Overexpression of S-adenosylmethionine decarboxylase.（SAMDC） in Xeno-pus embryos activates maternal program of apoptosm as a “fail-safe”mechanism of early embryogenesis

In Xenopus, injection of S-adenosylmethionine decarboxylase (SAMDC) mRNA into fertilized eggs or 2-cell stage embryos induces massive cell dissociation and embryo-lysis at the early gastrula stage due toactivation of the maternal program of apoptosis. We injected SAMDC mRNA into only one of the animalside blastomeres of embryos at different stages of cleavage, and examined the timing of the onset of theapoptotic reaction. In the injection at 4-and 8-cell stages, a considerable number of embryos developed intotadpoles and in the injection at 16-and 32-cell stages, all the embryos became tadpoles, although tadpolesobtained were sometimes abnormal. However, using GFP as a lineage tracer, we found that descendant cellsof the blastomere injected with SAMDC mRNA at 8-to 32-cell stages are confined within the blastocoel atthe early gastrula stage and undergo apoptotic cell death within the blastocoel, in spite of the continued development of the injected embryos. These results indicate that cells overexpressed with SAMDC undergo apoptotic cell death consistently at the early gastrula stage, irrespective of the timing of the mRNA injection.We assume that apoptosis is executed in Xenopus early gastrulae as a “fall-safe“ mechanism to eliminate physiologically-severely damaged cells to save the rest of the embryo.     

Overexpression of S-adenosylmethionine decarboxylase (SAMDC) activates the maternal program of apoptosis shortly after MBT in Xenopus embryos

Overexpression of S-adenosylmethionine decarboxylase (SAMDC) mRNA in 1- and 2-cell stage Xenopus embryos induces cell autonomous dissociation at the late blastula stage and developmental arrest at the early gastrula stage. The induction of cell dissociation took place "punctually" at the late blastula stage in the SAMDC-overexpressing cells, irrespective of the stage of the microinjection of SAMDC mRNA. When we examined the cells undergoing the dissociation, we found that they were TUNEL-positive and contained fragmented nuclei with condensed chromatin and fragmented DNA. Furthermore, by injecting Xenopus Bcl-2 mRNA together with SAMDC mRNA, we showed that SAMDC-overexpressing embryos are rescued completely by Bcl-2 and becometadpoles. These results indicatethat cell dissociation induced by SAMDC overexpression is due to apoptotic cell death. Since the level of S-adenosylmethionine (SAM) is greatly reduced in SAMDC-overexpressing embryos and this induces inhibition of protein synthesis accompanied by the inhibition of DNA and RNA syntheses, we conclude that deficiency in SAM induced by SAMDC overexpression activates the maternal program of apoptosis in Xenopus embryos at the late blastula stage, but not before. We propose that this mechanism serves as a surveillance mechanism to check and eliminate cells physiologically damaged during the cleavage stage.     

Maternal program of apoptosis activated shortly after midblastula transition by overexpression of S-adenosylmethionine decarboxylase in Xenopus early embryos

When we studied polyamine metabolism in Xenopus embryos, we cloned the cDNA for Xenopus S-adenosylmethionine decarboxylase (SAMDC), which converts SAM (S-adenosylmethionine), the methyl donor, into decarboxylated SAM (dcSAM), the aminopropyl donor, and microinjected its in vitro transcribed mRNA into Xenopus fertilized eggs. We found here that the mRNA injection induces a SAM deficient state in early embryos due to over-function of the overexpressed SAMDC, which in turn induces inhibition of protein synthesis. Such embryos developed quite normally until blastula stage, but stopped development at the early gastrula stage, due to induction of massive cell dissociation and cell autolysis, irrespective of the dosage and stage of the mRNA injection. We found that the dissociated cells were TUNEL-positive, contained fragmented nuclei with ladder-forming DNA, and furthermore, rescued completely by coinjection of Bcl-2 mRNA. Thus, overexpression of SAMDC in Xenopus embryos appeared to switch on apoptotic program, probably via inhibition of protein synthesis. Here, we briefly review our results together with those reported from other laboratories. After discussing the general importance of this newly discovered apoptotic program, we propose that the maternal program of apoptosis serves as a surveillance mechanism to eliminate metabolically severely-damaged cells and functions as a 'fail-safe' mechanism for normal development in Xenopus embryos.     

Occurrence of pre-MBT synthesis of caspase-8 mRNA and activation of caspase-8 prior to execution of SAMDC (S-adenosylmethionine decarboxylase)-induced, but not p53-induced, apoptosis in Xenopus late blastulae

Overexpression of S-adenosylmethionine decarboxylase (SAMDC) in Xenopus fertilized eggs activates caspase-9 and executes maternal program of apoptosis shortly after midblastula transition (MBT). We find that overexpression of caspase-8 and p53, like that of SAMDC, induces apoptosis in Xenopus late blastulae. The apoptosis induced by p53 was abolished by injection of mRNA for xdm-2, a negative regulator of p53, and by injection of a peptide inhibitor or a dominant-negative type mutant of caspase-9, but not caspase-8. The apoptosis induced by SAMDC was not abolished by injection of xdm-2 mRNA, but was abolished by injection of a peptide inhibitor or a dominant-negative type mutant mRNA of both caspase-9 and caspase-8. Unlike caspase-9 mRNA, caspase-8 mRNA did not occur as a maternal mRNA rather induced to be expressed during cleavage stage (pre-MBT stage) by overexpression of SAMDC but not p53. Furthermore, while activities to process procaspase-8 and procaspase-9 appeared in SAMDC-overexpressed apoptotic embryos, the activity to process procaspase-8 did not appear in p53-overexpressed apoptotic embryos. We conclude there are at least two pathways in the execution of the maternal program of apoptosis in Xenopus embryos; one being through do novo expression of caspase-8 gene during cleavage stage, and the other without involvement of caspase-8.     

The mitochondrial-apoptotic pathway is triggered in Xenopus mesoderm cells deprived of PDGF receptor signaling during gastrulation

Platelet-derived growth factor receptor (PDGFR) signaling is required for normal gastrulation in Xenopus laevis. Embryos deprived of PDGFR signaling develop with a range of gastrulation-specific defects including spina bifida, shortened anteroposterior axis, and reduced anterior structures. These defects arise because the involuting mesoderm fails to move appropriately. In this study, we determine that inhibition of PDGFR signaling causes prospective head mesoderm cells to appear in the blastocoel cavity at the onset of gastrulation, stage 10. These aberrant cells undergo apoptosis via the caspase 3 pathway at an embryonic checkpoint called the early gastrula transition (EGT). They are TUNEL-positive and have increased levels of caspase 3 activity compared to control embryos. Apoptotic death of these mesoderm cells can be prevented by co-injection of mRNA encoding Bcl-2 or by injection of either a general caspase inhibitor or a caspase 3-specific inhibitor. Prevention of cell death, however, is not sufficient to rescue gastrulation defects in these embryos. Based on these data, we propose that PDGFR signaling is necessary for survival of prospective head mesoderm cells, and also plays an essential role in the control of their cell movement during gastrulation.     

Involvement of caspase-9 in execution of the maternal program of apoptosis in Xenopus late blastulae overexpressed with S-adenosylmethionine decarboxylase

We previously demonstrated that overexpression of S-adenosylmethionine decarboxylase (SAMDC) in Xenopus early embryos induces execution of maternal program of apoptosis shortly after midblastula transition, which likely serves as a fail-safe mechanism of early development to eliminate physiologically damaged cells before they entering the gastrula stage. To determine how caspases are involved in this process, we microinjected peptide inhibitors and "dominant-negative forms" of caspase-9 and -1 into Xenopus fertilized eggs, and found that inhibitors of caspase-9, but not caspase-1, completely suppress SAMDC-induced apoptosis. The lysate of SAMDC-overexpressing late blastulae contained activity to cleave in vitro-synthesized [(35)S]procaspase-9, but not [(35)S]procaspase-1, and mRNA for caspase-9, but not caspase-1, occurred abundantly in the unfertilized egg as maternal mRNA. We also found that overexpression of caspase-9 and -1 equally executes the apoptosis, but the apoptosis executed by these mRNAs was only partially rescued by Bcl-2 and rescued embryos did not develop beyond neurula stage. These results indicate that activation of caspase-9 is a key step for execution of the maternally preset program of apoptosis in Xenopus early embryos.     

Gastrulation and larval pattern in Xenopus after blastocoelic injection of a Xenopus-derived inducing factor: experiments testing models for the normal organization of mesoderm

When a Xenopus XTC cell-derived mesoderm-inducing factor (MIF) is injected into the blastocoel of Xenopus embryos before gastrulation, they develop almost normally until just after the onset of mesoderm involution at the internal blastoporal lip. Cells from the entire lining of the blastocoel roof and inner marginal zone then undergo a synchronous, sudden change of contact and arrangement which resembles the transformation undergone by normal mesoderm at its time of involution at the vegetal edge of the marginal zone. We describe a dose-dependent spectrum of subsequent abnormalities in gastrulation and, in cases where gastrulation partially recovers, in the resulting larval pattern. Because of such recovery, embryos injected with widely different doses may appear equally abnormal at the early gastrula stage but very different by control larval stages. Extra spinocaudal axial patterns, in the area of ectopic mesoderm, are seen after MIF doses that just permit recovery of gastrulation. The sudden cellular transformation corresponding to involution, in the ectopically specified mesoderm, spreads throughout the animal cap within 15 min in individuals, at a time significantly later than the earliest normal transformation in the marginal zone. No systematic alteration could, however, be detected in its timing, in relation to a 250-fold range of injected MIF concentration or a 3.5-hr difference in time of injection. The severity of the effects on final embryonic pattern is largely independent of the blastular stage of injections. Splitting of the total injected dose into two, separated by 2 to 3 hr of blastular development, reveals that the degree of effect on gastrulation and patterning depends only upon the highest experienced concentration at any time before response. When fibroblast growth factor (bFGF), a different effective mesoderm inducer, is similarly injected, a similar abnormal cell behavior and ectopic mesoderm formation are seen, but beginning only at midgastrular stages some 1.5 hr beyond that characteristic of XTC-MIF. The findings are introduced and discussed in terms of models for the natural organization of the time course of gastrulation and mesodermal pattern.     

Effects of ethanol on the primitive streak stage mouse embryo

Recent studies of mouse models have suggested that malformations associated with the fetal alcohol syndrome (FAS) are caused by the effects of ethanol on early embryos during gastrulation and neurulation. A study of Xenopus laevis embryos showed that exposure of gastrula stage amphibian embryos to ethanol inhibits migration of the mesodermal cells, causes formation of small neural plates, and subsequently causes hypoplastic craniofacial malformations in tadpoles. We now report effects of ethanol on the primitive streak stage mouse embryos. An ethanol solution (25%) was injected intraperitonealy twice into mice of 6.5-7.0 days of pregnancy at a dose of 0.015 ml/gm of body weight. Histological and morphometric examinations of 7.5-day embryos, 20 hr after the second injection, showed that the epiblast layer was disorganized and shrunk with formation of many blebs. In addition, formation of the mesodermal cell layer was retarded in the ethanol-treated embryos, suggesting that exposure of gastrula stage embryos to ethanol causes similar abnormalities in mouse and Xenopus embryos. These results suggest that the inhibition of the morphogenetic movements during gastrulation may be the primary effect of ethanol in causing major craniofacial malformations of FAS.     

Chronological appearance of spontaneous and induced apoptosis during preimplantation development of rabbit and mouse embryos

This study was undertaken to obtain specific information on the characteristics of spontaneous and induced apoptosis during preimplantation development of rabbit in vivo and in vitro developed embryos and mouse in vitro embryos. After reaching appropriate developmental stages, embryos were transferred into culture media with or without apoptotic inductor (actinomycin D 500 ng/mL) and cultured for 10 h. The identification of apoptotic cells was based on morphological assessment of nuclei and on detection of specific DNA degradation, phosphatidylserine redistribution and active caspase-3 under fluorescence microscope. Our experiments proved that apoptosis is a frequent physiological event occurring during normal preimplantation development. A high number of untreated rabbit and mouse blastocysts contained at least one apoptotic cell. Rabbit embryos showed a lower incidence of spontaneous apoptosis. Treated blastocysts of both species responded to the presence of apoptotic inductor by significant decrease in the average number of blastomeres and significant increase in the incidence of apoptotic cell death. The occurrence of spontaneous apoptosis during earlier preimplantation development was sporadic and its presence was observed only at stages following embryonic genome activation (at 4-cell stage and later in mouse, at 16-cell and morula stage in rabbit). The susceptibility of embryos at early stages to the apoptotic inductor was much lower. The presence of actinomycin D did not increase the incidence of apoptotic embryos or apoptotic cells. Nevertheless, it slowed down embryo growth and triggered earlier appearance of some apoptotic features (at the 6-cell stage in rabbit). The results show that the occurrence of both spontaneous and induced apoptosis in preimplantation embryos is stage- and species-specific.     

Ectopic expression of a homeobox gene changes cell fate in Xenopus embryos in a position-specific manner.

We have injected XIHbox 6 mRNA together with the lineage tracer colloidal gold into individual dorso-anterior blastomeres of the 32-cell stage Xenopus embryo and analyzed their cell fate during embryogenesis. While the developing tadpoles appeared entirely normal, the fate of the progeny of the injected blastomere was altered. In the brain injected cells failed to differentiate terminally, as indicated by a loss of labeled cranial nerves. Differentiation of spinal nerves remained unaffected. Fate change in the CNS occurred at about the time of normal XIHbox 6 protein expression. In addition, progeny of injected blastomeres gained head epidermal fate and lost anterior notochord fate as a result of altered cell migrations during gastrulation. The results show that a homeodomain protein is capable of altering cell fate in a position-specific and cell-autonomous manner in Xenopus embryos. The experimental approach used here should be applicable to other molecules specifying cell fate.     

A study of cell interactions involved in Pleurodeles waltlii epidermal differentiation

Summary A polyclonal antibody (SP-2) has been produced, which recognizes antigens expressed in epidermal cells of Pleurodeles waltlii embryos. The antigens appear first at the end of gastrulation in the external surface of the embryo and are selectively expressed in ectodermally derived epidermal structures. Ectodermal commitment was investigated using cell cultures and blastocoel graft experiments. The four animal blastomeres of the 8-cell stage as well as the animal cap explants of the early gastrula stage cultured in vitro differentiate into epidermis, and SP-2 antigens are expressed. The expression of SP-2-defined antigens is inhibited both in vivo and in vitro by the inductive interaction of chordomesoderm. Once dissociated, ectodermal cells do not react with SP-2. Conversely, the aggregation of ectodermal cells may restore the expression of SP-2 antigens. Transplantation of animal cap explants or isolated ectodermal cells into the blastocoel of a host embryo at the early gastrula stage shows that only cells integrated into the epidermis express the marker antigens. When vegetal cells were dissociated from donor embryos before the mid-blastula stage and implanted into the blastocoel of host embryos at the early gastrula stage, their progeny were found in all germ layers, cells that were found in the host epidermis were stained with SP-2, whereas those contributing to mesoderm and endoderm were not. Thus the acquisition of cell polarity in epidermal differentiation and the organization of cells into epithelial structures are essential for SP-2-defined antigen expression.     

Chronological appearance of spontaneous and induced apoptosis during preimplantation development of rabbit and mouse embryos

This study was undertaken to obtain specific information on the characteristics of spontaneous and induced apoptosis during preimplantation development of rabbit in vivo and in vitro developed embryos and mouse in vitro embryos. After reaching appropriate developmental stages, embryos were transferred into culture media with or without apoptotic inductor (actinomycin D 500 ng/mL) and cultured for 10 h. The identification of apoptotic cells was based on morphological assessment of nuclei and on detection of specific DNA degradation, phosphatidylserine redistribution and active caspase-3 under fluorescence microscope.Our experiments proved that apoptosis is a frequent physiological event occurring during normal preimplantation development. A high number of untreated rabbit and mouse blastocysts contained at least one apoptotic cell. Rabbit embryos showed a lower incidence of spontaneous apoptosis. Treated blastocysts of both species responded to the presence of apoptotic inductor by significant decrease in the average number of blastomeres and significant increase in the incidence of apoptotic cell death. The occurrence of spontaneous apoptosis during earlier preimplantation development was sporadic and its presence was observed only at stages following embryonic genome activation (at 4-cell stage and later in mouse, at 16-cell and morula stage in rabbit). The susceptibility of embryos at early stages to the apoptotic inductor was much lower. The presence of actinomycin D did not increase the incidence of apoptotic embryos or apoptotic cells. Nevertheless, it slowed down embryo growth and triggered earlier appearance of some apoptotic features (at the 6-cell stage in rabbit). The results show that the occurrence of both spontaneous and induced apoptosis in preimplantation embryos is stage- and species-specific.     

A trial for induction of supernumerary primordial germ cells in Xenopus tadpoles by injecting RNA of Xenopus vasa homologue into germline cells of 32-cell embryos

Whether overexpression of Xenopus vasa homologue or Xenopus vasa-like gene 1 (XVLG1) in germline cells of Xenopus embryos can induce supernumerary primordial germ cells (PGC) at tadpole stage was investigated. XVLG1 RNA (0.1-2.0 ng) and beta-gal RNA (0.5 ng) were injected into one of, usually, four germ plasm-bearing cells (GPBC) of 32-cell embryos, with the beta-gal RNA (2.0 ng) serving as both lineage tracer and control for XVLG1 RNA. The total number of PGC, including X-gal-stained and unstained PGC of injected and uninjected GPBC origins respectively, was examined in the experimental tadpoles developed from the injected embryos. The injected RNA, XVLG1 and beta-gal RNA, were translated, resulting in a large amount of corresponding proteins in presumptive PGC (pPGC) as well as in somatic cells derived from the injected GPBC. Nevertheless, the average number of total PGC per tadpole found in the experimental tadpoles from the XVLG1 RNA-injected embryos was not significantly different from that of beta-gal RNA-injected ones, irrespective of the injected dose of XVLG1 RNA. This indicates that the extra XVLG1 protein in pPGC is not sufficient to increase the number of PGC in the tadpoles.     

Direct Evidence for the Presence of Germ Cell Determinant in Vegetal Pole Cytoplasm of Xenopus laevis and in a Subcellular Fraction of It

To test for the presence of germ cell determinant in Xenopus embryos, vegetal pole cytoplasm containing the "germ plasm", or a subcellular fraction of it, was microinjected into single somatic blastomeres isolated from 32-cell embryos. Injected or non-injected (control) blastomeres were cultured in ³H-thymidine until normal control embryos reached the neurula stage. The labeled explants were then implanted into unlabeled host neurulae, which were allowed to develop to the tadpole stage. Labeled PGCs of explant origin in the genital ridges of the experimental tadpoles were examined by autoradiography.
Isolated blastomeres were injected with vegetal pole cytoplasm of 32-cell embryos or with a 20,000 g pellet made from vegetal pole cytoplasm of 2-cell embryos. Labeled PGCs were found in 7.6% and 2.3% of the experimental tadpoles, respectively. No labeled PGCs were found in the control tadpoles, except for one tadpole in the first experiment. These results strongly suggest that the vegetal pole cytoplasm and its subcellular fractions act as germ cell determinant.     

Injected Xwnt-8 RNA acts early in Xenopus embryos to promote formation of a vegetal dorsalizing center.

Expression cloning from a pool of gastrula cDNAs identified the Wnt family member Xwnt-8 as having dorsal axis-inducing activity in Xenopus embryos. Microinjected Xwnt-8 mRNA was able to rescue the development of a dorsally complete anterior-posterior axis in embryos ventralized by exposure to UV light. Axis induction was observed in embryos injected in either marginal or vegetal blastomeres at the 32-cell stage. Vegetal blastomeres receiving Xwnt-8 mRNA contributed progeny not to the induced dorsal axis, but to the endoderm, a result consistent with Xwnt-8 causing cells to act as a Nieuwkoop center (the vegetal-inducing component of normal dorsal axis formation), rather than as a Spemann organizer (the induced dorsal marginal zone component that directly forms the dorsal mesoderm). Xwnt-8, which is normally expressed ventrally in midgastrula and neurula embryos, appears to mimic, when injected, maternally encoded dorsal mesoderm-inducing factors that act early in development.     

A method of microinjection: delivering monoclonal antibody 1223 into sea urchin embryos.

In this paper, a simpler method of microinjecting sea urchin embryos without using the conventional microinjection chamber designed by Kiehart is reported. A trough was made on a surface of 0.6% agarose gel dissolved in artificial sea water. Approximately fifty hatched embryos could be loaded in the trough and, consequently, swimming embryos were trapped in the trough. Monoclonal antibody (mAB) 1223 which blocks spiculogenesis in vitro was delivered into the blastocoels of sea urchin embryos to test whether this antibody inhibits spiculogenesis in vivo and also, whether this new technique is effective for the microinjection of the sea urchin embryos. The embryos were injected with mAB1223 at the hatched blastula, early mesenchyme blastula and early gastrula stages, and 63%, 90% and 97% of the embryos did not form spicules at the late gastrula stage, respectively. Therefore, mAB1223 was shown to also block spiculogenesis in vivo. From the fact that spiculogenesis occurred at a lower rate when mAB1223 was injected at the hatched blastula stage than at later stages, it may be speculated that endogenous proteases degraded the injected antibodies. Using this technique, extracellular events in the blastocoel or the function of certain molecules expressed in blastocoel can be easily investigated in vivo.     

Effect of antikeratin microinjection on the embryonic development of Xenopus laevis

Anti-keratin monoclonal antibody AF5 was introduced into fertilized eggs of Xenopus laevis.,and its effects on embryonic development were studied.Survival rate of the antikeratin-injected embryos was much lower(only 35.67% at gastrula)than that of the control(74.85% at gastrula),in which embryos were injected with mouse IgG.Most of survivors in the experimental series showed aberrant external appearance.On the other hand,in cleavage stage,ie 2-7h after fertilization,immunohistochemical staining of embryos showed that the expermental embryos were mostly keratin negative,while embryos of the control ones were keratin positive.When introducing this antikeratin into one cell of a 2-cell embryo,only the uninjected half of the embryo continued its development while the other half could not develop at all.These results suggested that intact keratin cytoskeleton in early embryos is indispensable to the embryonic development of Xenopus laevis.     

Chronology of apoptosis in bovine embryos produced in vivo and in vitro

The postimplantation developmental potential of embryos can be affected by various forms of cell death, such as apoptosis, at preimplantation stages. However, correct assessment of apoptosis is needed for adequate inference of the developmental significance of this process. This study is the first to investigate the independent chronological occurrence of apoptotic changes in nuclear morphology and DNA degradation (detected by the TUNEL reaction) and incidences of nuclei displaying these features at various preimplantation stages of bovine embryos produced both in vivo and in vitro. Different elements of apoptosis were observed at various developmental stages and appeared to be differentially affected by in vitro production. Nuclear condensation was observed from the 6-cell stage in vitro and the 8-cell stage in vivo, whereas the TUNEL reaction was first observed at the 6-cell stage in vitro and the 21-cell stage in vivo. Morphological signs of other forms of cell death were also observed in normally developing embryos produced both in vivo and in vitro. The onset of apoptosis seems to be developmentally regulated in a stage-specific manner, but discrete features of the apoptotic process may be differentially regulated and independently modulated by the mode of embryo production. Significant differences in indices of various apoptotic features were not evident between in vivo- and in vitro-produced embryos at the morula stage, but such differences could be observed at the blastocyst stage, where in vitro production was associated with a higher degree of apoptosis in the inner cell mass.     

Mediolateral cell intercalation in the dorsal, axial mesoderm of Xenopus laevis

The pattern of mediolateral cell intercalation in mesodermal tissues during gastrulation and neurulation of Xenopus laevis was determined by tracing cells labeled with fluorescein dextran amine (FDA). Patches of the involuting marginal zone (IMZ) of early gastrula stage embryos, labeled by injection of FDA at the one-cell stage, were grafted to the corresponding regions of unlabeled host embryos. The host embryos were fixed at several stages, serially sectioned, and examined with fluorescence microscopy and three-dimensional reconstruction. Patterns of mixing of labeled and unlabeled cells show that mediolateral cell intercalation occurs in the posterior, dorsal mesoderm as this region undergoes convergent extension and differentiates into somites and notochord. In contrast, it does not occur in any dorsoventral sector of the anterior, leading edge of the mesodermal mantle. These results, taken with other evidence, suggest that the mesoderm of Xenopus consists of two subpopulations, each with a characteristic morphogenetic movement, cell behavior, and tissue fate. The migrating mesoderm (1) does not show convergent extension; (2) migrates and spreads on the blastocoel roof; (3) is dependent on this substratum for its morphogenesis; (4) shows little mediolateral intercalation; (5) consists of the anterior, early-involuting region of the mesodermal mantle; and (6) differentiates into head, heart, blood island, and lateral body wall mesoderm. The extending mesoderm (1) shows convergent extension; (2) is independent of the blastocoel roof in its morphogenesis; (3) shows extensive mediolateral intercalation; (4) consists of the posterior, late-involuting parts of the mesodermal mantle; and (5) differentiates into somite and notochord.