ABSTRACT
The hypoblast of early-primitive-streak-stage chick blastoderms was partially removed. This experiment provokes a reaction in the epiblast which curls up and becomes even at its ventral surface. The basal lamina underlying the epiblast is also dependent upon the presence of hypoblast. During culture after partial hypoblast removal, active hypoblast wound healing is observed. Where the hypoblast underlies the epiblast again, the effects of the removal disappear and normal development proceeds. The results suggest that the normal epiblast morphology is dependent upon the presence of hypoblast. This influence of hypoblast on epiblast is thought to be concerned with the morphology of the epiblast and not directly with its morphogenesis.
INTRODUCTION
Several authors have studied the inductive action of the early hypoblast on the epiblast in the chick blastoderm (Waddington, 1932, 1933; Spratt & Haas, 1960a, b; Eyal-Giladi & Wolk, 1970; Azar & Eyal-Giladi, 1979, 1981). These experiments were concerned with the study of the role of the hypoblast in the morphogenesis of the epiblast.
In the present investigation, we will not deal with these morphogenetic influences, but with the immediate dependency of the normal epiblast morphology on the presence of hypoblast. To this end, the hypoblast was partially removed and the effects of the intervention on the epiblast were studied by light and electron microscopy. We compared an area of epiblast deprived of hypoblast (1) with a symmetric area in the same embryo, and (2) with the homologous areas of blastoderms of the same stage. These precautions were thought to be necessary since cell and tissue architecture are extremely stage- and area-dependent. The mechanism of wound healing of the hypoblast will not be considered here (see Mareel & Vakaet, 1977). The mechanism by which the cell morphology of the epiblast is influenced by the presence of hypoblast, is currently under investigation.
MATERIALS AND METHODS
Fertilized eggs from a commercial stock were incubated for 12 h at 38 °C to yield blastoderms of stage 3 (Hamburger & Hamilton, 1951, or Vakaet, 1970). All the blastoderms were explanted according to the technique of New (1955). The hypoblast was partially removed as shown in Fig. 1. The blastoderms were either (1) immediately fixed, or (2) further cultured on egg white at 38 °C for 20 min, 100 min, 3 to 5 h, and then fixed. As a control, some blastoderms were fixed without experimental manipulation. A total of 95 blastoderms was studied. .
Most blastoderms were fixed ‘on ring’ (New, 1955). Some others were fixed after detachment from the vitelline membrane. Fixation and tissue preparation were carried out as described earlier (Vanroelen & Vakaet, 1981). As a fixative, a solution containing 1% (w/v) tannic acid (mol.wt 1701) and 1% (w/v) glutaraldehyde in 0 ·1 M cacodylate buffer at pH 7 ·4 was used. Postfixation was carried out in a 1 % (w/v) osmium tetroxide solution in the same buffer. After embedding in Epon 812, 2 μm sections were stained according to the technique of Kühn (1970). Ultrathin sections were stained in a 2 % (w/v) aqueous solution of uranyl acetate for 30 min, and in lead citrate for 30 s (Reynolds, 1963). The ultrathin sections were examined in a JEOL 100B electron microscope.
RESULTS
In the following description, the area deprived of hypoblast will be termed experimental area, while control area refers to the area symmetric to the experimental area.
Partial removal of hypoblast and subsequent culture initiates active wound healing (Figs. 2 A and B). Although dependent upon its dimensions, closure of the wound is, in our experimental conditions, complete approximately 3 –4 h after the intervention. Further normal development of the blastoderm then occurs.
In toto photograph of (A) a blastoderm immediately after the hypoblast was partially removed, and of (B) the same blastoderm after 100 min of culture. Note the closure of the wound after culture (arrows). Scale bar is 1 mm.
Comparison of the cell morphology of the epiblast between the control area and the homologous areas (left and right) of blastoderms of the same stage and without any prior intervention does not reveal differences in morphology.
Light microscope observations
Immediately after hypoblast removal, i.e. within the time necessary for fixation, the epiblast in the experimental area tends to curl up (Fig. 3). This curling up, though not very pronounced if compared to the control area, is very constant. If the margin of overgrowth partially loses contact with the vitelline membrane during the intervention, the curling up is more pronounced. Because random folds in the epiblast can then be observed, and because the hypoblast wounds in these conditions will not heal normally, only results obtained from blastoderms firmly attached to the vitelline membrane and fixed ‘on ring’ will further be shown. Whereas in the experimental area the ventral surface is even (Fig. 4A), in the control area the ventral surface is uneven (Fig. 4B). After staining according to the technique of Kühn (1970), a purple line is observed in the control area on the ventral surface of the epiblast, where previous studies have shown the presence of a basal lamina (Low, 1967; Sanders, 1979; Vakaet, Vanroelen & Andries, 1980).
Besides the small undulations in control area (arrowheads), curling up of the epiblast in the experimental area can always be observed (arrows). Scale bar is 500 μm.
Transverse section of the epiblast of a blastoderm fixed immediately after partial hypoblast removal (A) in the experimental area, and (B) in the control area of the same germ. Note the even ventral surface in the experimental area (arrows) and the uneven ventral surface in the control area (arrowheads). Scale bar is 50 μm.
Transverse section of the epiblast of a blastoderm fixed immediately after partial hypoblast removal (A) in the experimental area, and (B) in the control area of the same germ. Note the even ventral surface in the experimental area (arrows) and the uneven ventral surface in the control area (arrowheads). Scale bar is 50 μm.
This positivity is still observed in the experimental area of blastoderms fixed immediately after the intervention (Fig. 4 A), but is less obvious after 100 min of culture (Fig. 5). During culture, the hypoblast wound closes, and epiblast that was deprived of hypoblast, gradually becomes underlain with hypoblast again. In these areas where hypoblast underlies the epiblast again, the morphology of the epiblast cannot be distinguished from the morphology of the epiblast in the control area (Fig. 5). Indeed, curling up of the epiblast and evenness of its ventral surface can no longer be observed. The purple line at the site of the basal lamina in the healed area is again present.
Transverse section of a blastoderm after partial hypoblast removal and 100 min of culture. The dark line on the ventral surface of the epiblast is absent in the experimental area (arrows) and this line is present in the healed area arrowheads Scale bar is 50 μm.
Electron microscope observations
After partial hypoblast removal and within the time necessary for fixation, a meshwork of microfilaments becomes visible close to the ventral inner surface of most epiblast cells in the experimental area (Fig. 6). Moreover, the ventral surface of some cells is irregular to a degree never observed in the control area (Fig. 7). In the latter area, the mesh work of microfilaments cannot be seen (Fig. 8). Sometimes sparse microfilament bundles can be observed at the basis of blebs or other cell extensions.
Fig. 6–8. Electron micrographs of transverse sections of the ventral surface of the epiblast of blastoderms fixed immediately after partial hypoblast removal. The basal lamina remains present during the intervention. Scale bar is 1 μm.
Fig. 6. The experimental area. A meshwork of microfilaments is present at the inner ventral surface of the epiblast cells (arrows).
Fig. 6–8. Electron micrographs of transverse sections of the ventral surface of the epiblast of blastoderms fixed immediately after partial hypoblast removal. The basal lamina remains present during the intervention. Scale bar is 1 μm.
Fig. 6. The experimental area. A meshwork of microfilaments is present at the inner ventral surface of the epiblast cells (arrows).
Fig. 6–8. Electron micrographs of transverse sections of the ventral surface of the epiblast of blastoderms fixed immediately after partial hypoblast removal. The basal lamina remains present during the intervention. Scale bar is 1 μm.
Fig. 7. The experimental area. The ventral plasma membrane of some epiblast cells is irregular.
Fig. 6–8. Electron micrographs of transverse sections of the ventral surface of the epiblast of blastoderms fixed immediately after partial hypoblast removal. The basal lamina remains present during the intervention. Scale bar is 1 μm.
Fig. 7. The experimental area. The ventral plasma membrane of some epiblast cells is irregular.
Fig. 6–8. Electron micrographs of transverse sections of the ventral surface of the epiblast of blastoderms fixed immediately after partial hypoblast removal. The basal lamina remains present during the intervention. Scale bar is 1 μm.
Fig. 8. The control area.
After culture, some disorganization at the level of the basal lamina (as suggested from the light microscope study) can be observed in the experimental area, but these changes are less obvious than in the light micrographs.
DISCUSSION
When the effects of the removal of hypoblast on the cell shape of the epiblast are studied, comparison with a reliable control area is necessary. It is a minimal prerequisite that the experimental and the control area belong to the same germ, because changes in cell shape occur during development. Comparison of the effects of the intervention between the experimental and the symmetric control area is, however, only valid if we assume (1) that during this period of development, the blastoderm is completely symmetric, and (2) that the removal of hypoblast on one side of the blastoderm does not alter the control area. Comparison of our control area with the two (left and right) homologous areas of blastoderms fixed without experimental manipulation does not show differences.
Our experiments revealed an influence of hypoblast removal on the overlying epiblast both at the light and electron microscope level. Curling up of the epiblast was also observed by Kohonen & Paranko (1980) after excision of the ectoderm of the gastrula of the newt. These observations may reflect the presence of differences in tension in the epiblast when the experimental area and the control area are compared. This idea is corroborated by the observation that the curling up is more obvious in blastoderms fixed after removal from the vitelline membrane, than in blastoderms fixed ‘on ring’ (New, 1955). Indeed, in blastoderms fixed after detachment from the vitelline membrane, the radial tension exerted by the margin of overgrowth is lost. In this respect, the visualization of a mesh work of microfilaments (see also Spooner & Wessels, 1970) may be correlated with the curling up and with the evenness of the ventral surface, if a contractile activity of microfilaments may be assumed. The reversibility of the effects at the healed edge of the wound and the normal further development after closure suggests that the normal cellular morphology depends upon the presence of hypoblast.
The effects observed in the basal lamina remain unclear. It is without doubt that the basal lamina also undergoes changes after hypoblast removal. These changes may reflect that a role is played by the basal lamina as an intermediate, or that the basal lamina changes are an accompanying event beside the described cell morphological changes. The observation that the basal lamina changes were more obvious in the light microscope after staining according to Kühn (1970), than in the electron microscope after routine staining, suggests that different products were visualized. In this respect, the hypoblast removal showed the disorganization of these basal lamina products stained with the technique of Kühn.
Concluding, besides the morphogenetic influences of the hypoblast on the young epiblast (see Introduction), which are assumed to be irreversible in their effect, the present study has demonstrated the existence of hypoblast influences on epiblast cell morphology. These were shown to be reversible and are suggested to differ from the morphogenetic influences.
ACKNOWLEDGEMENTS
The authors would like to thank Mrs G. Haest-Van Neuten and Mr Ch. De Schepper for technical assistance, Mr J. Van Ermengem for photographic processing of the micrographs, Mr F. De Bruyn for artwork and Mrs N. Van den hende-Bol for typing the manuscript.
This work was supported by grant no. 3.9001.81 of the Belgian Fonds voor Geneeskundig Wetenschappelijk Onderzoek.