We have determined the fate of presumptive mesodermal cells in the early Pleurodeles waltl gastrula. We labeled all cells in a gastrula with RLDx cell lineage tracer and superficial cells with 125I and then grafted small pieces of the marginal zone orthotopically into unlabeled host embryos. Labeled progeny were identified in sectioned embryos at the tail bud stage. The use of double-labeled grafts allowed us to study the relative contributions by superficial and deep cells to different derivatives. We found that the presumptive regions are generally distributed according to classical fate maps for urodeles but that the boundaries between presumptive regions are indistinct, due to extensive intermingling between cells at the edges of grafted regions. We have shown that there is a high dorsal to low ventral gradient of mixing between superficial and deep cells.

The early amphibian embryo is the classical embryological material for studying the interactions that lead to the formation of the three primary germ layers: ectoderm, mesoderm and endoderm. Induction of mesoderm is the first significant interaction during amphibian cleavage. Mesoderm arises as a result of an interaction between endodermal cells and animal pole cells (Nieuwkoop, 1969; Sudarwati and Nieuwkoop, 1971; Dale and Slack, 1987; Gurdon et al., 1989; Smith et al., 1989). After mesodermal induction, gastrulation begins. Gastrulation is a process during which extensive morphogenetic cell movements, including epiboly of the animal cap, bottle cell formation, cell migration on the basal surface of the blastocoel roof and convergent extension, lead to a complete rearrangement of the cells of the embryo (Slack, 1983; Keller, 1985; Gerhart and Keller, 1986; Cooke, 1987). The morphogenetic movements of gastrulation occur in a coordinated sequence so that movements in one part of the embryo have complementary counterparts in another region (Holtfreter, 1943, 1944; Keller and Schoenwolf, 1977; Boucaut et al., 1990).

A comprehensive study of gastrulation requires accurate fate maps of the early gastrula and concomitant knowledge of the paths of morphogenetic cell movements. Gastrulation in urodele embryos has been studied extensively by Vogt (1929) and Pasteels (1942) using the application of vital dyes to the outer surface of the embryo. Their methods were not refined enough to give detailed information concerning the mixing of cells and the relative contributions of deep and superficial cells in areas corresponding to the mesoderm in the early gastrula. Fate mapping in the anuran Xenopus revealed that mesodermal cells are derived from deep blastomeres with superficial ones forming exclusively ectoderm and endoderm (Keller, 1975, 1976; Lovtrup, 1975; Landström and Løvtrup, 1979). More recently, Smith and Malacinski (1983) labeled superficial cells with Bolton-Hunter reagent (Katz et al., 1982) and confirmed Keller’s observations on the deep origin of mesoderm in Xenopus but showed that in the urodele Amby stoma, both deep and superficial cells contribute to mesodermal derivatives. In the present study, we have obtained more detailed information of the fates of circumblastoporal cells in the Pleurodeles waltl early gastrula. We used orthotopic grafts of small fragments of tissue where all cells were labeled with a fluorescent lineage tracer and superficial cells were labeled with Bolton-Hunter reagent and grafts were placed in unlabeled host embryos. Injection of single superficial cells with lineage tracer showed that our grafting methods accurately define the boundary between suprablastoporal endoderm and mesoderm. We found that the fate maps published by Vogt (1929) are approximately accurate for mesodermal derivatives, although we discovered that there is cellular intermixing at the boundaries of graft and host. In addition, we found that superficial cells made an extensive contribution to mesodermal derivatives in the dorsal lip region with a gradual diminution of superficial cell contribution to mesodermal structures in the lateral and ventral marginal zones.

Embryos

Embryos of Pleurodeles waltl were obtained from natural matings in the laboratory. They were reared at 18°C and staged according to Gallien and Durocher (1957). Jelly and vitelline membranes were manually removed with forceps. Control and experimental embryos developed in 10% normal amphibian medium (NAM) (Slack, 1984). Gentamicin (50 μg/ml) was added to NAM to prevent bacterial growth.

Labelling procedures

Fertilized uncleaved eggs were microinjected in the animal half with 20 nl of rhodamine-lysine-dextran (RLDx) tracer (50 mg/ml in distilled water) (Gimlich and Gerhart, 1984). Microinjections were made with a Leitz micromanipulator using a PLI-100 microinjection system (Medical Systems, Greenvale, NY). After microinjection, embryos were transferred to 10% NAM and allowed to develop in the dark to avoid bleaching of fluorochrome. For superficial cell labeling, batches of RLDx-labeled late blastulae (stage 7) were incubated for 5 minutes at 18°C in Bolton-Hunter reagent (Amersham) (400 μCi/ml) in 10% NAM. Radioactively labeled blastulae were rinsed twice and allowed to develop further in 10% NAM. The specificity of surface labeling was determined by autoradiography on sections (Fig. 1A and B).

Fig. 1.

(A,B) Autoradiograph of a section of a blastula (Stage 7) labeled with Bolton-Hunter reagent. (A) Animal region. (B) Vegetal region. Notice that only superficial cells are radioactive, ap, animal pole; sc, superficial cells; de, deep cells; bl, blastocoel; vp, vegetal pole. (C) Fluorescent micrograph of an early gastrula (Stage 8a) showing an RLDx-labeled graft (arrow) in the dorsal marginal zone. Bars (A,B) 50 μm. Bar (C) 0.5 mm.

Fig. 1.

(A,B) Autoradiograph of a section of a blastula (Stage 7) labeled with Bolton-Hunter reagent. (A) Animal region. (B) Vegetal region. Notice that only superficial cells are radioactive, ap, animal pole; sc, superficial cells; de, deep cells; bl, blastocoel; vp, vegetal pole. (C) Fluorescent micrograph of an early gastrula (Stage 8a) showing an RLDx-labeled graft (arrow) in the dorsal marginal zone. Bars (A,B) 50 μm. Bar (C) 0.5 mm.

Cell lineage

Single suprablastoporal and sublastoporal cells in the early gastrula stage were microinjected with 0.5 nl of the RLDx tracer. The fate of the clone of cells derived from this single cell was determined. We injected bottle cells (be), sublastoporal cells 1 and 2 cells below the dorsal lip of the blastopore (endo 1 and endo 2) or suprablastoporal cells 1–4 cells removed from the dorsal lip of the blastopore (dll, dl2, dl3 and dl4) (Fig. 3A).

Fig. 2.

Schematic diagram showing the placement of grafts in early gastrulae (Stage 8a). Dark shading, RLDx- and 125l-labeled grafts. Light shading, RLDx-labeled grafts. The dorsal lip of the blastopore is indicated by the arrow. (A) Dorsal marginal zone. (B) Vegetal region.

Fig. 2.

Schematic diagram showing the placement of grafts in early gastrulae (Stage 8a). Dark shading, RLDx- and 125l-labeled grafts. Light shading, RLDx-labeled grafts. The dorsal lip of the blastopore is indicated by the arrow. (A) Dorsal marginal zone. (B) Vegetal region.

Fig. 3.

(A) Diagram of location of single cells injected with RLDx in the early gastrula (Stage 8a). be, bottle cells; endo 1 and 2, sublastoporal cells, 1 and 2 tiers below be; dll-dl4, suprablastoporal cells, 1–4 tiers above be. (B) Diagram showing the fates of singly-injected cells at the tail bud stage. Triangles, endo 1 and 2; circles, bottle cells; plus sign, dll; vertical dash, d12; horizontal dash, dl3, small dots, dl4. br, brain; end, endoderm; st, stomach; 1, liver, p, pharynx; nc, notochord. (C-E) Photomicrographs of transverse sections showing the fates of singly-injected cells. (C) dl4. Notice that the fluorescent cells are restricted to the cranial notochord. (D) dl3. Notice that the fluorescent cells are restricted to the cranial pharynx. (E) dl2. Notice that the fluorescent cells are restricted to the pharyngeal floor at the heart level. (F) dll. Notice that the fluorescent cells are restricted to the liver. (G) Bottle cells. Notice that the fluorescent cells are restricted to the ventral wall of endoderm. (H) endo 1 and endo 2. Notice that the fluorescent cells are restricted to the endodermal mass at the truncal level, br, brain; cv, cardiac vesicle; end, endoderm; 1, liver; nc, notochord; p, pharynx; st, stomach. Bars for A, 50 μm; B, 0.5 mm; C-H, 100 μm.

Fig. 3.

(A) Diagram of location of single cells injected with RLDx in the early gastrula (Stage 8a). be, bottle cells; endo 1 and 2, sublastoporal cells, 1 and 2 tiers below be; dll-dl4, suprablastoporal cells, 1–4 tiers above be. (B) Diagram showing the fates of singly-injected cells at the tail bud stage. Triangles, endo 1 and 2; circles, bottle cells; plus sign, dll; vertical dash, d12; horizontal dash, dl3, small dots, dl4. br, brain; end, endoderm; st, stomach; 1, liver, p, pharynx; nc, notochord. (C-E) Photomicrographs of transverse sections showing the fates of singly-injected cells. (C) dl4. Notice that the fluorescent cells are restricted to the cranial notochord. (D) dl3. Notice that the fluorescent cells are restricted to the cranial pharynx. (E) dl2. Notice that the fluorescent cells are restricted to the pharyngeal floor at the heart level. (F) dll. Notice that the fluorescent cells are restricted to the liver. (G) Bottle cells. Notice that the fluorescent cells are restricted to the ventral wall of endoderm. (H) endo 1 and endo 2. Notice that the fluorescent cells are restricted to the endodermal mass at the truncal level, br, brain; cv, cardiac vesicle; end, endoderm; 1, liver; nc, notochord; p, pharynx; st, stomach. Bars for A, 50 μm; B, 0.5 mm; C-H, 100 μm.

Grafting, histology and autoradiography

Embryos for grafting were transferred to 1% agarose-coated dishes containing 100% NAM. When donors and hosts reached the earliest gastrula stage (Stage 8a), i.e. when a faint trace of the blastopore was visible, we positioned embryos under a square ocular reticle in the Leitz stereomicroscope, removed a defined 0.2 mm × 0.2 mm graft, and transferred it to the same location in an unlabeled host embryo after removing the same region from the host (Fig. 1C). Our reticle was arbitrarily labeled: I, II or III for suprablastoporal regions progressively further from the dorsal lip of the blastopore, 1, 2, 3, 4, 5, 6 and 7 for sublast oporal regions progressively further from the dorsal lip; and A, B, C and D for regions progressively more lateral to the dorsal lip (Fig. 2). We grafted either single RLDx-labeled or double RLDx-125I-labeled grafts into unlabeled hosts. Grafts ranged in thickness from 70 μm thick near the animal pole to 120 μm thick near the blastopore. We made 216 different grafts into 32 different regions of hosts. In all, we grafted 152 different single-labeled grafts and 64 double-labeled grafts into unlabeled hosts. RLDx-injected embryos were examined histologically and we found that all cells were fluorescent. Double-labeled embryos develop normally up to the tail-bud stage (stage 28).

After healing for 20 minutes, host embryos were transferred to 10% NAM, allowed to develop until the tail-bud stage (stage 28) and fixed overnight at 4°C in 3.7% formaldehyde in 10% NAM. Fixed embryos were then dehydrated in graded ethanol, embedded in polyethylene glycol distearate 400 (Koch-Light), and sectioned at 7 μm. For autoradiography, sections were coated with 1:1 Ilford L4 emulsion, exposed for 3–5 days at 4°C, and developed in D 19 (Kodak). Sections were then mounted in Mowiol (Hoechst).

Notochordal map

The notochord is formed entirely from regions in the dorsal marginal zone, particularly those along the plane of bilateral symmetry (Fig. 4A, Table 1). Most of the notochord is derived from regions II A and III A. Region II A contributes mostly to the anterior notochord and only a little to the posterior notochord while region III A contributes mostly to the posterior notochord but also makes a minor contribution to the anterior notochord. Other regions make a minor contribution to the notochord. For example, I B forms cephalic notochord, II B and II C contribute a few scattered cells to anterior notochord, and III B contributes in a minor way to posterior notochord. Careful examination of sections of the notochord shows that fluorescent cells are grouped tightly together in the central portion of a derived structure but that there is considerable intermingling of fluorescent and non-fluorescent cells at the boundaries of a particular derivative (Fig. 4A, Fig. 5). This suggests a given region of the notochord is derived from a single explant but that there is considerable intermingling occurring between cells during morphogenesis, especially at the margins of the explant. When we studied hosts receiving double-labeled explants, it is clear that notochord is derived from both superficial and deep cells of explants II A-B, and III A-B. For example, in II A-B, about 30% of the notochord cells are derived from superficial cells while 70% are derived from deep cells.

Table 1.

Graft fates in tail-bud (stage 28) Pleurodeles waltl embryos

Graft fates in tail-bud (stage 28) Pleurodeles waltl embryos
Graft fates in tail-bud (stage 28) Pleurodeles waltl embryos
Fig. 4.

Summary diagram illustrating the fates of different regions of the early gastrula. Major contributions are show in dark shading and minor contributions are shown in light shading. (A) Notochord; (B) somites; (C) head mesenchyme; (D) pronephros; (E) lateral plate mesoderm; (F) heart; (G) blood islands; (H) caudal mesoderm.

Fig. 4.

Summary diagram illustrating the fates of different regions of the early gastrula. Major contributions are show in dark shading and minor contributions are shown in light shading. (A) Notochord; (B) somites; (C) head mesenchyme; (D) pronephros; (E) lateral plate mesoderm; (F) heart; (G) blood islands; (H) caudal mesoderm.

Fig. 5.

Photomicrographs of longitudinal sections at the anterior level of a tail-bud (Stage 28) embryo receiving a doublelabeled graft in region n A. These two photographs are congruent with identical points marked with arrows. (A) Fluorescent graft cells are distributed throughout the anterior notochord (nc) and head mesenchyme and are mixed with non-fluorescent host cells. (B) Radioactive cells are congruent with some fluorescent graft cells, indicating that superficial cells contribute to the notochord and head mesenchyme, end, endoderm; p, pharynx. Bar, 50 μm

Fig. 5.

Photomicrographs of longitudinal sections at the anterior level of a tail-bud (Stage 28) embryo receiving a doublelabeled graft in region n A. These two photographs are congruent with identical points marked with arrows. (A) Fluorescent graft cells are distributed throughout the anterior notochord (nc) and head mesenchyme and are mixed with non-fluorescent host cells. (B) Radioactive cells are congruent with some fluorescent graft cells, indicating that superficial cells contribute to the notochord and head mesenchyme, end, endoderm; p, pharynx. Bar, 50 μm

Somitic map

The somites are derived from regions in the dorsal and lateral marginal zones. The somites are derived from regions IC-D, IIA-D, III A-C, 1 C-D, 2 D and 3 D with most of the somites coming from I C-D, II B-C and III B-C (Fig. 4B, Table 1). In the myotomes, most cells are fluorescent although there is also considerable intermingling between fluorescent and non-fluorescent cells. Anterior myotomes are derived from I C-D and II B-C and posterior myotomes are derived from III B-C. Sclerotomes and dermatomes are derived from the lateral portions of regions I D, II C-D, 1 D, 2 D and 3 D. Our results employing double-labeled explants also show that somites are derived from both superficial and deep cells. Mixing of superficial and deep cells is particularly striking in myotomes (Fig. 6). The superficial cells contribute from 18.2% for trunk myotomes (region 1 D) to 53.4% for anterior myotomes (region II A).

Fig. 6.

Photomicrographs of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a doublelabeled graft in region n C. (A) Fluorescent cells are mainly found in the myotome (m). There are also fluorescent cells in the sclerotome (arrow) between the myotome and the neural tube (nf). nc, notochord. (B) Radioactive cells are mingled among fluorescent cells in the myotome and sclerotome. Bar, 10 μm.

Fig. 6.

Photomicrographs of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a doublelabeled graft in region n C. (A) Fluorescent cells are mainly found in the myotome (m). There are also fluorescent cells in the sclerotome (arrow) between the myotome and the neural tube (nf). nc, notochord. (B) Radioactive cells are mingled among fluorescent cells in the myotome and sclerotome. Bar, 10 μm.

Head mesenchymal map

Cells of the presumptive head mesoderm are localized in an extended area in the dorsal and lateral marginal zones, forming a loose collection of mesenchymal cells lying between the major cephalic organs. Head mesoderm is formed by minor contributions from regions I B-D, II A-D, III C, 1 C-D, 2 C-D and 3 D (Fig. 4C, Table 1). Head mesenchyme around the myelencephalon is derived from I B-D, II A-B, III C, 1 C, 2 D and 3 D while head mesenchyme around the telencephalon and mesencephalon is derived from I D, IIB-C, 1 C and 2 C. Head mesenchyme around optic and otic vesicles is derived from region I B. Once again, there is substantial mixing between superficial and deep cells (Fig. 7). Around the otic vesicles, we observed that superficial cells form medial derivatives while deep cells form lateral derivatives.

Fig. 7.

Photomicrographs of transverse sections at through the otic vesicle (ov) of a tail-bud (Stage 28) receiving a double-labeled graft in region I B. (A) Fluorescent cells are distributed in head mesenchyme (hm) in the spaces between the myelencephalon (my), pharynx (p) and otic vesicle. (B) Radioactive cells (arrows) are near the notochord (nc). Bar, 50 μm.

Fig. 7.

Photomicrographs of transverse sections at through the otic vesicle (ov) of a tail-bud (Stage 28) receiving a double-labeled graft in region I B. (A) Fluorescent cells are distributed in head mesenchyme (hm) in the spaces between the myelencephalon (my), pharynx (p) and otic vesicle. (B) Radioactive cells (arrows) are near the notochord (nc). Bar, 50 μm.

Pronephric map

The pronephros is derived from regions in the lateral marginal zone, i.e. major contributions from regions I D, II D, and 1 D (anterior glomeruli and pronephric tubules) with minor contributions derived from regions 2 D, 3 C (posterior glomeruli and pronephric tubules), 4 C and 5 D (pronephric duct) (Fig. 4D, Table 1). There is much less intermingling of fluorescent cells in pronephric structures (Fig. 8). Approximately 20% of the pronephric cells are derived from superficial cells and the rest are derived from deep cells.

Fig. 8.

Photomicrograph of transverse sections near the pronephros of a tail-bud (Stage 28) receiving a RLDx-labeled graft in region n D. Fluorescent cells are arranged in the wall of the pronephric duct (arrow) and other parts of the nephrotome (net), end, endoderm; s, somite. Bar, 50 μm.

Fig. 8.

Photomicrograph of transverse sections near the pronephros of a tail-bud (Stage 28) receiving a RLDx-labeled graft in region n D. Fluorescent cells are arranged in the wall of the pronephric duct (arrow) and other parts of the nephrotome (net), end, endoderm; s, somite. Bar, 50 μm.

Lateral plate mesodermal map

The lateral plate mesoderm arises from extended regions in the lateral and ventral marginal zones. Most of the cells in the lateral plate arise in regions 3 C, 4 C-D, 5 C and 6 C and the rest come from 2 D, 3 D, 5 D and 6 B (Fig. 4E, Table 1). Vascular endothelium in the dorsal aorta (3 C) and the cardinal vein (4 C, 5 C, and 6 C) are also derived from the regions that make a major contribution to the lateral plate mesoderm. Most of the lateral plate mesodermal cells are derived from deep cells (Fig. 9). For example, in region 4 C, only 13.8% of the cells are derived from the superficial layer while the remaining cells are derived from the deep layer.

Fig. 9.

Photomicrograph of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a RLDx-labeled graft in region 4 C. Fluorescent cells are mainly restricted to the lateral plate mesoderm (Ip), although a few are also found in the endoderm (end). Bar, 50 μm.

Fig. 9.

Photomicrograph of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a RLDx-labeled graft in region 4 C. Fluorescent cells are mainly restricted to the lateral plate mesoderm (Ip), although a few are also found in the endoderm (end). Bar, 50 μm.

Cardiac map

The endocardium, myocardium and pericardium all arise from highly restricted lateral regions from row D, (1 and 2 major; I, II and 3 minor) (Fig. 4F, Table 1). These presumptive regions also are closely related to the lateral plate mesodermal and pronephric regions. Double-label explants show that most of the cardiac cells are derived from deep cells (Fig. 10). For example, in regions 1 D and 2 D only 18.2% and 11.3%, respectively, of the cells are derived from superficial cells with the remaining cells arising from deep cells.

Fig. 10.

Photomicrographs of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a doublelabeled graft in region 1 D. (A) Fluorescent cells are found in the pericardium (pc), and endocardium (ec). (B) Only a few radioactive cells (arrows) are found among the fluorescent cells, me, myocardium. Bar, 50 μm.

Fig. 10.

Photomicrographs of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a doublelabeled graft in region 1 D. (A) Fluorescent cells are found in the pericardium (pc), and endocardium (ec). (B) Only a few radioactive cells (arrows) are found among the fluorescent cells, me, myocardium. Bar, 50 μm.

Blood island map

The blood islands are derived from several regions spread throughout the lateral and ventral marginal zones. Most of the blood islands come from regions 5 B and 6 A-B (Fig. 4G, Table 1) while the rest come from regions 3 C, 4 D, 5 C and 7 A (Fig. 11). We transplanted a doubled-labeled 6 A region and found very few superficial cells (1.3%) forming blood islands. Most of the superficial cells from region 6 A end up in either proctodeal endoderm or ventral epidermis. Most of the blood islands are derived from deep cells.

Fig. 11.

Photomicrograph of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a RLDx-labeled graft in region 6 A. Fluorescent cells are located mainly in the blood islands between ventral endoderm (end) and ventral epidermis (ep). Bar, 50 μm.

Fig. 11.

Photomicrograph of transverse sections at the trunk level of a tail-bud (Stage 28) receiving a RLDx-labeled graft in region 6 A. Fluorescent cells are located mainly in the blood islands between ventral endoderm (end) and ventral epidermis (ep). Bar, 50 μm.

Caudal mesodermal map

Presumptive caudal mesodermal cells are restricted to a small dorsal including regions III A-B and IV B (Fig. 4H, Table 1). These regions also make an important contribution to the notochord (III A), somites (III B) and neuroectoderm (IV B). We transplanted a doublelabeled III A graft and found 42.8% of the caudal mesodermal cells counted were derived from the superficial layer of the early gastrula.

Single-cell injections

When we injected single cells at the early gastrula stage with RLDx and then determined their fates at the tailbud stage (Fig. 3A and B), we found results that are as one would expect from Vogt’s (1929) fate map. All cells from dl3 to endo 2 make contributions to the endoderm (Fig. 3D-H) and they are arranged in a neat array with endo 2 derivative being most caudal and dl3 cells being most cranial. dl4 derivatives end up in the most cranial portion of the notochord (Fig. 3C). This single-cell injection technique also shows that the boundary between endodermal and notochordal derivatives in the early gastrula fate map lies between grafts I A and II A.

Dorsal-ventral diminution of superficial cell contribution to mesodermal derivatives

Using double-labeled grafts, we also observed a gradual dorsal to ventral diminution of the relative contribution by superficial cells to mesodermal derivatives (Fig. 12, Table 2). For example, regions I B, 2 D, 4 C, and 6 A have 39%, 27%, 12% and 1%, respectively, of their superficial cells contributing to mesodermal derivatives.

Table 2.

Relative contributions of superficial and deep cells to derivatives in different presumptive regions of the early Pleurodeles waltl gastrula

Relative contributions of superficial and deep cells to derivatives in different presumptive regions of the early Pleurodeles waltl gastrula
Relative contributions of superficial and deep cells to derivatives in different presumptive regions of the early Pleurodeles waltl gastrula
Fig. 12.

Diagram illustrating that the relative contribution of superficial cells to mesodermal derivatives decreases along the dorsal-ventral axis. The height of the bars are proportional to the percentage of superficial cells contributing to mesodermal derivatives in that region.

Fig. 12.

Diagram illustrating that the relative contribution of superficial cells to mesodermal derivatives decreases along the dorsal-ventral axis. The height of the bars are proportional to the percentage of superficial cells contributing to mesodermal derivatives in that region.

These results suggest that there is a strong dorsal-to-ventral gradient of mixing between superficial and deep cells during the morphogenetic movements of gastrulation. Our new mesodermal fate map for the Pleuro-deles waltl early gastrula is summarized in Fig. 13.

Fig. 13.

Summary diagram illustrating the fate map for presumptive mesodermal regions in the Pleurodeles waltl early gastrula. Red=regions forming only mesodermal derivatives. Yeilow=regions forming only endoderm. Orange=regions forming mesoderm and endoderm. Blue=regions forming only neuroectoderm. Purple=regions forming mesoderm and ectoderm.

Fig. 13.

Summary diagram illustrating the fate map for presumptive mesodermal regions in the Pleurodeles waltl early gastrula. Red=regions forming only mesodermal derivatives. Yeilow=regions forming only endoderm. Orange=regions forming mesoderm and endoderm. Blue=regions forming only neuroectoderm. Purple=regions forming mesoderm and ectoderm.

We have employed a new method of orthotopic transplantation of small double-labeled regions from Pleurodeles waltl donor embryos into unlabeled host embryos. This method not only allows us to construct a fate map, but also gives new information on cell rearrangement by intermingling, both by neighbor exchanges at the margin of grafts and by mixing of superficial and deep cells. We were surprised to discover how extensive cell rearrangement is during gastrulation in this urodele. We chose the early gastrula stage because the dorsal lip of the blastopore must form before we were able to make unequivocal localization of grafts in donors and hosts. Also, we chose this stage because it is a stage where mesoderm has been induced but has not yet undergone complex morphogenetic rearrangements.

The regions that contribute solely to mesodermal derivatives are situated in the dorsal and lateral marginal zones (I C-D, II A-D and 1-4 D). The other regions contribute to mesodermal derivatives and neuroectoderm, epidermis and endoderm. The boundaries between mesoderm and neuroectoderm lie in the regions III B-C and IV B-C. The boundaries between mesoderm and endoderm lie in the regions I B and 1-4 C. In the ventral marginal zone, there are regions that contribute to epidermis, mesoderm and endoderm (6 A-C). Cell lineage analysis in 16- and 32-cell stage Xenopus embryos (Moody, 1987a,b; Takasaki, 1987; Wetts and Fraser, 1989) and explant grafting in Ambystoma gastrulae (Cleine and Slack, 1985) also revealed that certain boundary regions in amphibian embryos contribute to multiple primary germ layers. These results indicate that the germ layer boundaries in fate maps should not be drawn as sharp Unes but rather as diffuse transition zones with considerable overlap.

The results of our mesodermal fate mapping allowed us to define different mesodermal regions. In the early gastrula, five of these regions are arranged on a neat dorsal-to-ventral polarity axis starting with the most dorsal and ending with the most ventral and include: notochord, somites, pronephros, lateral plate mesoderm and blood islands. These results are in general agreement with previously published urodele fate maps (Vogt, 1929, Pasteels, 1942, Cleine and Slack, 1985). Furthermore, these results are compatible with the three-signal model of mesodermal induction in Xenopus (Dale and Slack, 1987), which shows a dorsal specification established from the most dorsal noto-chordal area in the dorsal marginal zone to the most ventral blood islands in the ventral marginal zone. An anteroposterior specification of mesodermal territories may also be deduced from the anteroposterior distribution in organs at the tail-bud stage. For example, region II A gives rise to anterior notochord while III A forms posterior notochord. Similar observations have been made for somites, head mesenchyme, lateral plate mesoderm and blood islands. Generally speaking, in early gastrula, presumptive mesodermal cells nearest the blastopore end up in the anterior portion of the tailbud stage. Conversely, presumptive mesodermal cells farthest from the blastopore end up in the posterior portion of the tail-bud stage.

As in Xenopus development, where prospective anterior mesoderm involutes first, followed by more posterior mesoderm (Keller, 1976), the anteroposterior regionalization of Pleurodeles mesoderm may be related to the position of prospective mesodermal cells relative to the dorsal lip of the blastopore at the early gastrula stage. It is probable that some mesodermal anteroposterior regionalization occurs well before gastrulation.

Vogt (1929) illustrated sharp boundaries between presumptive mesodermal regions. Our results, showing that individual regions contribute to many different mesodermal organs and even organs in all three primary germ layers, show that these boundaries are quite indistinct. For example, many regions at the boundary between mesoderm and ectoderm or between mesoderm and endoderm exhibit extensive overlap. In many regions, major contributions are made to one organ and minor contributions are made to another. For example, region II A contributes mainly to notochord but also contributes in a minor way to somites and head mesenchyme. These results show that there is extensive intermingling among cells at the boundaries of presumptive regions.

The use of double-labeled explants-allows us to make estimates of the relative contributions made by deep and superficial cells to organs in the tail-bud stage. We have shown that superficial cells make a significant contribution to many different mesodermal organs. Our results are in agreement with Smith and Malacinski’s (1983) demonstration that urodele gastrulae have presumptive mesodermal cells in the deep and superficial portions of the marginal zone. In contrast, presumptive mesodermal cells are strictly limited to the deep marginal zone cells in Xenopus gastrulae (Keller, 1975, 1976; Smith and Malacinski, 1983). We have also discovered that there is a decreasing contribution to mesodermal derivatives from the superficial portions of presumptive mesodermal regions along the dorsal-to-ventral axis. The most dorsal regions contribute the largest number of superficial cells to mesodermal derivatives and the most ventral regions contribute the smallest number of superficial cells to mesodermal derivatives. The simplest interpretations of these results is that there is extensive mixing of superficial and deep cells among the first cells to involute but much less mixing in the cells that involute later from the lateral and ventral blastopore lips.

We have noted that the superficial cell population in the dorsal marginal zone makes an important contribution to mesodermal structures, especially in the notochord. Soon after involution, the multilayered dorsal marginal zone becomes a single layer of cells by mixing of deep and superficial cells. This radial intercalation process (Shi et al., 1987) varies considerably in different regions of the marginal zone, being important in the dorsal marginal zone, and minor in the ventral marginal zone. We have also observed a high rate of mixing between fluorescent and non-fluorescent cells. Several mechanisms could account for this mixing. Besides intercalation, autonomous migration of mesodermal cells across a fibronectin-rich substratum on the basal surface of the blastocoel roof is a major property of migration of mesodermal cells during urodele gastrulation (Boucaut et al., 1990) and could account for our observed mixing. Lundmark (1986) studied Ambystoma and found that somitic cells are removed from the archenteric surface by ingression followed by adhesion and migration along the blastocoel wall. During assembly of the notochord, cells crawl between one another to produce a longer, narrower array (Keller and Hardin, 1987; Wilson and Keller, 1991). Somitic cells rearrange during somitic rotation (Youn et al., 1980; Youn and Malacinski, 1981). All these distinct morphogenetic mechanisms may contribute to the cell mixing observed in our experiments. Intermixing of cells is also a common phenomenon during teleost (Kimmel and Law, 1985), avian (Selleck and Stern, 1991) and murine (Serbedzija et al., 1990) embryogenesis. It is clear from our experiments that there is a decreasing gradient of cell intermixing along the dorsal-to-ventral axis. We suggest that the lack of sharp boundaries between presumptive mesodermal regions in the early Pleurodeles gastrulae is due to this striking rearrangement that occurs not only at the boundaries of explants, but also between superficial and deep cells within individual regions. This more refined fate map will now allow us to proceed with a more sophisticated analysis of factors that regulate mesodermal specification and morphogenetic cell movements in Pleurodeles waltl development.

We wish to thank Dr G. Ville for preparing the fluorescent dextran, Drs J.F. Riou and D.L. Shi for critical reading of the manuscript, Mr P. Groue for maintenance of the amphibian colony and Mrs M.M. Tran Kim Lan for typing the manuscript. This work was supported by grants from the CNRS (URA-1135); MEN (France); P. and M. Curie University (Paris VI); and George Washington University. Dr S. Sanchez was an invited Professor at P. and M. Curie University.

Boucaut
,
J.-C.
,
Johnson
,
K. E.
,
Darribère
,
T.
,
Shi
,
D. L.
,
Riou
,
J. F.
,
Boulekbache
,
H.
and
Delarue
,
M.
(
1990
).
Fibronectin rich fibrillar extracellular matrix controls cell migration during amphibian gastrulation
.
Int. J. Devi Biol
.
34
,
139
147
.
Cleine
,
J. H.
and
Slack
,
J. M. W.
(
1985
).
Normal fates and states of specification of different regions in the axolotl gastrula
.
J. Embryol. exp. Morph
.
86
,
247
269
.
Cooke
,
J.
(
1987
).
Dynamics of the control of body pattern in the development of twin bodies after reorientation of eggs in gravity
.
Development
99
,
417
427
.
Dale
,
L.
and
Slack
,
J. M. W.
(
1987
).
Regional specification within the mesoderm of early embryos of Xenopus laevis
.
Development
100
,
279
295
.
Gallien
,
L.
and
Durocher
,
M.
(
1957
).
Table chronologique du développement chez Pleurodeles waltlii Michah
.
Bull. Biol. Fr. Belg
.
91
,
97
114
.
Gerhart
,
J. C.
and
Keller
,
R. E.
(
1986
).
Region-specific cell activities in amphibian gastrulation
.
Ann. rev. Cell Biol
.
2
,
201
229
.
Gimlich
,
R. L.
and
Gerhart
,
J. C.
(
1984
).
Early cellular interactions promote embryonic axis formation in Xenopus laevis
.
Devi Biol
.
104
,
117
130
.
Gurdon
,
J. B.
,
Mohun
,
T. J.
,
Sharpe
,
C. R.
and
Taylor
,
M. V.
(
1989
).
Embryonic induction and muscle gene activation
.
Trends in Genetics
5
,
51
56
.
Holtfreter
,
J.
(
1943
).
A study of the mechanics of gastrulation. Part I
.
J. exp. Zool
.
94
,
261
318
.
Holtfreter
,
J.
(
1944
).
A study of the mechanics of gastrulation, II
.
J. exp Zool
.
95
,
171
212
.
Katz
,
J.M.
,
Lasek
,
R.J.
,
Osdoby
,
P.
,
Whittaker
,
J.R.
and
Caplan
,
A. I.
(
1982
).
Bolton-Hunter reagent as a vital stain for developing systems
.
Devi Biol
.
90
,
419
429
.
Keller
,
R. E.
(
1975
).
Vital dye mapping of the gastrula and neurula of Xenopus laevis I. Prospective areas and morphogenetic movements in the superficial layer
.
Devi Biol
.
42
,
222
241
.
Keller
,
R. E.
(
1976
).
Vital dye mapping of the gastrula and neurula of Xenopus laevis II. Prospective areas and morphogenetic movements in the deep layer
.
Devi Biol
.
51
,
118
137
.
Keller
,
R. E.
(
1985
).
The function and mechanism of convergent extension during gastrulation of Xenopus laevis
. In
Molecular Determinants of Animal Form
(ed.
G.M.
Edelman
), pp.
111
141
.
New York
:
Alan R. Liss
.
Keller
,
R. E.
and
Hardin
,
J.
(
1987
).
Cell behavior during active cell rearrangement: evidence and speculation
.
J. Cell Sci. Supplement
8
,
369
393
.
Keller
,
R. E.
and
Schoenwolf
,
G. C.
(
1977
).
A SEM study of cellular morphology contact, and arrangement as related to gastrulation in Xenopus laevis
.
Wilhelm Roux Archiv devl Biol
.
182
,
165
186
.
Khnmel
,
C. B.
and
Law
,
R. D.
(
1985
).
Cell lineage of zebrafish blastomeres
.
Devi Biol
.
108
,
94
101
.
Landstrom
,
U.
and
Lovtrup
,
S.
(
1979
).
Fate map and cell differentiation in the amphibian embryo: An experimental study
.
J. Embryol. exp. Morphol
.
54
,
113
130
.
Lpvtrup
,
S.
(
1975
).
Fate map and gastrulation in Amphibia. A critic of current views
.
Cañad. J. Zool
.
53
,
473
479
.
Lundmark
,
C.
(
1986
).
Role of bilateral zones of ingressing superficial cells during gastrulation of Ambystoma mexicanum
.
J. Embryol. exp. Morph
.
97
,
47
62
.
Moody
,
S. A.
(
1987a
).
Fates of the blastomeres of the 16-cell stage Xenopus embryo
.
Devi Biol
.
119
,
560
578
.
Moody
,
S. A.
(
1987b
).
Fates of the blastomeres of the 32-cell stage Xenopus embryo
.
Devi Biol
.
122
,
300
319
.
Nieuwkoop
,
P. D.
(
1969
).
The formation of mesoderm in Urodelan amphibians. I. Induction by the endoderm
.
Wilhelm Roux Archiv. EntwMech. Org
.
162
,
341
373
.
Pasteels
,
J.
(
1942
).
New observations concerning the maps of presumptive areas of the young amphibian gastrula (Ambystoma and Discoglossus)
.
J. exp. Zool
.
89
,
255
281
.
Selleck
,
M. A. J.
and
Stern
,
C.
(
1991
).
Fate mapping and cell lineage analysis of Hensen’s node in the chick embryo
.
Development
116
,
615
626
.
Serbedzija
,
G. N.
,
Fraser
,
S. E.
and
Bronner-Fraser
,
M.
(
1990
).
Pathways of neural crest cell migration in the mouse embryo as revealed by vital dye mapping
.
Development
108
,
605
612
.
Shi
,
D.L.
,
Delarue
,
M.
,
Darribère
,
T.
,
Riou
,
J.-F.
and
Boucaut
,
J.-C.
(
1987
).
Experimental analysis of the extension of the dorsal marginal zone in Pleurodeles waldii gastrulae
.
Development
100
,
147
161
.
Slack
,
J. M. W.
(
1983
).
From Egg to Embryo. Determinative Events in Early Development. Cambridge: Cambridge University Press
.
Slack
,
J. M. W.
(
1984
).
Regional biosynthetic markers in the early amphibian embryo
.
J. Embryol. exp. Morph
.
80
,
289
319
.
Smith
,
J. C.
,
Cooke
,
J.
,
Green
,
J. B. A.
,
Horves
,
G.
and
Symes
,
K.
(
1989
).
Inducing factors and the control of mesodermal pattern in Xenopus laevis
.
Development
107
upp,
149
159
.
Smith
,
J. C.
and
Malaclnski
,
G. M.
(
1983
).
The origin of the mesoderm in the anuran, Xenopus laevis, and a urodele, Ambystoma mexicanum
.
Devi Biol
.
98
,
250
254
.
Sudarwati
,
S.
and
Nieuwkoop
,
P. D.
(
1971
).
Mesoderm formation in the anuran Xenopus laevis (Daudin)
.
Wilhelm Roux Archiv. EntwMech. Org
.
166
,
189
201
.
Takasakl
,
M.
(
1987
).
Fates and roles of presumptive organizer region in the 32-cell embryo in normal development of Xenopus laevis
.
Develop. Growth Differ
.
29
,
141
152
.
Vogt
,
W.
(
1929
).
Gestaltungsanalyse am Amphibienkeim mit drtlicher Vitalfàrbung. II. Teil. Gastrulation und Mesodermbildung bei Urodelen und Anuren
.
Wilhelm Roux Arch. EntwMech. Org
.
120
,
384
706
.
Wetts
,
R.
and
Fraser
,
S. E.
(
1989
).
Slow intermixing of cells during Xenopus embryogenesis contributes to the consistency of the blastomere fate map
.
Development
105
,
9
15
.
Wilson
,
P.
and
Keller
,
R. E.
(
1991
).
Cell rearrangement during gastrulation in Xenopus: direct observation of cultured explants
.
Development
112
,
289
300
.
Youn
,
B. W.
,
Keller
,
R. E.
and
Malaclnski
,
G. M.
(
1980
).
An atlas of notochord and somite morphogenesis in several anuran and urodelean amphibians
.
J. Embryol. exp. Morph
.
59
,
223
247
.
Youn
,
B. W.
and
Malaclnski
,
G. M.
(
1981
).
Somitogenesis in the amphibian Xenopus laevis: scanning electron microscopic analysis of intrasomitic cellular arrangements during somite rotation
.
J. Embryol. exp. Morph
.
64
,
23
43
.