We are searching data for your request:
Upon completion, a link will appear to access the found materials.
In many textbook, figures of embryo are drawn,but in reality how biologist know which zone is this one of an embryo in gastrulation stade? Except the dorsal lip here i can't localize other thing.
Arkadia enhances nodal-related signalling to induce mesendoderm
Nodal-related members of the transforming growth factor (TGF)-β family regulate the induction of mesoderm, endoderm, and mesendoderm, a tissue specific to the Spemann organizer 1,2,3,4,5,6,7 . How these different tissues form in response to the same signalling molecules is not completely understood. It has been suggested that concentration-dependent effects, mediated by extracellular cofactors and antagonists, are responsible for the differences 1,8,9,10 . Here we show that the nuclear protein Arkadia specifically potentiates the mesendoderm-inducing activity of a subset of TGF-β family members. The combined activities of Arkadia and Xenopus nodal-related-1 are sufficient to induce mesendoderm and suppress mesoderm. Arkadia dorsalizes ventral tissues, resulting in the induction of organizer-specific gene expression. Blocking nodal signalling extracellularly inhibits these effects. Arkadia influences nodal activity when co-expressed and can function in cells adjacent to those producing the nodal signal. Our findings, together with the observation that Arkadia mutant mice lack a node and node-derived mesendoderm, identify Arkadia as an essential modulator of the nodal signalling cascade that leads to induction of Spemann's organizer.
This year, 2019, marks the centennial of embryologist E. E. Just's discovery of what is known as the fast block to polyspermy. Just's observation of the subtle changes that occur at the egg's surface during fertilization (and in experimental parthenogenesis) led him to postulate that the egg, and indeed every cell, possesses a property he called independent irritability, which represents the cell's ability to respond in a physiologically-relevant way to a variety of signals or triggers. In this paper, I argue that Just's concept of independent irritability informed his contemporary Johannes Holtfreter as Holtfreter attempted to explain the phenomena of embryonic induction and competence and that Holtfreter, in turn, influenced Marc Kirschner and John Gerhart in their formulation of the theory of facilitated variation. Just's influence is especially evident in Gerhart and Kirschner's presentations of what they call weak linkage—a property of living systems that allows core processes and components to be mixed and matched in different ways to generate novel traits. Unfortunately, the connection between Holtfreter's work and Just's has remained hidden. This paper gives examples of phenomena that exhibit weak linkage, and it lays out the case that Just's concept of independent irritability, through Holtfreter, Gerhart, and Kirschner, has broadly infiltrated modern cell and developmental biology.
“Yet the recent great advances in [our understanding of the importance of the cell surface] will require us to go back to Just's book [The Biology of the Cell Surface] to reassess its contemporary relevance. Perhaps, just perhaps, he might be another Mendel or Miescher, doing fundamental work unappreciated for decades after his death” (Glass, 1984 ).
MATERIALS AND METHODS
Oocytes and embryos
Oocytes were manually defolliculated and cultured as described previously (Kofron et al., 1999). Oocytes were injected with oligos in oocyte culture medium (OCM) using two equatorial injections per oocyte for XTcf3 oligos, or one vegetal injection for VegT and Axin oligos, cultured at 18°C and fertilized using the host transfer technique as described previously (Zuck et al., 1998). Rescue experiments were carried out as described in the text either by injecting mRNA equatorially into oocytes, 24 hours after the oligo (Fig. 2F), or by injecting into 4-cell stage embryos (Fig. 2E). Eggs were stripped and fertilized using a sperm suspension and embryos were maintained in 0.2× MMR. For injections of mRNA after fertilization (Fig. 2E), embryos were dejellied, and transferred to 2% Ficoll in 0.5× MMR at the 1-cell stage. mRNAs were diluted with sterile distilled water and injected into blastomeres.
For animal cap assays, mid-blastula embryos were placed on 2% agarose dishes in 1× MMR and the animal caps were dissected using sharp forceps. The caps were then cultured in OCM until sibling embryos reached the mid- to late-gastrula stage. For equatorial explants, embryos were placed on 2% agarose dishes at the mid-blastula stage and the equatorial regions were dissected with tungsten needles (Xanthos et al., 2001) and cultured in OCM until sibling embryos reached the mid-neurula stage. For separation into dorsal and ventral halves at the gastrula stage, the dorsal side of embryos was marked at the four-cell stage using Nile blue crystals. The dorsoventral axis was recognized at the four-cell stage by the pigmentation differences of the dorsal and ventral sides. When wild-type embryos reached stage 10, all the batches were placed on 2% agarose dishes in 1× MMR pH 7.6 and bisected into dorsal and ventral halves, and frozen in groups of 4 half-embryos at 2-hour intervals through the gastrula stages.
Oligos and mRNAs
The antisense oligodeoxynucleotides used were HPLC purified phosphorothioate-phosphodiester chimeric oligonucleotides (Sigma/Genosys) with the base composition:
XTcf3 T1: 5′-C*G*A*G*GGATCCCAGTC*T*T*G*G-3′.
XTcf3 T2: 5′-G*A*G*ATAACTCTGA*T*G*G-3′.
The XTcf3 oligos are completely complementary to all four variants of XTcf3. Asterisks (*) represent phosphorothioate bonds. Oligos were resuspended in sterile, filtered water and injected in doses as described in the text. Full-length XTcf3 in the vector pGlomyc was linearized with XbaI and capped XTcf3 mRNA was synthesized using the T7 mMessage mMachine kit (Ambion). RNAs were phenol extracted, ethanol precipitated and then resuspended in sterile distilled water for injection.
Analysis of gene expression using real-time RT-PCR
Total RNA was prepared from oocytes, embryos and explants using proteinase K and then treated with RNase-free DNase as described previously (Zhang et al., 1998). Approximately one-sixth embryo equivalent of RNA was used for cDNA synthesis with oligo(dT) primers followed by real-time RT-PCR and quantitation using the LightCycler™ System (Roche) as described by Kofron et al. (Kofron et al., 2001). The primers and cycling conditions used are listed in Table 1. Relative expression values were calculated by comparison to a standard curve generated by serial dilution of uninjected control cDNA. Samples were normalized to levels of ornithine decarboxylase (ODC), which was used as a loading control. Samples of water alone or controls lacking reverse transcriptase in the cDNA synthesis reaction failed to give specific products in all cases.
GRADIENTS FOR THE PRIMARY BODY AXES IN VERTEBRATES
There are two primary body axes, anteroposterior (AP) and dorsoventral (DV). Usually, however, only one organizer is assumed to exist—the Spemann-type organizer. How can two orthogonal positional information systems emerge under the influence of a single organizer? Is there a second organizer, which has been so far overlooked? There are good arguments that the organizer for the AP pattern is not the Spemann-organizer but the entire marginal zone (Meinhardt 2006). In the early gastrula, Wnt is produced in the marginal zone except for the organizer region (Christian and Moon 1993). Wnt provides positional information for the separation into fore- and midbrain (Kiecker and Niehrs 2001 Nordström et al. 2002 Dorsky et al. 2003). This pattern-forming system is evolutionarily very old. A comparison of gene expressions in hydra and the early vertebrate gastrula shows a surprising correspondence, suggesting that patterning of the vertebrate brain and heart evolved from a system that was once responsible for the patterning of the body of a hydra-like ancestor (Fig. 2B) (Meinhardt 2002). In this view, the hydra organizer and the vertebrate blastopore, i.e., marginal zone, germ ring, etc., are homologous structures that are responsible for the AP patterning.
In contrast to the hydra organizer, the vertebrate blastopore evolved into a huge ring with Spemann organizer forming a small patch on this ring. The Spemann organizer is then assumed to pattern the DV axis, but it does so indirectly by giving rise to the dorsal midline, the notochord, and floor plate—a “high line” and not a “high spot” for the DV patterning. Both organizing regions, the blastopore for the AP and the midline for the DV axis, form a near Cartesian coordinate system that allows a combinatorial patterning along both axes (Fig. 2C). The generation of a single long-extended “high line” for the DV patterning is a subtle pattern-forming process. The vertebrate solution is not the only one. In insects, for instance, a dorsal organizer exerts a repressing influence, causing the midline to appear at the opposite ventral side (Meinhardt 2004, 2008), much in contrast to vertebrates in which the organizer initiates and elongates the midline dorsally. This model provides a rational for the dorsal or ventral location of the central nervous system in vertebrates and insects, respectively.
Chordin/Sizzled-dependent scaling model
We recently reported the molecular mechanism of scaling in bisected Xenopus embryos (Inomata et al. 2013 ). The dorsal half embryo should form a proper Chordin gradient according to the embryo size by regulating three factors, synthesis–diffusion–degradation. However, if the three factors dynamically change their value, it will be difficult to analyze the scaling system. To eliminate this complexity, we artificially created source cells, which synthesized a constant amount of Chordin protein without being affected by the embryo size or BMP activity (D–V axis reconstitution assay). First, embryos were injected with β-catenin-MO and chordin/noggin-MO (βCN-MO) to eliminate the source cells (organizer) and the expression of endogenous dorsalizing factors, respectively (Fig. 4) (Heasman et al. 2000 Oelgeschlager et al. 2003a ). Next, using these completely ventralized embryos that lacked a gradient, we locally injected chordin mRNA to exogenously create a Chordin gradient in the embryo. This resulted in the formation of three distinct regions, the dorsal (D), lateral (L), and ventral (V) regions, as in the control embryo. When the production rate of Chordin protein was enhanced by injecting fourfold higher chordin mRNA, embryos demonstrated a moderate change in the D–V axis formation. Therefore, the Chordin gradient shape appeared to be mainly regulated by degradation.
From these results, we focused on the association between Chordin and Sizzled, which controls the stability of the Chordin protein (Fig. 2B). To examine the diffusion rate, we performed fluorescence recovery after photobleaching (FRAP) or fluorescence correlation spectroscopy (FCS) assay and found that Chordin and Sizzled, as well as the secretory form of mEGFP, quickly diffused. In contrast, the degradation rate of each protein was distinctly different. The Sizzled protein was stable in the embryo, whereas the Chordin protein (26.7 fmol per embryo) degraded very fast with a half-life of approximately 30 min. This instability of Chordin protein was completely blocked by the excess amount of Sizzled, indicating that the majority of Chordin degradation depended on the metalloproteases BMP1 and Xlr. Furthermore, the Chordin gradient dynamically changed its shape from steep to shallow depending on the Sizzled concentration, even when the production rate of Chordin was fixed by the D–V axis reconstitution assay. These results indicated that the shape of the Chordin gradient was mainly regulated by degradation whose rate depended on the amount of Sizzled protein.
Based on the experimental observation, we proposed the Chordin/Sizzled-dependent scaling model mentioned below. Prior to gastrulation, bmps and its target gene sizzled were expressed in the whole embryo (Fig. 5A). However, when the organizer (source cells) was locally formed in the embryo at the early gastrula stage, synthesized Chordin diffused in the extracellular space and gradually suppressed the sizzled expression area by inhibiting the BMPs activity (Fig. 5A). During this process, the Sizzled protein was gradually accumulated in the embryo because of its low degradation rate (Fig. 5B). Considering the dorsal half embryo, the sizzled expression area was rapidly suppressed by Chordin diffusion (Fig. 5A bottom). The Sizzled accumulation became lower than that in the control embryo (Fig. 5B bottom). In this low-Sizzled dorsal half embryo, Chordin degradation was enhanced and formed a steep gradient suitable for the small embryo (Fig. 5C).
In this scaling model, we proposed that the embryo size regulates Chordin protein stability via the accumulation of the Sizzled protein. To address this possibility, Chordin production was fixed using the D–V axis reconstitution assay, and the embryo size was artificially changed by bisection. Despite fixed production, the reduction of Chordin protein was detected in the bisected embryos. Consistent with the Chordin/Sizzled-dependent scaling model, this change in Chordin protein stability was eliminated when Sizzled was depleted by morpholino. Furthermore, we constructed a mathematical model based on the experimental results: (i) identical diffusion rate of Chordin and Sizzled (ii) lower degradation rate of Sizzled than that of Chordin (iii) Sizzled-dependent regulation of Chordin degradation and (iv) suppression of the Sizzled expression area by Chordin diffusion. In this mathematical model, we confirmed that three distinct regions, dorsal, lateral, and ventral, could scale to the embryo size through the accumulation of Sizzled.
The Xenopus Spemann organizer has provided a fertile fishing ground for the discovery of secreted proteins that regulate development. It was expected that new growth factors might be isolated however, instead, it was found that the Spemann organizer mediates embryonic induction through the secretion of a mixture of growth factor antagonists (4, 5). In the present study, we used deep sequencing to investigate the choice between epidermis and neural tissue.
A Rich Transcriptomic Resource.
The transcriptome of animal cap cells that had been dissociated for several hours (causing neutralization), as well as that of ectodermal explants microinjected with a number of mRNAs that induce neural tissue, such as Chordin, Cerberus, and FGF8, was determined by RNA-seq. We also examined the effect of the endomesoderm inducer Xnr2, the epidermal inducer BMP4, and the mesoderm induction competence modifier xWnt8 (32). These data, which comprise a minimum of 45 × 10 9 sequenced nucleotides of RNA, are provided in Datasets S1–S3, which can be readily mined by the research community. This constitutes an important open resource for developmental biologists interested in germ layer differentiation.
Isolation of a Wnt Inhibitor.
By searching for neural induction genes activated by cell dissociation (which causes MAPK activation) (19) and by searching for Cerberus, Chordin, and xWnt8 mRNAs, we identified a protein that we designated as Bighead due to its overexpression phenotype. Unexpectedly, this molecule was not expressed in the ectoderm of late gastrula stage 12 when the RNA-seq libraries were prepared. At this stage, Bighead mRNA is expressed in the endoderm, particularly in the dorsal Spemann organizer. Organizer expression is found in the deep endoderm but does not overlap with the leading-edge anterior endoderm (which gives rise to the foregut and liver), which expresses Cerberus and Dkk1 (24, 41). In light of the requirement of Bighead for head development, it appears that Wnt antagonists must emanate also from the most posterior endoderm regions of the organizer to fully empower its head-inducing properties.
It is unlikely that dissociation of animal caps induces endoderm, since the pan-endodermal marker Sox17 is not expressed (Dataset S1). It seems likely that dissociation of animal caps leads to premature activation of the neural domains of Bighead expression, which, in the undisturbed embryo, are observed at later neurula stages. The identification of Bighead was fortunate, because it proved an interesting protein.
Since X. laevis is allotetraploid, Bighead is encoded by two genes from the S and L forms (20). Both encode proteins of about 270 aa with a signal peptide and are secreted. In overexpression experiments, Bighead caused phenotypes very similar to the archetypal Wnt antagonist Dkk1 (41). Bighead mRNA expanded the expression of a number of head markers, blocked expression of the En2 Wnt target gene, prevented secondary axis formation after a single injection of xWnt8 mRNA, and decreased induction of the early Wnt targets Siamois and Xnr3. Further, addition of Bighead protein inhibited canonical Wnt signaling in luciferase reporter gene assays. Thus, Bighead behaves as a canonical Wnt signaling antagonist, many of which are known to promote development of the head (48).
Extensive searches for homologs of Bighead in other organisms showed that it is only present in fish and amphibians. For example, in zebrafish, Bighead corresponds to LOC571755, a protein of unknown function. The protein evolved rapidly, but its six cysteines were conserved throughout many species. SWISS-MODEL prediction suggests that the C-terminal region of Bighead is compatible with the crystal structure of the prodomain of TGF-βs such as myostatin/GDF8 (38, 40) perhaps part of Bighead derived from a structural domain in the proregion of an ancient TGF-β.
No homologs were found in reptiles, birds, or mammals. Gene loss is very common during evolution. For example, we have described an ancient self-organizing network of Chordin/BMP/Tolloid that regulates D/V patterning in vertebrates and invertebrates (49). However, despite this deep conservation, some components of the network were lost. Anti-dorsalizing morphogenetic protein (ADMP) is a BMP that was lost in the platypus (Ornithorhynchus) (50). The sFRPs Crescent and Sizzled are present in birds and the platypus, but not in higher mammals, which have lost the egg yolk. In addition, sFRPs are not present in any invertebrates (51). It appears that the embryonic requirement for the level of regulation provided by Bighead was lost together with the invention of the amnion. Despite this, our studies with Bighead depletion by MOs demonstrate a remarkably strong requirement for this gene in head formation during frog development.
Why so Many Wnt Antagonists?
Bighead adds to a large list of secreted Wnt antagonists. These include the Dkk proteins (48), sFRPs, Wnt-inhibitory factor 1 (WIF-1) (52), SOST/Sclerostin (53), Notum (a hydrolase that removes palmitoleoylate from Wnt in the extracellular space) (54), and Angptl4 (6). In addition, transmembrane proteins such as Shisa (a protein involved in trafficking of Frizzled receptor to the cell surface) (55), Tiki (a protease that cleaves the amino terminus of Wnts) (56) and Znrf3/RNF43 (a ubiquitin ligase that targets Frizzled and Lrp6 receptors for lysosomal degradation) (57, 58) down-regulate Wnt signaling.
As shown in this study, Bighead binds to Lrp6, inducing its rapid endocytosis into lysosomes. As a result, Lrp6 is removed from the surface of the cell and degraded in endolysosomes. This molecular mechanism is very similar to that of the Wnt antagonists Dkk1 and Angptl4. Dkk1 binds to Lrp6 and Kremen1/2, and the complex is internalized. Angptl4 is a secreted protein best known for its role as an inhibitor of lipoprotein lipase (LPL), the key enzyme in the removal of triglycerides from blood plasma (59). Studies in Xenopus have shown that Angptl4 binds to cell surface syndecans (which are transmembrane proteoglycans) and that this interaction triggers endocytosis of Lrp6 (6). In the case of Bighead, it is not known whether a coreceptor is required for Lrp6 internalization. What is clear, however, is that these three Wnt antagonists lead to the internalization of Lrp6 into an endolysosomal population that is not involved in signal generation.
The existence of so many regulators underscores the rich complexity of the Wnt signaling pathway. We usually think of canonical Wnt as a signal that merely increases nuclear β-catenin levels to regulate transcription by T-cell factor/lymphoid enhancer-binding factor (TCF/LEF). However, Wnt has additional effects. For example, in Wnt-dependent stabilization of proteins, hundreds of cellular proteins become stabilized, leading to an increase in the size of the cell (60, 61). This is caused by the sequestration of GSK3 inside late endosomes/multivesicular bodies (MVBs) (62, 63), decreasing the phosphorylation of phosphodegrons in cytosolic proteins that normally lead to their degradation in proteasomes. In addition to GSK3, another important cytosolic enzyme, protein arginine methyltransferase 1 (PRMT1), is sequestered inside MVBs when the Wnt coreceptors are endocytosed together with their Wnt ligand (64). The recent realization that Wnt3a greatly stimulates non–receptor-mediated endocytosis of BSA-DQ from the extracellular medium (64) suggests that Wnt is a major regulator of membrane trafficking. We propose that Lrp5/6 is a major regulator not only of the trafficking of Wnts but also of the overall cellular fluid and nutrient uptake. Endocytosis is a universal cellular property that could be regulated by Dkk1, Angptl4, and Bighead. Much remains to be learned about the physiology of the remarkable Wnt signaling pathway (65, 66).
We thank F. Cong for providing the ZNRF3 constructs D. Koinuma for providing the ALK constructs C. Janda for providing the WNT surrogate construct R. Thomas for H1581 cells. We acknowledge G. Roth and Aska Pharmaceuticals Tokyo for generously providing hCG. We thank NXR (RRID: SCR_013731), Xenbase (RRID: SCR_004337), and EXRC for Xenopus resources. We thank Fabio da Silva for critical reading of the manuscript. Expert technical support by the DKFZ core facility for light microscopy and the central animal laboratory of DKFZ is gratefully acknowledged. This work was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – SFB1324 – project number 331351713.
SICKLE ENDOBLAST (SECONDARY HYPOBLAST)
Sickle endoblast forms a unicellular far-extending layer proliferating from the medial rim of the Rauber's sickle or junctional endoblast in a centripetal and cranial direction during early incubation (Callebaut and Van Nueten, 1994 Callebaut et al., 1999 Fig. 13). Principally, it contains γ ooplasm.
Sickle Endoblast Belongs to the Same Cell Lineage as Rauber's Sickle and Has a Similar Behavior but Is Dominated by Rauber's Sickle
If a quail sickle endoblast fragment is placed on the anti-sickle region of an unincubated chicken blastoderm from which Rauber's sickle has been selectively scraped away, then a whole embryo with a PS, definitive endoderm, Hensen's node, and a neural plate develops in a diametrically opposed direction, starting from the anti-sickle region (Callebaut et al., 2003a , b ). The same occurs when a fragment of quail sickle endoblast is placed on the isolated central part of an unincubated chicken blastoderm (Callebaut et al., 2002c ). Central subgerminal ooplasm artificially placed in contact with Rauber's sickle material or sickle endoblast in culture, can function as a substrate for cellular proliferation with again inducing and/or regenerating capacities in the neighboring upper layer (Callebaut et al., 2000c ). Even activation of embryo formation can occur by unfertilized quail blastodiscs (Callebaut et al., 2000d ).
When quail sickle endoblast is placed on the isolated anti-sickle region of an unincubated chicken blastoderm in culture, an early neural plate develops. By contrast, when a piece of quail sickle endoblast is placed on the anti-sickle region of a whole unincubated chicken in culture, it has no inducing effect. This finding indicates that Rauber's sickle dominates or inhibits ectopically placed sickle endoblast, which is derived from the same cell lineage. This sickle endoblast, if withdrawn from the influence of Rauber's sickle, has gastrulation- and/or neurulation-inducing potencies on the upper layer of the unincubated blastoderm, but it has no influence on blood island formation. The homeobox gene cHex is expressed in Rauber's sickle and sickle endoblast (Yatskievych et al., 1997 ). cHex transcripts were also detected within blood islands beginning at stage 4 (Hamburger and Hamilton, 1951 ) and in extraembryonic and intraembryonic vascular endothelial cells. Because we have shown that Rauber's sickle and junctional endoblast have an inducing effect on blood island formation, we can postulate an unknown relationship with the cHex gene.
Influence of Sickle Endoblast on Neurulation and Gastrulation
The molecular basis of neural induction has been extensively studied in Xenopus laevis, and it was found to be tightly coupled to the establishment of the dorsoventral axis (De Robertis and Sasai, 1996 Hemmati-Brivanlou and Thomsen, 1995 Hemmati-Brivanlou and Melton, 1997 ). In frogs, the prospective ectoderm is induced by BMPs. In contrast, a neural development requires the inactivation of BMPs and is achieved by direct complex formation between BMPs and neural-inducing factors such as chordin, noggin, or follistatin (Piccolo et al., 1996 Zimmermann et al., 1996 ). In the chick blastoderm at early stages, the prospective epidermis is characterized by the expression of the homeobox gene DLX5, which remains an epidermal marker during gastrulation and neurulation and enables it to be distinguished from the more central neural plate (Pera et al., 1999 ). That vertical signals from the lower layer are necessary for the establishment of the neural plate has been shown by the latter authors by repeated extirpations of the underlying endoblast. In the absence of the lower germ layers, the epidermis expanded into the region that normally forms the neural plate.
Knoetgen et al. ( 1999a , b ) analyzed the GANF (Gallus anterior neural fold) -inducing potential of various tissues at different stages during chick development by transplantation to the outer margin of the area pellucida, where the epiblast cells are fated to become epidermis (Spratt, 1952 Rosenquist, 1966 Schoenwolf and Sheard, 1990 Bortier and Vakaet, 1992 Garcia-Martinez et al., 1993 ). Transplants of Hensen's node (HH3+/HH4) on whole blastoderms led to the induction of a neuroectodermal structure with a strong expression of GANF in its cranial margin. Grafting of the young head process (HH4) to the lateral cranial area pellucida caused a thickening of the epiblast and an induction of GANF expression in juxtaposed cells.
A secreted molecule named “Cerberus,” which is expressed in anterior endoderm, has the property to induce ectopic head structures when microinjected into ventral regions of Xenopus embryos (Bouwmeester et al., 1996 Bouwmeester, 1997 ). The patterning of the chick forebrain anlage by the prechordal plate has been described by Pera and Kessel ( 1997 ). According to these authors also, the avian neural plate is evident before the first mesendodermal or axial mesodermal cells ingress, excluding the prechordal plate and the notochord as primary sources for neural induction. During early gastrulation, cells invaginate through the tip of the growing streak and spread radially to form the definitive (gut) endoderm (Vakaet, 1970 ). During this radial expansion, the latter definitive endoderm pushes the sickle endoblast also radially (Callebaut and Van Nueten, 1994 Fig. 13). The cranial hemicircular sickle endoblast slides under upper layer cells that transform into a hemicircular neural plate anlage (Bortier and Vakaet, 1992 ). The latter cells are localized close to the former anti-sickle region, exactly in the concavity of the cranially displaced endophilic crescent. The remaining more caudal sickle endoblast is localized under the upper layer, which will give rise to the PS-forming area centralis region. This different evolution in the cranial (anti-sickle) region vs. the central (area centralis) region probably can be explained by the different reactivity in these two upper layer regions (Callebaut et al., 2002c ).
The absence of neural induction after the grafting experiments with the deep layer on whole blastoderms by Gallera and Nicolet ( 1969 ) and by Knoetgen et al. ( 1999a , b ) can probably be explained by the full presence of Rauber's sickle material. This finding indicates also that the earlier conclusions from grafting experiments on whole unincubated blastoderms (containing Rauber's sickle, the primary major organizer, or on PS blastoderms, containing Hensen's node, a secondary major organizer) must be reconsidered. Therefore, we cannot agree with either of the conclusions of Knoetgen et al. ( 1999b ) that the endoblast on its own elicits any detectable change in the adjacent host ectoblast after transplantation or that the avian organizer is confined to Hensen's node only. Foley et al. ( 2000 ) studied the eventual role of the early deep layer (endophyll and/or sickle endoblast) on the expression of the molecular markers Sox3 (Uwanogho et al., 1995 ) and Otx2 (Bally-Cuif et al., 1995 ) in the upper layer. From Hamburger and Hamilton stage 6–7 on, Sox3 is specifically expressed in the entire chicken neural plate and Otx2 is expressed throughout the forebrain and midbrain. Foley et al. ( 2000 ) found that the early deep layer regulates an early transient phase of Otx2 and Sox3 expression in the adjacent upper layer. Therefore, they concluded that the early deep layer does not induce neural tissue or forebrain definitively. However, their transplantation experiments were not performed on Rauber's sickle—or junctional endoblast-free blastoderm fragments but on whole blastoderms. Recently, Knezevic and Mackem ( 2001 ) found evidence that two genes, later associated with the gastrula organizer (Gnot-1 and Gnot-2), are induced by the deep layer signals in prestreak embryos. According to the latter authors, these genes could perhaps regulate axis formation in the early embryo, which could also explain the induction of a streak in the isolated part of the area centralis by sickle endoblast (Callebaut et al., 2003b ).
Part 3: Early Frog Development: How to Make a Tadpole or a Twin
00:00:0708 I'm Richard Harland.
00:00:0808 I'm at UC Berkeley.
00:00:0926 And today I'm going to tell you about the signaling activities that give rise to
00:00:1322 the neural plate, the forerunner of the spinal cord and brain in vertebrate embryos.
00:00:1914 In previous talks, I've talked about the introduction to the Xenopus embryo, why it has some advantages
00:00:2509 for experiments.
00:00:2609 And I've talked about the cell shape changes that have happened in the. in the embryo
00:00:3015 that lead to the three-layered ectoderm/mesoderm/endoderm structure that is the case, the.
00:00:3625 the state at which neural tissue formation happens.
00:00:4120 Let's initially start by just reviewing the classic experiment, the organizer experiment
00:00:4520 done by Hilde Mangold and Hans Spemann.
00:00:4819 And this is the one that really set the stage for what I'm going to talk about.
00:00:5304 In this experiment, they used marked embryos.
00:00:5607 They used newts of different colors: a dark-colored, pigmented newt, and a light-colored newt.
00:01:0104 And what they were. asking the question, what happens if we move pieces of the embryo around?
00:01:0814 Do they differentiate according to how they normally were set up in the embryo?
00:01:1209 It's self-differentiating according to their original fate?
00:01:1502 Or do they adopt the fate of their surroundings?
00:01:1728 Are they told to become the fate of their surroundings?
00:01:2017 Well, there was one particular experiment that was quite spectacular,
00:01:2426 where not only did the cells self-differentiate, but they recruited cells from the rest of the embryo
00:01:3025 to make new structures, the phenomenon of embryonic induction.
00:01:3604 So here, what they did was to take this embryo, here. the dark embryo is the host, and there's
00:01:4212 this paler donor.
00:01:4328 And what they did was to cut out this dorsal lip region from the pale embryo, flip it around,
00:01:5013 and graft it in to the ventral side of the host.
00:01:5318 So, not only does this have its normal organizer side but it has a new piece of dorsal mesoderm
00:01:5904 stuck in the ventral side.
00:02:0028 And what they found was that this was enough to make the embryo twin.
00:02:0602 And that's shown down here in a representation, this particular one done by Andrea Wills,
00:02:1014 who was a student with me.
00:02:1204 So, you can see that there's a normal primary axis with a head with two eyes.
00:02:1619 Then down here, there's a secondary axis, which is fused down at the tail, but it's
00:02:2116 a complete, proper, organized secondary axis.
00:02:2524 Now, the important thing was, since they were using marked issues, they could tell
00:02:3006 what is the contribution of the graft and the host.
00:02:3228 And so here's a. a representation of the section that was made by Hilde Mangold.
00:02:3702 Here's the primary axis, with the tissues that should be familiar to us by now:
00:02:4121 the nervous system, the notochord, and the muscle.
00:02:4510 And here's the secondary axis.
00:02:4626 And there's the pale graft.
00:02:4801 The pale graft invariably contributes just to the midline tissues.
00:02:5215 Here, it's contributing to the notochord and a little bit of the somites.
00:02:5707 Sometimes it would contribute to a bit of the spinal cord.
00:02:5926 But the consistent observation is that it contributes to the midline tissues,
00:03:0420 whereas the bulk of these tissues are recruited from the host: most of the nervous system,
00:03:0810 the muscle, and so on.
00:03:1202 So this graft really must be instructing the surroundings to make this second axis and
00:03:1620 organize it properly.
00:03:1805 Here's a modern equivalent of the experiment, also done by Andrea Wills, where she's
00:03:2216 labeled the donor embryo with a red stain.
00:03:2515 And you can see in this twin, where the two axes have arranged themselves conveniently
00:03:3013 next to each other.
00:03:3117 Here is the red-labelled graft.
00:03:3308 And you can see above it the induced neural tissue.
00:03:3613 So, it's not a phenomenon just of the 1920s, but can be done in the current times.
00:03:4221 Now, I'm really going to emphasize that this induction mechanism was not obvious.
00:03:4714 And in fact, Warren Lewis, who is not as famous as Spemann and Mangold, tried a similar experiment
00:03:5313 earlier than they did.
00:03:5504 But what he concluded -- largely because the embryos were not marked --
00:03:5915 was that the result he was getting was exclusively the result of self-differentiation of the grafted tissue.
00:04:0521 And he couldn't see any induction, because the tissues were not marked.
00:04:0922 And so the bottom line is that he'd. because of this assumption of self-differentiation,
00:04:1506 he missed on the phenomenon of induction.
00:04:1800 And so we remember Spemann and Mangold, and less so Lewis.
00:04:2201 Okay, let's go back to the whole embryo and remind ourselves what we're looking at.
00:04:2713 So, we know from molecular mapping that the organizer is gonna be the top of this
00:04:3323 when it loops around again.
00:04:3428 We're going through gastrulation and neurulation.
00:04:3804 And when we loop around again. here's the organizer up here.
00:04:4013 It's going inside the embryo and is opposed to this neural plate, and is able to
00:04:4615 instruct it to make the neural tissue.
00:04:4907 So again, if we look at this MRI movie, we can see that dorsal mesoderm moving up
00:04:5604 against this overlying neural plate.
00:04:5811 And during this process, where it's opposed to the neural plate, it's in the right place
00:05:0207 to be inducing the neural tissue.
00:05:0407 So, that's the normal organizer that's going up there and is thought to be signaling
00:05:1005 to induce the neural plate.
00:05:1113 Okay, so we're going to discuss this more.
00:05:1501 But we first need to understand how we get to that position.
00:05:1724 And I'm not going to go through this in detail, but I'm going to give a brief summary of the
00:05:2126 initial events that happen to set up the organizer.
00:05:2517 And it comes down largely to the activity of two different signaling pathways: the Nodal/Smad2 pathway
00:05:3221 and the Wnt/beta-catenin pathway.
00:05:3512 We don't need to know about these pathways in detail, but what we do know is the way
00:05:4018 they're turned on in the embryo.
00:05:4224 So initially, when the egg is laid, it's got this axis from animal to vegetal,
00:05:4822 from the pigmented to the yolky side.
00:05:5122 And subsequently it was found that there are a number of pre-localized components in that
00:05:5515 polarized egg.
00:05:5717 And I'm going to talk about this red mRNA, messenger RNA, that's pre-localized,
00:06:0300 called vegt, first described by Mary Lou King's group.
00:06:0612 And there's also, slightly less well-characterized, activators of the Wnt/beta-catenin pathway
00:06:1119 down here.
00:06:1303 So initially, this is cylindrically symmetrical about the animal-vegetal axis.
00:06:1618 And an important process here, as is widespread in embryology, is the symmetry breaking.
00:06:2118 So, you have to go from a cylinder to a bilaterally symmetrical egg.
00:06:2523 And this is achieved during normal development because the sperm is going to enter on one side.
00:06:3213 That makes this giant aster.
00:06:3504 So, these astral microtubules extend throughout the egg cytoplasm during the first cell cycle.
00:06:4115 And not only do they serve to pull the maternal pronucleus towards it, but they also
00:06:4603 serve to bias the way that microtubules polymerize. polymerize in the outside cortex,
00:06:5101 the outer ten-micron layer.
00:06:5313 So as a result of this bias, there's an oriented array of microtubules that go around the embryo,
00:07:0028 And they act as tracks for carriage via kinesins of some of these purple components.
00:07:0622 There's a selectivity.
00:07:0822 The purple components that activate the Wnt/beta-catenin pathway get smeared out along the
00:07:1413 entire dorsal side of the embryo, whereas the red do not do this.
00:07:1826 They're sort of passively released from the vegetal cortex and spread in a graded way
00:07:2417 through the egg.
00:07:2517 So, we now have a broken symmetry, where we have the red going from vegetal to animal
00:07:2926 and the Wnt/beta-catenin concentrated from dorsal to ventral.
00:07:3504 Now, these two molecules get together and turn on the. the Nodal genes.
00:07:4000 The Nodal genes are signaling proteins in the TGF-beta superfamily.
00:07:4307 And they're turned on in the margin.
00:07:4515 And in normal development, there's a cooperativity between the purple Wnt signal and, now,
00:07:5026 this yellow protein produced from the vegt RNA.
00:07:5409 So, they get together and turn on these Nodal genes.
00:07:5802 And they turn them on at a higher level on the dorsal side and the ventral side.
00:08:0216 But they do turn them on everywhere.
00:08:0500 And these Nodal genes induce mesoderm.
00:08:0710 So, the cells that are initially naive get told, in this marginal zone, to become
00:08:1310 the prospective mesoderm.
00:08:1504 But where this interaction is the strongest -- the strongest interaction of beta-catenin
00:08:1922 and the Nodal gene expression -- that converges on the promoters of organizer gene and
00:08:2512 turns them on, especially in this dorsal region here.
00:08:2822 So, the marginal zone. the mesoderm goes all the way across in the equator,
00:08:3224 but the organizer is special in that it's only turned on at the convergence, the strongest convergence
00:08:3702 of these signals.
00:08:3804 Okay, so we've discussed that.
00:08:4024 And let's just contrast that with what happens if this cortical rotation doesn't occur.
00:08:4701 And so you can see here what. there are various tricks to. to cause this to happen.
00:08:5024 One is to irradiate the vegetal side of the embryo with ultraviolet light.
00:08:5413 And that prevents the polymerization of microtubules.
00:08:5726 The alternative is to eliminate beta-catenin production by using a reagent that
00:09:0209 blocks beta-catenin production.
00:09:0324 We'll come back to that.
00:09:0611 Either way, what happens is that we get the release of the vegt and we get the
00:09:1014 graded vegt protein, which turns on Nodal, but, in the case of the lack of cortical rotation,
00:09:1626 then this purple signal stays down here.
00:09:2001 And so as a result, there is no synergy on this side.
00:09:2225 There's no overlap between the signals.
00:09:2522 And so this whole marginal zone behaves like the ventral marginal zone.
00:09:3008 You get a. a ventralized type of embryo with no organizer.
00:09:3506 Okay, so that symmetry-breaking event back here was important.
00:09:3902 But we're gonna use this trick in the next experiment, that proves that we need organizer signal
00:09:4327 to get the neural plate to be formed.
00:09:4905 Before we go into that, we're just gonna discuss a little bit more about the graded Nodal signaling.
00:09:5327 Because there. a quite widespread view in the field is that this graded Nodal signaling
00:09:5823 is important in setting out the pattern of the marginal zone.
00:10:0217 It seems quite obvious that if there's going to be a graded signal with more on this side
00:10:0620 than that side, it should be used for something.
00:10:0922 So, we're going to get a graded response, and the phospho-Smad2 is the intracellular effector,
00:10:1515 which is known to be distributed like this.
00:10:1804 There really is graded expression of Smad2 going from dorsal to ventral.
00:10:2226 And then this proceeds as a wave across the embryo.
00:10:2505 So, it seems perfectly reasonable to think that in normal development what ought to be
00:10:3002 happening is that that signal will tell the embryo to make different kinds of mesoderm.
00:10:3618 And indeed, that whole idea is supported by this experiment, where we can take,
00:10:4127 from the blastula stage, naive ectoderm from this so-called animal cap and put it in culture.
00:10:4822 By itself, it will self-differentiate into epidermis.
00:10:5125 But if we add a signal, and we can use either Nodal or more conveniently, Activin,
00:10:5619 another member of the TGF-beta superfamily.
00:10:5908 If one adds increasing doses of that Activin signal, the mesoderm-inducing signal,
00:11:0416 one can get caps that develop in a more ventral way, making mesenchyme, whereas as one
00:11:0923 doses in the signal, more and more dorsal tissues, like muscle and ultimately a lot of notochord.
00:11:1519 So, those kinds of experiments, the description of the graded expression, as well as
00:11:2107 this result, where one's reconstructing what may be going on in the embryo, suggest that
00:11:2502 that graded signal may cause pattern.
00:11:2616 But I'm going to argue that's not true.
00:11:2913 So, just to sum up, this normal pattern in the marginal zone -- from notochord through
00:11:3502 muscle, kidney, and blood -- could in principle be set up by that graded Nodal signaling.
00:11:4204 But this was explicitly tested in a series of experiments by Ron Stewart and John Gerhard,
00:11:4724 and other very similar experiments by Jonathan Slack.
00:11:5102 And so what I want to review briefly is this experiment that shows that there's not enough
00:11:5617 information imparted by that early signal to give substantial pattern in the marginal zone.
00:12:0228 Now, what they did. they wanted to assess the effectiveness of organizer grafts.
00:12:0815 And they did this at the late blastula stage.
00:12:1021 So, this is a really early stage, before gastrulation goes on and before there's much to. there.
00:12:1522 there certainly is pattern is the marginal zone later on.
00:12:1822 So, they took normal embryos and cut them in half, vertically, so they got two hemispheres.
00:12:2418 And they used that UV irradiation trick.
00:12:2607 So, they had a graft. they were able to graft these -- labeled grafts, of course --
00:12:3220 onto UV-irradiated hosts.
00:12:3320 So, they made this recombinant.
00:12:3611 In this case, the organizer is schematically illustrated in red.
00:12:4005 So, you've got a half organizer here.
00:12:4214 One of their first questions was, if you only put in a right organizer, do you only get
00:12:4612 a right embryo?
00:12:4714 And there are. the answer was no.
00:12:4822 You get a bilaterally symmetrical embryo.
00:12:5101 But also, in many cases this graft will give you a normal tadpole.
00:12:5517 So, you rescue development also, as shown by the lineage tracing experiment,
00:13:0009 from this ventralized half.
00:13:0215 But this is really the key one in my exper. in my view.
00:13:0615 So here, they've cut just 30 degrees off the dorsal midline axis, so they've cut the organizer
00:13:1228 into the right piece, and this piece has no organizer.
00:13:1607 Now, by the model I was discussing earlier, there should be some graded Activin signaling
00:13:2212 or graded. graded Nodal signaling in here that's inducing things like muscle.
00:13:2707 Well, we're gonna ask that question.
00:13:2819 We're gonna take this side and, again, fuse it to a naive ventralized piece,
00:13:3418 put them together, and ask what happens.
00:13:3806 And the result in most cases is there's absolutely no dorsal pattern in the embryo.
00:13:4411 And just to nail this home, I want to stress that.
00:13:4713 So, they're taking this ventralized piece and putting on this piece from a normal embryo
00:13:5305 that lacks just the organizer, but still has that dorsolateral prospective mesoderm
00:13:5714 that would make muscle if it were left alone.
00:14:0002 But in the context of this recombinant, you just get this completely ventralized embryo,
00:14:0518 as opposed to something that would make a little muscle and so on.
00:14:0819 So, this experiment shows I think quite well that the pattern that's induced by
00:14:1423 that graded Activin/Nodal signaling is not enough to have any permanent effect on this tissue.
00:14:2021 And that you actually need the organizer signaling.
00:14:2226 So, that sort of loss of organizer function proves that you need organizer signaling
00:14:2901 in normal development.
00:14:3219 Another is a sort of descriptive view, where we've looked at gene expression at different phases.
00:14:3715 And here's a case where we see two. expression of two different genes.
00:14:4016 This is noggin expressed in the organizer -- we'll come back to that -- and this
00:14:4422 prospective muscle gene, myod, that's expressed in a complementary way, in the non-organizer tissue.
00:14:5020 When it's first turned on, it's turned on fairly uniformly around the rest of the marginal zone.
00:14:5508 Later on, this expression turns off, and this gets enhanced by signaling from the organizer.
00:15:0022 But when it first turns on, it looks like the marginal zone is organized in a binary way.
00:15:0513 Now, this very rapidly changes.
00:15:0625 So, we see expression of genes such as this one, lhx1, that's high in the dorsal marginal zone
00:15:1127 and then graded off to the side.
00:15:1408 So, that's a later stage.
00:15:1524 We also see that with this split, where the blue and the brown genes are expressed
00:15:2010 in complementary domains at the end of the blastula stage, but very quickly become elaborated
00:15:2600 so that the brown gene, Wnt8, is restricted away from the organizer and just in the marginal zone.
00:15:3027 So, things are very dynamic.
00:15:3228 And one really has to look at this early stage to see this binary difference.
00:15:3624 By this stage, this tissue has already been instructed by the organizer to make muscle.
00:15:4125 But anyway, this descriptive experiment does support the idea that initially the marginal zone
00:15:4715 is split in a binary way and is not graded in this induction.
00:15:5017 So again, arguing that you need a signal from the organizer and it's not enough to have
00:15:5500 that graded Nodal signaling.
00:15:5824 This just reinforces that.
00:16:0106 And as that slides up, I'll say that we need. need now to figure out, what are these other signals?
00:16:0801 And at the time, there were a lot of experiments that were done using both cell biology,
00:16:1223 using secreted signals from cells and assaying them in embryos, and many of these signals do have
00:16:1716 important embryonic functions: fibroblast growth factors Nodals, Activins, and so on.
00:16:2318 But the dorsalizing molecules that are made from the organizer were not understood.
00:16:2816 So, how do we find those?
00:16:3027 And here I give much credit to Bill Smith, who's now a professor at UC Santa Barbara,
00:16:3418 who in the early '90s joined me and decided to use an expression cloning approach
00:16:4013 to try to find these molecules.
00:16:4122 And again, he used this trick of ventralizing embryos.
00:16:4513 But at the four-cell stage, he then took synthetic messenger RNAs made from a library.
00:16:5112 This library was a library of gastrula-specific RNAs in a plasmid that could be transcribed
00:16:5618 with this synthetic phage polymerase, SP6 polymerase, so that we could get a library of,
00:17:0204 in the first instance, 100,000 colonies, extract the DNA, and then transcribe
00:17:0820 that whole library of plasmids to make a complex mixture of synthetic RNA that we hoped mimicked
00:17:1315 what was in the embryo.
00:17:1524 Remarkably, that first injection of the synthetic RNA, when it was injected back at the four-cell stage,
00:17:2107 instead of these embryos looking like this -- complete belly pieces as Spemann would
00:17:2525 have called them -- they looked more like this.
00:17:2723 They had some tail structures, some muscle, and spinal cord.
00:17:3107 So, that RNA conferred a morphological rescue.
00:17:3421 At that point, we knew there was an active ingredient in the library, and it was
00:17:3909 just a question of sub-selection, where we would split the library into smaller and smaller pools,
00:17:4400 assaying those pools as we go along, and ask, is there a pool in there that
00:17:4908 confers this ability to dorsalize embryos?
00:17:5219 And sure enough, we. he did this a couple of times.
00:17:5502 The first time, he isolated a Wnt signal, which I won't discuss but is thought to mimic
00:17:5907 that early Wnt signal in dorsalizing the embryos.
00:18:0128 But for the purposes of this presentation, the second one he isolated was really exciting
00:18:0623 because it was completely new.
00:18:0824 And as a single RNA, as you can see in this picture, with increasing dose of the gene
00:18:1324 we called noggin RNA, you get this progressive rescue of structures to the normal state.
00:18:1917 And then when one overdoses the embryo, ultimately you end up with these little noggins,
00:18:2420 these little heads alone.
00:18:2520 So, that. that single RNA is able to transform, from the four-cell stage, a ventralized embryo.
00:18:3023 And if it's put in at enough dose, you get just a big head.
00:18:3604 This has been a very useful assay to isolate a number of other embryological activities,
00:18:4006 but that's the one I want to concentrate on, noggin.
00:18:4310 And so this is an in situ hybridization where we're looking at the messenger RNA
00:18:4810 that's expressed from the noggin gene in early development.
00:18:5100 And so let's look at this stage.
00:18:5319 This is a late blastula.
00:18:5504 Noggin has already turned on.
00:18:5613 And as you can see, it's turned on in just one side of the embryo.
00:18:5909 And we know this is the dorsal side.
00:19:0109 So, this gene is expressed. it not only has the right activity in these messenger
00:19:0614 RNA injections, but it's also expressed in the right place -- here's a vegetal pole view,
00:19:1025 remarkably it's a 60-degree sector, just the same as the sector that Ron Stewart identified
00:19:1622 in his activity assay -- and then in the gastrula stage it continues to be expressed
00:19:2408 in the dorsal marginal zone, in this involuting dorsal mesoderm that is lying just underneath
00:19:2915 the neural plate, and in the right place to induce the neural plate.
00:19:3218 Now, here's a neural. neurula stage where it continues to be expressed in the head,
00:19:3626 mesoderm, and notochord, again, in the right place to continue inducing the neural plate.
00:19:4304 So, the activity is promising, the extra expression is promising, but could we show that it
00:19:4817 had the right properties?
00:19:4924 So, we initially used a protein that we made in CHO cells that Teresa Lamb had transformed
00:19:5523 with noggin plasmids.
00:19:5624 But later in the collaboration with Regeneron Pharmaceuticals, they made a human noggin,
00:20:0210 particularly, Aris Economides, and gave us that recombinant human noggin for our experiments.
00:20:0706 So we were able to ask, now, can noggin protein mimic what we know are the embryological effects
00:20:1317 of grafting this tissue?
00:20:1705 So, again, this is the normal case, where noggin is expressed in this red dorsal mesoderm,
00:20:2218 and potentially instructing this overlying ectoderm to become neural plate.
00:20:2606 But how do we assay that activity?
00:20:2809 Well, again, we turned to our animal cap assay.
00:20:3021 And actually, we can turn to that assay in the gastrula stage, where the sensitivity
00:20:3616 of these cells up here has changed, and they're no longer responsive to the Activin/Nodal signal.
00:20:4402 But we know from recombination experiments that if we graft an organizer onto here
00:20:4903 neural induction will occur.
00:20:5111 So now we can replace the graft of organizer by soaking this tissue in noggin protein.
00:20:5720 And so this is a schematic, which we can do in the late blastula or the gastrula,
00:21:0124 where we take this prospective ectoderm off into culture.
00:21:0503 Normally, just makes a hollow ball of epidermis when left alone.
00:21:0825 As I mentioned before, with. with Activin or Nodal, you get a complex induction of
00:21:1504 dorsal mesodermal cell types.
00:21:1717 And those in turn can secondarily induce neural tissue.
00:21:1926 But the induction is not direct.
00:21:2120 The important thing for our purposes is that, when we treat this just with recombinant noggin,
00:21:2707 we get a clean neural induction.
00:21:2820 There's a little epidermis that's on the outside and an epithelial layer that's not so responsive,
00:21:3405 but all of the underlying cells, here, are trans. transformed into neural cells.
00:21:3809 So, we get a cleaner. clean neural tissue, and that can't be induced by any mesoderm
00:21:4300 because there's no mesoderm in this explant.
00:21:4611 We can also do this as at a time when the mesoderm can no longer be induced.
00:21:5004 So, if we do this at the gastrula stage instead of the late blastula stage, this experiment
00:21:5511 wouldn't work.
00:21:5611 Activin would have no effect.
00:21:5714 And yet noggin can still induce neural tissue.
00:22:0002 So this, then, really identified noggin as an authentic neural inducer, that it's expressed
00:22:0421 in the right place and time, and has the right activity to be doing the job in the normal embryo.
00:22:1117 Here are some pictures of the kind of results that we get.
00:22:1417 These are the works of. these are done by Anne Knecht in the lab.
00:22:1722 And so, here we have a molecular assay for neural tissue.
00:22:2022 This is a gene that's expressed throughout the nervous system.
00:22:2413 And here are some of these explants, the explants that lack noggin, and they don't stain
00:22:2922 for this gene.
00:22:3022 But the parallel explants that were soaked in noggin protein, as you can see, are robustly
00:22:3514 induced to make this marker gene, and so we can say they're neural.
00:22:3805 And that contrasts. again, I'll draw the contrast with the Activin mesoderm inducer.
00:22:4413 Here, we're looking down on the top of the embryo with the muscle and notochord in the middle.
00:22:4819 These are mesodermal structures that we've lit up with this collagen probe.
00:22:5226 If we take explants just like these ones, but treat them with Activin at the late blastula stage,
00:22:5714 we get lots of this mesoderm induced.
00:22:5909 But down here, we see that noggin induces no mesoderm.
00:23:0220 So, we do get neural tissue in the absence of mesoderm, and hence a clean neural induction.
00:23:0804 Just as an interesting side point -- I won't be discussing this much today -- that we
00:23:1315 also have to account for production of the entire neural plate from brain to spinal cord.
00:23:1811 So, what kind of tissue does noggin induce?
00:23:2111 Here we can use regional markers like this cement gland, this very anterior marker,
00:23:2612 or this forebrain-midbrain marker, otx2, and then this engrailed-2 is expressed just
00:23:3123 at the border of the hindbrain.
00:23:3326 And the noggin-treated explants will make these very anterior marker genes.
00:23:3904 they'll turn them on, but they don't turn on the more posterior ones.
00:23:4304 And so noggin, exclusively in this explant situation, induces anterior brain-like tissue.
00:23:5114 So, we have a molecule that works very well, but of course we then had to figure out how it works.
00:23:5820 And so, for this experiment, we for a long time labored under the delusion that
00:24:0220 we may have invented a new kind of signal transducer.
00:24:0600 At the time, it wasn't really appreciated that development gets by with a remarkably
00:24:0914 limited number of pathways.
00:24:1125 And so, here, what eventually turned out, with some very useful information from
00:24:1627 Chip Ferguson's lab, it was suggested that it may be impacting the BMP pathway.
00:24:2201 And Lyle Zimmerman was able to show that, because we had all these reagents in the lab
00:24:2523 at the. at the time.
00:24:2725 And the way that noggin actually works is not by activating any new signal transduction pathway,
00:24:3304 but rather by interfering with the BMP signaling pathway.
00:24:3611 Normally, BMP binds to its two receptors, brings them together, and that has the consequence
00:24:4117 of ventralizing the embryo.
00:24:4325 In this case, when noggin is present, it binds tightly to BMP, prevents this interaction,
00:24:5009 and then by default, instead of being ventralized, the embryo is dorsalized.
00:24:5513 So I'll stress that, that it's the absence of this BMP signal that is instructive to the embryo
00:25:0104 and allows the embryonic structures to make dorsal structures rather than BMP-induced
00:25:0712 ventral structures.
00:25:0901 And satisfyingly, this crystal structure from Jay Groppe shows that noggin, as a dimer,
00:25:1423 sort of embraces the dimer of BMP.
00:25:1626 So, it's not surprising that it, as Lyle Zimmerman showed, prevents the BMP from binding their receptors.
00:25:2606 It's a very high-affinity reaction, as was shown here by this competition experiment.
00:25:2913 And this was done again by Lyle Zimmerman.
00:25:3126 Here took iodinated BMP4 and was able to bind it to a chimeric human noggin, which has an
00:25:3926 immunoglobulin tail which makes it easy to precipitate.
00:25:4303 And so if one simply mixes these together, you get a very active binding and precipitation.
00:25:4916 But by mixing in different doses of different kinds of other TGF-betas, we could work out
00:25:5510 the affinity of this interaction.
00:25:5708 And so, notably, if we take BMP4, the red one, and plot how that interferes
00:26:0214 with binding of iodinated BMP4, we get this nice curve.
00:26:0714 And if we plot the half-maximal inhibition, it comes out that the interaction there is.
00:26:1202 it has a remarkably tight Kd of about 20 picomolar.
00:26:1626 Other BMPs, like BMP7, in blue here, have a lower affinity.
00:26:2305 And then yet other TGF-beta family members, like TGF-beta itself, have no measurable affinity.
00:26:2820 So, there is a variation in affinity, but very tight affinity for the BMP2 and -4 class.
00:26:3500 So, we come up with this general model that in normal development you set up the mesoderm
00:26:4026 with a special dorsal territory, this naive and fairly uniform ventral-lateral territory,
00:26:4626 and the rest of the patterning is mediated during gastrulation by dorsalizing signals
00:26:5128 like noggin that come from the dorsal-marginal zone, instruct the overlying epidermis
00:26:5713 to instead become nervous system, and instruct this ventral mesoderm to become things like muscle.
00:27:0222 So, is that it?
00:27:0420 Really, we've done this by add-back, but what about loss of function?
00:27:0900 And in the interim, a large number of other antagonists were discovered.
00:27:1213 And a lot of these were discovered by Eddy De Robertis' lab, so chordin and cerberus.
00:27:1916 Noggin, we discovered, Bill Smith discovered, as well as this Nodal-3 molecule.
00:27:2422 And follistatin had already been known about, but its activity as a dorsalizing molecule
00:27:2928 was worked out by Ali Brivanlou in Doug Melton's lab.
00:27:3302 So, looking at all these, they're all expressed in the organizer, and they all have some similar
00:27:3911 anti-BMP activities.
00:27:4113 You can see that they turn on at different times.
00:27:4316 This is the early stage, so some are turned on very early.
00:27:4718 And then some are turned on in slightly different territories.
00:27:5004 And perhaps that's important in how they work in detail, but so far, as far as we can tell,
00:27:5424 they all work essentially the same way.
00:27:5617 But there are a lot of them, and they probably have overlapping activity.
00:28:0108 And so if we try to knock down their activity, do we. can we knock down one and get a result,
00:28:0517 or do we have to knock down many?
00:28:0824 To do this experiment, we use morpholino oligonucleotides.
00:28:1114 These are synthetic, uncharged, and very stable oligonucleotides that can hybridize
00:28:1526 to the messenger RNA and interfere with translation, in this case.
00:28:2000 So, we can specifically use the information from base pairing to specifically knock down
00:28:2602 these individual RNAs.
00:28:2720 So, Mustafa Khokha, when he was in the lab, did this experiment using mixtures of morpholinos
00:28:3227 against follistatin, chordin, and noggin.
00:28:3517 And just as a side note, to make this simpler, we did it in the related species to Xenopus laevis,
00:28:4020 Xenopus tropicalis, which now had a sequenced genome, and so we could identify
00:28:4514 and design these oligonucleotides easily to target just single genes rather than two genes
00:28:5003 in the paleotetraploid, Xenopus laevis, genome.
00:28:5419 So, we can inject these morpholinos and then ask what happens.
00:28:5925 And we're going to use, for example, in the neurula stage, this neural marker, which at
00:29:0323 the mid-neurula stage has this nice ability to light up the neural plate.
00:29:1008 Because of worries about specificity of these reagents, which always become toxic if you
00:29:1404 put enough of them in, we do a specificity assay of rescuing.
00:29:1716 So, by cloning the pufferfish noggin, we could use that different sequence to put back BMP-antagonist activity,
00:29:2417 as we'll find out, and rescue the whole process.
00:29:2728 So, these are the kinds of results.
00:29:3009 So, we're going to do single, double, and triple knockdowns.
00:29:3410 And the results are pretty simple.
00:29:3527 When comparing the uninjected to the morpholino-injected, when we knocked down noggin we see no effect.
00:29:4203 Essentially no effect with follistatin.
00:29:4402 Some mild effect, as reported by De Robertis' group, for chordin.
00:29:4716 But then, when we knocked down two -- follistatin/noggin, chordin/noggin, chordin/follistatin --
00:29:5220 we see a more extreme effect.
00:29:5324 There's a smaller neural plate.
00:29:5511 Well, when we knocked down all three, there's a spectacular result, where now, instead of
00:29:5919 making the neural plate, the neural plate is eliminate. eliminated.
00:30:0328 Not only is the neural plate eliminated, but the underlying muscle is almost completely gone.
00:30:0908 And by this hedgehog control, the notochord and the floor plate are also gone.
00:30:1428 So, this is important to rescue.
00:30:1627 We can rescue it with this mixture of morpholinos, rescuing it with the pufferfish noggin.
00:30:2303 And as you see, we get back all of these tissues, demonstrating specificity.
00:30:2817 We can also rescue it by knocking down additional BMPs.
00:30:3206 So, instead of this horrendously high mixture of morpholinos, we're putting in
00:30:3614 even more morpholinos, but also knocking down BMPs.
00:30:3925 And again, you can see that rescue.
00:30:4102 So, we're pretty satisfied that this is a specific manipulation.
00:30:4421 So, knocking down the BMP antagonists eliminates the neural plate.
00:30:4921 You need the antagonists to make the neural plate.
00:30:5327 As we would expect, as you can see on the right of this picture, the loss of those
00:30:5814 dorsal structures is accompanied by a gain in ventrolateral structures.
00:31:0302 So here, for instance, let's. let's look at msx1.
00:31:0522 It's expressed just in the flank.
00:31:0802 But in this manipulated embryo, there's just a narrow stripe left of nonexpressing tissue.
00:31:1306 So, all these ventral tissues are expanded.
00:31:1708 We can also ask, when does this dorsal identity fail?
00:31:2010 When we knock down those antagonists, we clearly lose dorsal structures, but at what step does
00:31:2417 this happen?
00:31:2517 And of course, we would predict that it should fail at the time that normal genes are expressed,
00:31:2906 at the late blastula stage.
00:31:3128 We can start to look at that, and contrast the situation where we interfere with antagonist function
00:31:3802 in normal development, with what happens when we ventralize the embryo from
00:31:4212 the get-go, either with UV light or by depleting beta-catenin, actually also with a beta-catenin
00:31:4727 specific morpholino.
00:31:4902 So, in that case, we lose the beta-catenin purple signal, and we get just this ventralized embryo.
00:31:5711 So, in looking at those, we can see that in the. in the. in the morpholino-knockdown cases,
00:32:0515 where we've knocked down follistatin, chordin, and noggin, we have normal mesoderm,
00:32:0905 as marked by this brachyury gene, but in the case where we look for dorsal identity
00:32:1521 with the goosecoid marker -- here's the control, here's the follistatin/chordin/noggin knockdown --
00:32:2013 there's still dorsal identity.
00:32:2322 Whereas if we contrast that with the beta-catenin knockdown, where we've knocked down
00:32:2705 the early dorsal signal, then of course we never get a dorsal identity.
00:32:3023 So, there's a contrast here, showing that in the absence of the antagonists
00:32:3528 we still get a dorsal identity.
00:32:3713 But then that dorsal identity fails to execute. execute its function.
00:32:4404 And indeed, we can also look at these embryos to ask what happens to BMP signaling.
00:32:4818 And in particular, we can use this useful mark, vent2, because that's normally expressed
00:32:5313 everywhere except the organizer.
00:32:5515 And we also know it's a direct target of BMP signaling.
00:32:5921 So again, when we knock down the follistatin, chordin, and noggin, we see that that gap
00:33:0323 is filled in.
00:33:0508 So in other words, in the absence of the antagonists, there's a sign pretty early on of excess BMP signaling,
00:33:1100 which is going to mess up dorsal developments and lead to a ventralized embryo.
00:33:1519 And again, we get the similar effect, as we would expect, if we eliminate the organizer
00:33:2000 completely with beta-catenin.
00:33:2104 The same result is found with this early muscle marker, which.
00:33:2528 muscle development used to be thought to be development mostly on the graded signal
00:33:2911 from Nodal.
00:33:3011 But you can see here, clearly, that we need that BMP antagonist in order to amplify the
00:33:3506 expression of this muscle determinant in the early embryo.
00:33:3928 So overall, we conclude now that this pathway, of knockdown of BMP activity by these BMP antagonists,
00:33:4726 by a cocktail of BMP antagonists, is crucial to get dorsal development,
00:33:5224 and in order to get neural induction to occur.
00:33:5508 We can show that these molecules by themselves, as protein, induce neural tissue.
00:33:5826 We can show that the combination is essential.
00:34:0127 And they are expressed in the right time and the right place to be ex. executing this function.
00:34:0704 So all in all, it's a comprehensive statement that these are crucial for neural induction
00:34:1303 and dorsal development.
00:34:1513 We can add on the additional observation from the end, that even in the absence of.
00:34:2200 well, in the absence of these antagonists, the early events go by perfectly normal. normally,
00:34:2725 and we get dorsal specification in the marginal zone.
00:34:3108 But in the absence of the antagonists, that organizer can no longer execute its function.
00:34:3621 So all in all, we can conclude then that BMP antagonists are essential for
00:34:4202 the Spemann organizer phenomenon.
00:34:4327 Thank you.
- Part 1: Early Frog Development: How to Make a Tadpole