A mouse macrophage factor induces head structures and organizes a body axis in Xenopus

Gastrulation morphogenesis in synthetic systems

Recent advances in pluripotent stem cell culture allow researchers to generate not only most embryonic cell types, but also morphologies of many embryonic structures, entirely in vitro. This recreation of embryonic form from naïve cells, known as synthetic morphogenesis, has important implications for both developmental biology and regenerative medicine. However, the capacity of stem cell-based models to recapitulate the morphogenetic cell behaviors that shape natural embryos remains unclear. In this review, we explore several examples of synthetic morphogenesis, with a focus on models of gastrulation and surrounding stages. By varying cell types, source species, and culture conditions, researchers have recreated aspects of primitive streak formation, emergence and elongation of the primary embryonic axis, neural tube closure, and more. Here, we describe cell behaviors within in vitro/ex vivo systems that mimic in vivo morphogenesis and highlight opportunities for more complete models of early development.

Chromatin accessibility and histone acetylation in the regulation of competence in early development

As development proceeds, inductive cues are interpreted by competent tissues in a spatially and temporally restricted manner. While key inductive signaling pathways within competent cells are well-described at a molecular level, the mechanisms by which tissues lose responsiveness to inductive signals are not well understood. Localized activation of Wnt signaling before zygotic gene activation in Xenopus laevis leads to dorsal development, but competence to induce dorsal genes in response to Wnts is lost by the late blastula stage. We hypothesize that loss of competence is mediated by changes in histone modifications leading to a loss of chromatin accessibility at the promoters of Wnt target genes. We use ATAC-seq to evaluate genome-wide changes in chromatin accessibility across several developmental stages. Based on overlap with p300 binding, we identify thousands of putative cis-regulatory elements at the gastrula stage, including sites that lose accessibility by the end of gastrulation and are enriched for pluripotency factor binding motifs. Dorsal Wnt target gene promoters are not accessible after the loss of competence in the early gastrula while genes involved in mesoderm and neural crest development maintain accessibility at their promoters. Inhibition of histone deacetylases increases acetylation at the promoters of dorsal Wnt target genes and extends competence for dorsal gene induction by Wnt signaling. Histone deacetylase inhibition, however, is not sufficient to extend competence for mesoderm or neural crest induction. These data suggest that chromatin state regulates the loss of competence to inductive signals in a context-dependent manner.

Isolation of nanobodies against Xenopus embryonic antigens using immune and non-immune phage display libraries

The use of Xenopus laevis as a model for vertebrate developmental biology is limited by a lack of antibodies specific for embryonic antigens. This study evaluated the use of immune and non-immune phage display libraries for the isolation of single domain antibodies, or nanobodies, with specificities for Xenopus embryonic antigens. The immune nanobody library was derived from peripheral blood lymphocyte RNA obtained from a llama immunized with Xenopus gastrula homogenates. Screening this library by immunostaining of embryonic tissues with pooled periplasmic material and sib-selection led to the isolation of several monoclonal phages reactive with the cytoplasm and nuclei of gastrula cells. One antigen recognized by a group of nanobodies was identified using a reverse proteomics approach as nucleoplasmin, an abundant histone chaperone. As an alternative strategy, a semi-synthetic non-immune llama nanobody phage display library was panned on highly purified Xenopus proteins. This proof-of-principle approach isolated monoclonal nanobodies that specifically bind Nuclear distribution element-like 1 (Ndel1) in multiple immunoassays. Our results suggest that immune and non-immune phage display screens on crude and purified embryonic antigens can efficiently identify nanobodies useful to the Xenopus developmental biology community.

RARγ is required for mesodermal gene expression prior to gastrulation in Xenopus

The developing vertebrate embryo is exquisitely sensitive to retinoic acid (RA) concentration, particularly during anteroposterior patterning. In contrast to Nodal and Wnt signaling, RA was not previously considered to be an instructive signal in mesoderm formation during gastrulation. Here, we show in Xenopus that RARγ is indispensable for the expression of early mesoderm markers and is, therefore, an obligatory factor in mesodermal competence and/or maintenance. We identified several novel targets upregulated by RA receptor signaling in the early gastrula that are expressed in the circumblastoporal ring and linked to mesodermal development. Despite overlapping expression patterns of the genes encoding the RA-synthesizing enzyme Aldh1a2 and the RA-degrading enzyme Cyp26a1, RARγ1 functions as a transcriptional activator in early mesoderm development, suggesting that RA ligand is available to the embryo earlier than previously appreciated. RARγ1 is required for cellular adhesion, as revealed by spontaneous dissociation and depletion of ncam1 mRNA in animal caps harvested from RARγ1 knockdown embryos. RARγ1 knockdown obliterates somite boundaries, and causes loss of Myod protein in the presomitic mesoderm, but ectopic, persistent expression of Myod protein in the trunk. Thus, RARγ1 is required for stabilizing the mesodermal fate, myogenic commitment, somite boundary formation, and terminal skeletal muscle differentiation.

Xenopus hairy2b specifies anterior prechordal mesoderm identity within Spemann's organizer

Spemann's organizer is a region of the gastrula stage embryo that contains future anterior endodermal and dorsal mesodermal tissues. During gastrulation, the dorsal mesoderm is divided into the prechordal mesoderm and the chordamesoderm. However, little is known regarding how this division is established. We analyzed the role of the anterior prechordal mesoderm-specific gene Xhairy2b in the regionalization of the organizer. We found that mesoderm-inducing transforming growth factor-beta signaling induced Xhairy2b expression. On the other hand, the ectopic expression of Xhairy2b induced the expression of organizer-specific genes and resulted in the formation of a secondary dorsal axis lacking head and notochord structures. We also showed that Xhairy2b down-regulated the expression of ventral mesodermal, anterior endodermal, and chordamesodermal genes. In Xhairy2b-depleted embryos, defects in the specification of anterior prechordal mesoderm identity were observed as the border between the prechordal mesoderm and the chordamesoderm was anteriorly shifted. These results suggest that Xhairy2b establishes the identity of the anterior prechordal mesoderm within Spemann's organizer by inhibiting the formation of neighboring tissues.

A Vertebrate Homolog of the Cell Cycle Regulator Dbf4 Is an Inhibitor of Wnt Signaling Required for Heart Development

Early stages of vertebrate heart development have been linked to Wnt signaling. Here we show in both gain- and loss-of-function experiments that XDbf4, a known regulator of Cdc7 kinase, is an inhibitor of the canonical Wnt signaling pathway. Depletion of endogenous XDbf4 protein did not disturb gastrulation movements or early organizer genes but resulted in embryos with morphologically defective heart and eyes and suppressed cardiac markers. These markers were restored by overexpressed XDbf4, or an XDbf4 mutant that inhibits Wnt signaling but lacks the ability to regulate Cdc7. This indicates that the function of XDbf4 in heart development is independent of its role in the cell cycle. Moreover, our data suggest that XDbf4 acts through the physical and functional interaction with Frodo, a context-dependent regulator of Wnt signaling. These findings establish an unexpected function for a vertebrate Dbf4 homolog and demonstrate the requirement for Wnt inhibition in early cardiac specification.

Elongation of axolotl tailbud embryos requires GPI-linked proteins and organizer-induced, active, ventral trunk endoderm cell rearrangements

Application of phosphatidylinositol-specific phospholipase C to early tailbud stage axolotl embryos reveals that a specific subset of morphogenetic movements requires glycosylphosphatidylinositol (GPI)-linked cell-surface proteins. These include pronephric duct extension, "gill bulge" formation, and embryonic elongation along the anteroposterior axis. The work of Kitchin (1949, J. Exp. Zool. 112, 393-416) led to the conclusion that extension of the notochord provided the motive force driving anteroposterior stretching in axolotl embryos, elongation of other tissues being a passive response. We therefore conjectured that axial mesoderm cells might display the GPI-linked proteins required for elongation of the embryo. However, we show here that removal of most of the neural plate and axial and paraxial mesoderm prior to neural tube closure does not prevent elongation of ventrolateral tissues. Tissue-extirpation and tissue-marking experiments indicate that elongation of the ventral trunk occurs via active, directed tissue rearrangements within the endoderm, directed by signals emanating from the blastopore region. Extension of both dorsal and ventral tissues requires GPI-linked proteins. We conclude that elongation of axolotl embryos requires active cell rearrangements within ventral as well as axial tissues. The fact that both types of elongation are prevented by removal of GPI-linked proteins implies that they share a common molecular mechanism.

Origins and functions of phagocytes in the embryo

To review the data on the origins, phenotype, and function of embryonic phagocytes that has accumulated over past decade. Most of the relevant articles were selected based on the PubMed database entries. In additional, the Interactive Fly database (http://sdb.bio. purdue.edu/fly/aimain/1aahome.htm), FlyBase (http://flybase.bio. indiana.edu:82/), and TBase (http://tbase.jax.org/) were used to search for relevant information and articles. Phagocytes in a vertebrate embryo develop in two sites (yolk sac and liver) and contribute to organogenesis in part through their ability to recognize and clear apoptotic cells. Yolk sac-derived phagocytes differ in differentiation pathway and marker gene expression from macrophages produced via classic hematopoietic progenitors in the liver. We argue that yolk sac-derived phagocytes constitute a separate cell lineage. This conclusion raises the question of whether primitive phagocytes persist into the adulthood.

Work in progress: the renaissance in amphibian embryology

Various historical eras in the distant as well as the recent past of amphibian embryology are briefly reviewed. The concepts which emerged from the early years matured, then were laid to rest for several decades. A resurgence, driven by key discoveries with peptide growth factors, and fueled by modern molecular biology methods, is underway. The future for several amphibian research projects should be promising since interest in basic concepts remains strong, and application of frontier methodologies is yielding novel findings.

More than 95% reversal of left-right axis induced by right-sided hypodermic microinjection of activin into Xenopus neurula embryos

In recent years, genes that show left-right (L-R) asymmetric expression patterns have been identified one after another in vertebrate gastrula-neurula embryos. However, we still have little information about when the irreversible L-R specification is established in vertebrate embryos. In this report, we show that almost 100% of the embryos develop to be L-R-inverted larvae after microinjection of activin molecules into the right lateral hypodermic space of Xenopus neurula embryos. After right-side injection of 10-250 pg activin protein, both early neurulae just after gastrulation movement (stage 13-14) and late neurulae just before neural tube closure (stage 17-18) showed almost 100% reversal of the heart and gut L-R axes. At higher doses of activin, more than 90% of the L-R-inverted embryos showed L-R reversal of both heart and gut. The survival ratio of the right-injected 4-day embryos was 90% on average. In the left-injected embryos, the occurrence of L-R inversion was less than 2% as observed in normal untreated siblings (1.7%). When the same amount of activin (1-50 pg) was microinjected into both sides of neurula embryos, the incidence of L-R inversion was reduced to 58%. The injection of activin along the dorsal midline in the trunk region also randomized the visceral L-R axis. Injection of activin into the right side changed normal left-handed expression of Xnr-1 to right-handed or bilateral expression. In contrast, left-handed expression of Pitx2 was switched to the right side by right activin injection. This is the first report of a method that achieves complete inversion of the visceral L-R axis by treatment of embryos at the neurula stage. Activin not only acts on the neurulae to cancel the original L-R specification up to the late neurula stage, but also rebuilds a new L-R axis whose left side coincides with the injection side. It is suggested that the left and right halves of neurulae have equal potential for L-R differentiation.

Temporal restriction of MyoD induction and autocatalysis during Xenopus mesoderm formation

In Xenopus, the activation of the myogenic determination factors MyoD and Myf-5 in the muscle-forming region of the embryo occurs in response to mesoderm-inducing factors (MIFs). Different members of the FGF, TGF-beta, and Wnt protein families have been implicated in this process, but how MIFs induce the myogenic regulators is not known. For MyoD, the induction process may serve to locally stabilize a transient burst of ubiquitous transcription at the midblastula transition, possibly by triggering MyoD's autocatalytic loop. Here we have sought to distinguish separate activating functions during MyoD induction by analyzing when MyoD responds to different MIF signaling or to MyoD autoactivation. We show that MyoD induction depends on the developmental age of the induced cells, rather than on the type or time point of inducer application. At the permissive time, de novo MyoD induction by Activin requires less than 90 min, arguing for an immediate response, rather than a series of inductive events. MyoD autoactivation is direct, but subject to the same temporal restriction as MyoD induction by MIF signaling. Further evidence implicating MyoD autocatalysis as an essential component of the induction process comes from the observation that both autocatalysis and induction of MyoD are selectively repressed by a dominant-negative MyoD mutant. In summary, our observations let us conclude that MyoD's expression domain in the embryo results from an interplay of timed changes in cellular competence, pleiotropic signaling pathways, and autocatalysis.

Notochord repression of endodermal Sonic hedgehog permits pancreas development

Notochord signals to the endoderm are required for development of the chick dorsal pancreas. Sonic hedgehog (SHH) is normally absent from pancreatic endoderm, and we provide evidence that notochord, in contrast to its effects on adjacent neuroectoderm where SHH expression is induced, represses SHH expression in adjacent nascent pancreatic endoderm. We identify activin-betaB and FGF2 as notochord factors that can repress endodermal SHH and thereby permit expression of pancreas genes including Pdx1 and insulin. Endoderm treatment with antibodies that block hedgehog activity also results in pancreatic gene expression. Prevention of SHH expression in prepancreatic dorsal endoderm by intercellular signals, like activin and FGF, may be critical for permitting early steps of chick pancreatic development.

A POU protein regulates mesodermal competence to FGF in Xenopus

XLPOU91, a POU-homeobox gene is expressed in a narrow window during early Xenopus development. We show that ectopic expression of XLPOU91 RNA causes severe posterior truncations in embryos without inhibiting the formation of Spemann's organizer. Ectopic XLPOU91 expression also inhibits mesoderm induction by fibroblast growth factor (FGF) and activin in animal cap explants. Using antisense RNA, we depleted endogenous XLPOU91 protein in animal caps. Gastrula-stage animal caps expressing XLPOU91 antisense RNA do not lose competence to FGF, unlike controls, these animal caps express XBra after FGF treatment. Endogenous XLPOU91 levels are peaking when FGF mesoderm-inducing competence is lost in animal caps. Thus XLPOU91 protein may act as a competence switch during early development, as XLPOU91 levels increase in the embryo, the mesoderm response to FGF is lost.

Somatic linker histones cause loss of mesodermal competence in Xenopus

In Xenopus, cells from the animal hemisphere are competent to form mesodermal tissues from the morula through to the blastula stage. Loss of mesodermal competence at early gastrula is programmed cell-autonomously, and occurs even in single cells at the appropriate stage. To determine the mechanism by which this occurs, we have been investigating a concomitant, global change in expression of H1 linker histone subtypes. H1 histones are usually considered to be general repressors of transcription, but in Xenopus they are increasingly thought to have selective functions in transcriptional regulation. Xenopus eggs and embryos at stages before the midblastula transition are deficient in histone H1 protein, but contain an oocyte-specific variant called histone B4 or H1M. After the midblastula transition, histone B4 is progressively substituted by three somatic histone H1 variants, and replacement is complete by early neurula. Here we report that accumulation of somatic H1 protein is rate limiting for the loss of mesodermal competence. This involves selective transcriptional silencing of regulatory genes required for mesodermal differentiation pathways, like muscle, by somatic, but not maternal, H1 protein.

Xenopus msx1 mediates epidermal induction and neural inhibition by BMP4

Epidermal fate in Xenopus ectoderm has been shown to be induced by a secreted growth factor, Bone Morphogenetic Protein 4 (BMP4). However, the molecular mechanism mediating this response is poorly understood. Here, we show that the expression of the homeobox gene, msx1, is an immediate early response to BMP4 in Xenopus embryos. The timing of expression and embryonic distribution of msx1 parallel those described for BMP4. Moreover, overexpression of msx1 in early Xenopus embryos leads to their ventralization as described for BMP4. Consistent with mediating a BMP type of signaling, overexpression of msx1 is sufficient to induce epidermis in dissociated ectoderm cells, which would otherwise form neural tissue. Finally, msx1 can also rescue neuralization imposed by a dominant negative BMP receptor (tBR) in ectodermal explants. We propose that Xenopus msx1 acts as a mediator of BMP signaling in epidermal induction and inhibition of neural differentiation.

Localization of MAP kinase activity in early Xenopus embryos: implications for endogenous FGF signaling

We have used a sensitive assay for MAP kinase activity to investigate the role of endogenous fibroblast growth factor (FGF)-activated MAP kinase in early Xenopus embryonic patterning. MAP kinase activity is low during cleavage stages and increases significantly during gastrulation. The temporal profile of this activity correlates well with the expression pattern of Xenopus eFGF. Spatially, MAP kinase activity is lowest in animal pole tissue and higher in vegetal pole cells and the marginal zone. Endogenous MAP kinase activity is FGF receptor-dependent, demonstrating that FGF signaling is active in all three germ layers of the early embryo. This activity is necessary for normal expression of Mix.1, a mesoendodermal marker, in the endoderm as well as in the mesoderm, indicating that MAP kinase plays a functional role in patterning of both of these germ layers. Spatial and temporal changes in MAP kinase activation during gastrulation also suggest a role for FGF signaling in this process. In addition, we find that embryonic wounding during dissection results in significant stimulation of this pathway, providing a possible explanation for earlier observations of effects of surgical manipulation on cell fate in early embryos.

Graded amounts of Xenopus dishevelled specify discrete anteroposterior cell fates in prospective ectoderm

Signals emitted from the prospective dorsal marginal zone (the organizer) are thought to specify neuroectodermal cell fates along the anteroposterior (AP) axis, but the mechanisms underlying this signaling event remain to be elucidated. To assess the effect of Xenopus Dishevelled (Xdsh), a proposed component of the Wnt, Notch and Frizzled signal transduction pathways, on AP axis determination, it was supplied in varying doses to presumptive ectodermal cells. Two-fold increments in levels of microinjected Xdsh mRNA revealed a gradual shift in cell fates along the AP axis. Lower doses of Xdsh mRNA activated anterior neuroectodermal markers, XAG1 and Xotx2, whereas the higher doses induced more posterior neural tissue markers such as En2, Krox20 and HoxB9. At the highest dose of Xdsh mRNA, explants contained maximal amount of HoxB9 transcripts and developed notochord and somites. When compared with Xdsh, Xwnt8 mRNA also activated anterior neuroectodermal markers, but failed to elicit mesoderm formation. Analysis of explants overexpressing Xdsh at the gastrula stage revealed activation of several organizer-specific genes which have been implicated in determination of neural tissue (Xotx2, noggin, chordin and follistatin). Whereas Goosecoid, Xlim1 and Xwnt8 were not induced in these explants, another early marginal zone marker, Xbra, was activated at the highest level of Xdsh mRNA. These observations suggest that the effects of Xdsh on AP axis specification may be mediated by combinatorial action of several early patterning genes. Increasing levels of Xdsh mRNA activate posterior markers, whereas increasing amounts of the organizer stimulate the extent of anterior development (Stewart, R.M. and Gerhart, J.C. (1990) Development 109, 363-372). These findings argue against induction of the entire organizer by Xdsh in ectodermal cells and implicate signal transduction pathways involving Xdsh in AP axis determination. Thus, different levels of a single molecule, Xdsh, can specify distinct cell states along the AP axis.

Partnership between DPC4 and SMAD proteins in TGF-beta signalling pathways

The TGF-beta/activin/BMP superfamily of growth factors signals through heteromeric receptor complexes of type I and type II serine/threonine kinase receptors. The signal originated by TGF-beta-like molecules appears to be transduced by a set of evolutionarily conserved proteins known as SMADs, which upon activation directly translocate to the nucleus where they may activate transcription. Five SMAD proteins have so far been characterized in vertebrates. These factors are related to the mediator of decapentaplegic (dpp) signalling, mothers against dpp (Mad), in Drosophila and to the Sma genes from Caenorhabditis elegans. Smad1 and Smad2 have been shown to mimic the effects of BMP and activin, respectively, both in Xenopus and in mammalian cells, whereas Smad3 (a close homologue of Smad2) and the related protein DPC4, a tumour-suppressor gene product, mediate TGF-beta actions. We report here that DPC4 is essential for the function of Smad1 and Smad2 in pathways that signal mesoderm induction and patterning in Xenopus embryos, as well as antimitogenic and transcriptional responses in breast epithelial cells. DPC4 associates with Smad1 in response to BMP and with Smad2 in response to activin or TGF-beta. DPC4 is therefore a regulated partner of SMADs that function in different signalling pathways of the TGF-beta family.

Short-range signaling by candidate morphogens of the TGF beta family and evidence for a relay mechanism of induction

The specification and patterning of cell fates by a morphogen gradient is a unifying theme of developmental biology, yet little evidence exists for the presence of gradients in vivo or to show how such putative gradients form. Vg1 and activin are candidate morphogens involved in Xenopus mesoderm induction. This study suggests that these TGF beta family members act on adjacent cells but do not travel through the intact extracellular space to induce distant cells directly. Moreover, we present evidence for the presence of secondary inducing signals that could be involved in relaying signals to distant cells. These results suggest that if a localized cellular source of an inducer acts to pattern mesodermal cells at a distance in Xenopus embryos, it does so by a relay mechanism.

The secreted product of Xenopus gene lunatic Fringe, a vertebrate signaling molecule

Signaling molecules are essential for vertebrate embryonic development. Here, two Xenopus homologs of the Drosophila gene fringe, lunatic Fringe (lFng) and radical Fringe (rFng), were identified and the protein product of lFng further characterized. The messenger RNA of lFng is supplied as a maternal message. Its product is a precursor protein consisting of pre-, pro-, and mature regions. The mature lunatic Fringe protein is secreted extracellularly, and it induced mesodermal tissue formation in animal cap assays. These results indicate that secreted lunatic Fringe can induce mesoderm and reveal that the Fringe proteins are a family of vertebrate signaling molecules.

Regulation of the zebrafish goosecoid promoter by mesoderm inducing factors and Xwnt1

Goosecoid is a homeobox gene that is expressed as an immediate early response to mesoderm induction by activin. We have investigated the induction of the zebrafish goosecoid promoter by the mesoderm inducing factors activin and basic fibroblast growth factor (bFGF) in dissociated zebrafish blastula cells, as well as by different wnts in intact embryos. Activin induces promoter activity, while bFGF shows a cooperative effect with activin. We have identified two enhancer elements that are functional in the induction of the goosecoid promoter. A distal element confers activin responsiveness to a heterologous promoter in the absence of de novo protein synthesis, whereas a proximal element responds only to a combination of activin and bFGF. Deletion experiments show that both elements are important for full induction by activin. Nuclear proteins that bind to these elements are expressed in blastula embryos, and competition experiments show that an octamer site in the activin responsive distal element is specifically bound, suggesting a role for an octamer binding factor in the regulation of goosecoid expression by activin. Experiments in intact embryos reveal that the proximal element contains sequences that respond to Xwnt1, but not to Xwnt5c. Furthermore, we show that the distal element is active in a confined dorsal domain in embryos and responds to overexpression of activin in vivo, as well as to dorsalization by lithium. The distal element is to our knowledge the first enhancer element identified that mediates the induction of a mesodermal gene by activin.

Activin-like signal activates dorsal-specific maternal RNA between 8- and 16-cell stages of Xenopus

In many animals the dorsal-ventral axis forms by an initial localization of maternal molecules, which then regulate the spatial location of signals that directly influence the expression of axis-specific fates. Several recent studies have demonstrated that dorsal-animal blastomeres of the Xenopus morula (8-32 cells) are biased toward dorsal fates prior to mesoderm inductive signaling. In this study we ask whether the dorsal bias is the result of autonomous expression of maternal molecules specifically localized within dorsal cells or of early activating signals. It was found that although 16-cell dorsal-animal blastomeres (D1.1) can differentiate into dorsal tissues when cultured alone, the 8-cell mothers (D1) can not. Likewise, although RNA extracted from D1.1 can induce an extra dorsal axis when injected into vegetal blastomeres, RNA extracted from D1 can not. However, D1 does express dorsal tissues if co-cultured with dorsal-vegetal cells or with culture medium containing a mixture of activins (PIF-medium). Furthermore, short-term culture of D1 in PIF-medium enables the D1 RNA to induce an ectopic dorsal axis. Ventral-animal blastomeres also can express dorsal axial tissues when co-cultured with dorsal-vegetal blastomeres or in PIF-medium, but the RNA from the activin-treated ventral cells cannot induce ectopic dorsal axes. These studies demonstrate that there are maternal RNAs that, shortly after fertilization, are present only in the dorsal-animal region. They do not act cell autonomously, but require an activin-like signal. These RNAs may function by increasing the responsiveness of dorsal-animal blastomeres to the mesoderm inductive signals present in both the morula and the blastula.

Competition between noggin and bone morphogenetic protein 4 activities may regulate dorsalization during Xenopus development

Bone morphogenetic protein 4 (BMP-4) induces ventral mesoderm but represses dorsal mesoderm formation in Xenopus embryos. We show that BMP-4 inhibits two signaling pathways regulating dorsal mesoderm formation, the induction of dorsal mesoderm (Spemann organizer) and the dorsalization of ventral mesoderm. Ectopic expression of BMP-4 RNA reduces goosecoid and forkhead-1 transcription in whole embryos and in activin-treated animal cap explants. Embryos and animal caps overexpressing BMP-4 transcribe high levels of genes expressed in ventral mesoderm (Xbra, Xwnt-8, Xpo, Mix.1, XMyoD). The Spemann organizer is ventralized in these embryos; abnormally high levels of Xwnt-8 mRNA and low levels of goosecoid mRNA are detected in the organizer. In addition, the organizer loses the ability to dorsalize neighboring ventral marginal zone to muscle. Overexpression of BMP-4 in ventral mesoderm inhibits its response to dorsalization signals. Ventral marginal zone explants ectopically expressing BMP-4 form less muscle when treated with soluble noggin protein or when juxtaposed to a normal Spemann organizer in comparison to control explants. Endogenous BMP-4 transcripts are downregulated in ventral marginal zone explants dorsalized by noggin, in contrast to untreated explants. Thus, while BMP-4 inhibits noggin protein activity, noggin downregulates BMP-4 expression by dorsalizing ventral marginal zone to muscle. Noggin and BMP-4 activities may control the lateral extent of dorsalization within the marginal zone. Competition between these two molecules may determine the final degree of muscle formation in the marginal zone, thus defining the border between dorsolateral and ventral mesoderm.

Anti-dorsalizing morphogenetic protein is a novel TGF-beta homolog expressed in the Spemann organizer

We have identified a novel growth factor in Xenopus, which is most closely related to human Bone Morphogenetic Protein-3. Its expression peaks during gastrulation, most prominently in the Spemann organizer, and persists in the posterior neural floor plate and prechordal plate during neurulation. Injection of the corresponding mRNA into dorsal blastomeres results in dose-dependent suppression of dorsal and anterior structures, even in the presence of lithium chloride. Overexpression of the gene downregulates the dorsalizing factors noggin, goosecoid and follistatin, as well as the dorsal markers NCAM, muscle actin and MyoD; conversely, ventral markers are induced. We therefore designate this gene product Anti-Dorsalizing Morphogenetic Protein (ADMP). Though development of dorsoanterior structures is suppressed when exogenous ADMP is injected, the gene is induced by lithium chloride treatment or activin, both of which are known to produce the opposite effect. Thus, the expression of ADMP resembles that of several dorsalizing signals, but its product exerts dorsal-suppressing activity. This suggests that ADMP may moderate organizer-associated dorsalizing influences. These findings are also consistent with the recently advanced proposal of dorsally expressed inhibitory activin-like signals.

Control of the embryonic body plan by activin during amphibian development

Embryonic induction plays an important role in establishing the fundamental body plan during early amphibian development. The factors mediating this embryonic induction have, however, only recently been discovered. In the mid-1980's, certain peptide growth factors belonging to the FGF and TGF-beta families were found to have a mesoderm-inducing effect on isolated Xenopus blastula ectoderm. The study of embryonic induction subsequently expanded rapidly and knowledge at the molecular level has gradually accumulated. One of these peptide growth factors, activin, a member of the TGF-beta superfamily, is present maternally in the Xenopus early embryo and induces various mesodermal and endodermal tissues in isolated presumptive ectoderm. After exposure of presumptive ectoderm to activin, many genes are expressed in the same manner as in normal embryogenesis. Ectoderm treated with activin can induce a complete secondary embryo, the same as the organizer does in transplantation experiments. These findings suggest that activin is one of the first induction signals responsible for establishing the embryonic body plan in early amphibian development. In this article we shall review to what extent we can control the embryonic body plan in vitro, referring to some significant findings in this field.

FGF is a prospective competence factor for early activin-type signals in Xenopus mesoderm induction

Normal pattern formation during embryonic development requires the regulation of cellular competence to respond to inductive signals. In the Xenopus blastula, vegetal cells release mesoderm-inducing factors but themselves become endoderm, suggesting that vegetal cells may be prevented from expressing mesodermal genes in response to the signals that they secrete. We show here that addition of low levels of basic fibroblast growth factor (bFGF) induces the ectopic expression of the mesodermal markers Xbra, MyoD and muscle actin in vegetal explants, even though vegetal cells express low levels of the FGF receptor. Activin, a potent mesoderm-inducing agent in explanted ectoderm (animal explants), does not induce ectopic expression of these markers in vegetal explants. However, activin-type signaling is present in vegetal cells, since the vegetal expression of Mix.1 and goosecoid is inhibited by the truncated activin receptor. These results, together with the observation that FGF is required for mesoderm induction by activin, support our proposal that a maternal FGF acts at the equator as a competence factor, permitting equatorial cells to express mesoderm in response to an activin-type signal. The overlap of FGF and activin-type signaling is proposed to restrict mesoderm to the equatorial region.

Induction of dorsal mesoderm by soluble, mature Vg1 protein

Mesoderm induction during Xenopus development has been extensively studied, and two members of the transforming growth factor-beta family, activin beta B and Vg1, have emerged as candidates for a natural inducer of dorsal mesoderm. Heretofore, analysis of Vg1 activity has relied on injection of hybrid Vg1 mRNAs, which have not been shown to direct efficient secretion of ligand and, therefore, the mechanism of mesoderm induction by processed Vg1 protein is unclear. This report describes injection of Xenopus oocytes with a chimeric activin-Vg1 mRNA, encoding the pro-region of activin beta B fused to the mature region of Vg1, resulting in the processing and secretion of mature Vg1. Treatment of animal pole explants with mature Vg1 protein resulted in differentiation of dorsal, but not ventral, mesodermal tissues and dose-dependent activation of both dorsal and ventrolateral mesodermal markers. At high doses, mature Vg1 induced formation of ‘embryoids’ with a rudimentary axial pattern, head structures including eyes and a functional neuromuscular system. Furthermore, truncated forms of the activin and FGF receptors, which block mesoderm induction in the intact embryo, fully inhibited mature Vg1 activity. To examine the mechanism of inhibition, we have performed receptor-binding assays with radiolabeled Vg1. Finally, follistatin, a specific inhibitor of activin beta B which is shown not to block endogenous dorsal mesoderm induction, failed to inhibit Vg1. The results support a role for endogenous Vg1 in dorsal mesoderm induction during Xenopus development.

Dorsalizing and neuralizing properties of Xdsh, a maternally expressed Xenopus homolog of dishevelled

Signaling factors of the Wnt proto-oncogene family are implicated in dorsal axis formation during vertebrate development, but the molecular mechanism of this process is not known. Studies in Drosophila have indicated that the dishevelled gene product is required for wingless (Wnt1 homolog) signal transduction. We demonstrate that injection of mRNA encoding a Xenopus homolog of dishevelled (Xdsh) into prospective ventral mesodermal cells triggers a complete dorsal axis formation in Xenopus embryos. Lineage tracing experiments show that cells derived from the injected blastomere contribute to anterior and dorsal structures of the induced axis. In contrast to its effect on mesoderm, overexpression of Xdsh mRNA in prospective ectodermal cells triggers anterior neural tissue differentiation. These studies suggest that Wnt signal transduction pathway is conserved between Drosophila and vertebrates and point to a role for maternal Xdsh product in dorsal axis formation and in neural induction.

The SH2-containing protein-tyrosine phosphatase SH-PTP2 is required upstream of MAP kinase for early Xenopus development

SH-PTP2, the vertebrate homolog of Drosophila corkscrew, associates with several activated growth factor receptors, but its biological function is unknown. We assayed the effects of injection of wild-type and mutant SH-PTP2 RNAs on Xenopus embryogenesis. An internal phosphatase domain deletion (delta P) acts as a dominant negative mutant, causing severe posterior truncations. This phenotype is rescued by SH-PTP2, but not by the closely related SH-PTP1. In ectodermal explants, delta P blocks fibroblast growth factor (FGF)- and activin-mediated induction of mesoderm and FGF-induced mitogen-activated protein (MAP) kinase activation. Our results indicate that SH-PTP2 is required for early vertebrate development, acting as a positive component in FGF signaling downstream of the FGF receptor and upstream of MAP kinase.

Translational control of activin in Xenopus laevis embryos

Activin is a potent mesoderm inducing factor present in embryos of Xenopus laevis. Recent evidence has implicated activin in the inhibition of neural development in addition to the well-established induction of mesoderm in ectodermal explants. These diverse effects are critically dependent on the concentration of activin yet little is known about the mechanisms regulating the level of activin in the embryo. We report that the 3' untranslated region (3' UTR) of activin beta B mRNA inhibits the translation of activin in embryos. Micro-injection of activin mRNA from which the 3' UTR has been deleted is 8-10-fold more potent in inducing mesoderm than mRNA containing the 3' UTR. Truncation of the 3' UTR also leads to a marked enhancement of activin protein levels in embryos but has no effect when the truncated mRNA is translated in vitro. The 3' UTR also confers translational inhibition on a heterologous mRNA. These data show that a maternal factor(s) present in X. laevis regulates the translation of injected activin beta B mRNA. This factor(s) could be responsible for regulating the levels of endogenous activin beta B protein during mesoderm induction and the specification of ectodermal derivatives such as neural and epidermal tissues.

Glial differentiation: a review with implications for new directions in neuro-oncology

Major advances in cell culture techniques, immunology, and molecular biology during the last 10 years have led to significant progress in understanding the process of normal glial differentiation. This article summarizes our current understanding of the cellular and molecular basis of glial differentiation based on data obtained in cell culture and reviews current hypotheses regarding the transcriptional control of the gene switching that controls differentiation. Understanding normal glial differentiation has potentially far-reaching implications for developing new forms of treatment for patients with glial neoplasms. If oncogenesis truly involves a blockage or a short circuiting of the differentiation process in adult glial progenitor cells, or if it results from dedifferentiation of previously mature cells, then a clear understanding of differentiation may provide a key to understanding and potentially curtailing malignancy. Differentiation agents represent a relatively new class of drugs that effect cellular gene transcription at the nuclear level, probably through alterations in chromatin configuration and/or differential gene induction. These exciting new agents may provide a means of preventing the dedifferentiation of low-grade gliomas or inducing malignant glioma cells to differentiate with minimal toxicity. In the future, genetic therapy has the potential of more specifically rectifying the defect in genetic control that led to oncogenesis in any given tumor.

Vertebrate embryonic induction: mesodermal and neural patterning

Within the fertilized egg lies the information necessary to generate a diversity of cell types in the precise pattern of tissues and organs that comprises the vertebrate body. Seminal embryological experiments established the importance of induction, or cell interactions, in the formation of embryonic tissues and provided a foundation for molecular studies. In recent years, secreted gene products capable of inducing or patterning embryonic tissues have been identified. Despite these advances, embryologists remain challenged by fundamental questions: What are the endogenous inducing molecules? How is the action of an inducer spatially and temporally restricted? How does a limited group of inducers give rise to diversity of tissues? In this review, the focus is on the induction and patterning of mesodermal and neural tissues in the frog Xenopus laevis, with an emphasis on families of secreted molecules that appear to underlie inductive events throughout vertebrate embryogenesis.

Heparan sulfate proteoglycans are required for mesoderm formation in Xenopus embryos

Mesoderm forms in the vertebrate embryo as a result of inductive interactions involving secreted growth factors and cell surface molecules. Proteoglycans have recently been implicated in the control of cell adhesion, migration and growth factor responsiveness. We have found that removal of glycosaminoglycan chains of proteoglycans from Xenopus ectodermal explants by heparinase, but not by chondroitinase, results in inhibition of elongation and mesodermal differentiation in response to signaling factors: activin, FGF and Wnt. Heparinase treatment differentially affected expression of early general and region-specific mesodermal markers, suggesting that mesodermal cell fates become specified in the early embryo via at least two signaling pathways which differ in their requirements for heparan sulfate proteoglycans. Addition of soluble heparan sulfate restored activin-mediated induction of muscle-specific actin gene in heparinase-treated explants. Finally, heparinase inhibited autonomous morphogenetic movements and mesodermal, but not neural, differentiation in dorsal marginal zone explants, which normally give rise to mesoderm in the embryo. These results directly demonstrate that heparan sulfate proteoglycans participate in gastrulation and mesoderm formation in the early embryo.

Mesodermal patterning by an inducer gradient depends on secondary cell-cell communication

BACKGROUND

Gradients of inducing molecules, or morphogens, could impose pattern on early embryos. Although there are candidates for morphogens in several systems, it is not well understood how cells might translate differences in extracellular inducer concentration into an orderly arrangement of cell types. With this question in mind, we have re-examined mesodermal patterning in Xenopus in response to the secreted growth factor activin. Previous work has shown that activin can initiate the formation of a variety of mesodermal tissues in a concentration-dependent fashion. We have sought to disentangle the roles played by individual cell responses to activin and subsequent interactions among induced cells in producing this outcome.

RESULTS

We find that the initial response of dispersed cells to activin concentration is unexpectedly simple, showing neither the thresholds of activin concentration nor the distinct domains of gene expression that characterize the later response. The eventual emergence of an ordered series of coherent differentiation steps requires the reaggregation of the induced cells, implying that secondary interactions occur. Furthermore, when cells induced at different doses of activin are mixed, the final response apparently represents a consensus, rather than a mosaic, of the mixed populations.

CONCLUSIONS

We conclude that communication among responding cells underlies much of the remarkable patterning influence of activin. Moreover, we suggest that these findings can inform thinking about how inducer gradients might act in other systems, shifting emphasis from the initial response of cells to inducer concentration toward the elaboration of complex pattern by secondary interactions.

Neural induction in Xenopus

Although induction of neural tissue in vertebrates has been recognized since the experiments of Spemann and Mangold seventy years ago, only recently has the phenomenon been put onto a molecular footing. Three molecules, noggin, follistatin and fibroblast growth factor, have been shown to have neuralizing activity in various assays. These assays for neural induction and the molecular mechanisms for the action of neural inducers are reviewed here.

Heparitinase inhibition of mesoderm induction and gastrulation in Xenopus laevis embryos

We have examined the involvement of proteoglycan molecules in the induction of mesodermal tissue in Xenopus laevis embryos. Blastocoelic injections of the enzyme heparitinase at early blastula stages lead to gastrulation defects and to failures in the development of anterior embryonic structures. The period of sensitivity of embryos to this treatment suggests a possible role for these molecules during mesoderm induction. We show that heparan sulfate proteoglycans (HSPGs) and chondroitin sulfate proteoglycans are the predominant sulfated glycoconjugates synthesized in early Xenopus embryos and that HSPGs are degraded by blastocoelic injections of heparitinase. Further, bFGF induction of mesoderm in explants of Xenopus stage 8 embryonic animal cap tissue is blocked by heparitinase but not by Chondroitinase ABC, using three separate criteria of mesoderm induction. Since HSPGs present in blastula animal cap cells are digested by heparitinase under the culture conditions used in the mesoderm-induction assay, we suggest that cell-surface heparan sulfate proteoglycans are required for basic fibroblast growth factor-mediated mesoderm induction.

Inhibition of activin receptor signaling promotes neuralization in Xenopus

Expression of a truncated activin type II receptor, which blocks signaling by activin, neuralizes explants of embryonic cells that would otherwise become epidermal cells. This neuralization is direct and does not require the presence of mesoderm. The induced neural tissue expresses general molecular markers of the central nervous system as well as an array of neural markers along the anteroposterior axis. In the context of the whole embryo, expression of this truncated activin receptor diverts prospective ectoderm and endoderm to a neural fate. We propose that inhibition of the activin type II receptor signaling causes the cells of Xenopus embryos to adopt a neural fate. These results, along with previous experiments performed in Drosophila, suggest that the formation of the nervous system in vertebrates and invertebrates occurs by a common strategy.

Follistatin, an antagonist of activin, is expressed in the Spemann organizer and displays direct neuralizing activity

In the accompanying paper, we show that the expression of a dominant negative activin receptor can convert prospective ectoderm into neural tissue, which suggests that activin is an inhibitor of neuralization. Here we report the isolation and characterization of an activin antagonist, follistatin, that can induce neural tissue directly in vivo. Follistatin RNA is localized in the Spemann organizer and notochord, tissues known to be potent neural inducers. We demonstrate that follistatin RNA and protein are able to block the activity of activin in embryonic explants. Furthermore, we show that follistatin RNA directly neuralizes ectodermal explants in the absence of detectable mesoderm. Thus, follistatin is present at the correct time and location to play a role in neural induction in vivo.

Conversion of a mesodermalizing molecule, the Xenopus Brachyury gene, into a neuralizing factor

It has been shown previously that a Xenopus homolog of the mouse gene Brachyury, Xbra, can initiate mesodermal differentiation. Here, I report that a Xbra mutant truncated at the carboxyl terminus, B304, has lost the mesodermalizing activity and can block the activity of the wild-type Xbra. Injection of B304 mRNA led to formation of neural structures in animal cap explants. Examination of molecular markers in B304-injected explants shows expression of anterior neural markers in the absence of mesodermal markers, indicating that B304 can cause neuralization without the mediation of mesoderm. Implications of these findings on intracellular mechanisms underlying the initiation of neural differentiation in the ectodermal cells are discussed.

Differential expression of a Distal-less homeobox gene Xdll-2 in ectodermal cell lineages

Neural induction in Xenopus requires the activation of new sets of genes that are necessary for cellular and regional specification of the neural tube. It has been reported earlier that members of the Distal-less homeobox gene family are specifically activated in distinct regions of the central nervous system (CNS) of Xenopus embryos (Dirksen et al., 1993; Papalopulu and Kintner, 1993). In this paper we describe in detail a Xenopus homeobox containing gene Xdll-2, which belongs to the Distal-less gene family. In contrast to other previously described Xenopus family members, Xdll-2 is expressed in the embryonic ectoderm and is specifically repressed in the CNS. This repression can be mimicked in isolated animal caps by treatment with activin. Expression of Xdll-2 persists in the epidermis and some neural crest cells. Because of its spatial and temporal expression pattern this gene is a good candidate to have a regulatory function in the initial formation of the epidermis. Its high level of expression in adult skin indicates that its function is continuously required in this tissue.

Fibroblast growth factor, but not activin, is a potent activator of mitogen-activated protein kinase in Xenopus explants

Isolated explants from the animal hemisphere of Xenopus embryos were incubated with Xenopus basic fibroblast growth factor (XbFGF) or human activin A. XbFGF incubation resulted in the rapid activation of mitogen-activated protein kinase (MAPK) and ribosomal S6 protein kinase (pp90rsk) in a dose-dependent manner with the highest levels of activation occurring at 50 ng/ml. Maximal activation occurred within 6-10 min after the addition of growth factor, and the activity of both kinases declined to unstimulated levels after 30 min. Activin was unable to activate either MAPK or pp90rsk in the Xenopus explants to a substantial level, although it induced dorsal mesoderm better than XbFGF under the same experimental conditions. The regulatory protein Xwnt-8 did not activate MAPK, nor did it enhance the activation of MAPK by XbFGF. XbFGF was able to activate MAPK through at least the midgastrula stage, suggesting that this family of growth factors may have a role in gastrula-stage events.

Mesoderm induction by activin requires FGF-mediated intracellular signals

We have examined the role of FGF signaling during activin-mediated mesoderm induction in Xenopus. Using dominant inhibitory mutants of FGF signal transducers to disrupt the FGF-signaling pathway at the plasma membrane or in the cytosol prevents animal cap blastomeres from expressing several mesodermal markers in response to exogenous activin. Dominant inhibitory mutants of the FGF receptor, c-ras or c-raf inhibit the ability of activin to induce molecular markers of both dorsal and ventral mesoderm including Xbra, Mix1 and Xnot. Some transcriptional responses to activin such as goosecoid and Xwnt8 are inhibited less effectively than others, however, suggesting that there may differing requirements for an FGF signal in the responses of mesoderm-specific genes to activin induction. Despite the requirement for this signaling pathway during activin induction, downstream components of this pathway are not activated in response to activin, suggesting that activin does not signal directly through this pathway.

Wnt-3a regulates somite and tailbud formation in the mouse embryo

Amphibian studies have implicated Wnt signaling in the regulation of mesoderm formation, although direct evidence is lacking. We have characterized the expression of 12 mammalian Wnt-genes, identifying three that are expressed during gastrulation. Only one of these, Wnt-3a, is expressed extensively in cells fated to give rise to embryonic mesoderm, at egg cylinder stages. A likely null allele of Wnt-3a was generated by gene targeting. All Wnt-3a-/Wnt-3a- embryos lack caudal somites, have a disrupted notochord, and fail to form a tailbud. Thus, Wnt-3a may regulate dorsal (somitic) mesoderm fate and is required, by late primitive steak stages, for generation of all new embryonic mesoderm. Wnt-3a is also expressed in the dorsal CNS. Mutant embryos show CNS dysmorphology and ectopic expression of a dorsal CNS marker. We suggest that dysmorphology is secondary to the mesodermal and axial defects and that dorsal patterning of the CNS may be regulated by inductive signals arising from surface ectoderm.

Zinc, iron, and copper contents of Xenopus laevis oocytes and embryos

Zinc is essential for vertebrate development; its deficiency results in multiple congenital malformations. Knowledge of the zinc biochemistry that underlies embryologic development is very limited. This has led us to investigate the zinc, iron, and copper contents of Xenopus laevis oocytes and embryos. Stage 1-6 oocytes, isolated from ovaries, and stage 1-40 embryos, obtained by in vitro fertilization techniques, were washed in metal-free water prior to digestion by 70% ultrapure HNO3. The metal content of the digests was analyzed by atomic absorption spectrometry. Stage 6 oocytes contain 65.8 +/- 4, 31.1 +/- 3, and 0.68 +/- 0.2 ng of zinc, iron and copper, respectively. The corresponding concentrations are 1, 0.5, and 0.01 mM in 1 microliter eggs. The metal content varies as a function of egg maturation. The zinc content increases from 3-7 to > 60 ng by stages 3 and 6, respectively. A similar pattern is noted for iron, which increases from 2-5 to 30 ng at analogous stages. In contrast, the copper content remains virtually unchanged in oocytes undergoing maturation. Importantly, the total of all three metals does not vary throughout the first 50 stages of development, when all tadpole organs are forming. Hence, the full complement of zinc, iron, and copper needed for incorporation into apoproteins during development is already present at a time when oocyte maturation is completed. The specific metalloproteins that store, donate, and accept these metals during induction and organogenesis and the alterations caused by metal deficiency can now be identified.