THE JOURNAL OF EXPERIMENTAL ZOOLOGY 263:414-422 (1992)
Epibolic Extension of the Presumptive Ectodermal Layer of Embryos of the Newt Cynops pyrrhogaster Before and During Gastrulation SHINJI KOMAZAKI Department of Anatomy, Saitama Medical School, Saitama, 350-04 Japan ABSTRACT Epibolic extension of the presumptive ectodermal layer (PEL) was investigated in embryos of the newt Cynops pyrrhogaster before and during gastrulation. The PEL was composed of only one layer of columnar cells a t all stages examined. The cells of the PEL became elongated from the blastula to the early gastrula stage. They were most elongated at the early gastrula stage and then shortened during gastrulation. Present observations suggest that changes in cell shape of the PEL play an important role in the control of the epibolic extension of the newt embryos. The morphology and movement of the isolated cells from the PEL were examined in an attempt to elucidate the role of cell movement in epibolic extension of the PEL. Blebbing and vermiform cells which showed active cell movement appeared at the early blastula stage. The blebbing cells, which formed large hyaline blebs that moved around the circumference of each cell, appeared in large numbers at the early blastula stage. The frequency of the blebbing cells decreased from the early blastula to the early gastrula stage and increased again during gastrulation. The vermiform cells, which had an elongated cell body and moved in a worm-like manner, increased in frequency from the early blastula to the early gastrula stage. The relative number of such vermiform cells was maximal at the early gastrula stage and decreased abruptly during gastrulation. These results suggest that the elongation of the cells of the PEL is controlled by the active cell movement which resembles that of a worm. 0 1992 Wiley-Liss, Inc.
Amphibian gastrulation is accomplished principally by three types of morphogenetic movement: epiboly of the presumptive ectodermal layer (PEL), involution of the marginal zone, and migration of the presumptive mesodermal cells (Keller, '86). In such morphogenetic movements, cell-to-celland cellto-extracellular-matrixinteractions play important roles in the control of the movement of individual cells. Mechanisms of cell-to-cell and cell-to-extracellular-matrix interactions have been investigated in terms of specific molecules, such as cell adhesion molecules (Boucaut et al., '85; Choi and Gumbiner, '89; Johnson et al., '90) and a cell-surfacereceptor for the adhesion molecule (Darribere et al., '88, '91). However, even though an important role of cell movement is very probable during morphogenetic movement, the mechanisms by which cell movements control morphogenetic movements are not well understood. In the present study, the roles of cell movement in the epibolic extension of the PEL before and during gastrulation were investigated. The epibolic extension of the PEL has been examined in detail in embryos of Xenopus laevis (Keller, '80).The PEL of Xenopus is composed of several layers of cells during early development. The 01992 WILEY-LISS, INC.
epibolic extension of the PEL is accomplished via a decrease in the number of layers of cells and a flattening of the cells in the superficial layer. This series of events is widely applicable in the case of epibolic extension of a PEL that is composed of several layers of cells. However, such a phenomenon is not applicable in the case of the epibolic extension of a PEL that is composed of only a single layer of cells, as in embryos of the newt Cynops pyrrhoguster. Therefore, in the present study, the mechanism of epibolic extension was investigated in the PEL of Cynops embryos.
MATERIALS AND METHODS Embryos Fertilized eggs of the Japanese newt Cynops pyrrhogaster were obtained after artificial inducement of spawning by human chorionic gonadotropin. The eggs were collected and kept at 20 or 25°C until embryos reached appropriate stages. Embryonic stages were according t o Okada and Ichikawa ('47). Embryos were collected from a single batch Received June 5,1991;revision accepted February 17,1992.
EPIBOLY OF THE NEWT EMBRYO
415
for measurements of diameter after removal of jelly capsules.
Light and scanning electron microscopy The jelly capsule of the embryos was removed manually with forceps. Penetration of fixative solution into the blastocoel was facilitated by puncturing the lateral side of the blastocoel roof with sharp tungsten needles in 2.5% glutaraldehyde, 3%paraformaldehyde in 0.1 M cacodylate buffer, pH 7.0. The embryos were kept in fixative for 16-18 h at room temperature (24-27'0. They were washed briefly in the same buffer solution and postfixed in 1%Os04 in 0.1 M cacodylate buffer, pH 7.0, for 6-8 h a t room temperature. After washing with the same buffer solution, they were dehydrated in a graded series of ethanol and acetone and embedded in epoxy resin. Semithin sections (1-5 pm thick) were cut and stained with 0.1% toluidine blue. For scanning electron microscopy, the same fixation and dehydration methods were used and then specimens were critical point dried in a liquid COz,coated with platinum, and examined in a Hitachi S-550 scanning electron microscope. The thickness of the PEL was measured under the light microscope in the epoxy-resinsections. The thickness of the PEL and ratio of height to width of the cells in the PEL at animal-pole region were measured. The area of apical surfaces of the cells was measured using a TV image processor (Excel-11,TVIP4100; Nippon Avionics Co.). The scanning electron microscope photographs at a magnification of 1,000x were used for measurement. Cell culture The cells of the PEL were collected from animalpole regions of morulae, blastulae, and gastrulae. The cells of the presumptive ectoderm (the same as the PEL), mesoderm, and endoderm of embryos at stage 12a (early gastrula stage) were collected from the regions shown in Figure 1.The cell masses were removed from these regions with tungsten needles and disaggregated in dissociation medium (60 mM NaC1,0.7 mM KC1,0.2%EDTA, buffered with 3 mM HEPES at pH 8.0) for 10-20 min at room temperature. The dissociated cells were transferred to culture dishes that had been filled with 5 ml of culture medium (60 mM NaC1, 0.7 mM KC1, 0.8 mM MgS04, 0.3 mM Ca(NO&, buffered with 3 mM HEPES at pH 7.4). Three different substratapolystyrene culture dishes (LUX, No. 5221; Miles, IL, USA), coverglasses (Matsunami Glass, Osaka, Japan), and 3%agar-were used. Dissociated cells
Fig. 1. Schematic representation of a mid-sagittal section of an embryo at stage 12a showing the three regions from which cells were isolated. PE, presumptive ectoderm (the same as with the PEL); PEn, presumptive endoderm; PM, presumptive mesoderm.
were distributed uniformly on the various substrata and incubated in the culture medium for 1h at room temperature. The cells were observed under an inverted phase-contrast microscope after incubation. The numbers of cells of various shapes as percentages of the total number of cells were calculated in each case. Cells were treated with 5 pg/ml cytochalasin B (Aldrich, Milwaukee, WI, USA), 10 mM colchicine (Wako Pure Chem., Osaka, Japan), 100 pM H-7 (Seikagaku Kogyo, Tokyo, Japan), and 0.2 mg/ml bovine serum albumin (Cappel, PA, USA) by incubation in culture media supplemented with these agents. Effects of divalent cations were examined in Ca2+-and Mg2+-freeculture medium.
RESULTS Scanning electron microscopy The PEL of morulae (stages 7 and 8) was composed of a single layer of columnar cells (Fig. 2a). The ratio of height to width of the cells of the PEL at stage 8 was 2.1 ? 0.4 (mean SD for 60 cells from 6 embryos). The cells of the PEL were separated by wide intercellular spaces and had many filamentous processes on their basal surfaces which face the blastocoel (Fig. 2b). The PEL of blastulae (stages 9 and 10) and of initial and early gastrulae (stages 11and 12a)were also composed of a single layer of cells. The cells of the PEL were elongated and columnar in shape at these stages and were especially elongated at stages 11 and 12a (Fig. 2c). The ratio of height to width of the cells of the PEL at stage 12a was 4.4 -+ 0.6 (mean ? SD for 60 cells from 6 embryos). The area of apical surface of individual cells at stage 12a was 484.1 k 57.3 pm2 (mean SD for 563 cells from 7
*
*
416
S.KOMAZAKI
Fig. 2. Scanning electron micrographs showing the changes in shape of cells in the PEL. a: Cells at stage 8. b: Basal surface of the PEL at stage 8. Many filamentous processes are seen at margins of the cells. c: Cells of the PEL at stage 12a. d: High magnification of the intercellular spaces between the cells of the PEL at stage 12a. e: Basal surface of cells at stage 12a. Large bulbous processes are visible. fi The archenteric roof at
stage 13. The presumptive ectoderm (upper layer) and endoderm (lower layer). g: The basal surface of the presumptive ectodermal cells at stage 13. The surface was exposed after removal of the presumptive endodermal cells. Many filamentous and lamellar processes are visible. Bars indicate 100 km (a),50 pm (b,e,D,and 10 km (c,d,g).
EPIBOLY OF THE NEWT EMBRYO
417
embryos). The cells of the PEL at these stages were t IJm) separated by wide intercellular spaces and had large bulbous processes on their basal surfaces (Fig. 2d,e). The large bulbous processes were also prominent on the cells of the embryos at stages 10 and 11. The PEL of middle gastrulae (stage 13) was also 200composed of one layer of columnar cells (Fig. 20. The ratio of height to width of the cells of the PEL at stage 13b was 2.3 & 0.3 (mean & SD for 72 cells from 6 embryos). The area of apical surface of indi\T vidual cells at stage 13bwas449.6 2 52.2 pm2(mean -t SD for 563 cells from 7 embryos). The invagination of the archenteron was almost complete at this loo-stage. The cells of the PEL were in close contact with one another and there were many filamentous and lamellar processes on their basal surfaces (Fig. 2g). Changes in the thickness of the PEL and in the diameter of the embryos before and during gastrulation are shown in Figure 3. The thickness of the PEL at stage 8 was reduced by 50% during stages 0 1 : : : : : : : : I 9 10 11 12 13 14 15 16 11-12 and by 75%at stage 13. By contrast, the diam9 10 11 12 13 8 eter of embryos increased less (by less than 1.3% from stages 8 t o 12a and by 5%from stages 8 to 13) Fig. 3. Changes of the thickness of the PEL and the diamduring these stages. eters of embryos before and during gastrulation. Horizontal axis:
TT
!\
i
\f/
p \Q
Behavior ofisolated cells The isolated cells of the PEL could be classified into several classes after 1h in culture. From their morphology and the nature of their movements, the cells of the PEL can be clearly divided into four classes as follows (Fig. 4): 1)blebbing cells, with hyaline blebs that extend around each cell's periph-
Fig. 4. Cultured cells isolated from embryos at stage 12a. Four types of cell are distinguishable after 1h of culture on a coverglass. a: A sphere cell, which is spherical in shape. b: A blebbing cell with a hyaline bulbous process that moves around
the upper numbers indicate the timing of cleavages and the lower numbers indicate the stages of embryos. The relationship between the timing of cleavages and the stages of embryos was estimated from the results of Suzuki et al. ('76). Vertical axis: the numbers indicate the thickness of the PEL (left) and the diameter of embryos (right). Eleven embryos collected from the same batch were used for measurements of the diameters of embryos. Changes in the thickness of the PEL were calculated from the results from 7,15,16,15,30,and 12 embryos at stages 8,9,10, 11,12, and 13, respectively.
its circumference. c: A vermiform cell. d A spreading cell that adheres to the substratum via filamentous and lamellar processes. Bars indicate 50 Fm.
S.KOMAZAKI
418
TABLE 1 . Effects of substrata on relative numbers, as percentage, of various types of cell' Substratum Coverglass Normal culture medium
Ca2+-and Mg2+-free culture medium
Culture medium containing 0.2 mg/ml albumin Polystyrene dish
Agar
Cell type Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading SDhere
Percentages of various types of cell2
No. of embryos examined
65.7 f 7.4 23.9 i 9.5 3.8 f 2.1 6.5 f 4.1 58.1 f 9.4 29.7 i 9.7 3.0 3.5 9.2 ? 7.1
12
12
*
0.4 f 0.5 96.0 f 3.5 0.3 ? 0.3 3.4 t 3.3 0.2 k 0.2 90.3 f 5.3 6.4 t 5.4 3.1 f 1.6 0.1 f 0.1 91.2 ? 3.6 0.0 f 0.0 8.7 3.7
16
14
11
*
'The cells isolated from the PEL of early gastrulae (stage 12a) were used. 2Valuesare means ? SD.
ery; 2) vermiform cells, which are elongated and show active movement, namely expansion, contraction, and bending of the cell body; 3) spreading cells, which adhere to the substratum via filamentous and lamellar processes; and 4) sphere cells, which are spherical in shape and remain on the substratum. The blebbing and the vermiform cells moved actively, whereas the spreading and the sphere cells remained still. The effects of various substrata on relative frequencies of these four types of cell were investigated using cells isolated from the PEL of embryos at stage 12a (Table 1).Almost all of the isolated cells were blebbing just after the isolation. The vermiform and the spreading cells appeared within 30 min of incubation and the frequency of these cells gradually increased for a further 30 min. The vermiform cells appeared at high frequency only on the coverglasses among the substrata examined. Omission of Ca2+ and Mg2+ ions from the culture medium had little inhibitory effect on the formation of the vermiform cells. Addition of bovine serum albumin blocked the cells from sticking to the glass substratum and completely suppressed the formation of the vermiform cells, resulting in an increased number of blebbing cells. The cells stuck to the polystyrene substratum but very few cells changed to vermiform cells. The cells did not stick to the agar substratum and very few cells changed to vermiform cells. The relative numbers of various types of cell in
the different embryonic regions were examined in cells isolated from embryos at stage 12a (Table 2). Cells were isolated from three different regions, the presumptive ectoderm (the same as the PEL), the mesoderm, and the endoderm, and they were cultured on coverglasses. The vermiform and blebbing cells appeared among cells from the presumptive ectodermal region at higher frequency than among cells from the other two regions. By contrast, sphere cells appeared at higher frequency among cells from both the presumptive mesodermal and endodermal regions than among cells from the presumptive ectodermal region. The relative numbers of spreading cells were low among cells from all three regions. Changes in the relative numbers of variously shaped cells before and during gastrulation were examined using cells from the PEL (Figs. 5, 6). Almost all of the cells were sphere cells at stages 7 and 8. The blebbing and the vermiform cells appeared at stage 9. The relative numbers of vermiform cells increased gradually duringstages 9- 12a and reached a maximum at stage 12a. The relative numbers of blebbing cells decreased during stages 9-12a. The relative numbers of vermiform cells decreased abruptly at stage 13, while those of the blebbing cells increased again at stage 13. The relative numbers of spreading cells were low during stages 7-12a but increased markedly a t stage 13. The effects of several drugs, known t o influence
EPIBOLY OF THE NEWT EMBRYO
419
TABLE 2. Relative numbers, as percentages, of various types of cell in three differentregions of early gastrulae' Embryonic region Presumptive ectoderm
Presumptive mesoderm
Presumptive endoderm
Cell type Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere
Percentages of various types ofcel12
*
65.7 7.4 23.9 L 9.5 3.8 2 2.1 6.5 4.1 43.0 2 9.5 8.2 L 6.2 1.6 L 0.9 47.1 IT 11.7 30.4 2 8.4 6.6 f 4.5 1.7 L 2.8 61.4 L 8.5
No. of embryos examined
12
*
15
16
'Coverglass substratum and the cells isolated from the PEL of early gastrulae (stage 12a) were used. 'Values are means ? SD.
the shape and movement of cells, and of bovine serum albumin on cultured cells from the PEL were examined (Table 3). Cytochalasin B completely inhibited the appearance of the vermiform and the blebbing cells. Almost all of the cells were sphere cells in the presence of this drug. Colchicine had little inhibitory effect on the appearance of the vermiform and the blebbing cells. H-7, an inhibitor of cyclic nucleotide-dependent protein kinase C, had a remarkable inhibitory effect on the appearance of the vermiform and the blebbing cells. The relative numbers of spreading cells and sphere cells increased with a decrease in the relative numbers of vermiform and blebbing cells.
DISCUSSION In Xenopus embryos, division of cells, a decrease in the number of layers of cells, and flattening of the cells are the main morphological events related t o the epibolic extension of the PEL (Keller, '80). Cell division is also important in the epibolic extension of the PEL of Cynops embryos. However, the details of the epibolic extension of the PEL differ between Xenopus and Cynops embryos. Changes in the shapes of the cells of the PEL play an important role in the control of the epibolic extension of Cynops embryos. As shown in the present study, the PEL of Cynops embryos decreases in thickness without any close correlation with changes in diameter of embryos and the timing of cleavages. The PEL did not decrease much in thickness during stages 8-12a and then its thickness decreased abruptly during gastrulation. Elongation and shortening of the cells would modify the decrease in thickness of the PEL. Elongation of the cells before gastrulation will suppress the decrease in thickness of the PEL. Conversely, the shortening of the cells
during gastrulation will accelerate the decrease in the thickness of the PEL. As for the epibolic extension of the PEL, elongation of the cells will decrease the area of apical surface of the cells and consequently suppress the epibolic extension of the PEL before gastrulation. On the other hand, shortening of the cells will increase the area of apical surface of the cells and consequently accelerate the epibolic extension of the PEL during gastrulation. In fact, though the number of the cells of the PEL nearly doubled between stage 12a and 13b (Suzuki et al., ,761,the area of apical surface of the cells decreased only 7.1% during these stages. These facts confirm the conclusion that shortening of the cells during gastrulation accelerates the epibolic extension of the PEL. Such mechanisms for controlling the epibolic extension of the PEL have not been demonstrated in Xenopus embryos. It seems that the differences in the mechanisms of epibolic extension of the PEL between Cynops and Xenopus embryos are mostly the result of differences in the structure of the PEL. However, it is unclear whether or not the epibolic extension of the PEL of both types of embryo is controlled by intrinsically the same mechanisms. Morphological studies of Cynops embryos showed that changes in cell shape play an important role in control of the epibolic extension of the PEL. However, observations by scanning electron microscopy alone can never define whether or not the changes in cell shape are caused by the active movement of the cells themselves. Therefore,changes in the movement of isolated cells were examined in relation to changes in cell shape before and during gastrulation. Isolated cells showed two types of cell movement, blebbing and venniform movement. These two types of cell movement are common in isolated cells
420
S. KOMAZAKI
Fig. 5. Changes in types of cell before and during gastrulation. Isolated cells were cultured for 1 h on a coverglass. a: Cells from the PEL at stage 8. Almost all ofthe cells are sphere cells. b: Cells from the PEL at stage 9. Blebbing cells are abundant. c: Cells from the PEL at stage 12a. Most of the cells are
vermiform cells. d: Presumptive mesodermal cells at stage 12a. e: Presumptive endodermal cells at stage 12a. fi Cells from the PEL at stage 13. Spreading cells are abundant. Bars indicate 200 pm (a)and 100 pm (b-0.
EPIBOLY OF THE NEWT EMBRYO
421
of early amphibian embryos (Holtfreter, '46;Satoh et al., '76). The blebbing cells appeared abruptly at stage 9. The appearance of the blebbing cells is dependent on the timing of cleavages and on structural changes in the submembranous cortical layer, a thick layer of cytoskeleton just beneath the cell membrane. The blebbing cells appear after the 12th cleavage in Xenopus (Newport and Kirschner, '82) and after the 11th cleavage in Cynops (Komazaki, '90). In Cynops embryos, the structure of the submembranous cortical layer changes after the 11th cleavage, and artificial changes in the struc5( ture of this layer induce the formation of blebs (Komazaki,'91).The frequency of the blebbing cells gradually decreased from stages 10 to 12a and increased again at stage 13. By contrast, the frequency of vermiform cells changed in inverse proportion to that of the blebbing cells. It seems that the blebbing cells change into the vermiform cells during the course of development since both forms are interconvertible in culture (Holtfreter, '46).The present observations showed a close correlation between the elongation of the cells of the PEL and the appearance of the venniform cells during develFig. 6. Changes in frequencies of various types of isolated opment course,implying that the elongationof these embryonic cell before and during gastrulation. Cells were cul- cells is caused by active cell movement, which resemtured for l h on coverglasses. Horizontal axis: the numbers indicate the stages of the embryos. The vertical axis shows the bles that of a worm. The formation of large hyaline percentages of various cells. Relative numbers, as percentages, blebs in the cells of the PEL at stages 10 and 11 of vermiform cells (solid triangles), blebbing cells (open circles), may be indicative of the active movement of spreading cells (open triangles), and sphere cells (solidcircles) are these cells. shown. The results are from 11,13,16,15,12,11,and 11embryos Among the substrata tested, only coverglasses at stages 7,8, 9, 10, 11,12, and 13, respectively. The numbers of induced the formation of the vermiform cells at high cellsexaminedwere 20-40,50-100,150-250,200-300, and300400 from embryos at stages 7,8,9,10, and 11-13, respectively. frequency. Contact of cells with a coverglass sub-
1O(
TABLE 3. Effects of inhibitors on relative numbers, as percentage, of various types of cell'
Inhibitor Control
Cytochalasin B (5 pglml)
Colchicine (10 mM)
H-7 (100 FM)
Cell type
Percentages of various types of cell2
No. of embryos examined
Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere
65.7 + 7.4 23.9 k 9.5 3.8 2 2.7 6.5 f 4.1 0.2 f 0.3 0.3 -t 0.3 1.1 f 1.4 98.3 ? 1.4 55.1 f 11.4 14.6 + 7.0 3.7 5 2.3 26.7 f 12.9
12
Vermiform Blebbing Spreading Sphere
1.2 t 2.2 0.2 ? 0.4 39.5 ? 13.1 59.1 f 12.8
13
'Coverglass substratum and the cells isolated from the PEL of early gastrulae (stage 12a)were used. 2Values are means t SD.
13
12
S.KOMAZAKI
422
stratum may be important for the formation of the vermiform cells since disturbance of the cell-tosubstratum contacts by addition of bovine serum albumin to the culture medium completely suppressed the formation of the vermiform cells. Divalent cations may not be significantly involved in the cell-to-substratum contacts during the formation of the vermiform cells since omission of divalent cations from the culture medium had little effect on the formation of the vermiform cells. Why can only a coverglass substratum induce the formation of the vermiform cells? The answer to this question remains to be determined. However, from the present study, it is clear that actin filaments and protein kinase C play important roles in the formation of the vermiform and the blebbing cells because cytochalasin B, an inhibitor of the polymerization of actin filaments, and H-7, a potent inhibitor of cyclic nucleotide-dependent protein kinase C (Hidaka et al., '84),completely suppressed the formation of the vermiform and the blebbing cells.
ACKNOWLEDGMENTS The author thanks Professor Reiji Hirakow for his interest in this study. This study was supported in part by a grant-in-aid from the Institute of Space and Astronautical Science, U-18 (901037).
LITERATURE CITED Boucaut, J.C., T. Darribere, D.L. Shi, H. Boulekbache, K.M. Yamada, and J.P. Thiery (1985) Evidence for the role of fibronectin in amphibian gastrulation. J. Embryol. Exp. Morphol., 89(S~ppL):211-227. Choi, Y.-S., and B. Gumbiner (1989) Expression of cell adhesion molecule E-Cadoherin in Xenopus embryos begins at gastrulation and predominates in the ectoderm. J . Cell Biol., 108:2449-2458. Darribere, T., J.-F.Riou, K. Guida, A.-M. Duprat, J.-C. Boucaut,
and J.-C. Beetschen (1991) A maternal-effect mutation disturbs extracellular matrix organization in the early Pleurodeles waltlii embryo. Cell Tissue Res., 263:507-514. Darribhre, T., K.M. Yamada,K.E. Johnson, and J.-G. Boucaut (1988) The 140-kDa fibronectin receptor complex is required for mesodermal cell adhesion during gastrulation in the amphibian Pleurodeles waltlii. Dev. Biol., 126:182-194. Hidaka, H., M. Inagaki, S. Kawamoto, and Y. Sasaki (1984) Isoquinolinesulfonamides, novel and potent inhibitors of cyclic nucleotide-dependent protein kinase C. Biochemistry, 23: 5036-5041. Holtfreter, J. (1946) Structure, motility and locomotion in isolated embryonic amphibian cells. J. Morphol., 79:27-62. Johnson, K.E., N. Nakatsuji, and J.-C. Boucaut (1990) Extracellular matrix control of cell migration during amphibian gastrulation. In: Cytoplasmic Organization Systems: A Primer in Developmental Biology. G.M. Malacinski, ed. McGraw Hill, New York, pp. 349-374. Keller, R.E. (1980) The cellular basis of epiboly: An SEM study of deep-cell rearrangement during gastrulation in Xenopus laeuis. J. Embryol. Exp. Morphol., 60:201-234. Keller, R.E. (1986) The cellular basis of amphibian gastrulation. In: Developmental Biology. The Cellular Basis of Morphogenesis. L.W. Browder, ed. Plenum Press, New York, Vol. 2, pp. 241-327. Komazaki, S. (1990) Development of cell surface activity and cell surface adhesiveness in early embryos of the newt, Cynops pyrrhogaster. Cell Differ. Dev., 29:195-204. Komazaki, S. (1991) A morphological study on regulation of hyaline bleb formation in early embryonic cells of Cynops pyrrhogaster. Cell Tissue Res., 263:337-344. Newport, J., and M. Kirschner (1982) A major developmental transition in early Xenopus embryos: I. Characterization and timing of cellular changes at the midblastula stage. Cell, 30:675-686. Okada, Y.K., and M. Ichikawa (1947) Revised normal table of thedevelopment ofl)-ituruspyrrhogasteer.J. Exp. Morph.,3:1-6. Satoh, N., T. Kageyama, and K.-I. Shirakami (1976) Motility of dissociated embryonic cells in Xenopus laeuis: Its significance to morphogenetic movements. Dev. Growth, Differ., 18:55-67. Suzuki, A,, Y. Kuwabara, and T. Kuwana (1976) Analysis of cell proliferation during early embryogenesis. Dev. Growth, ~Differ., 18:447-455. -
Epibolic Extension of the Presumptive Ectodermal Layer of Embryos of the Newt Cynops pyrrhogaster Before and During Gastrulation SHINJI KOMAZAKI Department of Anatomy, Saitama Medical School, Saitama, 350-04 Japan ABSTRACT Epibolic extension of the presumptive ectodermal layer (PEL) was investigated in embryos of the newt Cynops pyrrhogaster before and during gastrulation. The PEL was composed of only one layer of columnar cells a t all stages examined. The cells of the PEL became elongated from the blastula to the early gastrula stage. They were most elongated at the early gastrula stage and then shortened during gastrulation. Present observations suggest that changes in cell shape of the PEL play an important role in the control of the epibolic extension of the newt embryos. The morphology and movement of the isolated cells from the PEL were examined in an attempt to elucidate the role of cell movement in epibolic extension of the PEL. Blebbing and vermiform cells which showed active cell movement appeared at the early blastula stage. The blebbing cells, which formed large hyaline blebs that moved around the circumference of each cell, appeared in large numbers at the early blastula stage. The frequency of the blebbing cells decreased from the early blastula to the early gastrula stage and increased again during gastrulation. The vermiform cells, which had an elongated cell body and moved in a worm-like manner, increased in frequency from the early blastula to the early gastrula stage. The relative number of such vermiform cells was maximal at the early gastrula stage and decreased abruptly during gastrulation. These results suggest that the elongation of the cells of the PEL is controlled by the active cell movement which resembles that of a worm. 0 1992 Wiley-Liss, Inc.
Amphibian gastrulation is accomplished principally by three types of morphogenetic movement: epiboly of the presumptive ectodermal layer (PEL), involution of the marginal zone, and migration of the presumptive mesodermal cells (Keller, '86). In such morphogenetic movements, cell-to-celland cellto-extracellular-matrixinteractions play important roles in the control of the movement of individual cells. Mechanisms of cell-to-cell and cell-to-extracellular-matrix interactions have been investigated in terms of specific molecules, such as cell adhesion molecules (Boucaut et al., '85; Choi and Gumbiner, '89; Johnson et al., '90) and a cell-surfacereceptor for the adhesion molecule (Darribere et al., '88, '91). However, even though an important role of cell movement is very probable during morphogenetic movement, the mechanisms by which cell movements control morphogenetic movements are not well understood. In the present study, the roles of cell movement in the epibolic extension of the PEL before and during gastrulation were investigated. The epibolic extension of the PEL has been examined in detail in embryos of Xenopus laevis (Keller, '80).The PEL of Xenopus is composed of several layers of cells during early development. The 01992 WILEY-LISS, INC.
epibolic extension of the PEL is accomplished via a decrease in the number of layers of cells and a flattening of the cells in the superficial layer. This series of events is widely applicable in the case of epibolic extension of a PEL that is composed of several layers of cells. However, such a phenomenon is not applicable in the case of the epibolic extension of a PEL that is composed of only a single layer of cells, as in embryos of the newt Cynops pyrrhoguster. Therefore, in the present study, the mechanism of epibolic extension was investigated in the PEL of Cynops embryos.
MATERIALS AND METHODS Embryos Fertilized eggs of the Japanese newt Cynops pyrrhogaster were obtained after artificial inducement of spawning by human chorionic gonadotropin. The eggs were collected and kept at 20 or 25°C until embryos reached appropriate stages. Embryonic stages were according t o Okada and Ichikawa ('47). Embryos were collected from a single batch Received June 5,1991;revision accepted February 17,1992.
EPIBOLY OF THE NEWT EMBRYO
415
for measurements of diameter after removal of jelly capsules.
Light and scanning electron microscopy The jelly capsule of the embryos was removed manually with forceps. Penetration of fixative solution into the blastocoel was facilitated by puncturing the lateral side of the blastocoel roof with sharp tungsten needles in 2.5% glutaraldehyde, 3%paraformaldehyde in 0.1 M cacodylate buffer, pH 7.0. The embryos were kept in fixative for 16-18 h at room temperature (24-27'0. They were washed briefly in the same buffer solution and postfixed in 1%Os04 in 0.1 M cacodylate buffer, pH 7.0, for 6-8 h a t room temperature. After washing with the same buffer solution, they were dehydrated in a graded series of ethanol and acetone and embedded in epoxy resin. Semithin sections (1-5 pm thick) were cut and stained with 0.1% toluidine blue. For scanning electron microscopy, the same fixation and dehydration methods were used and then specimens were critical point dried in a liquid COz,coated with platinum, and examined in a Hitachi S-550 scanning electron microscope. The thickness of the PEL was measured under the light microscope in the epoxy-resinsections. The thickness of the PEL and ratio of height to width of the cells in the PEL at animal-pole region were measured. The area of apical surfaces of the cells was measured using a TV image processor (Excel-11,TVIP4100; Nippon Avionics Co.). The scanning electron microscope photographs at a magnification of 1,000x were used for measurement. Cell culture The cells of the PEL were collected from animalpole regions of morulae, blastulae, and gastrulae. The cells of the presumptive ectoderm (the same as the PEL), mesoderm, and endoderm of embryos at stage 12a (early gastrula stage) were collected from the regions shown in Figure 1.The cell masses were removed from these regions with tungsten needles and disaggregated in dissociation medium (60 mM NaC1,0.7 mM KC1,0.2%EDTA, buffered with 3 mM HEPES at pH 8.0) for 10-20 min at room temperature. The dissociated cells were transferred to culture dishes that had been filled with 5 ml of culture medium (60 mM NaC1, 0.7 mM KC1, 0.8 mM MgS04, 0.3 mM Ca(NO&, buffered with 3 mM HEPES at pH 7.4). Three different substratapolystyrene culture dishes (LUX, No. 5221; Miles, IL, USA), coverglasses (Matsunami Glass, Osaka, Japan), and 3%agar-were used. Dissociated cells
Fig. 1. Schematic representation of a mid-sagittal section of an embryo at stage 12a showing the three regions from which cells were isolated. PE, presumptive ectoderm (the same as with the PEL); PEn, presumptive endoderm; PM, presumptive mesoderm.
were distributed uniformly on the various substrata and incubated in the culture medium for 1h at room temperature. The cells were observed under an inverted phase-contrast microscope after incubation. The numbers of cells of various shapes as percentages of the total number of cells were calculated in each case. Cells were treated with 5 pg/ml cytochalasin B (Aldrich, Milwaukee, WI, USA), 10 mM colchicine (Wako Pure Chem., Osaka, Japan), 100 pM H-7 (Seikagaku Kogyo, Tokyo, Japan), and 0.2 mg/ml bovine serum albumin (Cappel, PA, USA) by incubation in culture media supplemented with these agents. Effects of divalent cations were examined in Ca2+-and Mg2+-freeculture medium.
RESULTS Scanning electron microscopy The PEL of morulae (stages 7 and 8) was composed of a single layer of columnar cells (Fig. 2a). The ratio of height to width of the cells of the PEL at stage 8 was 2.1 ? 0.4 (mean SD for 60 cells from 6 embryos). The cells of the PEL were separated by wide intercellular spaces and had many filamentous processes on their basal surfaces which face the blastocoel (Fig. 2b). The PEL of blastulae (stages 9 and 10) and of initial and early gastrulae (stages 11and 12a)were also composed of a single layer of cells. The cells of the PEL were elongated and columnar in shape at these stages and were especially elongated at stages 11 and 12a (Fig. 2c). The ratio of height to width of the cells of the PEL at stage 12a was 4.4 -+ 0.6 (mean ? SD for 60 cells from 6 embryos). The area of apical surface of individual cells at stage 12a was 484.1 k 57.3 pm2 (mean SD for 563 cells from 7
*
*
416
S.KOMAZAKI
Fig. 2. Scanning electron micrographs showing the changes in shape of cells in the PEL. a: Cells at stage 8. b: Basal surface of the PEL at stage 8. Many filamentous processes are seen at margins of the cells. c: Cells of the PEL at stage 12a. d: High magnification of the intercellular spaces between the cells of the PEL at stage 12a. e: Basal surface of cells at stage 12a. Large bulbous processes are visible. fi The archenteric roof at
stage 13. The presumptive ectoderm (upper layer) and endoderm (lower layer). g: The basal surface of the presumptive ectodermal cells at stage 13. The surface was exposed after removal of the presumptive endodermal cells. Many filamentous and lamellar processes are visible. Bars indicate 100 km (a),50 pm (b,e,D,and 10 km (c,d,g).
EPIBOLY OF THE NEWT EMBRYO
417
embryos). The cells of the PEL at these stages were t IJm) separated by wide intercellular spaces and had large bulbous processes on their basal surfaces (Fig. 2d,e). The large bulbous processes were also prominent on the cells of the embryos at stages 10 and 11. The PEL of middle gastrulae (stage 13) was also 200composed of one layer of columnar cells (Fig. 20. The ratio of height to width of the cells of the PEL at stage 13b was 2.3 & 0.3 (mean & SD for 72 cells from 6 embryos). The area of apical surface of indi\T vidual cells at stage 13bwas449.6 2 52.2 pm2(mean -t SD for 563 cells from 7 embryos). The invagination of the archenteron was almost complete at this loo-stage. The cells of the PEL were in close contact with one another and there were many filamentous and lamellar processes on their basal surfaces (Fig. 2g). Changes in the thickness of the PEL and in the diameter of the embryos before and during gastrulation are shown in Figure 3. The thickness of the PEL at stage 8 was reduced by 50% during stages 0 1 : : : : : : : : I 9 10 11 12 13 14 15 16 11-12 and by 75%at stage 13. By contrast, the diam9 10 11 12 13 8 eter of embryos increased less (by less than 1.3% from stages 8 t o 12a and by 5%from stages 8 to 13) Fig. 3. Changes of the thickness of the PEL and the diamduring these stages. eters of embryos before and during gastrulation. Horizontal axis:
TT
!\
i
\f/
p \Q
Behavior ofisolated cells The isolated cells of the PEL could be classified into several classes after 1h in culture. From their morphology and the nature of their movements, the cells of the PEL can be clearly divided into four classes as follows (Fig. 4): 1)blebbing cells, with hyaline blebs that extend around each cell's periph-
Fig. 4. Cultured cells isolated from embryos at stage 12a. Four types of cell are distinguishable after 1h of culture on a coverglass. a: A sphere cell, which is spherical in shape. b: A blebbing cell with a hyaline bulbous process that moves around
the upper numbers indicate the timing of cleavages and the lower numbers indicate the stages of embryos. The relationship between the timing of cleavages and the stages of embryos was estimated from the results of Suzuki et al. ('76). Vertical axis: the numbers indicate the thickness of the PEL (left) and the diameter of embryos (right). Eleven embryos collected from the same batch were used for measurements of the diameters of embryos. Changes in the thickness of the PEL were calculated from the results from 7,15,16,15,30,and 12 embryos at stages 8,9,10, 11,12, and 13, respectively.
its circumference. c: A vermiform cell. d A spreading cell that adheres to the substratum via filamentous and lamellar processes. Bars indicate 50 Fm.
S.KOMAZAKI
418
TABLE 1 . Effects of substrata on relative numbers, as percentage, of various types of cell' Substratum Coverglass Normal culture medium
Ca2+-and Mg2+-free culture medium
Culture medium containing 0.2 mg/ml albumin Polystyrene dish
Agar
Cell type Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading SDhere
Percentages of various types of cell2
No. of embryos examined
65.7 f 7.4 23.9 i 9.5 3.8 f 2.1 6.5 f 4.1 58.1 f 9.4 29.7 i 9.7 3.0 3.5 9.2 ? 7.1
12
12
*
0.4 f 0.5 96.0 f 3.5 0.3 ? 0.3 3.4 t 3.3 0.2 k 0.2 90.3 f 5.3 6.4 t 5.4 3.1 f 1.6 0.1 f 0.1 91.2 ? 3.6 0.0 f 0.0 8.7 3.7
16
14
11
*
'The cells isolated from the PEL of early gastrulae (stage 12a) were used. 2Valuesare means ? SD.
ery; 2) vermiform cells, which are elongated and show active movement, namely expansion, contraction, and bending of the cell body; 3) spreading cells, which adhere to the substratum via filamentous and lamellar processes; and 4) sphere cells, which are spherical in shape and remain on the substratum. The blebbing and the vermiform cells moved actively, whereas the spreading and the sphere cells remained still. The effects of various substrata on relative frequencies of these four types of cell were investigated using cells isolated from the PEL of embryos at stage 12a (Table 1).Almost all of the isolated cells were blebbing just after the isolation. The vermiform and the spreading cells appeared within 30 min of incubation and the frequency of these cells gradually increased for a further 30 min. The vermiform cells appeared at high frequency only on the coverglasses among the substrata examined. Omission of Ca2+ and Mg2+ ions from the culture medium had little inhibitory effect on the formation of the vermiform cells. Addition of bovine serum albumin blocked the cells from sticking to the glass substratum and completely suppressed the formation of the vermiform cells, resulting in an increased number of blebbing cells. The cells stuck to the polystyrene substratum but very few cells changed to vermiform cells. The cells did not stick to the agar substratum and very few cells changed to vermiform cells. The relative numbers of various types of cell in
the different embryonic regions were examined in cells isolated from embryos at stage 12a (Table 2). Cells were isolated from three different regions, the presumptive ectoderm (the same as the PEL), the mesoderm, and the endoderm, and they were cultured on coverglasses. The vermiform and blebbing cells appeared among cells from the presumptive ectodermal region at higher frequency than among cells from the other two regions. By contrast, sphere cells appeared at higher frequency among cells from both the presumptive mesodermal and endodermal regions than among cells from the presumptive ectodermal region. The relative numbers of spreading cells were low among cells from all three regions. Changes in the relative numbers of variously shaped cells before and during gastrulation were examined using cells from the PEL (Figs. 5, 6). Almost all of the cells were sphere cells at stages 7 and 8. The blebbing and the vermiform cells appeared at stage 9. The relative numbers of vermiform cells increased gradually duringstages 9- 12a and reached a maximum at stage 12a. The relative numbers of blebbing cells decreased during stages 9-12a. The relative numbers of vermiform cells decreased abruptly at stage 13, while those of the blebbing cells increased again at stage 13. The relative numbers of spreading cells were low during stages 7-12a but increased markedly a t stage 13. The effects of several drugs, known t o influence
EPIBOLY OF THE NEWT EMBRYO
419
TABLE 2. Relative numbers, as percentages, of various types of cell in three differentregions of early gastrulae' Embryonic region Presumptive ectoderm
Presumptive mesoderm
Presumptive endoderm
Cell type Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere
Percentages of various types ofcel12
*
65.7 7.4 23.9 L 9.5 3.8 2 2.1 6.5 4.1 43.0 2 9.5 8.2 L 6.2 1.6 L 0.9 47.1 IT 11.7 30.4 2 8.4 6.6 f 4.5 1.7 L 2.8 61.4 L 8.5
No. of embryos examined
12
*
15
16
'Coverglass substratum and the cells isolated from the PEL of early gastrulae (stage 12a) were used. 'Values are means ? SD.
the shape and movement of cells, and of bovine serum albumin on cultured cells from the PEL were examined (Table 3). Cytochalasin B completely inhibited the appearance of the vermiform and the blebbing cells. Almost all of the cells were sphere cells in the presence of this drug. Colchicine had little inhibitory effect on the appearance of the vermiform and the blebbing cells. H-7, an inhibitor of cyclic nucleotide-dependent protein kinase C, had a remarkable inhibitory effect on the appearance of the vermiform and the blebbing cells. The relative numbers of spreading cells and sphere cells increased with a decrease in the relative numbers of vermiform and blebbing cells.
DISCUSSION In Xenopus embryos, division of cells, a decrease in the number of layers of cells, and flattening of the cells are the main morphological events related t o the epibolic extension of the PEL (Keller, '80). Cell division is also important in the epibolic extension of the PEL of Cynops embryos. However, the details of the epibolic extension of the PEL differ between Xenopus and Cynops embryos. Changes in the shapes of the cells of the PEL play an important role in the control of the epibolic extension of Cynops embryos. As shown in the present study, the PEL of Cynops embryos decreases in thickness without any close correlation with changes in diameter of embryos and the timing of cleavages. The PEL did not decrease much in thickness during stages 8-12a and then its thickness decreased abruptly during gastrulation. Elongation and shortening of the cells would modify the decrease in thickness of the PEL. Elongation of the cells before gastrulation will suppress the decrease in thickness of the PEL. Conversely, the shortening of the cells
during gastrulation will accelerate the decrease in the thickness of the PEL. As for the epibolic extension of the PEL, elongation of the cells will decrease the area of apical surface of the cells and consequently suppress the epibolic extension of the PEL before gastrulation. On the other hand, shortening of the cells will increase the area of apical surface of the cells and consequently accelerate the epibolic extension of the PEL during gastrulation. In fact, though the number of the cells of the PEL nearly doubled between stage 12a and 13b (Suzuki et al., ,761,the area of apical surface of the cells decreased only 7.1% during these stages. These facts confirm the conclusion that shortening of the cells during gastrulation accelerates the epibolic extension of the PEL. Such mechanisms for controlling the epibolic extension of the PEL have not been demonstrated in Xenopus embryos. It seems that the differences in the mechanisms of epibolic extension of the PEL between Cynops and Xenopus embryos are mostly the result of differences in the structure of the PEL. However, it is unclear whether or not the epibolic extension of the PEL of both types of embryo is controlled by intrinsically the same mechanisms. Morphological studies of Cynops embryos showed that changes in cell shape play an important role in control of the epibolic extension of the PEL. However, observations by scanning electron microscopy alone can never define whether or not the changes in cell shape are caused by the active movement of the cells themselves. Therefore,changes in the movement of isolated cells were examined in relation to changes in cell shape before and during gastrulation. Isolated cells showed two types of cell movement, blebbing and venniform movement. These two types of cell movement are common in isolated cells
420
S. KOMAZAKI
Fig. 5. Changes in types of cell before and during gastrulation. Isolated cells were cultured for 1 h on a coverglass. a: Cells from the PEL at stage 8. Almost all ofthe cells are sphere cells. b: Cells from the PEL at stage 9. Blebbing cells are abundant. c: Cells from the PEL at stage 12a. Most of the cells are
vermiform cells. d: Presumptive mesodermal cells at stage 12a. e: Presumptive endodermal cells at stage 12a. fi Cells from the PEL at stage 13. Spreading cells are abundant. Bars indicate 200 pm (a)and 100 pm (b-0.
EPIBOLY OF THE NEWT EMBRYO
421
of early amphibian embryos (Holtfreter, '46;Satoh et al., '76). The blebbing cells appeared abruptly at stage 9. The appearance of the blebbing cells is dependent on the timing of cleavages and on structural changes in the submembranous cortical layer, a thick layer of cytoskeleton just beneath the cell membrane. The blebbing cells appear after the 12th cleavage in Xenopus (Newport and Kirschner, '82) and after the 11th cleavage in Cynops (Komazaki, '90). In Cynops embryos, the structure of the submembranous cortical layer changes after the 11th cleavage, and artificial changes in the struc5( ture of this layer induce the formation of blebs (Komazaki,'91).The frequency of the blebbing cells gradually decreased from stages 10 to 12a and increased again at stage 13. By contrast, the frequency of vermiform cells changed in inverse proportion to that of the blebbing cells. It seems that the blebbing cells change into the vermiform cells during the course of development since both forms are interconvertible in culture (Holtfreter, '46).The present observations showed a close correlation between the elongation of the cells of the PEL and the appearance of the venniform cells during develFig. 6. Changes in frequencies of various types of isolated opment course,implying that the elongationof these embryonic cell before and during gastrulation. Cells were cul- cells is caused by active cell movement, which resemtured for l h on coverglasses. Horizontal axis: the numbers indicate the stages of the embryos. The vertical axis shows the bles that of a worm. The formation of large hyaline percentages of various cells. Relative numbers, as percentages, blebs in the cells of the PEL at stages 10 and 11 of vermiform cells (solid triangles), blebbing cells (open circles), may be indicative of the active movement of spreading cells (open triangles), and sphere cells (solidcircles) are these cells. shown. The results are from 11,13,16,15,12,11,and 11embryos Among the substrata tested, only coverglasses at stages 7,8, 9, 10, 11,12, and 13, respectively. The numbers of induced the formation of the vermiform cells at high cellsexaminedwere 20-40,50-100,150-250,200-300, and300400 from embryos at stages 7,8,9,10, and 11-13, respectively. frequency. Contact of cells with a coverglass sub-
1O(
TABLE 3. Effects of inhibitors on relative numbers, as percentage, of various types of cell'
Inhibitor Control
Cytochalasin B (5 pglml)
Colchicine (10 mM)
H-7 (100 FM)
Cell type
Percentages of various types of cell2
No. of embryos examined
Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere Vermiform Blebbing Spreading Sphere
65.7 + 7.4 23.9 k 9.5 3.8 2 2.7 6.5 f 4.1 0.2 f 0.3 0.3 -t 0.3 1.1 f 1.4 98.3 ? 1.4 55.1 f 11.4 14.6 + 7.0 3.7 5 2.3 26.7 f 12.9
12
Vermiform Blebbing Spreading Sphere
1.2 t 2.2 0.2 ? 0.4 39.5 ? 13.1 59.1 f 12.8
13
'Coverglass substratum and the cells isolated from the PEL of early gastrulae (stage 12a)were used. 2Values are means t SD.
13
12
S.KOMAZAKI
422
stratum may be important for the formation of the vermiform cells since disturbance of the cell-tosubstratum contacts by addition of bovine serum albumin to the culture medium completely suppressed the formation of the vermiform cells. Divalent cations may not be significantly involved in the cell-to-substratum contacts during the formation of the vermiform cells since omission of divalent cations from the culture medium had little effect on the formation of the vermiform cells. Why can only a coverglass substratum induce the formation of the vermiform cells? The answer to this question remains to be determined. However, from the present study, it is clear that actin filaments and protein kinase C play important roles in the formation of the vermiform and the blebbing cells because cytochalasin B, an inhibitor of the polymerization of actin filaments, and H-7, a potent inhibitor of cyclic nucleotide-dependent protein kinase C (Hidaka et al., '84),completely suppressed the formation of the vermiform and the blebbing cells.
ACKNOWLEDGMENTS The author thanks Professor Reiji Hirakow for his interest in this study. This study was supported in part by a grant-in-aid from the Institute of Space and Astronautical Science, U-18 (901037).
LITERATURE CITED Boucaut, J.C., T. Darribere, D.L. Shi, H. Boulekbache, K.M. Yamada, and J.P. Thiery (1985) Evidence for the role of fibronectin in amphibian gastrulation. J. Embryol. Exp. Morphol., 89(S~ppL):211-227. Choi, Y.-S., and B. Gumbiner (1989) Expression of cell adhesion molecule E-Cadoherin in Xenopus embryos begins at gastrulation and predominates in the ectoderm. J . Cell Biol., 108:2449-2458. Darribere, T., J.-F.Riou, K. Guida, A.-M. Duprat, J.-C. Boucaut,
and J.-C. Beetschen (1991) A maternal-effect mutation disturbs extracellular matrix organization in the early Pleurodeles waltlii embryo. Cell Tissue Res., 263:507-514. Darribhre, T., K.M. Yamada,K.E. Johnson, and J.-G. Boucaut (1988) The 140-kDa fibronectin receptor complex is required for mesodermal cell adhesion during gastrulation in the amphibian Pleurodeles waltlii. Dev. Biol., 126:182-194. Hidaka, H., M. Inagaki, S. Kawamoto, and Y. Sasaki (1984) Isoquinolinesulfonamides, novel and potent inhibitors of cyclic nucleotide-dependent protein kinase C. Biochemistry, 23: 5036-5041. Holtfreter, J. (1946) Structure, motility and locomotion in isolated embryonic amphibian cells. J. Morphol., 79:27-62. Johnson, K.E., N. Nakatsuji, and J.-C. Boucaut (1990) Extracellular matrix control of cell migration during amphibian gastrulation. In: Cytoplasmic Organization Systems: A Primer in Developmental Biology. G.M. Malacinski, ed. McGraw Hill, New York, pp. 349-374. Keller, R.E. (1980) The cellular basis of epiboly: An SEM study of deep-cell rearrangement during gastrulation in Xenopus laeuis. J. Embryol. Exp. Morphol., 60:201-234. Keller, R.E. (1986) The cellular basis of amphibian gastrulation. In: Developmental Biology. The Cellular Basis of Morphogenesis. L.W. Browder, ed. Plenum Press, New York, Vol. 2, pp. 241-327. Komazaki, S. (1990) Development of cell surface activity and cell surface adhesiveness in early embryos of the newt, Cynops pyrrhogaster. Cell Differ. Dev., 29:195-204. Komazaki, S. (1991) A morphological study on regulation of hyaline bleb formation in early embryonic cells of Cynops pyrrhogaster. Cell Tissue Res., 263:337-344. Newport, J., and M. Kirschner (1982) A major developmental transition in early Xenopus embryos: I. Characterization and timing of cellular changes at the midblastula stage. Cell, 30:675-686. Okada, Y.K., and M. Ichikawa (1947) Revised normal table of thedevelopment ofl)-ituruspyrrhogasteer.J. Exp. Morph.,3:1-6. Satoh, N., T. Kageyama, and K.-I. Shirakami (1976) Motility of dissociated embryonic cells in Xenopus laeuis: Its significance to morphogenetic movements. Dev. Growth, Differ., 18:55-67. Suzuki, A,, Y. Kuwabara, and T. Kuwana (1976) Analysis of cell proliferation during early embryogenesis. Dev. Growth, ~Differ., 18:447-455. -
