Fig. 2. Vegetal turnover of SoxB1GFP requires nuclear ß-catenin and can be driven by Pmar1, but is not affected by Krl. (A) Embryos at the 16-cell stage derived from fertilized eggs that had been injected with mRNA encoding SoxB1-GFP. Images were obtained with a Nikon inverted microscope equipped with epifluoresence optics and Kodak Elite 35 mm slide film. The embryo on the left is shown in a slightly tilted, vegetal pole view that shows the four micromeres, eight macromeres and one mesomere. The embryo is oriented so that mesomeres at the animal pole are up and micromeres at the vegetal pole are down. (B-G) Zygotes were injected with mRNA encoding SoxB1-GFP and the indicated proteins, allowed to develop to the temporal equivalent of mesenchyme blastula stage, deciliated, and fluorescence images were captured from live embryos by laser confocal microscopy. (B) Control embryo injected with SoxB1-GFP mRNA and glycerol, demonstrating that the SoxB1-GFP fusion protein mimics the vegetal turnover exhibited by the endogenous SoxB1. The arrow indicates PMCs that have ingressed from the vegetal plate. (C) Embryos in which nuclearization of ß-catenin is blocked by co-injection of C-cadherin mRNA do not clear SoxB1GFP from vegetal blastomeres. These embryos lack PMCs (Logan et al., 1999). (D) By contrast, upregulation of nuclear ß-catenin activity with mRNA encoding stabilized ß-catenin vegetalizes the embryo and also expands the vegetal domain of SoxB1-GFP degradation. (E,F) Mis/overexpression of Krl (E) also appears to expand the SoxB1-GFP degradation domain, consistent with its strong vegetalizing effect (Howard et al., 2001), but knockdown of Krl by means of Krl MO injection (F) does not detectably alter SoxB1 turnover. (G) Mis/overexpression of Pmar1 converts most of the cells to a PMC-like fate and upregulates SoxB1-GFP turnover throughout the embryo. Scale bar: 20 µm.