Group B1 Sox genes encode HMG domain transcription factors that play major roles in neural development. We have identified six zebrafish B1 sox genes, which include pan‐vertebrate sox1a/b, sox2, and sox3, and also fish‐specific sox19a/b. SOX19A/B proteins show a transcriptional activation potential that is similar to other B1 SOX proteins. The expression of sox19a and sox3 begins at approximately the 1,000‐cell stage during embryogenesis and becomes confined to the future ectoderm by the shield stage. This is reminiscent of the epiblastic expression of Sox2 and/or Sox3 in amniotes. As development progresses, these six B1 sox genes display unique expression patterns that overlap distinctly from one region to another. sox19a expression is widespread in the early neuroectoderm, resembling pan‐neural Sox2 expression in amniotes, whereas zebrafish sox2 shows anterior‐restricted expression. Comparative genomics suggests that sox19a/b and mammalian Sox15 (group G) have an orthologous relationship and that the B1/G Sox genes arose from a common ancestral gene through two rounds of genome duplication. It seems likely, therefore, that each B1/G Sox gene has gained a distinct expression profile and function during vertebrate evolution. Developmental Dynamics 235:811–825, 2006.

To obtain an insight into the transcription factor (TF)-dependent regulation of epiblast stem cells (EpiSCs), we performed ChIP-seq analysis of the genomic binding regions in EpiSCs of five major TFs, ZIC2, OTX2, SOX2, POU5F1, and POU3F1, using biotinylated TFs, identifying the following new features: (1) The genomic domains of megabase scale rich in ZIC2 peaks and genes, and those rich in POU3F1 but sparse in genes alternate in EpiSCs, reflecting the clustering of regulatory regions that act in short- and long-ranges, which involve binding of ZIC2 and POU3F1, respectively. (2) The enhancers bound by ZIC2 and OTX2 prominently regulate TF genes in EpiSCs. (3) The binding sites for SOX2 and POU5F1 in mouse embryonic stem cells (ESCs) and EpiSCs are divergent, reflecting the shift of major-acting TFs from SOX2/POU5F1 in ESCs to OTX2/ZIC2 in EpiSCs. (4) This shift in the major-acting TFs appears to be primed by binding of ZIC2 in ESCs at relevant genomic positions that later function as enhancers following the disengagement of SOX2/POU5F1from the major regulatory functions and subsequent binding by OTX2.

The canonical Wnt signaling pathway plays a major role in the regulation of embryogenesis and organogenesis, where signal strength‐dependent cellular responses are of particular importance. To assess Wnt signal levels in individual cells, and to circumvent the integration site‐dependent bias shown in previous Wnt reporter lines, we constructed a new Wnt signal reporter mouse line R26‐WntVis. Heptameric TCF/LEF1 binding sequences were combined with a viral minimal promoter to confer a graded response to the reporter depending on Wnt signal strengths. The histone H2B‐EGFP fusion protein was chosen as the fluorescent reporter to facilitate single‐cell resolution analyses. This WntVis reporter gene was then inserted into the ROSA26 locus in an orientation opposite to that of the endogenous gene. The R26‐WntVis allele was introduced into Wnt3a^−/− and Wnt3a^vt/− mutant mouse embryos and compared with wild‐type embryos to assess its performance. The R26‐WntVis reporter was activated in known Wnt‐dependent tissues and responded in a graded fashion to signal intensity. This analysis also indicated that the major Wnt activity early in embryogenesis switched from Wnt3 to Wnt3a around E7.5. The R26‐WntVis mouse line will be widely useful for the study of Wnt signal‐dependent processes.