Distinct mesodermal regions underlying the neural plate appear to possess distinct roles in neural development. During gastrula stages, the most dorsal mesoderm, comprising Spemann’s organizer or the node, depending on the species, is essential for the induction of the neural fate in the epidermal ectoderm, via attenuation of BMP signaling by various means (see Chapter 1, “Introduction”). Cells from this region involute during gastrulation forming the notochord, with the prechordal plate preceding its progress to the most anterior end of the embryo.

Lateral to the organizer mesoderm, on both sides, lays the paraxial-fated mesoderm. Cells from this region also involute during gastrulation but give rise to the paraxial mesoderm that will later form the somites. A long line of evidence from zebrafish, Xenopus, and chick embryos suggests a key role for this region in neural patterning and caudalization (Bang et al., 1999; Bonstein et al., 1998; Elkouby et al., 2010; Erter et al., 2001; Gould et al., 1998; Grapin-Botton et al., 1997; Itasaki et al., 1996; Monsoro-Burq et al., 2003; Muhr et al., 1999; Muhr et al., 1997; Nordstrom et al., 2002; Woo and Fraser, 1997). Several candidate Wnt ligand molecules are expressed in the paraxial-fated mesoderm that could account for its posterior inducing activity in vertebrates (Figure 10.1).

FIGURE 10.1

Expression of various Wnt ligands in the paraxial mesoderm. The paraxial-fated mesoderm expresses Wnt3a in Xenopus (see Elkouby and Frank, unpublished), Wnt8c and Wnt11 in chick (Nordstrom et al., 2002), and Wnt3a and ORFs 1 and 2 of Wnt8 in zebrafish (more...)

In zebrafish transplantation experiments, paraxial-fated mesoderm transformed forebrain to more posterior cell types (Figure 10.2A) (Woo and Fraser, 1997). Grafting and explant experiments in chick embryos showed that paraxial-fated mesoderm caudalized the neural plate (Figure 10.2B) (Grapin-Botton et al., 1997; Itasaki et al., 1996; Muhr et al., 1999; Muhr et al., 1997; Nordstrom et al., 2002). This endogenous chick caudalizing ligand was not identified, but neither FGF nor RA activity could replace paraxial tissue in caudalizing assays (Muhr et al., 1999; Muhr et al., 1997), thus arguing for the involvement of Wnt/β-catenin signaling in this process. Moreover, in zebrafish mutants lacking paraxial-fated mesoderm, posterior neural tissue failed to form, and this was attributed to the lack of Wnt3a and Wnt8 ligands normally expressed in this region (Erter et al., 2001). In Xenopus, ablation of the paraxial-fated mesoderm in vivo results in embryos lacking posterior neural tissue (Elkouby et al., 2010), whereas ablation of the dorsal organizer mesoderm did not perturb posterior neural development. In recombinant explants composed of paraxial-fated mesoderm and anterior neural ectoderm, posterior neural cell fates were induced in the neural tissue (Figure 10.2C). In contrast, Wnt3a-deficient paraxial-fated mesoderm completely failed to induce posterior cell fates in anterior neural ectoderm, thus identifying Wnt3a as the crucial secreted ligand that caudalizes this tissue (Figure 10.2C) (Elkouby et al., 2010). This Wnt3a ligand, derived from the paraxial-fated mesoderm, was shown to induce hindbrain cell fates via the direct activation of the hindbrain expressed homeobox gene, Meis3, which interprets this Wnt morphogenic activity to induce hindbrain specific Hox gene expression (Figure 10.2D) (see Chapter 5, “Induction of the Hindbrain”).

FIGURE 10.2

Paraxial mesoderm caudalizes the neural plate via Wnt signals. (A) The capability of paraxial mesoderm to induce neural caudalization was shown, when cells from the paraxial margins, 90^o from the shield (brown staining), were transplanted into the animal (more...)

In chick embryos, Wnt activity was suggested to mediate paraxial-fated mesoderm induction of posterior neural cell fates, but no specific ligand was identified (Nordstrom et al., 2002). In zebrafish and Xenopus, Wnt3a and Wnt8 were identified as potential neural caudalizers expressed in the mesoderm. Studies in zebrafish morphants and mutants suggest that Wnt8 and Wnt3a may be functionally redundant and mesodermal sources of either molecule could act as a neural caudalizer. However, these embryos also suffered severe mesoderm perturbations (Erter et al., 2001; Lekven et al., 2001), so it is difficult to conclude if neural patterning defects are specific or the indirect result of losing paraxial mesoderm fates. In zebrafish, the Wnt8 MO consistently gave a stronger mesoderm perturbation phenotype versus the Wnt3a MO, and coinjection of both gave a more severe additive phenotype, which included neural mis-patterning (Shimizu et al., 2005). Hence, the potential for two overlapping but nonidentical roles for Wnt8 and Wnt3a in mesodermal and/or neural patterning is not ruled out.

In Xenopus and zebrafish, the disruption of Wnt8 activity perturbs mesoderm pattern as dorsal mesoderm is expanded and more ventral–lateral regions, including the paraxial-fated mesoderm, are reduced (Christian and Moon, 1993; Hoppler et al., 1996; Hoppler and Moon, 1998; Lekven et al., 2001; Ramel et al., 2005; Ramel and Lekven, 2004). These observations are not seen for Wnt3a in Xenopus (Elkouby et al., 2010). In Wnt3a morphant embryos, the organizer was not expanded, ventrolateral markers were expressed normally, and paraxial mesoderm formed muscle. Moreover, the Wnt3a MO blocked the induction of posterior-neural cell fates without altering either Wnt8 expression or activity (Elkouby et al., 2010). The timing of expression in both Xenopus and zebrafish shows that Wnt8 is detected before Wnt3a (Christian and Moon, 1993; Kelly et al., 1995; Krauss et al., 1992; McGrew et al., 1997; Shimizu et al., 2005; Smith and Harland, 1991). Altogether, these observations suggest that Wnt8 patterns the mesoderm and Wnt3a could act downstream to supplement this process by patterning neuroectoderm. Therefore, Wnt3a seems to be the crucial Wnt ligand, secreted from the paraxial-fated mesoderm required for neural patterning.

Nieuwkoop’s “activation–transformation” principle could thus be executed separately by two mesodermal domains. The activation step is mediated by dorsal mesoderm via its attenuation of BMP signaling in the overlying ectoderm. The following transformation step is then executed by the paraxial-fated mesoderm, probably via its secretion of Wnt3a ligands to the overlying neural plate. Because the neural fate is a prerequisite for posterior neural patterning, the activation step may provide the cellular competence for the inducer of the following transformation. Indeed, it was shown that while BMP antagonism and its mediating Zic protein activity are not sufficient for the expression of the Wnt/β-catenin direct target, Meis3, they are strongly required for its optimal activation by Wnt3a (Elkouby et al., 2010; Gutkovich et al., 2010).

As discussed previously (see Chapter 9, “Anti-Wnt Anterior Determinants”), organizer mesoderm and its subsequent prechordal plate mesoderm fated cells express a wide range of secreted and cell-autonomously acting factors to maintain its anterior character. These factors include an arsenal of Wnt antagonists, as well as CyP26, an RA degrading enzyme that restricts RA’s functional distribution. On the other hand, paraxial-fated and paraxial mesoderm express Wnt3a, Wnt8, FGF8, and RA caudalizing signals. Therefore, a balance between the organizer mesoderm and the paraxial-fated mesoderm is required to regulate proper A–P patterning in the developing nervous system.