Date Published: May 11, 2004
Publisher: Public Library of Science
Author(s): Igor B Dawid
Abstract: The current understanding of organizer formation and neural induction in vertebrate embryos is discussed.
Partial Text: In what is usually referred to as the most famous experiment in embryology, Hans Spemann and Hilde Mangold (1924) showed that a specific region in early frog embryos called the blastopore lip can induce a second complete embryonic axis, including the head, when transplanted to a host embryo. Most of the axis, including the nervous system, was derived from the host, whose cells were induced to form an axis by the graft, therefore named the organizer. Induction refers to the change in fate of a group of cells in response to signals from other cells. The signal-receiving cells must be capable of responding, a property termed competence. The Spemann–Mangold organizer. which—as the transplantation experiment shows—is able to turn cells whose original fate would be gut or ventral epidermis into brain or somites, is the prototypical inducing tissue. And neural induction has for a long time been regarded as a process by which organizer signals, in their normal context, redirect ectodermal cells from an epidermal towards a neural fate. The nature of the neural inducer or inducers and the mechanism of neural induction have been and remain hot topics in developmental biology.
The frog egg is radially symmetrical around the animal–vegetal axis that has been established during oogenesis. Fertilization triggers a rotation of the cortex relative to the cytoplasm that is associated with the movement of dorsal determinants from the vegetal pole to the future dorsal region of the embryo (Gerhart et al. 1989). (A brief parenthetical point is in order here. Conventionally, the side of the amphibian and fish embryo where the organizer forms has been called dorsal, with the opposite side labeled as ventral. This axis assignment does not project unambiguously onto the clearly defined dorsal–ventral polarity of the larva, as pointed out forcefully in recent publications [Lane and Smith 1999; Lane and Sheets 2000, 2002]. In these papers, a new proposal is made for polarity assignments in the gastrula that, I believe, has some merit, but also presents some difficulties. As the conventional approach of equating organizer side with dorsal seems to remain in wide use at present, I shall apply this convention, albeit with the reservation above.)
As gastrulation starts, the Spemann–Mangold organizer, which includes mostly axial mesodermal precursors, was classically believed to instruct naïve ectoderm to convert to neural tissue. In transplant or explant studies, animal ectoderm that forms epidermis, when undisturbed, is susceptible to neural induction by the organizer. This fact prompted a search for neural inducers that eventually led to the identification of several substances with the expected properties—organizer products that can neuralize ectoderm. Their molecular properties were at first surprising: they proved to be antagonists of other signaling factors, mostly of bone morphogenetic proteins (BMPs) and also of WNT (a secreted protein homologous to the Drosophila Wingless protein) and Nodal factors (Sasai and De Robertis 1997; Hibi et al. 2002). These observations led to the formulation of a “default” model of neural induction (Weinstein and Hemmati-Brivanlou 1997), which states that ectodermal cells will differentiate along a neural pathway unless induced to a different fate. The heuristic simplicity and logical cogency of this model facilitated its wide acceptance, although it did not explain the processes that set the “default.” Some of these processes have been the subject of subsequent studies that were conducted in several different species, and this has led to a more refined (and probably more accurate) picture.
For example, additional signaling pathways are now known to operate. Recent work on neural induction comes to two major conclusions: (i) the fibroblast growth factor (FGF) signaling pathway plays a major role in this process, and (ii) neural specification starts well before gastrulation and thus before the formation and function of the organizer. Studies on the role of FGF in early Xenopus development initially discovered its role in mesoderm induction and the formation of posterior tissues (Kimelman et al. 1992). And while the involvement of FGF in neuralization was observed early in this system (Lamb and Harland 1995; Launay et al. 1996; Hongo et al. 1999; Hardcastle et al. 2000), in view of the impressive effects seen with Chordin and other BMP pathway antagonists, the relevance of FGF in neural specification in amphibians and fish was slow to be recognized. It took elegant studies, mostly in chick embryos (Streit et al. 2000), and their eloquent exposition (Streit and Stern 1999; Wilson and Edlund 2001; Stern 2002) to turn the tide, but there is now no doubt that the FGF signaling pathway plays a major role in the specification and early development of the neural ectoderm in chordates.
Not surprisingly, then, attention has turned to the target tissues and to the prepatterns that might already exist. In Xenopus, it was long known that the animal region or pre-ectoderm is not uniform or naïve, in that the dorsal, organizer-proximal region is predisposed towards a neural fate (Sharpe et al. 1987). The paper by Kuroda et al. (2004) adds much information about neural specification before gastrulation in Xenopus and the factors involved in this process. The authors identify a region in the dorsal ectoderm of the blastula that they name the “blastula Chordin- and Nogginexpressing” (or BCNE) region (Figure 2). They show that this region, which I prefer to simply call dorsal ectoderm, expresses siamois, chordin, and Xnr3, another β-catenin target. The dorsal ectoderm or BCNE is fully specified as anterior neural ectoderm, as excision of this region led to headless embryos, and explants differentiated into neural tissue in culture, even when the formation of any mesodermal cells was blocked by interference with nodal signaling (Kuroda et al. 2004).
The formation of the vertebrate nervous system thus depends on multiple signaling pathways, such as the FGF, BMP, and WNT signaling cascades, that interact in complex ways (e.g., Pera et al. 2003). In contrast to the classical view, neural induction is not exclusively promoted by organizer-derived signals, in that earlier signals and intrinsic processes that determine ectodermal competence are prominently involved. Whether inductive signals or competence of responding tissue is more important in embryology has been debated, much like the nature–nurture controversy in the behavioral arena. Current work has given some boost to the competence side of the argument, but, as in behavior, the truth lies somewhere in between, though not necessarily at the halfway mark. Studies such as those discussed here bring us closer to finding the answer to this question.