Reorganizing the Organizer 75 Years On
1999; Cell Press; Volume: 98; Issue: 4 Linguagem: Inglês
10.1016/s0092-8674(00)81971-6
ISSN1097-4172
Autores Tópico(s)Language, Metaphor, and Cognition
ResumoSeventy-five years ago Spemann and Mangold published the results of an experiment that was to mark a turning point in experimental embryology. They transplanted the blastopore lip of a newt gastrula into another newt embryo leading to the generation of a second embryo. A more detailed analysis revealed that the additional embryo was in part derived from the graft and, more importantly, also from the host. In particular and apart from its contribution to the floor plate, the secondary neural tube was derived from host ectoderm. This experiment demonstrated that the grafted tissue contained an "organization center" capable of influencing the fate of the surrounding tissues. On the occasion of the 75th anniversary of this celebrated publication (37Spemann, H., and Mangold, H. (1924). Ueber Induktion von Embryonalanlagen durch Implantation Artfremder Organisatoren. Wilhem Roux Arch. Entw. Mech. 100, 599–638. English translation in Willier and Oppenheimer (1974). Foundations on Experimental Embryology (New York: Haffner Press), pp. 38–50.Google Scholar), E. De Robertis (UCLA) and J. Aréchaga (Universidad del Pais Vasco) organized a workshop that was held on May 24–26 at the Instituto Juan March in Madrid. The workshop, entitled "Molecular Nature of the Gastrula Organizer," provided a general overview regarding our current understanding of the "Spemann Organizer." I apologize for not including an exhaustive reference list in this review when referring to published work due to size constraints (for excellent recent reviews of the field, see18Harland R Gerhart J Formation and function of Spemann's organizer.Annu. Rev. Gen. Dev. Biol. 1997; 13: 611-617Crossref PubMed Scopus (683) Google Scholar, 36Schier A Talbot W.S The zebrafish organizer.Curr. Opin. Gen. Dev. 1998; 8: 464-471Crossref PubMed Scopus (50) Google Scholar, 6Beddington R.S.P Robertson E.J Axis development and early asymmetry in mammals.Cell. 1999; 96: 195-209Abstract Full Text Full Text PDF PubMed Scopus (609) Google Scholar, 38Streit A Stern C.D Neural induction a bird's eye view.Trends Genet. 1999; 15 (a): 20-24Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar). At the turn of the century, Spemann was conducting experiments in the two-cell newt embryo, separating the two blastomeres by constriction. He realized that in some cases, the two independent parts were able to form two fairly normal embryos, while in others only one part developed as an embryo, the rest giving rise to what he called the "belly piece." He considered that the different planes of division might be responsible for the different outcomes, and he deduced that only the dorsal part could develop into a complete embryo in response to a differentiation signal. Thus, as early as 1903, he foresaw the existence of an organizer. Nine years later, from his studies on lens development, he launched the concept of embryonic induction, by which certain tissues can respond to instructive signals emanating from another nearby tissue. It is, however, fair to mention that C. Herbst was the first to describe the concept of induction in 1895 (see 30Oppenheimer, J.M. (1991). Curt Herbst's contributions to the concept of embryonic induction. In Developmental Biology: A Comprehensive Synthesis, S.F. Gilbert, ed. (New York: Plenum Press), 7, 63–89.Google Scholar). Nevertheless, it was clearly Spemann and Mangold who designed and executed the critical experiment that allowed the dorsal lip of the blastopore to be identified as the organizer, the culmination of some 20 years of work. In these 75 years, structures homologous to the amphibian organizer have been described in all vertebrates: the embryonic shield in the fish, Hensen's node in the chick, and the node in the mouse. In particular, great progress has been made in the last 10 years or so with respect to the molecular mechanisms underlying the formation and functions of the organizer. The organizer is considered a structure characteristic of the chordate phylum (J. Gerhart, University of California, Berkeley) and can be defined as an embryonic region capable of inducing a secondary axis when transplanted to an ectopic site. In addition, the organizer self-differentiates to give rise to different axial tissues such as the notochord, prechordal mesoderm, floor plate, and dorsal endoderm. It is important to stress that the organizer not only instructs neighboring cells to differentiate into several tissues but it also influences time, place, and orientation within the embryo (Gerhart). The current model for induction of the Xenopus organizer implies the existence of two maternal components that act as a mesoendodermal inducer and as a dorsal modifier (see 18Harland R Gerhart J Formation and function of Spemann's organizer.Annu. Rev. Gen. Dev. Biol. 1997; 13: 611-617Crossref PubMed Scopus (683) Google Scholar). In molecular terms, these factors correspond to the activation of the TGFβ- and Wnt-signaling pathways, respectively. The mesoendoderm inducer is secreted by all vegetal cells, while those located in tiers 3 and 4 on the dorsal side of the 32-cell stage blastula embryo also secrete the dorsal modifier and constitute the Nieuwkoop center. Mesoendoderm induction sets up the competence groups to form ectoderm, mesoderm, and endoderm, and the dorsal modifier sets up the dorsoventral axis of the embryo that confers information for the formation of the organizer (Figure 1). Once the gastrula organizer has been formed, the three germ layers are responsive to the signals emanating from it that coordinate neural induction and the patterning of the mesoderm and endoderm. The organizer produces and releases signals capable of influencing both dorsal and anterior development in surrounding tissues. In molecular terms, this implies the generation of an area where BMP and Wnt signals are inhibited. The cells in the ventral region of the embryo that will normally develop ventral and posterior structures produce both ligands (see Figure 2). The mesoderm of the Xenopus blastula is located around the equator and it is patterned along the dorsoventral axis to give rise to axial (notochord and prechordal mesoderm), paraxial (somites), intermediate (pronephros), and lateral (blood) derivatives. This patterning is carried out by an interplay between ventralizing (BMP-mediated) and dorsalizing signals (BMP antagonists secreted by the organizer) that generate a gradient of BMP activity. The more BMP activity, the more ventral the phenotypes obtained (Figure 2). This has been confirmed in animal cap explants by a titration of the levels of BMP activity through the use of different concentrations of its antagonists Chordin, Noggin, and Follistatin (see 18Harland R Gerhart J Formation and function of Spemann's organizer.Annu. Rev. Gen. Dev. Biol. 1997; 13: 611-617Crossref PubMed Scopus (683) Google Scholar). These antagonists are secreted by the organizer and sequester BMP from the extracellular space, preventing binding to its receptors and, thus, blocking its activity. Spemann and Mangold's experiment also showed that signals from the organizer were able to induce neural tissue from host cells, generating a secondary neural tube in cells that otherwise would be epidermis. Thus, the ectoderm has to make a choice between becoming neural or nonneural after sensing signals that act as neural inducers. Recently, evidence has accumulated to suggest that neural induction in Xenopus is the result of the derepression of a neural default state (20Hemmati-Brivanlou A Melton D Vertebrate embryonic cells will become nerve cells unless told otherwise.Cell. 1997; 88: 13-17Abstract Full Text Full Text PDF PubMed Scopus (280) Google Scholar). Ectodermal cells are in principle fated to become neural but this process is inhibited by BMP4. In particular regions of the embryo, BMP signaling is abolished by the same BMP antagonists that pattern the mesoderm (Chordin, Noggin, and Follistatin), which in this case act as neural inducers. This implies that the organizer makes use of the same molecules for the patterning of the ectoderm and the mesoderm. In spite of the amount of information accumulated regarding our knowledge of the organizer, many pivotal questions are still open, proving that we are far from understanding the complex mechanisms that underlie the morphogenesis of an early vertebrate embryo. Among the relevant questions approached at the meeting that I will try to address below are: the effectors that establish embryonic polarity, the interplay of the signaling pathways that cooperate to form the Nieuwkoop center and the Spemann organizer, the real nature of the cells that constitute the organizer, the ins and outs of the molecules that together with BMP pattern the mesoderm and the ectoderm, how the cells interpret the concentration of a signaling molecule, and even the fact that Spemann's organizer is not the only organizing center in early embryos. Following fertilization, Xenopus eggs undergo a rotation of the cortex relative to the cytoplasm that positions a dorsal inducing activity at the region where the Nieuwkoop center and organizer will form. This cortical rotation leads to the dorsal accumulation of β-catenin (23Larabell C.A Torres M Rowning B.A Yost C Miller J.R Wu M Kimelman D Moon R.T Establishment of the dorso-ventral axis in Xenopus embryos is presaged by early asymmetries in β-catenin that are modulated by the Wnt signaling pathway.J. Cell. Biol. 1997; 136: 1123-1136Crossref PubMed Scopus (351) Google Scholar), a downstream effector in Wnt signaling. The importance of this signaling pathway can be seen when Wnt is ectopically activated in the ventral side of the frog. This manipulation dorsalizes the ventral side and leads to the formation of a secondary axis (25McMahon A.P Moon R.T Ectopic expression of the proto-oncogene int-1 in Xenopus embryos leads to duplication of the embryonic axis.Cell. 1989; 58: 1074-1084Abstract Full Text PDF Scopus (392) Google Scholar). While there is no evidence yet that a Wnt is required for axis specification in Xenopus, there is clear evidence that besides being dorsally localized, nuclear localization of β-catenin is necessary for axis formation (see 28Moon R.T Kimelman D From cortical rotation to organizer gene expression toward a molecular explanation of axis specification in Xenopus.Bioessays. 1998; 20: 536-545Crossref PubMed Scopus (274) Google Scholar). What has been missing, therefore, are data addressing how β-catenin becomes enriched on the prospective dorsal side. New evidence (27Miller, J.R., Yang-Schneider, J.A., Rowning, B.A., Larabell, C.A., Bates, R.L., and Moon, R.T. (1999). Cortical rotation promotes a dorsal accumulation of Dishevelled that may regulate the specification of dorsal fates in Xenopus. J. Cell Biol., in press.Google Scholar) indicates that Dishevelled is vectorially transported to the dorsal side of the embryo along microtubule tracks during cortical rotation (R. T. Moon, University of Washington, Seattle). On the dorsal side Dishevelled likely interacts with a complex formed by several molecules and results in the downregulation of GSK3 activity. Since GSK3 is required for the degradation of β-catenin (45Yost C Torres M Miller J.R Huang E Kimelman D Moon R.T The axis inducing activity, stability and subcellular distribution of β-catenin is regulated in Xenopus by glycogen synthase kinase 3.Genes Dev. 1996; 10: 1443-1454Crossref PubMed Scopus (1017) Google Scholar), inhibiting its activity stabilizes β-catenin in the dorsal side, while it is degraded in the ventral side where GSK3 activity is high. This mechanism seems to be fairly well conserved in other vertebrates since although fish eggs do not exhibit cortical rotation, vesicle transport results in dorsal accumulation of β-catenin (34Schneider S Steinbeisser H Warga R.M Hausen P Beta-catenin translocation into nuclei demarcates the dorsalizing centers in frog and fish embryos.Mech. Dev. 1996; 57: 191-198Crossref PubMed Scopus (442) Google Scholar). Since β-catenin acts as a transcription factor in Wnt signaling, it requires nuclear localization (45Yost C Torres M Miller J.R Huang E Kimelman D Moon R.T The axis inducing activity, stability and subcellular distribution of β-catenin is regulated in Xenopus by glycogen synthase kinase 3.Genes Dev. 1996; 10: 1443-1454Crossref PubMed Scopus (1017) Google Scholar and see below). Recently, nuclear localization of β-catenin has also been observed in the chick blastoderm (M. Kessel, MPI, Götingen). In addition to the β-catenin-mediated Wnt signaling pathway operating in the dorsal side of the embryo, recent evidence points to the existence of another β-catenin-independent Wnt pathway triggered in the ventral side of the embryo (Moon). This pathway works through the same receptors (Frizzled homologs), but is dependent upon pertussis toxin–sensitive G proteins, leading to the elevation of intracellular calcium concentrations. The signaling through this Wnt/calcium pathway involves the activation of two kinases, Cam-kinase II and Protein kinase C (Moon). Thus, at early cleavage stages, two different Wnt pathways may promote antagonistic dorsal and ventral fates in Xenopus embryos generating polarity and presaging the localization of the Spemann's organizer. Although the dorsal restriction of β-catenin localization occurs in Xenopus during the first cell cycle after fertilization, its activity as a transducer of the Wnt pathway requires it to be translocated to the nucleus, which only occurs at the midblastula stage. This signal probably constitutes the proposed dorsal modifier released by the Nieuwkoop center which, together with the mesoendoderm inducer (a TGFβ-related signal), leads to the formation of the organizer in the dorsal side of the embryo. Loss-of-function experiments with antisense β-catenin support this hypothesis (19Heasman J Crawford A Goldstone K Garner-Hamrick P Gumbiner B McCrea P Kintner C Noro C.Y Wylie C Overexpression of cadherins and overexpression of β-catenin inhibit dorsal mesoderm induction in early Xenopus embryos.Cell. 1994; 79: 791-803Abstract Full Text PDF PubMed Scopus (589) Google Scholar). But how is it that these two signaling pathways converge to form the organizer? On the future dorsal side of the embryo, nuclear β-catenin acts as a transcription factor that when bound to Tcf3 can activate siamois expression, which in turn, will activate organizer genes such as goosecoid (Moon; also see 28Moon R.T Kimelman D From cortical rotation to organizer gene expression toward a molecular explanation of axis specification in Xenopus.Bioessays. 1998; 20: 536-545Crossref PubMed Scopus (274) Google Scholar). On the future ventral side, Tcf3 serves as a repressor of siamois expression, in part through interaction with the corepressor XCTBP (Moon). Thus, the dorsal accumulation and transcriptional activity of β-catenin provides a molecular explanation for induction in Xenopus from the stages of cortical rotation. However, the story is not complete because efficient activation of goosecoid also requires TGFβ signaling. Indeed, a requirement for a TGFβ-mediated molecular pathway for axis specification in Xenopus has been well established. Several TGFβ-related molecules have been proposed as candidates to trigger this pathway, including Vg1, Activin, and Nodal. These molecules are present in Xenopus and zebrafish eggs and their inhibition blocks mesoderm induction (18Harland R Gerhart J Formation and function of Spemann's organizer.Annu. Rev. Gen. Dev. Biol. 1997; 13: 611-617Crossref PubMed Scopus (683) Google Scholar, 36Schier A Talbot W.S The zebrafish organizer.Curr. Opin. Gen. Dev. 1998; 8: 464-471Crossref PubMed Scopus (50) Google Scholar). However, the majority of the experiments aimed to test their effects in Xenopus have been carried out with dominant-negative versions of the activin receptor, which are rather nonspecific in terms of their association to either receptors or ligands of this family. A more specific approach has been taken by the use of "Cerberus short" (a mutated form of the secreted factor Cerberus that acts as an antagonist of Nodal). Blocking Nodal activity by this means inhibited the formation of a functional Nieuwkoop center and, consequently, the formation of the organizer and of the mesoderm (De Robertis; see 32Piccolo S Agius E Leyns L Bhattacharayya S Grunz H Bouwmeester T De Robertis E.M The head inducer Cerberus is a multifunctional antagonist of Nodal, BMP and Wnt signals.Nature. 1999; 397: 707-710Crossref PubMed Scopus (670) Google Scholar). In the zebrafish, Nodal-related genes have been shown to be required for organizer and mesoderm formation (12Feldman B Gates M.A Egan E.S Dougan S.T Rennebeck G Sirotkin H.I Schier A.F Talbot W.S Zebrafish organizer development and germ layer formation require nodal-related signals.Nature. 1998; 395: 181-185Crossref PubMed Scopus (555) Google Scholar), acting with One-eyed pinhead (Oep, a member of the EGF-CFC family of signaling factors, which contain an e pidermal g rowth f actor–like motif and a novel c ysteine-rich motif) as an extracellular cofactor and Antivin (a Lefty protein) as a feedback inhibitor (A. F. Schier, Skirball Institute, New York; 15Gritsman K Zhang J Cheng S Heckscher E Talbot W.S Schier A The EGF-CFC protein One-eyed pinhead is essential for Nodal signaling.Cell. 1999; 97: 121-132Abstract Full Text Full Text PDF PubMed Scopus (601) Google Scholar). Similarly, in the mouse, nodal mutant embryos display reduced mesoderm formation (8Conlon F.L Lyons K.M Takaesu N Barth K.S Kispert A.K Herrmann B Robertson E.J A primary requirement for nodal in the formation and maintenance of the primitive streak.Development. 1994; 120: 1919-1928Crossref PubMed Google Scholar) and lefty2 mutants have an expanded domain of mesoderm (26Meno C Gritsman K Ohishi S Ohfuji Y Heckscher E Mochida K Shimono A Kondoh H Talbot W.S Robertson E.J et al.Mouse Lefty2 and zebrafish Antivin are feedback inhibitors of Nodal signaling during vertebrate gastrulation.Molecular Cell. 1999; in pressGoogle Scholar). The connection between all the above-mentioned TGFβ family members and the activation of goosecoid is provided by their induction of Lim1, a homeobox transcription factor expressed in the dorsal marginal zone at blastula stages and, later on, restricted to the organizer. Lim1, through its cooperation with Ldb (LIM domain binding protein), directly induces goosecoid expression, the step where the Wnt- (see above) and TGFβ-signaling pathways converge. The fact that goosecoid also seems to be regulated by Otx2 (I. Dawid, NIHCH, Bethesda and M. Taira, University of Tokyo) places it at a central point in the regulatory cascade leading to organizer formation and function. In the chick, the Nieuwkoop center probably corresponds to the posterior marginal zone at preprimitive streak stages, which also releases an activity compatible with the induction of the organizer (Hensen's node). As in Xenopus, this is a region where Wnt and TGFβ signals overlap (Kessel) and the candidate inducers are Wnt8c and Vg1 (4Bachvarova R.F Skromne I Stern C.D Induction of primitive streak and Hensen's node by the posterior marginal zone in the early chick embryo.Development. 1998; 125: 3521-3534PubMed Google Scholar). This convergence of signals may also occur in the mouse (Figure 1). Based on fate mapping studies and experiments showing that the organizer can be regenerated in embryos in which it has been previously ablated, Stern (Columbia, New York) proposed that the organizer cannot be defined as a static population, but rather as a state, which is acquired and lost by cells as they pass through a specific region of the embryo. In the chick, the organizer property is acquired as a result of induction involving the same signals as those that define the Nieuwkoop center (Vg1 and Wnt8c), but during gastrulation they emanate from different cells, in the middle of the primitive streak (Stern; 22Joubin K Stern C.D Molecular interactions continuously define the organizer during the cell movements of gastrulation.Cell. 1999; in pressGoogle Scholar). In addition to the gradient of BMP activity generated by its binding to the inhibitors, a higher degree of complexity is generated during mesodermal patterning by the activity of specific proteases that cleave inactive Chordin/BMP complexes to release active BMP (31Piccolo S Agius E Lu B Goodman S Dale L De Robertis E.M Cleavage of Chordin by Xolloid metalloprotease suggests a role for proteolytic processing in the regulation of Spemann organizer activity.Cell. 1997; 91: 407-416Abstract Full Text Full Text PDF PubMed Scopus (341) Google Scholar), a mechanism homologous to that described in Drosophila for Tolloid cleaving the Sog/Dpp complex. Xolloid, the Xenopus tolloid homolog, cleaves Chordin at specific sites that are also present in proteins such as collagens, which also bind BMP. This suggests that the mechanism of inactivation by binding and the posterior proteolysis to reactivate the signal could be more general than previously perceived (De Robertis). Recent analyses of zebrafish mutants have greatly helped our understanding of mesodermal induction and patterning. Zebrafish minifin embryos, which are mutants for the tolloid homolog gene, show an expanded domain of Chordin expression, as may be expected in the absence of protease activity (M. Mullins, University of Pennsylvania, Philadelphia). S. Schulte-Merker (MPI, Tübingen) presented evidence that the zebrafish yolk cell is responsible for mesoderm induction at early stages through two distinct activities: one that induces the notochord and is independent of BMP2 (swirl), and another one involved in the formation of the trunk region and the neuroectoderm, which is repressed by BMP2. Additional signals that can modify the pattern of the mesoderm were proposed by R. Harland (University of California, Berkeley). A neural plate peeled off a stage 11.5 Xenopus embryo was able to induce muscle markers (MyoD) when placed in the ventral marginal zone, suggesting the existence of signals from the ectoderm to the underlying mesoderm. Thus, in the neurula embryo, reciprocal signaling between the mesoderm and the neural plate may occur to generate and stabilize patterning. Regarding the differential patterning of the trunk and tail mesoderm, D. Kimelman (University of Washington, Seattle) showed that synergistic interactions must occur between Nodal and FGF signals to regulate the expression of trunk- and tail-specific T box transcription factors, with different regulatory relationships occurring at the two anteroposterior levels. Xenopus is an extremely useful system to overexpress or ectopically express different molecules, a strategy that led to the identification of BMP antagonists as neural inducers. Those experiments, however, had the limitation that what they really showed was the ability of those factors to elicit a function, but they did not unequivocally demonstrate that they were necessary and sufficient for neural induction in the normal embryo. Loss-of-function experiments are needed to clarify this matter, something that is not so easy to perform in amphibia. In this respect, the fish and mouse embryos are proving to be highly informative owing to the possibility of generating and analyzing mutants. Interestingly, mice mutant for chordin, noggin, or follistatin, as well as fish lacking chordin function, show complete early neural tubes. Furthermore, mice double mutant for chordin and noggin present severe reductions of the prosencephalon but a fairly complete neural tube (De Robertis). They do however, show forebrain defects, compatible with a function in maintenance and/or differentiation, rather than in initial neural induction. This is in agreement with results obtained in chick embryos, where ectopic expression of Chordin and/or Noggin is not able to induce neural tissue in nonneural ectoderm (40Streit A Lee K.J Woo I Roberts C Jessell T.M Stern C.D Chordin regulates primitive streak development and the stability of induced neural cells, but is not sufficient for neural induction in the chick embryo.Development. 1998; 125: 507-519PubMed Google Scholar). Furthermore, the expression patterns of the three neural inducers do not correlate with their putative role in neural induction in different vertebrates. Noggin and Follistatin, although able to elicit neural induction in overexpression experiments, are not expressed in the zebrafish organizer (5Bauer H Meier A Hild M Stachel S Economids A Hazelett D Harland R.M Hammerschmidt M Follistatin and noggin are excluded from the zebrafish organizer.Dev. Biol. 1998; 204: 488-507Crossref PubMed Scopus (84) Google Scholar); Follistatin is not expressed in the mouse node; and the three of them show a temporal pattern of expression in the chick node that is not suggestive of a central role in neural induction (38Streit A Stern C.D Neural induction a bird's eye view.Trends Genet. 1999; 15 (a): 20-24Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar, 39Streit A Stern C.D Establishment and maintenance of the border of the neural plate in the chick involvement of FGF and BMP activity.Mech. Dev. 1999; 82 (b): 51-66Crossref PubMed Scopus (190) Google Scholar). The reduced expression of neural markers in the zebrafish chordin mutant suggests that, although possibly in conjunction with other factors, it may act as a neural inducer (35Schulte-Merker S Lee K.J McMahon A.P Hammerschmidt M The zebrafish organizer requires chordino.Nature. 1997; 387: 862-863Crossref PubMed Scopus (332) Google Scholar). However, the possibility of this being a consequence of prior effects on the organizer and/or mesoderm, cannot be discarded. In spite of the caveats imposed in the action of Chordin, Noggin, and Follistatin as qualified neural inducers in different vertebrates, it is clear that the formation of neural tissue occurs in the absence of BMP signaling. These BMP inhibitors may act in conjunction with other molecules that have also been shown to inhibit BMP signaling. As discussed by 7Chang C Hemmati-Brivanlou A Cell fate determination in embryonic ectoderm.J. Neurobiol. 1998; 36: 128-151Crossref PubMed Scopus (68) Google Scholar, the Nodal-related protein Xnr3 may function as a BMP inhibitor by competitive binding to its receptors without eliciting a productive signal. FGF may also interfere with BMP signaling through the phosphorylation and subsequent inactivation of the BMP transducer Smad1 by the FGF-activated Erk MAP kinase. If the absence of BMP signaling is a consistent finding in neural induction, it is not clear whether its inhibition is a prerequisite. In relation to this, BMPs are expressed in neither chick nor mouse ectoderm at the stages preceding neural induction. Furthermore, mice mutant for different BMP family members that develop to this stage show a normal early neural plate; mice mutant for BMP inhibitors do not provide much more information either, since they also have a nervous system (this review and 38Streit A Stern C.D Neural induction a bird's eye view.Trends Genet. 1999; 15 (a): 20-24Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar). The possibility of additional unidentified BMP inhibitors playing a role in neural induction cannot be excluded at the moment, and it is also plausible that other molecular pathways might sensitize the ectoderm to respond to BMP antagonists (Harland, Stern). One important question now would be to know when neural induction really starts, and which are the factors involved. The precise establishment of the border between neural and nonneural ectoderm in the early embryo is important for several reasons: first, because its position determines the territory that will give rise to the central nervous system and second, because it is the region where several tissues originate. The cement gland forms at the anterior border of the neural plate and epidermis, and the neural crest forms at the border between the neural plate and the nonneural ectoderm. Both in Xenopus (R. Mayor, Universidad de Chile) and zebrafish (Mullins), a gradient model of BMP activity has been proposed to pattern the ectoderm in a similar manner to that proposed for the dorsoventral patterning of the mesoderm. By changing the level of BMP signaling in tissue recombination experiments in explants or in different dorsalized embryos, which are mutant for components of the BMP pathway, a threshold concentration of BMP has been found to induce neural crest (24Marchant L Linker C Ruiz P Guerrero N Mayor R The inductive properties of mesoderm suggest that the neural crest cells are specified by a BMP gradient.Dev. Biol. 1998; 198: 319-329Crossref PubMed Google Scholar, 29Nguyen V.H Schmid B Trout J Connors S.A Ekker M Mullins M.C Ventral and lateral regions of the zebrafish gastrula, including neural crest progenitors, are established by a bmp2b/swirl pathway of genes.Dev. Biol. 1998; 199: 93-110Crossref PubMed Scopus (364) Google Scholar). A lower concentration will give rise to neural plate and a higher one will induce epidermis. This model is reminiscent of that proposed for the formation of the border between the anterior neural plate and epidermis, where intermediate levels of BMP activity define the border and give rise to the induction of the cement gland in this region (44Wilson P.A Lagna G Suzuki A Hemmati-Brivanlou A Concentration-dependent patterning of the Xenopus ectoderm by BMP4 and its signal transducer Smad1.Development. 1997; 124: 3177-3184Crossref PubMed Google Scholar). In the chick embryo, the neural crest also forms at the border between neural and nonneural ectoderm, although in this case, it does not coincide with an intermediate level of BMP signaling. However, it does coincide with a point of confrontation between BMP-expressing and non-expressing
Referência(s)