ABSTRACT
At the 64-cell-stage embryos of Patella develop a prototroch consisting of four groups of four cilia-bearing cells. Ciliogenesis of isolated blastomeres and trochoblasts was studied, as well as the effect on it of cleavage arrest caused by cytochalasin B treatment. Isolation of blastomeres or trochoblast cells has no influence on ciliogenesis; neither has arrest of cleavage in whole embryos after the third cleavage. However, cleavage arrest before third cleavage completely prevents ciliogenesis. Thus, third cleavage is decisive for the expression of the developmental potential of the primary trochoblasts. Impairment of DNA synthesis by aphidicolin in the S-phase preceding third cleavage also prevents ciliogenesis. It is concluded that a determinant for ciliogenesis as well as certain nuclear factors must be segregated into the micromeres at third cleavage for ciliogenesis to occur in the prototroch cells.
INTRODUCTION
Ciliated prototroch cells of Molluscs appear early in the development. The early development of Patella has been described by Wilson (1904). The first quartet of micromeres gives rise to sixteen primary trochoblasts (Figs 1 and 2). These cells do not divide further and develop cilia within 1 h after their last cleavage. The primary trochoblasts derive from the micromere daughter cells la2-ld2, which each divide twice. In this way four interradial groups of four primary trochoblasts each are formed (Fig. 2). Cilia are also produced by descendants of the apical rosette cells la1–ld1 (Fig. 2), which form ciliated accessory prototroch cells and apical tuft cells. The second quartet of micromeres produces secondary trochoblasts.
Cell lineage of the first micromere quartet of Patella. Pr = prototroch cell. The end-cells in the lines terminating in→ go on dividing.
58-cell Patella embryo seen from the upper pole, after division of the rosette cells (la11-ld11) and the basal cells (la12-ld12). Primary trochoblasts stippled (after Wilson, 1904).
58-cell Patella embryo seen from the upper pole, after division of the rosette cells (la11-ld11) and the basal cells (la12-ld12). Primary trochoblasts stippled (after Wilson, 1904).
Wilson (1904) showed that isolated primary trochoblasts differentiate in exactly the same way as they would have done in the intact embryo. It is generally accepted that the capacity for self-differentiation of cells in early Molluscan development depends on localized determinants which are generated during oogenesis. Localized determinants seem to be basic to cell specification in many developing systems, but despite many efforts an exact understanding of their nature, distribution and functional significance is still lacking (Davidson, 1976; Freeman, 1979; Whittaker, 1979).
In this paper we show at which time in development the determinants for ciliogenesis in the primary trochoblasts are segregated and that their capacity to evoke ciliation depends on their allocation to the first quartet of micromeres.
MATERIALS AND METHODS
Patella vulgata collected at the coast of Normandy were kept in the laboratory in circulating sea water (16°C). To obtain gametes the animals are opened laterally along the shell (van den Biggelaar, 1977). Parts of the testis or ovary are removed and placed in a dish containing filtered sea water. The eggs and sperm then separate naturally into the water. Eggs exposed to sea water round off and lose their chorion. After washing the eggs twice with sea water they are fertilized by adding a few drops of a highly diluted sperm solution. Eggs are about 160 μm in diameter. The embryos were reared at 18 °C in a constant temperature room.
Synchronously cleaving eggs were obtained by selecting eggs which started first cleavage within lmin. Embryos were dissociated at the 2-, 4-, 8-, 16- or 32-cell stage by treatment with Ca++-free sea water starting before cleavage.
Cytochalasin B (CCB) solutions were made by diluting a standard solution of 10 μg CCB/ml and 1 % DMSO with sea water. Aphidicolin solutions used were dilutions with sea water of a standard solution of 50μg/ml plus 1 % DMSO in distilled water. For observation of nuclei whole mounts of Feulgen-stained embryos (van den Biggelaar, 1971) were made.
For scanning electron microscopy embryos were fixed for 30 min in 2·5 % glutaraldehyde in 0·1 M-cacodylate buffer (pH 7·3) and postfixed for 1 h in a 1 % OsO4 solution in the same buffer (temperature 20°C). After dehydration in acetone and cyclohexane the embryos were dried over P2O5. After mounting the eggs on aluminium stubs they were sputter-coated with gold and examined in a Cambridge 600M scanning electron microscope.
RESULTS
Development of isolated blastomeres
Cleavages in Patella embryos are synchronous up to the 32-cell stage. According to our measurements the four cleavage cycles took 27, 27, 31 and 41 min, respectively, at a constant temperature of 18 °C. Primary trochoblasts isolated after fifth cleavage went through their final divisions 75 min later. About 1 h after this final cleavage moving cilia could be observed.
Blastomeres were isolated by embryo dissociation as described under Materials and Methods. The isolation procedure had a retarding effect on cell cycle duration. The mean intercleavage time was 10–15 % longer than the intercleavage time of the respective cleavage cycles in intact embryos.
All blastomeres isolated at the 2- or 4-cell stage produced ciliated trochoblasts; so did all the micromeres isolated at the 8-cell stage. At fourth cleavage each micromere divides into a smaller l1 cell (apical rosette cell) and a larger I2 cell (primary trochoblast). Each primary trochoblast divides twice and forms four cilia-bearing cells (Fig. 2). Isolated primary trochoblasts divided and formed cilia as well (Fig. 3). The 11, so-called apical rosette cells go on cleaving beyond the sixth cleavage round (Fig. 1). Ciliogenesis-determining factors are also present in these apical rosette cells. A number of their descendants differentiate into ciliated accessory prototroch cells and into apical tuft cells. Moreover, the second quartet of micromeres produces ciliated secondary trochoblasts (Wilson, 1904). The secondary and accessory prototroch cells become ciliated several hours after the primary trochoblasts have developed cilia.
Four prototroch cells reared from a 12 trochoblast isolated after fourth cleavage. All four cells have produced cilia. Scanning electron micrograph. Magn. ×1000.
Effects of cytochalasin B on cleavage and ciliogenesis
We used the lowest concentration of cytochalasin B (0·1μg/ml sea water) which inhibited cleavage in Patella but had no visible effect on mitosis. Mitosis was not impaired in eggs continuously incubated from before first cleavage till after the normal time of sixth cleavage. Treated embryos and controls were compared with respect to number and synchrony of mitoses up to 2 h after sixth cleavage in the controls. The duration of the mitotic cycles in treated embryos was slightly shorter than in controls; the treated embryos were one mitosis ahead at the normal time of sixth cleavage, but we were unable to express this in terms of cell-cycle duration. The nuclei in the treated embryos were localized near the egg surface and all went through mitosis at the same time, thus failing to show the mitotic desynchronization that occurs in normal embryos.
Isolated trochoblasts (la2-ld2) and 16-cell embryos were incubated with cytochalasin B continuously from 10min after fourth cleavage. Divisions were no longer observed. In both groups cilia developed on the undivided I2 cells in synchrony with untreated controls.
Intact 8-cell embryos were incubated continuously from 10 min after third cleavage. Divisions were no longer observed; cilia developed on all four micromeres (Fig. 4). At the time of cilia formation the uncleaved micromeres contained the number of nuclei expected if mitotic activity was unimpaired. Continuous incubation with cytochalasin B of 1-cell, 2-cell and 4-cell embryos again prevented cleavage, mitosis being unimpaired. However, cilia never developed on these embryos (Table 1).
Cleavage inhibition at the 8-cell stage with cytochalasin B. Embryo shown 2 h after 64-cell stage in control embryos was reached. The four cleavage-inhibited micromeres produced cilia at approximately the same time as in the controls. Magn. ×500.
Figs 4–7. Scanning electron micrographs of cleavage inhibited embryos.
Cleavage inhibition at the 8-cell stage with cytochalasin B. Embryo shown 2 h after 64-cell stage in control embryos was reached. The four cleavage-inhibited micromeres produced cilia at approximately the same time as in the controls. Magn. ×500.
Figs 4–7. Scanning electron micrographs of cleavage inhibited embryos.
In order to study the exact stage from which cytochalasin B prevents ciliogenesis, we treated eggs just before third cleavage, at third cleavage, 5 min after the start of third cleavage, and 10 min after third cleavage. When treatment started before or at third cleavage, no cleavage furrow was formed or the furrow was retracted, respectively; no cilia developed. Treatment starting at 10 min after third cleavage inhibited further cleavage but had no effect on ciliogenesis; as expected ciliation was observed on all four micromeres (Fig. 4).
Treatment starting at 5 min after the onset of third cleavage gave variable results, even within one and the same embryo. Sometimes the cleavage furrow was retracted, sometimes it was not. Thus embryos were obtained consisting of 8, 7, 6, 5 or 4 cells with 4, 3, 2,1 or 0 micromeres, respectively. Figs 4,5,6 and 7 show such embryos. Cilia only formed on micromeres that were present as such. It is clear from these data that determinants for ciliation will only be expressed if they are segregated into micromeres at third cleavage. Thus third cleavage is decisive for the expression of the developmental potential of the primary trochoblasts. In this respect it is truly asymmetrical.
Third cleavage partially inhibited with cytochalasin B. Three micromeres were formed, which then produced cilia. Magn. ×500.
Third cleavage partially inhibited with cytochalasin B. Two micromeres were formed, which then produced cilia. Magn. ×500.
Third cleavage partially inhibited with cytochalasin B. Only one micromere was formed, which then produced cilia. Magn. ×500.
In another series of experiments the effect of cytochalasin B treatment during one cell cycle only was studied. Embryos were treated from 10 min after second cleavage to 10 min after third cleavage and then washed in sea water. The embryos resumed cleavage, but in not one case did the embryos form cilia at the time appropriate for primary trochoblasts, nor during the 3h to follow. Overnight the embryos formed an apical tuft and cells with short cilia typical of accessory and secondary trochoblasts. Apparently suppression of ciliogenesis is irreversible for primary trochoblasts, whereas in the other cells ciliogenesis is not suppressed. In conclusion it can be said that the asymmetry of third cleavage with respect to irreversible effects on ciliogenesis only pertains to the primary trochoblasts.
Effects of aphidicolin on ciliogenesis
Aphidicolin, a known inhibitor of DNA polymerase-α activity, was used to study the role of DNA replication in ciliogenesis. Aphidicolin concentrations of 1 and 0·5μg/ml sea water were tested in continuous incubation experiments starting at stages shortly before second, third, fourth, fifth and sixth cleavages. One or two cleavages occurred after the start of treatment. After some hours, the embryos disaggregated into single cells. These never developed cilia. Feul-gen staining of whole embryos showed fragmented nuclear material. These experiments show that aphidicolin, in addition to its effect on the nucleus, affects cleavage and cell adhesion.
Embryos continuously incubated in 0·1 μg aphidicolin/ml went on cleaving. If treatment started shortly before fifth or sixth cleavage no effect on development was observed; normal trochophores developed in 24h. A slight effect on cell adhesion was noted, however. This effect was much stronger if treatment began one cleavage cycle earlier; treatment starting shortly before the fourth cleavage caused the embryos to dissociate into single cells. Ciliation of the dissociated cells was not impaired, however.
Embryos treated from before third cleavage disaggregated and the cells showed a marked retardation of ciliation. With treatment starting before second cleavage the embryos again disaggregated but no ciliation of the cells was observed at all.
In conclusion it can be said that aphidicolin affects cell adhesion in early embryos of Patella and may inhibit ciliation, possibly by impairment of DNA synthesis following second cleavage.
Effects of actinomycin D
In order to study the possible role of nuclear activity in the ciliation of the prototroch cells we used treatment with actinomycin D, a known inhibitor of RNA synthesis. Actinomycin D at a concentration of 5 μg/ml suppressed 90 % or more of uridine incorporation into trochoblast RNA during cilia formation (to be published). This concentration did not inhibit ciliogenesis. Even a tenfold higher concentration had no effect whatsoever on cilia formation, even if the eggs were continuously treated from 1 h before first cleavage.
DISCUSSION
The differentiation of primary trochoblasts of Patella into prototroch cells comprises a limited number of S-phases and cleavages, and the formation of cilia.
The Patella egg provides for this differentiation process in two steps. The first step is at third cleavage, when the micromeres are separated from the macromeres. The second step is the first division of the micromeres, which separates the primary trochoblasts from the apical rosette cells.
With respect to the differentiation of trochoblasts into cilia-bearing cells, the experiments with cytochalasin B clearly show that the determination of ciliogenesis occurs between second and third cleavage. Third cleavage as such makes the determination definitive as it separates the nuclei, cytoplasm and cortex of the micromeres from those of the macromeres.
As shown by a strong increase in tubulin synthesis in the primary trochoblasts shortly before cilia formation (results to be published), factors promoting ciliation override other synthetic activities during the final step in the differentiation process.
One could think that the specialization of the primary trochoblasts is entirely regulated by the localization of cytoplasmic or cortical factors in the future prototroch cells. However, localization cannot be the only factor involved. Aphidicolin in very low concentrations inhibits cilia formation if applied at the time of second cleavage. Application after the S-phase following second cleavage no longer has an effect on ciliation. This strongly suggests the participation in ciliogenesis of nuclear activity at the 4-cell stage. We argue that during the S-phase in question the nuclei are being prepared for specific activities, so that upon segregation into the correct cytoplasmic environment, they will be able to support cellular activities leading to ciliogenesis. The latter activities are apparently limited in time, because, if the separation of the micromeres is postponed for the duration of one cell cycle, cilia no longer develop. Clearly a nuclear change is obligatory and as important as the localization phenomenon. After the nuclear change the segregation of certain localized factors into a micromere must occur within a definite time span.
Experimental data favouring this interpretation have been obtained by Gather (1973), who observed that in Ilyanassa polar lobe material inhibits ciliation of cells in the pretrochal region of the larva. Brachet, de Petrocelis & Alexandre (1981) found that ciliogenesis could be suppressed in Chaetopterus embryos by aphidicolin. The explanatory description by Freeman (1979) of the dynamics of the localization process in Cerebratulus and Mnemiopsis fits the observations in Patella. It cannot, however, explain the results of our experiments in which 1, 2 or 3 ciliated micromeres were formed, while blastomeres in which the third cleavage furrow retracted never produced cilia.
Cytochalasin-arrested Clona embryos develop tyrosinase activity in pairs of blastomeres if cleavages are inhibited after the 8-cell stage, but not at earlier stages (Whittaker, 1979). These results are clearly similar to the present findings in Patella.
ACKNOWLEDGEMENTS
We wish to thank the members of the staff of the Laboratoire Arago for their hospitality and for the opportunity to use their facilities, and particularly Dr J. Marthy for his inspiring presence. We thank Dr J. Faber for his critical comments. Thanks are due to Dr W. Berendsen for his help with the scanning electron microscopy.