Fig. 7. Relationship between pou2 and fgf8. All embryos are depicted dorsally with the exception of the embryo in F, which is depicted laterally. (A-D) Gain of fgf8 function by unilateral misexpression of fgf8 mRNA into one cell of two-cell stage embryos. To determine the effect caused by fgf8 overexpression, gbx2 (A,B) and spry4 (C,D), both markers for the prospective hindbrain, were used. The activity of misexpressed fgf8 can be judged from dorsalization of the embryos indicated by lateral expansion of endogenous gbx2 and spry4 expression, indicated by arrows (A-E) at the injected side of the embryo. Deposition of co-injected lacZ mRNA is visualized by staining for anti-ß-gal antibody (brown), reflecting the location of injected fgf8 mRNA (not visualized). Distribution of injected mRNA is restricted to one half of the embryo, allowing for comparison with the contralateral side as a control. In spg embryos, neither expression of gbx2 (B) nor spry4 (D) could be rescued or upregulated, respectively, by fgf8 overexpression. (E,F) In a reversed experiment, pou2 mis-expression into ace embryos (carried out in the same unilateral fashion described for fgf8 injection above), pou2 overexpression and fgf8 itself can provoke dorsalization of the injected half of the embryo (obviously seen in the wild-type embryo in E, but not in the ace embryo in F, owing to complete loss of the readout marker gbx2) but cannot rescue expression of gbx2 in ace mutant embryos. (H) A bead soaked with Fgf8 protein can not rescue the morphology of the isthmic constriction at the MHB but can evoke ectopic spry4 expression in wild-type and spg embryos (G,H; white circles indicate the implanted bead).