We have investigated the autonomous migration of marginal cells and their interactions with extracellular matrix (ECM) located on the inner surface of the blastocoel roof in the urodele amphibian, Pleurodeles waltl, using a novel in vitro migration assay. Animal hemispheres containing equatorial cells removed at different cleavage stages and dorsal marginal zone (DMZ) expiants of early gastrula stage were cultured either on fibronectin (FN)-coated or ECM-conditioned substrata. In explanted animal hemispheres, dorsal marginal cells showed autonomous migration on FN-coated substratum at the same time as the onset of gastrulation in control embryos. They acquired this capacity at least at the 32-cell stage, whereas lateral and ventral marginal cells acquired it after the 64-cell stage. DMZ outgrowths of early gastrula stage exhibited auton-omous spreading on both substrata. In addition, we showed that they spread preferentially toward the ani-mal pole when deposited on substratum conditioned by the dorsal roof of the blastocoel. By culturing dissociated marginal cells on ECM-conditioned substratum, we also found that increased spreading capacity of marginal cells was related to the initiation of their migration. A comparative study of the migration of marginal cells in ultraviolet (u.v.)-irradiated and normal embryos was also made. The results indicate that dorsal marginal cell migration was absent or dramatically reduced by u.v.-irradiation. These results suggest that the differential acquisition in the spreading capacity both in timing and in intensity around the marginal zone was correlated with the sequential involution of mesodermal cells in the course of gastrulation.

The basic body plan of the amphibian embryo arises from sequentially programmed cellular activities. The dorsoventral axis is established after fertilization as a result of cortical/cytoplasmic rotation (Vincent & Gerhart, 1987). The cytoplasmic localizations responsible for the formation of axial structures are initially localized in the subequatorial region of uncleaved egg (Gurdon et al. 1985b). Alternatively, during cleavage stages, mesoderm induction which involves cell interactions between the presumptive endoderm and the adjacent animal hemisphere leads to the formation of an equatorial zone of mesoderm (Nieuwkoop, 1969a; Sudarwati & Nieuwkoop, 1971; Nieuwkoop, 1977; Gurdon et al. 1985a). Mesoderm induction is known to have regional specificity because there are different signals to specify the dorsal and ventral mesodermal structures (Nieuwkoop, 1969b; Smith & Slack, 1983; Dale et al. 1985; Dale & Slack, 1987b).

The most immediate consequence of mesoderm induction is the extensive morphogenetic cell movements that occur during gastrulation. These involve various region-specific cellular activities (reviewed by Gerhart & Keller, 1986). The dorsal mesoderm plays an important role in this process in both anuran and urodele amphibians. In Xenopus laevis, cells of the DMZ undergo convergent extension by the mechanism of active cell intercalation which alone can account for its involution and blastopore closure (Keller et al. 1985). Unlike Xenopus, the DMZ of the Pleurodeles waltl (Shi et al. 1987), and also other urodele amphibians such as Cynops pyrrhogaster (Nakatsuji, 1975) and Ambystoma mexicanum (Lundmark, 1986), consists of two or three layers of presumptive mesodermal cells between the late blastula and the early gastrula stages. The DMZ becomes a single layer of cells soon after involution has begun. We have previously found that migration of the involuted cells on the inner surface of the blastocoel roof ensures complete gastrulation, whereas cell intercalation plays only a limited role in this process.

Recently, several lines of evidence have demonstiated the presence of ECM fibrils covering the inner surface of the blastocoel roof both in anuran and urodele amphibians (Nakatsuji & Johnson, 1983b). These fibrils have been shown to contain both FN (Boucaut & Darribère, 1983; Lee et al. 1984; Darribère et al. 1985; Nakatsuji et al. 1985a) and laminin (LM) (Nakatsuji et al. 1985b; Darribère et al. 1986). ECM-fibrils serve as a contact-guidance system for mesoder-mal cell migration toward the animal pole (Nakatsuji et al. 1982; Nakatsuji & Johnson, 1983a, 1984a,b; Nakatsuji, 1984). The role of FN in this contact-guidance system has been well established (Boucaut et al. 1984a,b, 1985). Also, it has been shown that Xenopus gastrula mesodermal cells adhere to and migrate on LM-coated substratum (Nakatsuji, 1986). More recently, integrin (INT), a 140×103Mr glycoprotein complex involved in cell adhesion to FN and LM, has been reported to be present and play a functional role during gastrulation in Pleurodeles embryos (Darribère et al. 1988; reviewed by Johnson et al. 1988). Thus studies of mesodermal cell interactions with ECM components are important to our understanding of the control of gastrulation movements. The search for intrinsic factors governing mesodermal cell migration is equally important because it can provide information about the temporal-spatial regulation of cellular movements as well as the relationship between the time of onset and the level of morphogenetic cell movements and their directed migration toward the animal pole during gastrulation.

Here we report on the initiation and temporalspatial regulation of mesodermal cell migration in Pleurodeles gastrulae. The study was undertaken primarily to examine the idea that intrinsic factors govern the initiation of mesodermal cell migration. For that purpose, we devised an in vitro migration assay by culturing animal hemispheres containing the equatorial cells from different cleavage stages and DMZ expiants from early gastrulae on FN-coated substratum. We also investigated the migratory autonomy of mesodermal cells and their interactions with ECM components. Our results indicate that marginal zone cells show autonomous migration and spreading on ECM-conditioned substratum.

Embryos

Embryos of the urodele amphibian Pleurodeles waltl Michah were obtained and staged as described previously (Shi et al. 1987).

U.v. irradiation

Eggs were collected from spawning females. Jelly was manually removed with forceps under a dissecting microscope. The time period between fertilization and the first cleavage division is about 6 h at 21–23 °C. Irradiation was carried out at 2h after fertilization. Dejellied eggs were transferred into a quartz chamber placed 7·5cm above a 254 nm u.v. lamp (15 W; Vilbert Lourmat, France) and irradiated for 1 min. The energy was measured using a 254 nm u.v. dosimeter (Vilbert Lourmat, France) as l·8–2·0×104ergsmm−2. The development of axial structures was determined according to the descriptions of Scharf & Gerhart (1980).

For FN-coated substratum, plastic culture dishes (35 mm; Nunc, Denmark) were incubated overnight at 4°C with human plasma FN (Sigma Chemical Co.) at 50 μg ml−1 in sterile distilled water. ECM-conditioned substratum was prepared according to Nakatsuji & Johnson (1983a, 1984a,b). Blastocoel roofs were removed from early gastrula embryos using a tungsten needle and a platinum loop. They were then explanted onto culture dishes containing Steinberg’s solution (SS) with their basal surface against the plastic. After 2 h, the expiants were flushed away with a Pasteur pipette. For both kinds of substrata, residual protein-binding sites on the culture dishes were blocked by incubation with 5% bovine serum albumin (BSA) (Fraction V; Sigma Chemical Co.) in SS for 1 h. The culture dishes were rinsed with SS before use.

Migration assays

Dissection was carried out in 1 % agar-coated culture dishes containing SS. A rectangular explant of DMZ and adjacent ectoderm was removed from the early gastrula stage and then deposited onto FN-coated or ECM-conditioned substrata. Animal hemispheres containing equatorial cells from different cleavage stages were obtained by cutting circumferentially below the equatorial line. Next, they were deposited onto the culture dishes with their basal surface against the substratum. Both kinds of expiants were cultured in SS supplemented with penicillin (100i.u. ml−1), streptomycin (1 μg ml−1) and gentamycin (25 μg ml−1) for appropriate periods.

Cell culture

Dissociated cell spreading was performed on ECM-conditioned substratum. Expiants of DMZ and ventral marginal zone (VMZ) were dissociated in Ca2+-/Mg2+-free Barth’s medium (88mm-NaCl, lmm-KCl, 2 · 4mm-NaHCO3, 2mm-Na2HPO4, 0 ·1 mm-KH2PO4, 0 · 5mm-EDTA, pH 7 · 8). Completely dissociated cells were rinsed in SS for 10 min and then seeded onto the ECM-conditioned areas (300 – 400 cells for each area). Spread cells were flattened, formed lamellipodial attachments and assumed a bipolar, triangular or polygonal shape. The percentage of spread cells was determined at 30, 60, 90 and 120min in culture.

Inhibition of mesodermal cell migration

The interactions of spreading mesodermal cells with ECM components were investigated using different antibodies and synthetic peptides. Antibodies to Ambystoma mexicanum plasma FN have been previously described (Boucaut & Darribère, 1983). Antibodies to avian INT that cross-react with Pleurodeles (Darribère et al. 1988), the FN cell-binding peptide Gly-Arg-Gly-Asp-Ser (GRGDS), FN collagen-binding peptide Cys-Gln-Asp-Ser-Glu-Thr-Arg-Thr-Phe-Tyr (CQNSETRTFY) and LM cell-binding peptide Tyr-Iso-Gly-Ser-Arg (YIGSR) were kind gifts from Dr Kenneth M. Yamada (NIH, Bethesda, MD) and Dr Jean-Paul Thiery (ENS, Paris). Affinity-purified antibodies to highly purified LM from mouse Engelbreth-Holm-Swarm sarcoma have been obtained and characterized previously (Riou et al. 1987). Monovalent Fab’ fragments were prepared according to Brackenbury et al. (1977). Their final concentrations in the culture medium were: antibodies to FN, 0 · 5 mg ml−1; antibodies to INT, I mg ml−1; antibodies to LM, I mg ml−1; GRGDS, 0 · 2 mg ml−1; YIGSR, I mg ml−1 and CQNSETRTFY, I mg ml−1. A control experiment was performed using antibodies to FN absorbed with the antigen in a ratio of 1:2.

Immunofluorescence

ECM-conditioned substratum was monitored for the presence of FN and LM. After removing the blastocoel roofs, conditioned substrata were fixed in 3 · 7 % formaldehyde in SS for 30 min. They were then rinsed in SS and incubated sequentially with antibodies to FN (5 μg ml−1) or LM (20 μg ml−1), and fluorescein isothiocyanate (FITC)-conjugated sheep IgG anti-rabbit IgG (5 μg ml−1; Biosys, France). After rinsing in SS, conditioned substrata were covered by a coverslip in the presence of Mowiol (Hoeshst, RFG). They were examined with a Leitz Dialux-20 epifluorescent microscope and photo-graphed using Kodak Tri-X film.

Scanning electron microscopy

Control embryos at the midgastrula stage and u.v.-irradiated embryos that displayed a circular blastopore at the vegetal pole were selected. Vitelline membranes were removed using sharpened forceps and, at the same time, a small area of animal pole cells was removed to make a hole for penetration of fixatives. Embryos were then processed as described previously (Shi et al. 1987).

Temporal-spatial acquisition of cell migration in the marginal zone

The developmental capacity for autonomous migration of mesodermal cells was investigated by culture of animal hemispheres containing equatorial cells from the 32-cell stage onward on the FN-coated substratum. At the 32-cell stage, the 8 vegetal cells were removed and the remaining 24 animal and equatorial cells were cultured. From the 64-cell stage onward until the 512-cell stage, dissection of embryos was performed below the equatorial region in order to discard only the presumptive endodermal cells (see Fig. 1 and Fig. 2A –D). As judged by shaking the culture dishes, 1 –2 h was necessary for such expiants to attach firmly to the FN substratum.

Fig. 1.

Schematic representation of the dissection of embryos for cell migration assay. At the 32-cell stage, the 24-animal and equatorial blastomeres were isolated. From the 64-cell to the late blastula stages, expiants included the entire blastocoel roof. Expiants of DMZ and the adjacent ectoderm were excised at the early gastrula stage when the dorsal blastopore was just visible (stage 8a).

Fig. 1.

Schematic representation of the dissection of embryos for cell migration assay. At the 32-cell stage, the 24-animal and equatorial blastomeres were isolated. From the 64-cell to the late blastula stages, expiants included the entire blastocoel roof. Expiants of DMZ and the adjacent ectoderm were excised at the early gastrula stage when the dorsal blastopore was just visible (stage 8a).

Fig. 2.

Culture of animal hemispheres containing ectoderm and equatorial cells on FN-coated substratum. (A-D) Animal hemispheres isolated from the 32- to the 256-cell stages, 10min after explantation onto FN-coated substratum. (E-H) When control embryos for expiants of each stage reached the early gastrula stage, progeny cells of the dorsal equatorial blastomeres become less cohesive and start to migrate away from the expiants. (I –L) Control embryos for expiants of each stage were at the end of gastrulation. Mesodermal cell migration is restricted to the dorsal region in expiants removed at the 32- and the 64-cell stages (I,J). At the 128- and the 256-cell stages (K,L), mesodermal cell migration can be observed all around the margin of expiants. Scale bar, 1mm.

Fig. 2.

Culture of animal hemispheres containing ectoderm and equatorial cells on FN-coated substratum. (A-D) Animal hemispheres isolated from the 32- to the 256-cell stages, 10min after explantation onto FN-coated substratum. (E-H) When control embryos for expiants of each stage reached the early gastrula stage, progeny cells of the dorsal equatorial blastomeres become less cohesive and start to migrate away from the expiants. (I –L) Control embryos for expiants of each stage were at the end of gastrulation. Mesodermal cell migration is restricted to the dorsal region in expiants removed at the 32- and the 64-cell stages (I,J). At the 128- and the 256-cell stages (K,L), mesodermal cell migration can be observed all around the margin of expiants. Scale bar, 1mm.

When the respective control embryos reached the early gastrula stage (8a), i.e. about 30h after fertiliz-ation at 21 –23°C, where involution is initiated in the dorsal region, cell migration begins at the margin of the expiants. Cells migrated away from the expiants largely as a cohesive cell sheet but with some isolated pioneer cells (Fig. 2E – H). At each stage, the results show that the average culture time required for the initiation of cell migration corresponds approximately to that needed for control embryos to start gastrulation (Table 1) except that a 3 – 4h delay was observed. This delay may be due to the healing process and the initial retardation of cell adhesion to the substratum. There are significant differences in cell migration in the margin of expiants at different developmental stages (Fig. 2I – L). In expiants derived from the 32-cell stage, cell migration was restricted to a limited zone (Fig. 2I). The area covered by the migrating mesodermal cell sheet was enlarged in expiants derived from the 64-cell stage (Fig. 2J). A contracted region developed between migrating mesodermal cells and the adjacent ectoderm was observed in expiants cultured from early embryos (32- and 64-cell stages). In expiants derived from the 128-cell stage onward, mesodermal cell migration begins first in a small region but later extends all around the explant margin (Fig. 2K,L).

Table 1.

Time of the initiation of DMZ cell migration in explanted animal hemispheres from different cleavage stages on FN-coated substratum

Time of the initiation of DMZ cell migration in explanted animal hemispheres from different cleavage stages on FN-coated substratum
Time of the initiation of DMZ cell migration in explanted animal hemispheres from different cleavage stages on FN-coated substratum

In order to see if the region where cell migration first occurs corresponds to the future dorsal blastoporal lip of the embryos, vital dye staining was performed. In Pleurodeles waltl, a discordance between the position of grey crescent and the future dorsal region of the embryos could be observed frequently. For this reason. 80 eggs from the same batch were vital-dye marked in the grey crescent region. When they reached the 32-cell stage, at which time the dye markers were localized in the tier-3 equatorial blastomeres, 30 were used for the in vitro migration assay as described above, the remaining 50 were allowed to develop until the early gastrula stage in order to score the percentage of embryos with dorsal region stained. In 62% of the controls, staining was localized in the dorsal blastoporal lip. In other control embryos, vital dye staining was observed on both sides of the forming blastopore. In 60% of the expiants excised at 32-cell stage and cultured on FN-coated substratum, the initiation of cell migration occurred in the stained region. As a result, we can conclude that DMZ cells begin to migrate first in both cultured expiants and in intact embryos.

In vitro migration of early gastrula DMZ cells

Expiants of DMZ and the adjacent ectoderm were dissected just when the blastoporal depression was visible (stage 8a), they were then cultured on the FN-coated substratum (see Fig. 1). Both DMZ and the adjacent ectoderm adhered to the substratum within 30 min (Fig. 3A) as assessed by gently shaking the culture dishes. One hour later, migration of cells was occurring. Mesodermal cells that were initially near the dorsal blastoporal lip became less cohesive and some of them detached from the expiants to migrate as isolated cells (Fig. 3B). In contrast, expiants deposited onto BSA-coated or uncoated plastic culture dishes showed neither adhesion nor migration. The actively migrating pioneer mesodermal cells are bipolar and extend lamellipodia and filopodia at opposing ends of the cell body. These cells migrate both forward and backward. Cells behind these pioneer cells also undergo spreading migration. In the explant, a contracted region was observed between migrating mesodermal cells and the adjacent ectodermal cells (Fig. 3B,C) just as observed in expiants cultured from early embryos. The appearance of this contracted region is concurrent with the initiation of DMZ spreading as can be discerned in Fig. 3B. By the end of gastrulation in control embryos, the spreading of mesodermal cells reached its maximal level as a fan-shaped cohesive cell sheet (Fig. 3C). We estimated the extent of outgrowth of mesodermal cells from the dorsal lip of DMZ expiants to the outer edge of outgrowth. Isolated cells were not taken in account. During the 24 h period from the time of initiation of cultures until the end of gastrulation in control embryos, we found that there was a migration of 1 · 4mm ± 0 · 1 S.D. (n = 28), giving a migration rate of 1 μm min−1. After the gastrula stage, the rate of DMZ spreading slowed down. There was no obvious increase of this distance when the expiants were cultured until control embryos reached the late neurula stage. More-over, these cells begin to differentiate according to their prospective fate such as notochord and somitic cells (results not shown).

Fig. 3.

In vitro mesodermal cell migration from DMZ on FN-coated substratum. Explant (1 · 0 mm in length, 0 · 8mm in width) of DMZ together with adjacent ectoderm was cultured for 24h. (A) 30 min after explantation, the explant has healed and adhered to the substratum. (B) 2h after explantation, cells initially near the dorsal lip of the blastopore detach from the explant and migrate. A contracted region (arrows) appears concurrently with the initiation of DMZ spreading. (C) 24 h of culture, DMZ has fully expanded forming a fan-shaped cohesive cell sheet with isolated cells which migrate randomly. The region between migrating DMZ and the adjacent ectoderm becomes more contracted (arrows). Scale bar, 0 · 5mm.

Fig. 3.

In vitro mesodermal cell migration from DMZ on FN-coated substratum. Explant (1 · 0 mm in length, 0 · 8mm in width) of DMZ together with adjacent ectoderm was cultured for 24h. (A) 30 min after explantation, the explant has healed and adhered to the substratum. (B) 2h after explantation, cells initially near the dorsal lip of the blastopore detach from the explant and migrate. A contracted region (arrows) appears concurrently with the initiation of DMZ spreading. (C) 24 h of culture, DMZ has fully expanded forming a fan-shaped cohesive cell sheet with isolated cells which migrate randomly. The region between migrating DMZ and the adjacent ectoderm becomes more contracted (arrows). Scale bar, 0 · 5mm.

DMZ migration was also assayed on ECM sub-stratum conditioned by the blastocoel roof of early gastrulae. A cohesive cell sheet developed after 24h of culture. However, unlike on FN-coated substratum where isolated cells migrated randomly (see Fig. 3C), we rarely observed isolated migrating cells on ECM-conditioned substratum (see Fig. 6). In fact, on the later substratum, a few cells of the dorsal blastoporal lip detached from the expiants and migrated as isolated cells during the first 3 – 4 h of culture. Later, all migrating cells fused together as a cohesive cell sheet. This phenomenon is observed during normal development where isolated pioneer mesodermal cells appeared only at the early gastrula stage (results not shown). The extent of DMZ outgrowth was the same on both substrata.

Interactions of migrating mesodermal cells with ECM components

In order to ascertain that the in vitro spreading and migration of the DMZ cell sheet depends on specific interactions between mesodermal cells and ECM components, antibodies and synthetic peptides were added to the culture medium. We cultured DMZ expiants on FN-coated and also on ECM-conditioned substrata hoping to mimic the conditions found in vivo. On the latter substratum, immunofìuorescent studies showed that, at the front of migration, mesodermal cells extend lamellipodia and filopodia interacting with the ECM components present in the deposited fibrils (Fig. 4A,B). Each reagent was tested for 2h on duplicate cultures including ten DMZ expiants.

Fig. 4.

Mesodermal cell interactions with ECM-conditioned substratum. (A) Mesodermal cells at the front of migration extend lamellipodia and filopodia which interact with ECM fibrils (arrows) revealed by rabbit anti-FN antibodies (5 μg ml−1) followed by F1TC sheep anti-rabbit IgG (5 μg ml−1). (B) Mesodermal cells interaction with ECM fibrils (arrows) revealed by rabbit anti-LM antibodies (20 μg ml−1followed by FITC sheep anti-rabbit IgG (5 μg ml>−l). (C) Effect of monovalent Fab’ fragments of anti-FN on DMZ cell migration. Explant of DMZ and adjacent ectoderm of the early gastrula was cultured on ECM-conditioned substratum, monovalent Fab’ fragments of anti-FN were added after 24 h of culture as a final concentration of 0·5 mg ml, DMZ cell sheet is completely disintegrated into clusters of cells which become rounded. The adjacent ectoderm cell sheet is not affected. (D) Monovalent Fab’ fragments of anti-LM at the final concentration of 1 mg ml−1 has no effect on mesodermal cell migration. (E) Phase contrast corresponding to C showing the rounding up of disrupted DMZ cells. (F) Phase contrast of the leading edge of DMZ outgrowth in D. Cells remain cohesive and adherent to the substratum, m, Mesodermal cells. Scale bars: (A.B), 50 μm; (C,D), 0·5mm; (E,F), 100 μm.

Fig. 4.

Mesodermal cell interactions with ECM-conditioned substratum. (A) Mesodermal cells at the front of migration extend lamellipodia and filopodia which interact with ECM fibrils (arrows) revealed by rabbit anti-FN antibodies (5 μg ml−1) followed by F1TC sheep anti-rabbit IgG (5 μg ml−1). (B) Mesodermal cells interaction with ECM fibrils (arrows) revealed by rabbit anti-LM antibodies (20 μg ml−1followed by FITC sheep anti-rabbit IgG (5 μg ml>−l). (C) Effect of monovalent Fab’ fragments of anti-FN on DMZ cell migration. Explant of DMZ and adjacent ectoderm of the early gastrula was cultured on ECM-conditioned substratum, monovalent Fab’ fragments of anti-FN were added after 24 h of culture as a final concentration of 0·5 mg ml, DMZ cell sheet is completely disintegrated into clusters of cells which become rounded. The adjacent ectoderm cell sheet is not affected. (D) Monovalent Fab’ fragments of anti-LM at the final concentration of 1 mg ml−1 has no effect on mesodermal cell migration. (E) Phase contrast corresponding to C showing the rounding up of disrupted DMZ cells. (F) Phase contrast of the leading edge of DMZ outgrowth in D. Cells remain cohesive and adherent to the substratum, m, Mesodermal cells. Scale bars: (A.B), 50 μm; (C,D), 0·5mm; (E,F), 100 μm.

When monovalent Fab’ fragments of anti-FN were added to culture of fully spread DMZ cell sheet, solated migrating cells no longer adhered to the substratum. The cohesive DMZ cell sheet was disrupted into isolated cultures of cells (Fig. 4C). These cells became rounded but still attached to the substratum (Fig. 4E). In contrast, the coherence of the ectodermal sheet was not affected (Fig. 4C). Antibodies to INT and also the FN cell-binding peptide GRGDS gave the same results (not shown). On the other hand, neither the monovalent antibodies to LM nor the LM cell-binding peptide YIGSR had such effects (Fig. 4D). As can be seen in Fig. 4F, the cohesive cell sheet remains intact and the filopodia of cells at the leading edge of the outgrowth are still adherent to the substratum. Control experiments using anti-FN absorbed with the antigen and the FN collagen-binding peptide CQNSETRTFY, indicated that these reagents had no inhibitive effects on mesodermal cell migration. All DMZ expiants gave the same results for a given reagent tested.

Furthermore, if antibodies to FN, INT and the peptide GRGDS were added to the medium at the beginning of DMZ culture, cell spreading was prevented. Removal of the antibodies or peptide from the culture before gastrulation had ended in control embryos, restores mesodermal cell migration and allows the development of DMZ cell sheet. Antibodies to LM and the peptide YIGSR did not prevent DMZ from spreading on ECM-conditioned substratum. It appears that the autonomous spreading and migration of DMZ cells involves specific interactions with FN.

Involvement of ECM in the orientation of DMZ outgrowths

In experiments involving cultures of dissociated mesodermal cells on ECM-conditioned substratum, it was shown that these cells migrated preferentially toward the animal pole (Nakatsuji & Johnson, 1983a, 1984a,b). We made use of the in vitro migration assay to see if ECM provides a similar guidance for DMZ outgrowth. ECM substratum was conditioned by rectangular expiants excised in the area extending between the dorsal blastoporal lip and the animal pole of the early gastrula (stage 8a). After conditioning, a small explant corresponding to DMZ and the adjacent ectoderm was excised from another early gastrula and then deposited on the ECM-conditioned substratum. The DMZ explant orientation was perpendicular to that of the ECM-depositing explant (Fig. 5). After a culture period of 24 h, the orientations of DMZ outgrowths were scored.

Fig. 5.

Experimental design to test the involvement of ECM in the orientation of DMZ outgrowths. To make the conditioned substratum, rectangular expiants (l · 2nim in length, 0 · 8 mm in width) between the blastopore (BP) and the animal pole (AP) were dissected from the early gastrula stage. They were then deposited onto a culture dish with the blastocoel roof against the plastic (upper part). The edges of the explant were marked with forceps. After 2 h in culture, the explant was flushed away. A smaller explant (0 · 6 mm in length, 0 · 4 mm in width) of DMZ and adjacent ectoderm was dissected from another gastrula (lower part) and deposited in the centre of the ECM substratum (represented by crosses) with its animal pole to blastopore axis (arrow) perpendicular to that of the substratum. It was then cultured for 24 h.

Fig. 5.

Experimental design to test the involvement of ECM in the orientation of DMZ outgrowths. To make the conditioned substratum, rectangular expiants (l · 2nim in length, 0 · 8 mm in width) between the blastopore (BP) and the animal pole (AP) were dissected from the early gastrula stage. They were then deposited onto a culture dish with the blastocoel roof against the plastic (upper part). The edges of the explant were marked with forceps. After 2 h in culture, the explant was flushed away. A smaller explant (0 · 6 mm in length, 0 · 4 mm in width) of DMZ and adjacent ectoderm was dissected from another gastrula (lower part) and deposited in the centre of the ECM substratum (represented by crosses) with its animal pole to blastopore axis (arrow) perpendicular to that of the substratum. It was then cultured for 24 h.

Fig. 6.

Orientation of DMZ outgrowth toward the animal pole (AP+) on ECM-conditioned substratum. Explant containing DMZ and adjacent ectoderm was initially deposited in the centre of the substratum (see Fig. 5). Its original animal pole to blastopore axis (thick arrow) was perpendicular to that of the substratum which is marked by fine arrows in the corners of the micrograph. After 24h of culture, DMZ outgrowth has fully spread toward the animal pole, a cohesive cell sheet developed on this side. The front of migration reached the edge of animal pole and stopped there because ECM-conditioned substratum was no longer available. Scale bar, 0 · 5mm.

Fig. 6.

Orientation of DMZ outgrowth toward the animal pole (AP+) on ECM-conditioned substratum. Explant containing DMZ and adjacent ectoderm was initially deposited in the centre of the substratum (see Fig. 5). Its original animal pole to blastopore axis (thick arrow) was perpendicular to that of the substratum which is marked by fine arrows in the corners of the micrograph. After 24h of culture, DMZ outgrowth has fully spread toward the animal pole, a cohesive cell sheet developed on this side. The front of migration reached the edge of animal pole and stopped there because ECM-conditioned substratum was no longer available. Scale bar, 0 · 5mm.

Results based on 78 expiants are summarized in Table 2. 68 % of expiants had DMZ outgrowths toward the animal pole (AP+); in such cases, a cohesive DMZ cell sheet was deflecting 90° with respect to its animal pole to blastopore axis (Fig. 6). In this direction, DMZ cells migrated only until they reached the edges of the conditioned area. By contrast, only two expiants (3%) had DMZ outgrowths oriented toward the blastopore (BP+). Spreading in both directions on the substratum (Neutral) was observed in 19% (15 out of 78) of the expiants, with outgrowth toward the animal pole not significantly more marked. Finally, total absence of outgrowths was observed in 10 % (8 out of 78) of the expiants (no migration). Control experiments were carried out by coating the same area as in experiments with ECM-conditioned substratum with plasma FN instead. Under these conditions, DMZ outgrowths always spread regularly at the periphery of the expiants without preferential orientation (results not shown).

Table 2.

Orientation of DMZ outgrowths on conditioned substrata

Orientation of DMZ outgrowths on conditioned substrata
Orientation of DMZ outgrowths on conditioned substrata

Increase of the spreading capacity on ECM by marginal cells at the onset of their involution

The in vitro migration assays have indicated that spreading capacity was acquired differentially around the marginal zone. Dorsal cells have acquired this capacity as early as 32-cell stage. In expiants, they begin to migrate only at the beginning of gastrulation in control embryos. These observations suggest that the initiation of migration involves some regulative process that is intrinsic to mesodermal cells. Examination of the spreading capacity of dissociated DMZ and VMZ cells on ECM-conditioned substratum provides a way to test this hypothesis.

DMZ cells were dissociated at early gastrula stage (stage 8a), and VMZ cells dissociated at early gastrula, early midgastrula (stage 9) and late midgastrula (stage 10) stages, respectively. Once spread on ECM-conditioned substratum, cells flattened and extended lamellipodia, assuming a bipolar, triangular or polygonal shape (see Fig. 3C). On unconditioned plastic culture dishes (control), cells had a rounded shape and did not extend lamellipodia (not shown). Significant differences in cell spreading were noticed between DMZ and VMZ cells dissociated at early gastrula stage (Fig. 7). A high percentage of spread cells (58 % ± 8 %) was recorded for DMZ whereas only 10 % ± 2 % VMZ cells spread. As gastrulation proceeds, the percentage of spread VMZ cells increased progressively, it attained 15 % ± 2 % at the early midgastrula stage and reached 26% ±3% at the late midgastrula stage, but never reached the same level as the early gastrula DMZ cells, even at the stage where involution would normally take place (stage 10). These results suggest that DMZ possesses a stronger inherent spreading capacity than its ventral counterpart in the course of gastrulation. It is also true that there are fewer mesodermal cells in the VMZ than in the DMZ. These results were confirmed by the observation that VMZ expiants cultured on either FN-coated or ECM-conditioned substrata exhibited the same behaviour as DMZ expiants except that spreading of mesodermal cell sheet was less extensive (results not shown).

Fig. 7.

Spreading of dissociated DMZ and VMZ cells on ECM-conditioned substratum. DMZs were dissociated at the early gastrula stage (stage 8a); VMZs dissociated from the early gastrula stage, the early midgastrula stage (stage 9) and late midgastrula stage (stage 10). They were then seeded on ECM substratum conditioned by early gastrula blastocoel roof, 300 – 400 cells for each conditioned area. Attached cells with a bipolar, triangular or polygonal shape were counted at 30, 60, 90 and 120min in culture. The values represent the mean of triplicate determinations ± S.D.

Fig. 7.

Spreading of dissociated DMZ and VMZ cells on ECM-conditioned substratum. DMZs were dissociated at the early gastrula stage (stage 8a); VMZs dissociated from the early gastrula stage, the early midgastrula stage (stage 9) and late midgastrula stage (stage 10). They were then seeded on ECM substratum conditioned by early gastrula blastocoel roof, 300 – 400 cells for each conditioned area. Attached cells with a bipolar, triangular or polygonal shape were counted at 30, 60, 90 and 120min in culture. The values represent the mean of triplicate determinations ± S.D.

Reduced dorsal cell migration in u. v.-irradiated embryos

U.v. irradiation was performed to investigate further the dorsal-ventral polarity of mesodermal cell migration in relation to gastrulation. In each experiment, half of the irradiated embryos were used for migration assays and the other half was allowed to develop in order to control the alteration of dorsal polarity induced by u.v. irradiation. We observed that, at the gastrula stage, 80% of irradiated embryos bear a circular blastopore in the vegetal pole. Later, they developed as aneural or acephalic embryos.

For comparison with unirradiated control embryos, the cellular organization of irradiated embryos that displayed a circular blastopore was first observed by scanning electron microscopy (Fig. 8A,B). Fractured sections show clearly a regular distribution of bottle cells all around the vegetal pole (Fig. 8B). They were radially oriented with their contracted apices bordering the blastopore. No cells adhere to the inner surface of the blastocoel roof. Endodermal cells were pushed upward by the invaginated bottle cells. The same cellular organization was observed in all sagittal sections made through the animal-vegetal axis of u.v.irradiated embryos forming a circular blastopore. The entire marginal zone of these irradiated embryos was similar to the ventral region of unirradiated control embryos (Fig. 8A). This homology between VMZ of normal embryo and the entire marginal zone of u.v.irradiated embryos was also reflected by the in vitro migratory behaviour of equatorial cells. When animal hemispheres of u.v.-irradiated 256-cell-stage embryos were cultured on FN-coated substratum, in most cases, mesodermal cell migration was observed simultaneously in multiple regions (Fig. 8C). Later, when control embryos were at the late gastrula stage, a discontinuous mesodermal cell sheet had developed in u.v.-irradiated expiants. In contrast to control cultures where the spreading is regular around the margin of expiants, the margin of irradiated expiants had multiple regions devoid of migrating cells (Fig. 8D).

Fig. 8.

Cellular organization and mesodermal cell migration in u.v.-irradiated embryos. (A,B) Scanning electron microscopy of ceil rearrangement at the midgastrula stage. (A) Normal embryo fractured midsagittally, involuted cells attach to the inner surface of the blastocoel roof and migrate toward the animal pole. (B) Embryo irradiated at l·8–2·0×104 ergsmm−2 at 2h after fertilization and fractured when control embryos were at the midgastrula stage. A circular blastopore was formed at the vegetal pole, endodermal cells were pushed upward by the invaginated bottle cells. No cells attached to the inner surface of the blastocoel roof. (C,D) Pattern of mesodermal cell migration in animal hemispheres of irradiated embryo of the 256-cell stage on FN-coated substratum. (C) Mesodermal cell migration can be observed to occur initially at multiple sites. (D) When control embryos were at the late gastrula stage, the irradiated explant has a discontinuous mesodermal cell sheet with multiple regions devoid of migrating cells (arrowheads), bl, Blastocoel; bp, blastopore; m, mesodermal cells. Scale bars: (A,B), 250 μm; (C,D) 1mm.

Fig. 8.

Cellular organization and mesodermal cell migration in u.v.-irradiated embryos. (A,B) Scanning electron microscopy of ceil rearrangement at the midgastrula stage. (A) Normal embryo fractured midsagittally, involuted cells attach to the inner surface of the blastocoel roof and migrate toward the animal pole. (B) Embryo irradiated at l·8–2·0×104 ergsmm−2 at 2h after fertilization and fractured when control embryos were at the midgastrula stage. A circular blastopore was formed at the vegetal pole, endodermal cells were pushed upward by the invaginated bottle cells. No cells attached to the inner surface of the blastocoel roof. (C,D) Pattern of mesodermal cell migration in animal hemispheres of irradiated embryo of the 256-cell stage on FN-coated substratum. (C) Mesodermal cell migration can be observed to occur initially at multiple sites. (D) When control embryos were at the late gastrula stage, the irradiated explant has a discontinuous mesodermal cell sheet with multiple regions devoid of migrating cells (arrowheads), bl, Blastocoel; bp, blastopore; m, mesodermal cells. Scale bars: (A,B), 250 μm; (C,D) 1mm.

Therefore, u.v. irradiation has transformed the spreading migration of dorsal cells into ventral type. This effect can be observed more clearly if we repeat the same experiment as shown in Fig. 2 while using u.v.-irradiated animal hemispheres. Spreading of mesodermal cells in explanted animal hemispheres excised from u.v.-irradiated and unirradiated embryos was compared. A dramatic reduction and/or disappearance of the capacity of mesodermal cells to migrate were observed in irradiated expiants (Fig. 9). We distinguished three kinds of expiants for each stage: (1) no migrating cells at all; (2) reduced cell migration where irradiated expiants presented either few migrating cells or a discontinuous mesodermal cell sheet at the margin of explant and (3) normal cell migration. A significant percentage of the irradiated expiants derived from the 32(77%) and the 64-cell (52%) stages were totally devoid of migrating cells (Fig. 9). This result indicates that u.v. irradiation considerably reduces dorsal mesodermal cell migration. There were progressively fewer expiants devoid of migrating cells when they were removed from the 128-(36%) and the 256-cell (26%) stages (Fig. 9), however, most expiants derived from these two stages (45 % and 52 %, respectively) exhibited abnormal marginal cell migration as described above and as illustrated in Fig. 8C,D.

Fig. 9.

Histogram of the effect of u.v. irradiation on mesodermal cell migration. Embryos were irradiated at 2h after fertilization at l·8–2·0×104 ergs mm−2. Animal hemispheres were excised from the 32-cell (n = 41), the 64-cell (n = 46), the 128-cell (n = 22) and the 256-cell (n = 24) stages and were cultured on FN-coated substratum. The pattern of mesodermal cell migration was scored when control embryos were at late gastrula stage using the following criteria: (1) expiants without any migrating cells at all; (2) abnormal cell migration, expiants with either few migrating cells or an interrupted mesodermal cell sheet at multiple regions and (3) expiants with normal cell migration with respect to control expiants.

Fig. 9.

Histogram of the effect of u.v. irradiation on mesodermal cell migration. Embryos were irradiated at 2h after fertilization at l·8–2·0×104 ergs mm−2. Animal hemispheres were excised from the 32-cell (n = 41), the 64-cell (n = 46), the 128-cell (n = 22) and the 256-cell (n = 24) stages and were cultured on FN-coated substratum. The pattern of mesodermal cell migration was scored when control embryos were at late gastrula stage using the following criteria: (1) expiants without any migrating cells at all; (2) abnormal cell migration, expiants with either few migrating cells or an interrupted mesodermal cell sheet at multiple regions and (3) expiants with normal cell migration with respect to control expiants.

In a previous study, we showed that mesodermal cell migration plays a major role in gastrulation of urodele amphibian embryos (Shi et al. 1987). The DMZ of Pleurodeles waltl gastrula is initially composed of two or three layers of mesodermal cells and becomes onelayered soon after involution has begun (see also Fig. 8A). In the course of gastrulation, involuted cells migrate toward the animal pole and form the archenteron roof as one sheet of cohesive cells. In the present study, we have observed that, in vitro, the spreading and migration of mesodermal cells as a cohesive cell sheet mimics the conditions found in vivo, and thus provides a useful model to study mesodermal cell behaviour in amphibian gastrulation. This paper deals with four new aspects of this process by using an in vitro migration assay. First, the autonomous spreading and migration of early gastrula marginal zones was estab-lished using animal hemispheres taken from the 32-cell stage onward. Second, the involvements of FN in cell migration and of ECM in the orientation of early gastrula DMZ outgrowths were investigated. Third, the spreading capacity of dissociated cells on ECM from DMZ and VMZ was analysed at gastrula stages. Finally, the migration of mesodermal cells in normal and u.v.-irradiated embryos was compared.

We have found that the capacity for mesodermal cells to undergo autonomous migration on FN-coated sub-stratum was acquired as early as the 32-cell stage for DMZ, and between the 64-to the 128-cell stages for lateral and ventral marginal zones. A similar difference in the time at which DMZ and VMZ become specified to form their appropriate mesodermal cell types was also noticed in Xenopus laevis (Nakamura, 1978). When animal hemispheres containing equatorial cells were cultured, mesodermal cells derived from the dorsal equatorial blastomeres began to migrate at the time of initiation of gastrulation in control embryos if one allows for a delay in healing and attachment of expiants. These results are consistent with those reported by Gimlich (1986), who showed that, in Xenopus, dorsal equatorial cells acquired their developmental autonomy at the 32-cell stage. When these cells were transplanted to u.v.-irradiated recipients, they promoted normal gastrulation and later formed axial structures. Consequently, on the basis of in vitro migration assays in Pleurodeles and transplantation experiments in Xenopus, it seems that DMZ cells of the 32-cell embryo have already acquired the capacity to migrate autonomously at gastrulation. Nevertheless, the ability of both DMZ and VMZ cells to migrate may also be related to an increased number of specified mesodermal cells. In this respect, the possibility that vegetal inducing cells are included in the progeny of the dorsal equatorial blastomeres and may cause mesoderm induction cannot be completely excluded. It can be argued that there is little or no mesoderm on the ventral side of the explanted animal hemisphere from 32-cell stage. At this stage, mesoderm induction has probably not yet occurred (Jones & Woodland, 1987). Examination of the fate map of the 32-cell stage of Xenopus embryo shows that, on the dorsal side, the subequatorial cells give rise to a significant amount of endoderm, which may be able to cause mesoderm induction on this side. However, much less endoderm is derived from ventral subequatorial cells (Dale & Slack, 1987a). Therefore, there were probably few mesodermal cells present in the ventral side at gastrulation. Further experiments should be done to test this important point.

In other respects, the initiation of mesodermal cell migration is probably due to a programmed developmental ‘clock’ because dorsal equatorial cells of different developmental stages begin to migrate at the same time both in vitro and in control embryos. Symes & Smith (1987) reported that, in Xenopus, animal pole expiants exposed to mesoderm-inducing activity at different developmental stages exhibit gastrulation-like movements (convergent extension) at about the same time. They also showed that the gastrulation clock is not dependent upon the number of cell divisions, number of DNA replication cycles or the ratio of nuclear to cytoplasmic volume. In this study, we found that the spreading capacity of VMZ cells on ECM-conditioned substratum increases as gastrulation proceeds. Thus a correlation exists between the rate of spreading and the amplitude of mesodermal cell migration. Work on other amphibian species has shown a change in cell contact during gastrulation (Johnson, 1970) and an increasing ability of gastrula cells to adhere to FN (Johnson, 1985; Johnson & Silver, 1986; Komazaki, 1988). Therefore, the migratory capacity displayed by marginal cells at the onset of their involution is attributable primarily to an increase in the adherence and spreading on ECM-substratum.

While this paper was in preparation, Keller & Danilchik (1988) reported that, in Xenopus, the degree of extension of marginal zone is greater on the dorsal side of the gastrula. They also showed that convergent extension of the involuting marginal zone (IMZ) is an autonomous process and begins at a stage when the IMZ has involuted. These results agree well with the observations made in this study. This consistency in results suggests that expression of regional cellular activities during gastrulation in Xenopus and Pleurodeles shares common characteristics. However, marginal cells of these two species may use different mechanisms to accomplish gastrulation. An important observation made in this study is that DMZ cells of Pleurodeles need an adequate substratum to exercise autonomous migration. It may explain the previous observation that sandwich expiants of DMZ showed little convergent extension until the late gastrula stage (Shi et al. 1987).

We showed also that the marginal mesodermal cells possess the capacity to undergo autonomous attachment, spreading and migration on both FN-coated and ECM-conditioned substrata. If antibodies to FN, INT or the cell-binding peptide GRGDS were added to the medium, spreading of DMZ cells was inhibited. These results support an important role of FN and INT in mesodermal cell migration, which has been established both in vivo (Boucaut et al. 1984a,b, 1985; Darribère et al. 1988) and in vitro using other experimental systems (Johnson, 1985; Johnson & Silver, 1986; Nakatsuji, 1986). Although antibodies to LM or the cell-binding peptide YIGSR failed to inhibit migration and spreading of DMZ cells on ECM-conditioned substratum, differences exist between outgrowth from DMZ expiants on FN-coated and ECM-conditioned sub-strata. On the latter substratum, we rarely observed any isolated migrating cells as we did on the former substratum. These observations may suggest that the assemblage of ECM components such as LM and FN into ECM fibrils suppresses the random migration of mesodermal cells, and thus serves as a contact-guidance system. However, we cannot exclude the possibility that other extracellular molecules not identified presently can also fulfil this function.

The preferential spreading of DMZ outgrowths toward the animal pole observed in this study suggests that their strong autonomous spreading capacity associated with the orientation modulated by ECM are major mechanisms for gastrulation movements in Pleurodeles. These observations have two important implications. First, they may imply that differences exist in the ECM along the blastopore-animal pole axis. The substratum is more favourable for migrating mesodermal cells approaching the animal pole. Second, concerning the DMZ expiants perse, they are capable of responding to the guidance of ECM in spite of the discrepancy between the direction of migration and their original axis. They also reinforce our previous experiments showing that grafted rotated DMZ expiants were able to migrate according to the animal-vegetal axis of host embryos (Shi et al. 1987). Both in vivo and in vitro studies have shown that ECM fibrils are predominantly oriented along the blastopore-animal pole axis and provide a contact-guidance system for migrating mesodermal cells (Nakatsuji et al. 1982; Nakatsuji & Johnson, 1983a, 1984α,b; Nakatsuji, 1984). Moreover, during the gastrulation process, epibolic movement of the prospective ectoderm could be involved in the orientation of such ECM fibrils (Nakatsuji, 1984). However, this process does not rule out the possibility that regional differences in ECM components might exist and play a role in directing the oriented movement of migrating mesodermal cells. Further microsurgical experiments and biochemical analysis are needed to understand this important mechanism of cell guidance during amphibian gastrulation.

The u.v. irradiation of fertilized amphibian eggs interferes with the determination of axial structures (Grant & Wacaster, 1972; Malacinski et al. 1975, 1977; Scharf & Gerhart, 1980, 1983; Youn & Malacinski, 1981). It also inhibits grey crescent formation. This early effect may be involved in the subsequent developmental abnormalities observed in u.v.-irradiated embryos (Manes & Elinson, 1980). In Pletuodeles, we have found that u.v. irradiation was associated with a complete inhibition or a pronounced decrease in DMZ cell migration in animal hemispheres excised at either the 32- or the 64-cell stage. In later stages, there is an obvious synchrony in the migratory capacity of mesodermal cells everywhere in the margins of expiants. Thus, changes in DMZ cell migration may explain the absence of dorsal blastoporal lip formation and the development of the circular blastopore observed in u.v.-irradiated embryos (see Fig. 8B). The results from scanning electron microscopic analysis and in vitro migration assays indicate that the entire marginal zone of u.v.-irradiated embryos behaves like the VMZ of normal embryos. Moreover, it has been shown that all meridians of an irradiated embryo produce mesoderm founder cells in a fashion that is typical of the slowestdeveloping posterior/ventral sector of the normal embryo (Cooke & Smith, 1987). An alternative example is the dorsoanterior-enhanced embryos produced by lithium treatment. In this case, the entire mesodermal mantle is composed of chordomesoderm and represents the long axis of the embryo (Kao & Elinson, 1988). In contrast to u.v.-irradiated embryos which form rudimentary gut (grade 5), the invagination in lithiumtreated embryos leads to the elongation of a symmetrical archenteron. Therefore, our present results support the proposition that the predominance of cell migration in DMZ over lateral and ventral zones accounts for the graded times of onset of gastrulation found in normal embryos.

In conclusion, we have provided evidence that mi-gration of mesodermal cells appears to be an autonomous process, their interactions with FN and ECM are a crucial factor causing dorsal cells to migrate toward the animal pole. The initiation of mesodermal cell migration is correlated with an increase in the adherence and subsequent spreading on the ECM. Thus, it will be of interest to investigate the developmental expression of molecules involved in these processes in order to understand the molecular control of mesodermal cell migration in amphibian embryos.

We are grateful to Dr J. P. Thiery and Dr K. Yamada for generous gifts of antibodies and synthetic peptides. We would like to thank Dr H. Boulekbache and F. Meury for SEM observations, J. Dérosiers for illustrations and P. Grevé for his help. This work was supported by grants from the CNRS and FRM.

Boucaut
,
J. C.
&
Darribère
,
T.
(
1983
).
Fibronectin in early amphibian embryos: migrating mesodermal cells contact fibronectin established prior to gastrulation
.
Cell Tissue Res
.
234
,
135
145
.
Boucaut
,
J. C.
,
Darribère
,
T.
,
Boulekbache
,
H.
&
Thiery
,
J. P.
(
1984a
).
Prevention of gastrulation but not neurulation by antibodies to fibronectin in amphibian embryos
.
Nature, Lond
.
307
,
364
367
.
Boucaut
,
J. C.
,
Darribère
,
T.
,
Poole
,
T. J.
,
Aoyama
,
H.
,
Yamada
,
K. M.
&
Thiery
,
J. P.
(
1984b
).
Biologically active synthetic peptides as probe of embryonic development: a competitive peptide inhibitor of fibronectin function inhibits gastrulation in amphibian and neural crest cell migration in avian embryos
.
J. Cell Biol
.
99
,
1822
1830
.
Boucaut
,
J. C.
,
Darribère
,
T.
,
Shi
,
D. L.
,
Boulekbache
,
H.
,
Yamada
,
K. M.
&
Thiery
,
J. P.
(
1985
).
Evidence for the role of fibronectin in amphibian gastrulation
.
J. Embryol. exp. Morph
.
89
,
Supplement
,
211
227
.
Brackenbury
,
R.
,
Thiery
,
J. P.
,
Ruthishauser
,
U.
&
Edelman
,
G. M.
(
1977
).
Adhesion among neural crest cells of the chick embryo. I. An immunological assay for molecules involved in cell-cell binding
.
J. biol. Chem
.
252
,
6835
6840
.
Cooke
,
J.
&
Smith
,
J. C.
(
1987
).
The midblastula cell cycle transition and the character of mesoderm in u.v.-induced nonaxial Xenopus development
.
Development
99
,
197
210
.
Dale
,
L.
&
Slack
,
J. M. W.
(
1987a
).
Fate map for the 32-cell stage of Xenopus laevis
.
Development
99
,
527
551
.
Dale
,
L.
&
Slack
,
J. M. W.
(
1987b
).
Regional specification within the mesoderm of early embryos of Xenopus laevis
.
Development
100
,
279
295
.
Dale
,
L.
,
Smith
,
J. C.
&
Slack
,
J. M. W.
(
1985
).
Mesoderm induction in Xenopus laevis; a quantitative study using a cell lineage label and tissue-specific antibodies
.
J. Embryol. exp. Morph
.
89
,
289
312
.
Darribère
,
T.
,
Boulekbache
,
H.
,
Shi
,
D. L.
&
Boucaut
,
J. C.
(
1985
).
Immuno-electron microscopic study of fibronectin in gastrulating amphibian embryos
.
Cell Tissue Res
.
239
,
75
80
.
Darribère
,
T.
,
Riou
,
J. F.
,
Shi
,
D. L.
,
Delarue
,
M.
&
Boucaut
,
J. C.
(
1986
).
Synthesis and distribution of laminin-related polypeptides in early amphibian embryos
.
Cell Tissue Res
.
246
,
45
51
.
Darribère
,
T.
,
Yamada
,
K. M.
,
Johnson
,
K. E.
&
Boucaut
,
J. C.
(
1988
).
The 140-kDa fibronectin receptor complex is required for mesodermal cell adhesion during gastrulation in the amphibian Pleurodeles waltl
.
Devi Biol
.
126
,
182
194
.
Gerhart
,
J.
&
Keller
,
R. E.
(
1986
).
Region-specific cell activities in amphibian gastrulation
.
Ann. Rev. Cell Biol.
2
,
201
229
.
Gimlich
,
R. L.
(
1986
).
Acquisition of developmental autonomy in the equatorial region of the Xenopus embryo
.
Devi Biol
.
115
,
340
352
.
Grant
,
P.
&
Wacaster
,
J. F.
(
1972
).
The amphibian grey crescent region - A site of developmental information?
Devi Biol
.
28
,
454
471
.
Gurdon
,
J. B.
,
Fairman
,
S.
,
Mohun
,
T. J.
&
Brennan
,
S.
(
1985a
).
Activation of muscle-specific actin genes in Xenopus development by an induction between animal and vegetal cells of a blastula
.
Cell
41
,
913
922
.
Gurdon
,
J. B.
,
Mohun
,
T. J.
,
Fairman
,
S.
&
Brennan
,
S.
(
1985b
).
All components required for the eventual activation of muscle-specific actin genes are localized in the subequatorial region of an uncleaved amphibian egg
.
Proc. Natl. Acad. Sci. USA
82
,
139
143
.
Johnson
,
K. E.
(
1970
).
The role of changes in cell contact behavior in amphibian gastrulation
.
J. Exp. Zool
.
175
,
319
428
.
Johnson
,
K. E.
(
1985
).
Frog gastrula cells adhere to fibronectin-Sepharose beads
.
In Molecular Determinants of Animal Form
(ed.
G. M.
Edelman
), pp.
291
306
.
New York
:
A. R. Liss
.
Johnson
,
K. E.
&
Silver
,
M. H.
(
1986
).
Cells from Xenopus laevis gastrula adhere to FN-Sepharose beads and other lectin coated beads
.
Scanning Electron. Microsc
.
2
,
671
678
.
Johnson
,
K. E.
,
Nakatsuji
,
N.
&
Boucaut
,
J. C.
(
1988
).
Extracellular matrix control of cell migration during amphibian gastrulation
.
In Cytoplasmic Information Systems
(ed.
G. M.
Malacinski
).
New York
:
Macmillan Pub. Co. (In press
).
Jones
,
E. A.
&
Woodlond
,
H. R.
(
1987
).
The development of animal cap cells in Xenopus: a measure of the start of animal cap competence to form mesoderm
.
Development
101
,
557
563
.
Kao
,
K. R.
&
Elinson
,
R. P.
(
1988
).
The entire mesodermal mantle behaves as Spemann’s organizer in dorsoanterior enhanced Xenopus laevis embryos
.
Devi Biol
.
127
,
64
77
.
Keller
,
R. E.
,
Danilchik
,
M.
,
Gimlich
,
R.
&
Shih
,
J.
(
1985
).
The function and mechanism of convergent extension during gastrulation of Xenopus laevis
.
J. Embryol. exp. Morph
.
89
,
Supplement
,
185
209
.
Keller
,
R. E.
&
Danilchik
,
M.
(
1988
).
Regional expression, pattern and timing of convergence and extension during gastrulation of Xenopus laevis
.
Development
103
,
193
209
.
Komazaki
,
S.
(
1988
).
Factors related to the initiation of cell migration along the inner surface of the ?astocoelic wall during amphibian gastrulation
.
Cell Differ
.
24
,
25
32
.
Lee
,
G.
,
Hynes
,
R.
&
Kirschner
,
M.
(
1984
).
Temporal and spatial regulation of fibronectin in early Xenopus development
.
Cell
36
,
729
740
.
Lundmark
,
C.
(
1986
).
Role of bilateral zones of ingressing superficial cells during gastrulation of Ambystoma mexicanum
.
J. Embryol. exp. Morph
.
97
,
47
62
.
Malacinski
,
G. M.
,
Benford
,
H.
&
Chung
,
H. M.
(
1975
).
Association of an ultraviolet irradiation sensitive cytoplasmic localization with the future dorsal side of the amphibian egg
.
J. exp. Zool
.
191
,
97
110
.
Malacinski
,
G. M.
,
Brothers
,
J.
&
Chung
,
H. M.
(
1977
).
Destruction of components of the neural induction system of the amphibian egg with ultraviolet irradiation
.
Devi Biol
.
56
,
24
39
.
Manes
,
E. M.
&
Elinson
,
R. P.
(
1980
).
Ultraviolet light inhibits grey crescent formation on the frog egg
.
Wilhelm Roux’s Arch, devl Biol
.
189
,
73
76
.
Nakamura
,
O.
(
1978
).
Epigenetic formation of the organizer
.
In Organizer. A Milestone of a Half Century from Spemann
(ed.
O.
Nakamura
&
S.
Toivonen
), pp.
179
220
.
Amsterdam
:
Elsevier/North-HolLond
.
Nakatsuji
,
N.
(
1975
).
Studies on the gastrulation of amphibian embryos: Light and electron microscopic observation of an urodele Cynops pyrrhogaster
.
J. Embryol. exp. Morph
.
34
,
669
685
.
Nakatsuji
,
N.
(
1984
).
Cell locomotion and contact guidance in amphibian gastrulation
.
Am. Zool
.
24
,
615
627
.
Nakatsuji
,
N.
(
1986
).
Presumptive mesoderm cells from Xenopus laevis gastrulae attach to and migrate on substrata coated with fibronectin or laminin
.
J. Cell Sci
.
18
,
109
118
.
Nakatsuji
,
N.
,
Gould
,
A. C.
&
Johnson
,
K. E.
(
1982
).
Movements and guidance of migrating mesodermal cells in Ambystoma maculatum gastrulae
.
J. Cell Sci
.
56
,
207
222
.
Nakatsuji
,
N.
&
Johnson
,
K. E.
(
1983a
).
Conditioning of a culture substratum by the ectodermal layer promotes attachment and oriented locomotion by amphibian gastrula mesodermal cells
.
J. Cell Sci
.
59
,
43
60
.
Nakatsuji
,
N.
&
Johnson
,
K. E.
(
1983b
).
Comparative study of extracellular fibrils on the ectodermal layer in gastrulae of five amphibian species
.
J. Cell Sci
.
59
,
61
70
.
Nakatsuji
,
N.
&
Johnson
,
K. E.
(
1984a
).
Ectodermal fragments from normal frog gastrulae condition substrata to support normal and hybrid mesodermal cell migration in vitro
.
J. Cell Sci
.
68
,
49
67
.
Nakatsuji
,
N.
&
Johnson
,
K. E.
(
1984b
).
Experimental manipulation of a contact guidance system in amphibian gastrulation by mechanical tension
.
Nature, Lond
.
307
,
453
455
.
Nakatsuji
,
N.
,
Smolira
,
M. A.
&
Wylie
,
C. C.
(
1985a
).
Fibronectin visualized by scanning electron microscopy immunocytochemistry on the substratum for cell migration in Xenopus laevis gastrula
.
Devi Biol
.
107
,
264
268
.
Nakatsuji
,
N.
,
Hashimato
,
K.
&
Hayashi
,
M.
(
1985b
).
Laminin fibrils in the newt gastrula visualized by the immunofluorescent staining
.
Dev Growth Differ
.
27
,
639
643
.
Nieuwkoop
,
P. D.
(
1969a
).
The formation of the mesoderm in the urodelean amphibians. I. Induction by the endoderm
.
Wilhelm Roux’s Arch. EntwMech. Org
.
162
,
341
373
.
Nieuwkoop
,
P. D.
(
1969b
).
The formation of the mesoderm in the urodelean amphibians. II. The origin of the dorso-ventral polarity of the mesoderm
.
Wilhelm Roux Arch. EntwMech. Org
.
163
,
298
315
.
Nieuwkoop
,
P. D.
(
1977
).
Origin and establishment of embryonic polar axes in amphibian development
.
Curr. Top. devl. Biol
.
11
,
115
.
Rjou
,
J. F.
,
Darribère
,
T.
,
Shi
,
D. L.
,
Richoux
,
V.
&
Boucaut
,
J. C.
(
1987
).
Synthesis of laminin-related polypeptides in oocytes, eggs and early embryos of the amphibian Pleurodeles waltl
.
Wilhelm Roux’s Arch, devl Biol
.
196
,
328
332
.
Scharf
,
S. R.
&
Gerhart
,
J. C.
(
1980
).
Determination of the dorso-ventral axis in eggs of Xenopus laevis: Complete rescue of uv-impaired eggs by oblique orientation before first cleavage
.
Devl Biol.
79
,
181
198
.
Scharf
,
S. R.
&
Gerhart
,
J. C.
(
1983
).
Axis determination in eggs of Xenopus laevis: A critical period before first cleavage, identified by the common effects of cold, pressure and ultraviolet irradiation
.
Devl Biol
.
99
,
75
87
.
Shi
,
D. L.
,
Delarue
,
M.
,
Darribère
,
T.
,
Rjou
,
J. F.
&
Boucaut
,
J. C.
(
1987
).
Experimental analysis of the extension of the dorsal marginal zone in Pleurodeles waltl gastrula
.
Development
100
,
147
161
.
Smith
,
J. C.
&
Slack
,
J. M. W.
(
1983
).
Dorsalization and neural induction: Properties of the organizer in Xenopus laevis
.
J. Embryol. exp. Morph
.
78
,
299
317
.
Sudarwati
,
S.
&
Nieuwkoop
,
P. D.
(
1971
).
Mesoderm formation in the anuran Xenopus laevis (Daudin)
.
Wilhelm Roux’s Arch. EntwMech. Org
.
166
,
189
201
.
Symes
,
K.
&
Smith
,
J. C.
(
1987
).
Gastrulation movements provide an early marker of mesoderm induction in Xenopus laevis
.
Development
101
,
339
349
.
Vincent
,
J. P.
&
Gerhart
,
J. C.
(
1987
).
Subcortical rotation in Xenopus egg: an early step in embryonic axis specification
.
Devl Biol
.
123
,
526
539
.
Youn
,
B. W.
&
Malacinski
,
G. M.
(
1981
).
Axial structure development in ultraviolet-irradiated (notochord defective) amphibian embryos
.
Devl Biol
.
83
,
339
352
.