians, at least, the invagination of the presumptive pharyngeal endoderm and mesoderm seems mainly due to the active migration of the individual cells along the inner surface of the ectodermal layer. It is possible that the active migration of these cells causes the elongation of the archenteron.

The author wishes to thank Professor K. I. Sirakami for his helpful discussions, Dr H. Mayahara for the continuing guidance of the techniques and Dr A. Hagiwara for his encouragement. He also thanks all of these and Dr P. Baur of the Texas University for critical reading of the manuscript. The author also wishes to thank Mr M. Matsui of the laboratory of animal phylogeny, who kindly provided the toads used in this study.