Fig. 3.

Direct activation of Dll4 expression by Foxn4 and Ascl1, but not Neurog1, through a phylogenetically conserved enhancer. (A) Schematics of truncated CR1 DsRed reporter constructs. Corresponding relative activities by Ascl1 are indicated to the right (ND, not determined). The green and red ovals in the CR1 fragment indicate the E-box motif CANNTG, with the red being the critical one. The four cyan ovals show the clustered ACGC Foxn4 binding motifs. In the CR1c/mE construct, the E-box is mutagenized to ATGATG, and in CR1c/m(1-3), the first three Foxn4 binding motifs are mutated to AAAA. The vertical dashed lines outline the critical 26-bp region containing the critical E-box and a ACGC motif. (B-Q′) DsRed expression in spinal cords co-electroporated with the CR1 reporter construct and indicated expression plasmids. Reporter activity was visualized in whole-mount spinal cords (B,B′,D,D′,F,F′,H,H′,J,J′,L,L′,N,N′,P,P′) or their cross sections (C,C′,E,E′,G,G′,I,I′,K,K′,M,M′,O,O′,Q,Q′). GFP alone showed only minimal endogenous enhancer activation (B-C′), whereas both Foxn4 and Ascl1 could strongly induce DsRed expression (D-G′). With Neurog1 barely any induction of DsRed expression was observed (H-I′). This repression was maintained even when Neurog1 was co-electroporated with Ascl1 (J-K′). Foxn4 and Ascl1 together, by contrast, caused synergistic increase of enhancer activation (L-M′), and electroporation of all three factors together could rescue Neurog1 inhibition to some extent (N-O′). Notably, Neurog1 and Foxn4 co-electroporation also resulted in significant reduction of ectopic enhancer activation observed with Foxn4 alone (P-Q′). (R-W′) Ascl1 regulation of truncated CR1 DsRed reporter constructs. As visualized in whole-mount embryos (R,R′,U,U′,V,V′) and transverse sections (R′,U′,V′), CR1d and CR1f exhibited high levels of DsRed expression, albeit less than full-length CR1. By contrast, CR1a, CR1b and CR1g had little or no activity (S-T′,W,W′). (X-AA′) E-box mutation nearly abolished DeRed reporter activation from CR1c by Ascl1 (X-Y′) while mutating Foxn4 binding motifs abrogated activation by Foxn4 (Z-AA′). (BB,CC) ChIP assay showing enrichment of the critical E-box region by anti-Ascl1 and anti-Neurog1 antibodies on chromatin of neural tubes pooled from E10.5, E11.5 and E12.5 wild-type embryos (BB,CC). Control DNA region from Dll4 3′ UTR was not significantly enriched (BB). Statistical significance was determined by Student’s t-test: ^*P<0.05. Scale bars: in Q′, 70 μm for B,B′,D,D′,F,F′,H,H′,J,J′,L,L′,N,N′,P,P′; in Q′, 30 μm for C,C′,E,E′,G,G′,I,I′,K,K′,M,M′,O,O′,Q,Q′; in W′, 54 μm for R-W′; in V′, 30 μm for R′,U′,V′; in AA′, 20 μm for X-AA′.