Neural differentiation is induced by inhibition of BMP signaling. Secreted inhibitors of BMP such as Chordin from the Spemann organizer contribute to the initial step of neural induction. Xenopus Smad-interacting protein-1 gene (XSIP1) is expressed in neuroectoderm from the early gastrula stage through to the neurula stage. XSIP1 is able to inhibit BMP signaling and overexpression of XSIP1 induces neural differentiation. To clarify the function of XSIP1 in neural differentiation, we performed a loss-of-function study of XSIP1. Knockdown of XSIP1 inhibited SoxD expression and neural differentiation. These results indicate that XSIP1 is essential for neural induction. Furthermore, loss-of-function experiments showed that SoxD is essential for XSIP1 transcription and for neural differentiation. However, inhibition of XSIP1 translation prevented neural differentiation induced by SoxD; thus, SoxD was not sufficient to mediate neural differentiation. Expression of XSIP1 was also required for inhibition of BMP signaling. Together, these results suggest that XSIP1 and SoxD interdependently function to maintain neural differentiation.

In vertebrates, nodal-related genes are crucial for specifying mesendodermal cell fates. Six nodal-related genes have been identified in Xenopus, but only one, nodal, has been identified in the mouse. The Xenopus nodal-related gene 3 (Xnr3), however, lacks the mesoderm-inducing activity of the other five nodal-related genes in Xenopus, and can directly induce neural tissue in animal caps by antagonizing BMP signals. In this study, we isolated three clones of the Xenopus (Silurana) tropicalis nodal-related gene 3 (Xtnr3) and analyzed their function. The Xtnr3 genes show high homology to Xnr3 and have the same activity. Southern blot and genomic PCR analyses indicate that the X. tropicalis genome has duplications in the Xtnr3 gene sequences and our three clones represent separate gene loci. We also found a partial clone of Xtnr3 that coded for the N-terminal part of its pro-region. Surprisingly, this sequence also induced neural tissue by antagonizing BMP signals, and its coded protein physically associated with BMP4 mature protein. Furthermore, we showed that the pro-region of Xnr5 has the same activity. Together, these findings indicate that the pro-region of nodal-related genes acts antagonistically towards BMP signals, which identifies a novel mechanism for the inhibition of BMP signaling.

In a screening for activin-responsive genes, we isolated a Xenopus lefty/antivin-related gene, called Xantivin (Xatv). In the animal cap assay, the expression of Xatv was induced by activin signaling, and in the embryo, by nodal-related genes. Overexpression of Xatv in the marginal zone caused suppression of mesoderm formation and gastrulation defects, and inhibited the secondary axis formation induced by Xnr1 and Xactivin, suggesting that Xatv acted as a feedback inhibitor of activin signaling. However, in the animal cap, Xatv failed to antagonize Xnr1 and Xactivin. This result suggested that Xatv has different responses in the marginal zone and in the animal region, and antagonizes to a higher degree activin signaling in the marginal zone.

Smad-interacting protein-1 (SIP1), also known as δEF2, ZEB2 and zfhx1b, is essential for the formation of the neural tube and the somites. Overexpression of Xenopus SIP1 causes ectopic neural induction via inhibition of bone morphogenetic protein (BMP) signaling and inhibition of Xbra expression. Here, we report the functional analyses of 4 domain-deletion mutants of XSIP1. Deletion of the N-terminus zinc finger domain suppressed neural induction and BMP inhibition, but these were not affected by deletion of the other domains (the Smad binding domain, the DNA-binding homeodomain together with the CtBP binding site and the C-terminus zinc finger). Therefore SIP1 does not inhibit BMP signaling by binding to Smad proteins. In contrast, all of the deletion constructs inhibited Xbra expression. These results suggest that the N-terminus zinc finger domain of XSIP1 has an important role in neural induction and that Xbra suppression occurs via a mechanism separate from the neural inducing activity.

Recent studies indicate an essential role for the EGF-CFC family in vertebrate development, particularly in the regulation of nodal signaling. Biochemical evidence suggests that EGF-CFC genes can also activate certain cellular responses independently of nodal signaling. Here, we show that FRL-1, a Xenopus EGF-CFC gene, suppresses BMP signaling to regulate an early step in neural induction. Overexpression of FRL-1 in animal caps induced the early neural markers zic3, soxD and Xngnr-1, but not the panmesodermal marker Xbra or the dorsal mesodermal marker chordin. Furthermore, overexpression of FRL-1 suppressed the expression of the BMP-responsive genes, Xvent-1 and Xmsx-1, which are expressed in animal caps and induced by overexpressed BMP-4. Conversely, loss of function analysis using morpholino-antisense oligonucleotides against FRL-1 (FRL-1MO) showed that FRL-1 is required for neural development. FRL-1MO-injected embryos lacked neural structures but contained mesodermal tissue. It was suggested previously that expression of early neural genes that mark the start of neuralization is activated in the presumptive neuroectoderm of gastrulae. FRL-1MO also inhibited the expression of these genes in dorsal ectoderm, but did not affect the expression of chordin, which acts as a neural inducer from dorsal mesoderm. FRL-1MO also inhibited the expression of neural markers that were induced by chordin in animal caps, suggesting that FRL-1 enables the response to neural inducing signals in ectoderm. Furthermore, we showed that the activation of mitogenactivated protein kinase by FRL-1 is required for neural induction and BMP inhibition. Together, these results suggest that FRL-1 is essential in the establishment of the neural induction response.