Neural Induction

The developmental significance of the dorsal lip of the blastopore, which is derived from the gray crescent, was initially demonstrated by the organizer transplant experiments conducted by Spemann and Mangold. They demonstrated that dorsal lip material transplanted to the ventral side of a host gastrula would induce a secondary embryo. The transplanted dorsal lips induced the overlying host ectoderm to develop as neural tissue through vertical signaling.

As a refresher, you should review the following aspects of neural induction (Browder et al., 1991, pages 485-496): Kessler and Melton (1994; see Fig. 3) have summarized very nicely the interactions that are involved in induction and regionalization of the central nervous system in Xenopus. As shown in Fig. 3, the involuting dorso-anterior mesoderm induces the adjacent mesoderm to form anterior neural tissue. As the mesoderm migrates toward the former animal pole, it contacts progressively more overlying ectoderm, which is also induced to form anterior neural tissue. As involution proceeds, ectoderm is progressively contacted by more posterior mesoderm, which induces more posterior neural elements. According to this model, the anteroposterior character of the neural ectoderm is dependent upon corresponding differences in the underlying dorsal mesoderm.

The discovery of neural induction by Spemann triggered a decades-long search for the "inductive factor". The recent discovery of the involvement of growth factor signaling in mesoderm induction has facilitated progress in understanding the molecular and cellular basis for neural induction. Some of the factors implicated in mesoderm induction may also play a role in neural induction.

As we have previously discussed, noggin is a secreted factor that has the capability of dorsalizing the mesoderm (see Mesoderm Induction II). Noggin protein can also induce anterior and general neural markers (NCAM and XIF3 mRNA; XAG-1 and otx2 [formerly otxA] protein; Figs. 1, 3 and 4, Lamb et al., 1993) in animal pole explants in the absence of detectable mesoderm. These dual properties are characteristics to be expected of the Spemann Organizer. Recall that noggin transcripts are synthesized in the organizer at the gastrula stage and in the notochord at later stages; both of these tissues have been implicated in neural induction. Noggin protein, however, had little or no ability to induce posterior neural markers (e.g., ß-tubulin mRNA, En-2, Krox20 and XlHbox6 protein) or antigens that are characteristic of differentiated neural cells. Likewise, noggin-treated animal pole explants failed to grow neuronal processes. Thus, Lamb et al. (1993) concluded that noggin can apparently induce anterior neural tissue, but it can neither induce posterior neural tissue nor cause neural differentiation (Lamb et al., 1993). Recently, however, Knecht et al. (1995) have reported that some neural differentiation can occur in response to noggin induction. This paper will be the subject of a student seminar.

There is also evidence to support the hypothesis that inhibition of activin signaling is involved in neural induction. This evidence is derived from experiments using activin inhibitors and may help to explain the enigma that neural tissue can form from dissociated animal cap isolates in the absence of mesoderm. One set of experiments has utilized the dominant-negative inhibitor of an activin receptor. Although expression of this receptor in animal pole explants blocks mesoderm induction (see Mesoderm Induction I), anterior and general neural markers are expressed (Fig. 2, Hemmati-Brivanlou and Melton, 1992; Figs. 1, 2, 3 and Table 1, Hemmati-Brivanlou and Melton, 1994). This result suggests that neural development can only proceed when a signal transduced by an activin receptor is inhibited. This possibility would require the presence of an antagonist of activin-like activity in the embryo. A native antagonist of activin signaling has, indeed, been identified: follistatin. Follistatin is normally expressed in the organizer at the gastrula stage and later in the notochord and anterior nervous system (Fig. 3, Hemmati-Brivanlou et al., 1994). Hence, it is expressed at the right time and in the right tissues to be involved in neural induction. Expression of follistatin in animal pole explants can cause expression of anterior and general neural markers in the absence of mesoderm (Figs. 6, 7, 9, Table 1, Hemmati-Brivanlou et al., 1994). At the present time, it is uncertain which TGF-ß family member signals are perturbed by follistatin; it is conceivable that follistatin perturbs the Vg-1 signal. We are left with two very strong candidates for neural induction: noggin and follistatin. Although there is no reason to assume that they are mutually exclusive candidates for the neural inducer, any interaction between them should be examined. Does noggin operate by activating follistatin expression? To examine this, Hemmati-Brivanlou et al. (1994) injected noggin mRNA into embryos at the 2-cell stage. Animal caps were isolated at the blastula stage, and half of them were assayed for expression of follistatin mRNA. As shown in Fig. 10, follistatin was not expressed. The other group of animal caps did express N-CAM, which demonstrates that noggin caused neural induction in the absence of follistatin expression. Activin, on the other hand, can induce the expression of both follistatin and noggin (Fig. 10, Hemmati-Brivanlou et al., 1994). Thus, the effects of activin (or a related TGF-ß family member) may be mediated through either follistatin or noggin - or both.

The localization of noggin in the Spemann organizer makes it a serious candidate to be one of the endogenous mediators of neural induction. Recently, another TGF-ß homolog (Anti-Dorsalizing Morphogenetic Protein; ADMP) has been detected in the organizer (Moos et al., 1995). This factor, which is in the BMP group of TGF-ß proteins, is unique in that it appears to be responsible for suppressing dorsalization. ADMP mRNA is first detected during development in the late blastula stage and peaks during gastrulation (Fig. 2, Moos et al., 1995). In situ staining revealed that the ADMP mRNA accumulates in the Spemann Organizer region during gastrulation (Fig. 3, Moos et al., 1995). Evidence concerning the potential role of ADMP during development was revealed by microinjecting ADMP mRNA into dorsal blastomeres at the 4- to 8- cell stage. As shown in Figure 4 and 5 (Moos et al., 1995), these embryos showed defects in dorso-anterior structures. Interestingly, these defects did not become evident until neurulation. The results suggest that mesodermal fate was diverted from dorsal/anterior to ventral/posterior. Injection of the messenger into ventral blastomeres had less pronounced effects.

The paradox of an expression pattern similar to that of dorsalizing agents and the ventralizing effect of overexpression of ADMP led Moos et al. to examine the effects of dorsalizing and ventralizing factors on ADMP expression. The dorsalizing agent LiCl produced a 4- to 8-fold increase in ADMP mRNA, whereas UV irradiation had the converse effect: a 2- to 3-fold decrease (Fig. 6, Moos et al., 1995). The opposing effects of lithium and ADMP are demonstrated by an experiment in which embryos were injected with ADMP mRNA before lithium treatment; the dorsalizing effects of lithium were reduced and in some cases ventralized embryos resulted. Activin treatment of animal caps resulted in increased ADMP expression (Fig. 6C and D). The molecular consequences of ADMP expression are shown by monitoring the expression of dorsalizing signals and dorsal markers in embryos that had been injected with ADMP mRNA. As shown in Figure 7, the expression of organizer-specific markers noggin, goosecoid and follistatin was suppressed, as was the activin receptor XAR1 and dorsal markers XMyoD, NCAM and muscle actin. Expression of XHox-3, XWnt-8 and Xbra was increased.

The results of these experiments suggest that ADMP suppresses dorsalization signals. The apparent existence of antagonizing factors in the organizer suggests that a counterbalancing of positive and negative signals is necessary for proper development.

In addition to the signaling factors, their receptors and transducers, homeobox-class transcription factors play important roles in patterning the early embryo. Some of these factors have patterns of expression restricted to the organizer. Of special interst is goosecoid (gsc). This homeobox gene is initially expressed in the organizer at late blastula to early gastrula stages, followed by expression in anterior dorsal mesoderm. Figures 7, 8, and 9 in Vodicka and Gerhart (1995) document beautifully the presence of this and other transcripts in the organizer. If it is involved in regulating gene expression in the organizer, it is essential to determine which genes are targets of its activity.

Although the factors that we have discussed appear to have central roles to play in neural induction, a number of additional factors apparently have additional roles to play in induction and patterning. Among these are Xotch, dorsalin1, hedgehog, Xash, and brachyury.

The progress in understanding the sequence of events from the initial cortical rotation that follows sperm entry through neural induction and patterning has been rapid in the past few years. Spemann's search for the inducing principle has blossomed into a growth industry that has produced a long list of factors that may be involved in induction and patterning. Identification of the actual players and elucidation of their specific roles in induction, however, remain to be resolved. Nevertheless, we are close to knowing in molecular terms how the Xenopus embryo generates a bilaterally symmetric embryo with three germ layers and a central nervous system with anterior-posterior patterning from a radially symmetric egg.

Have you ever lain awake at night asking yourself why vertebrates have dorsal nerve cords, whereas insects have a ventral nerve cord? This conundrum will be the subject of a student seminar.

Hot off the Press:

A common plan for dorsoventral patterning in Bilateria. E M De Robertis & Y Sasai. Nature 380, 37-40 (March 7, 1996)


References

Browder, L.W., C.A. Erickson and W.R. Jeffery. 1991. Developmental Biology. Third ed. Saunders College Publishing. Philadelphia.

Hemmati-Brivanlou, A. and D.A. Melton. 1992. A truncated activin receptor inhibits mesoderm induction and formation of axial structures in Xenopus embryos. Nature 359: 609-614.

Hemmati-Brivanlou, A., O.G. Kelly and D.A. Melton. 1994. Follistatin, an antagonist of activin, is expressed in the Spemann organizer and displays direct neuralizing activity. Cell 77: 283-295.

Kessler, D.S. and D. A. Melton. 1994. Vertebrate embryonic induction: mesodermal and neural patterning. Science 266: 596-604.

Moos, M., Jr., S. Wang and M. Krinks. 1995. Anti-dorsalizing morphogenetic protein is a novel TGF-ß homolog expressed in the Spemann organizer. Development 121: 4293-4301.

Hemmati-Brivanlou, A. and D.A. Melton. 1994. Inhibition of activin signalling promotes neuralization in Xenopus laevis.. Cell 77: 273-281.

Vodicka, M.A. and J.C. Gerhart. 1995. Blastomere derivation and domains of gene expression in the Spemann Organizer of Xenopus laevis. Development 121: 3505-3518.

Lamb, T.M., Knecht, A.K., Smith, W.C., Stachel, S.E., Economides, A.N., Stahl, N., Yancopolous, G.D., and Harland, R.M. 1993. Neural induction by the secreted polypeptide noggin. Science 262: 713-718.



References for Student Seminars:
Knecht, A.K., P.J. Good, I.B. Dawid and R.M. Harland. 1995. Dorsal-ventral patterning and differentiation of noggin-induced neural tissue in the absence of mesoderm. Development 121: 1927-1936. (Note: Check out the expression patterns for sybII, cpl-1 , etr-1 and nrp-1, which are discussed in this paper.)

Holley, S.A., Jackson, P.D., Sasai, Y., Lu, B., De Robertis, E.M., Hoffmann, F.M., and Ferguson, E.L. 1995. A conserved system for dorsal-ventral patterning in insects and vertebrates involving sog and chordin. Nature 376: 249-253. (See also News and Views: Hogan. Nature 376: 210.)

Jones, C.M. and J.C. Smith. 1995. Revolving vertebrates. Current Biology 5: 574-576.


RETURN TO ADVANCED DEVELOPMENTAL BIOLOGY