Vertebrate anteroposterior patterning: the Xenopus neurectoderm as a paradigm

This review discusses formation of the vertebrate anteroposterior (AP) axis, focusing on the dorsal ectoderm, which gives rise to the nervous system, using the frog Xenopus as a model. After summarizing classical models of AP neural patterning, we describe recent molecular studies that are encouraging re-examination of these models. Such studies have shown that AP ectodermal patterning occurs by the onset of gastrulation, much earlier than previously thought. The identity of tissues that determine AP pattern is discussed, and the definition of the Organizer is reconsidered. The activity of factors secreted by inducing tissues in early patterning decisions is assessed and formulated into a revised model for Xenopus AP neural patterning. Finally, AP ectodermal patterning in Xenopus dorsal ectoderm is compared to that of other germ layers, and to other vertebrates.     

Misexpression of the RNA-binding protein ELRB in Xenopus presumptive neurectoderm induces proliferation arrest and programmed cell death.

Proteins of the ELAV/Hu family share the presence of three RNA binding domains. In Xenopus, three nervous system-specific elav/Hu related genes, elrB, elrC and elrD, have been identified so far. The temporally regulated expression patterns of elrB, elrC and elrD suggest their involvement at different steps of neural differentiation. In the present study we misexpressed elrB by RNA injection in early Xenopus embryos and analyzed morphologically and molecularly its effects on neural development. We showed that heterochronous expression of elrB in presumptive neurectoderm down-regulates the expression of neural markers, such as N-tubulin, as well as that of other Xenopus elav-like genes, elrC and elrD, whereas ectopic expression of elrB in presumptive mesoderm has no effect on MyoD. Misexpression of elrB also induces severe defects in neural tube development, associated with massive cell loss resulting from early cell cycle arrest and programmed cell death. Our results are discussed in the context of early neural differentiation.     

Xenopus p63 expression in early ectoderm and neurectoderm

The tumor-suppressor protein p53 belongs to a small gene family that includes p63 and p73. While p53 and p73 regulate cell cycle progression and apoptosis, the major role of p63 appears to be in promoting ectodermal proliferation and differentiation. In this report we describe the cloning of a Xenopus orthologue of mammalian p63 that is extraordinarily conserved in sequence. The major sites of expression of Xenopus p63 mRNA are the epidermis and some neural crest and crest derivatives such as the branchial arches and tail fin. Expression is also observed in the neural plate and in the stomodeal-hypophyseal anlage. Antibodies against p63 detect a nuclear protein that is distributed in a manner similar to that of Xp63 mRNA. Both mRNA and protein are conspicuously absent from regions of the epidermal sensorial layer that are induced to form a number of (but not all) ectodermal placodes and Xp63 protein levels are particularly dynamic in the epidermis of the eye as the lens forms.     

Progressive determination during formation of the anteroposterior axis in Xenopus laevis

The cement gland is an ectodermal organ in the head of frog embryos, lying anterior to any neural tissue. As analyzed by specific RNA expression, cement gland, like neural tissue, was induced by the dorsal mesoderm. Interestingly, mesoderm with the highest cement gland-inducing potential lay posterior to the ectoderm fated to form this organ, indicating that its induction occurred at a distance from the inducer source. Cement gland induction first occurred during early gastrulation. However, most initially induced cells did not contribute to the mature cement gland, but instead formed part of the neural plate. This change in fate could be reconstituted in vitro. These results suggest that determination of part of the anteroposterior axis occurs progressively, where future neural ectoderm is first induced to a cement glandlike state. As gastrulation proceeds, further induction by mesoderm may override this state, which persists only in the extreme anterior of the embryo.     

The control of Xenopus embryonic primary neurogenesis is mediated by retinoid signalling in the neurectoderm

In Xenopus, the primary neurons form in three domains either side of the midline in the posterior neurectoderm. At the late neurula stage there are approximately 120 primary sensory neurons on each side of the embryo. Co-injecting synthetic mRNA encoding retinoic acid receptor alpha (NR1B1) and retinoid X receptor beta (NR2B2) results in an increase in the number of primary neurons and this is further enhanced by the addition of retinoic acid indicating that elevated retinoid signalling promotes an increase in the number of cells undergoing primary neurogenesis. However, primary neurogenesis remains confined to the three domains that normally give rise to primary neurons indicating that not all regions of the neurectoderm respond equivalently to elevated retinoid signalling. The inhibition of retinoid signalling with a dominant negative retinoid receptor or treatment with citral, an inhibitor of retinoid metabolism, inhibits the formation of primary neurons. However, the lateral extent of the neurectoderm does not differ following these experimental manipulations suggesting that changes in primary neuron cell number, in response to changes in retinoid signalling, cannot be accounted for by significant gains or losses of neurectoderm. In addition, two lines of evidence are presented to suggest that retinoid signalling affects primary neurogenesis by acting directly on the neurectoderm. First, animal caps neuralized by noggin undergo primary neurogenesis in response to retinoid signalling and second primary neurogenesis is elevated in neural conjugates in which the ectodermal, but not the mesodermal, component has been co-injected with RAR/RXR mRNA.     

Mapping of the presumptive brain regions in the neural plate of Xenopus laevis

Two cell autonomous fluorescent labels (DiI and Hoechst) were used as vital markers in a fate map study of the Xenopus neural plate and ridge. Most areas of the brain derive from the neural plate in a fate map that is consistent with the topology of a sheet rolling into a tube, i.e., neighboring areas are maintained as neighbors. This has enabled us not only to plot the fates of larval brain structures, but also to suggest their primordial orientation in the neural plate. Since overlapping areas of the plate gave rise to overlapping regions of the central nervous system (CNS), we have been able to construct a space-filling model of the neural plate, whereby the number of founder cells for each brain region fate-mapped may be estimated roughly. Much of the telencephalon, ventral forebrain, and dorsal brain stem derives from the neural ridge and not the neural plate in the stage 15 Xenopus embryo. The structures of the forebrain were examined in detail because there were indications of substantial cell movements in this region. The anterior pituitary arises from the mid-anterior ridge, while hypothalamic structures arise from the midline regions of the anterior neural plate. Consistent groups of ventral hypothalamic structures were labeled when fluorescent markers were applied to these parts of the neural plate, indicating stereotyped cell movements. Detailed comparisons were made between the fate map of the Ambystoma neural plate (Jacobson, 1959) and that of Xenopus.     

Long-term culture of Xenopus presumptive ectoderm in a nutrient-supplemented culture medium

Animal cap assay is a useful experimental model for investigating the activity of inducers in amphibian development. This assay has revealed that activin A is a potent mesoderm-inducing factor. However, it has been very difficult to induce highly differentiated tissues such as cartilage in a 3-4 day culture period. It was recently reported that jaw cartilage was induced in vitro in an animal cap that had been cultured for 14 days in Steinberg's solution using the sandwich culture method and activin A. Under these conditions, necrosis was occasionally observed in the explants. In this study, we have achieved long-term animal cap cultures in a nutrient-supplemented culture medium designated RDX. This medium was made by modifying the saline concentration of the RD medium previously developed as a basal medium for the serum-free culture of various kinds of mammalian cells. The explants cultured in RDX grew more vigorously compared with those in Steinberg's solution. RDX medium promoted a wider variety of tissue induction and gene expression in the animal caps than Steinberg's solution, and also increased the frequency of cartilage induction. Therefore, the supplemental nutrients may support and promote the differentiation of cartilage. This long-term culture method using RDX medium is useful for studying the differentiation of tissues or organs such as cartilage in vitro.     

Xenopus oocytes and the biochemistry of cell division

The control of cell proliferation involves both regulatory events initiated at the plasma membrane that control reentry into the cell cycle and intracellular biochemical changes that direct the process of cell division itself. Both of these aspects of cell growth control can be studied in Xenopus oocytes undergoing meiotic maturation in response to mitogenic stimulation. All mitogenic signaling pathways so far identified lead to the phosphorylation of ribosomal protein S6 on serine residues, and the biochemistry of this event has been investigated. Insulin and other mitogens activate ribosomal protein S6 kinase II, which has been cloned and sequences in oocytes and other cells. This enzyme is activated by phosphorylation on serine and threonine residues by an insulin-stimulated protein kinase known as MAP-2 kinase. MAP kinase itself is also activated by direct phosphorylation on threonine and tyrosine residues in vivo. These results reconstitute one step of the insulin signaling pathway evident shortly after insulin receptor binding at the membrane. Several hours after mitogenic stimulation, a cell cycle cytoplasmic control element is activated that is sufficient to cause entry into M phase. This control element, known as maturation-promoting factor or MPF, has been purified to near homogeneity and shown to consist of a complex between p34cdc2 protein kinase and cyclin B2. In addition to apparent phosphorylation of cyclin, regulation of MPF activity involves synthesis of the cyclin subunit and its periodic degradation at the metaphase----anaphase transition. The p34cdc2 kinase subunit is regulated by phosphorylation/dephosphorylation on threonine and tyrosine residues, being inactive when phosphorylated and active when dephosphorylated. Analysis of phosphorylation sides in histone H1 for p34cdc2 has revealed a consensus sequence of (K/R)S/TP(X)K/R, where the elements in parentheses are present in some but not all sites. Sites with such a consensus are specifically phosphorylated in mitosis and by MPF in the protooncogene pp60c-src. These results provide a link between cell cycle control and cell growth control and suggest that changes in cell adhesion and the cytoskeleton in mitosis may be regulated indirectly by MPF via protooncogene activation. S6 kinase II is also activated upon expression of MPF in cells, indicating that MPF is upstream of S6 kinase on the mitogenic signaling pathway. Further study both of the signaling events that lead to MPF activation and of the substrates for phosphorylation by MPF should lead to a comprehensive understanding of the biochemistry of cell division.     

Pattern and morphogenesis of presumptive superficial mesoderm in two closely related species, Xenopus laevis and Xenopus tropicalis

The mesoderm, comprising the tissues that come to lie entirely in the deep layer, originates in both the superficial epithelial and the deep mesenchymal layers of the early amphibian embryo. Here, we characterize the mechanisms by which the superficial component of the presumptive mesoderm ingresses into the underlying deep mesenchymal layer in Xenopus tropicalis and extend our previous findings for Xenopus laevis. Fate mapping the superficial epithelium of pregastrula stage embryos demonstrates ingression of surface cells into both paraxial and axial mesoderm (including hypochord), in similar patterns and amounts in both species. Superficial presumptive notochord lies medially, flanked by presumptive hypochord and both overlie the deep region of the presumptive notochord. These tissues are flanked laterally by superficial presumptive somitic mesoderm, the anterior tip of which also appears to overlay the presumptive deep notochord. Time-lapse recordings show that presumptive somitic and notochordal cells move out of the roof of the gastrocoel and into the deep region during neurulation, whereas hypochordal cells ingress after neurulation. Scanning electron microscopy at the stage and position where ingression occurs suggests that superficial presumptive somitic cells in X. laevis ingress into the deep region as bottle cells whereas those in X. tropicalis ingress by "relamination" (e.g., [Dev. Biol. 174 (1996) 92]). In both species, the superficially derived presumptive somitic cells come to lie in the medial region of the presumptive somites during neurulation. By the early tailbud stages, these cells lie at the horizontal myoseptum of the somites. The morphogenic pathway of these cells strongly resembles that of the primary slow muscle pioneer cells of the zebrafish. We present a revised fate map of Xenopus, and we discuss the conservation of superficial mesoderm within amphibians and across the chordates and its implications for the role of this tissue in patterning the mesoderm.     

Neuronal determination without cell division in Xenopus embryos

Cell division in the Xenopus CNS was blocked by incubating embryos in a mixture of the DNA synthesis inhibitors hydroxyurea and aphidicolin. Surprisingly, embryos treated at the beginning of gastrulation proceeded normally through neurulation, neural tube closure, and CNS subdivision. Thus, cell division is not critical for neural induction or early morphogenetic events in the CNS. Neuroblasts in treated embryos differentiated into neurons of many classes, indicating that cellular determination in the CNS can be dissociated from lineage and birth date. Axonal tracts and embryonic reflexes also developed. The remarkable amount of normal CNS development that occurs in these animals may be explained by a series of sequential inductions that are largely independent of cell proliferation.     

The inducing capacity of the presumptive endoderm of Xenopus laevis studied by transfilter experiments

Summary The inducing capacity of the vegetal hemisphere of early amphibian blastulae was studied by placing a Nucleopore filter (pore size 0.4 m) between isolated presumptive endoderm and animal (ectodermal) caps. The inducing effect was shown to traverse the Nucleopore membrane. The reacting ectoderm differentiated into mainly ventral mesodermal derivatives. Expiants consisting of five animal caps also formed dorsal mesodermal and neural structures. Those results together with data published elsewhere suggest that, in addition to a vegetalizing factor, different mesodermal factors must be taken into consideration for the induction of either the ventral or the dorsal mesodermal derivatives. The neural structures are thought to be induced by the primarily induced dorsal mesodermal tissue. Electron microscopic (TEM) examination did not reveal any cell processes in the pores of the filter. The results indicate that transmissible factors rather than signals via cytoplasmic contacts or gap junctions are responsible for the mesodermal induction of ectodermal cells. The data support the view that in normogenesis the mesoderm is determined by the transfer of inducing factors from vegetal blastomeres to cells of the marginal zone (presumptive mesodermal cells).     

Different activities of the frizzled-related proteins frzb2 and sizzled2 during Xenopus anteroposterior patterning

In a search for factors that regulate patterning of the Xenopus anteroposterior (A/P) axis, particularly the anterior ectoderm, we isolated two members of the Frizzled-related protein (FRP) gene family that are thought to encode antagonists of Wnt signaling. frzb2 is expressed in head mesoderm while sizzled2 is expressed in ventral ectoderm and mesoderm, tissues that modulate anterior fates. Consistent with a role for these genes in A/P patterning, ectopically expressed frzb2 inhibited head formation, while sizzled2 dorsalized embryos, causing expansion of the head. The different activities of frzb2 and sizzled2 may be explained by their interaction with distinct proteins since frzb2 is an inhibitor of Xwnt8 activity, while sizzled2 is unable to inhibit the activity of Xwnt8 or any other Xwnt tested. The data suggest that anteroposterior patterning is modulated by multiple components of the Wnt signaling pathway.     

Growth-arrest-specific protein 2 inhibits cell division in Xenopus embryos

Background

Growth-arrest-specific 2 gene was originally identified in murine fibroblasts under growth arrest conditions. Furthermore, serum stimulation of quiescent, non-dividing cells leads to the down-regulation of gas2 and results in re-entry into the cell cycle. Cytoskeleton rearrangements are critical for cell cycle progression and cell division and the Gas2 protein has been shown to co-localize with actin and microtubules in interphase mammalian cells. Despite these findings, direct evidence supporting a role for Gas2 in the mechanism of cell division has not been reported.

Methodology and Principal Findings

To determine whether the Gas2 protein plays a role in cell division, we over-expressed the full-length Gas2 protein and Gas2 truncations containing either the actin-binding CH domain or the tubulin-binding Gas2 domain in Xenopus laevis embryos. We found that both the full-length Gas2 protein and the Gas2 domain, but not the CH domain, inhibited cell division and resulted in multinucleated cells. The observation that Gas2 domain alone can arrest cell division suggests that Gas2 function is mediated by microtubule binding. Gas2 co-localized with microtubules at the cell cortex of Gas2-injected Xenopus embryos using cryo-confocal microscopy and co-sedimented with microtubules in cytoskeleton co-sedimentation assays. To investigate the mechanism of Gas2-induced cell division arrest, we showed, using a wound-induced contractile array assay, that Gas2 stabilized microtubules. Finally, electron microscopy studies demonstrated that Gas2 bundled microtubules into higher-order structures.

Conclusion and Significance

Our experiments show that Gas2 inhibits cell division in Xenopus embryos. We propose that Gas2 function is mediated by binding and bundling microtubules, leading to cell division arrest.     

A morphogen gradient of Wnt/beta-catenin signalling regulates anteroposterior neural patterning in Xenopus.

Anteroposterior (AP) patterning of the vertebrate neural plate is initiated during gastrulation and is regulated by Spemann's organizer and its derivatives. The prevailing model for AP patterning predicts a caudally increasing gradient of a 'transformer' which posteriorizes anteriorly specified neural cells. However, the molecular identity of the transforming gradient has remained elusive. We show that in Xenopus embryos (1) dose-dependent Wnt signalling is both necessary and sufficient for AP patterning of the neuraxis, (2) Wnt/beta-catenin signalling occurs in a direct and long-range fashion within the ectoderm, and (3) that there is an endogenous AP gradient of Wnt/beta-catenin signalling in the presumptive neural plate of the Xenopus gastrula. Our results indicate that an activity gradient of Wnt/beta-catenin signalling acts as transforming morphogen to pattern the Xenopus central nervous system.     

Forming and Interpreting Gradients in the Early Xenopus Embryo

The amphibian embryo provides a powerful model system to study morphogen gradients because of the ease with which it is possible to manipulate the early embryo. In particular, it is possible to introduce exogenous sources of morphogen, to follow the progression of the signal, to monitor the cellular response to induction, and to up- or down-regulate molecules that are involved in all aspects of long-range signaling. In this article, I discuss the evidence that gradients exist in the early amphibian embryo, the way in which morphogens might traverse a field of cells, and the way in which different concentrations of morphogens might be interpreted to activate the expression of different genes.The idea that a morphogen gradient activates the expression of different genes at different concentrations was perhaps stated most clearly by Wolpert''s French flag model, in which a graded signal activates the expression of “blue,” “white,” and “red” genes at high, intermediate, and low concentrations (Wolpert 1969). Since that original work, great progress has been made in identifying morphogens and their target genes and it is now clear that the spatial pattern of gene expression in the developing embryo is frequently established by graded signals of this sort. But many questions remain, and in particular little is known about how gradients are established in the embryo with the necessary precision and how cells interpret different concentrations of morphogen to activate different genes. I discuss these issues with respect to mesoderm induction in the developing amphibian embryo.     

The mode and molecular mechanisms of the migration of presumptive PGC in the endoderm cell mass of Xenopus embryos

We investigated the mode of migration of presumptive primordial germ cells (pPGC) in the endoderm cell mass of Xenopus embryos at stages 7-40. The molecules underlying the migration were also studied cytochemically and immunocytologically. By examining the relative positions of pPGC and somatic cells derived from the single, fluorescein-dextran lysine (FDL)-injected, germ plasm-bearing cells of stage 6 embryos, pPGC in embryos at stages 7-23 and those at stages later than 24 were assumed to passively and actively migrate in the endoderm cell mass, respectively. This assumption was supported by the observation that F-actin, essential for active cell migration, was recognized on pPGC of the latter stages, but never on those of the former ones. In addition, the molecule like CXC chemokine receptor 4 (CXCR4) found on directionally migrating PGC in mouse and zebrafish, probably Xenopus CXCR4 (xCXCR4), was detected on pPGC only at latter stages. Accordingly, F-actin and xCXCR4, and probably beta1-integrin and collagen type IV, which are indispensable for the formation of F-actin, are thought to be involved in the active migration of pPGC in the endoderm cell mass.