Fig. 3

RA synthesis is required for initiation of posterior hindbrain markers. (A–D) Control (A, C) and 5 µM DEAB-treated (B, D) embryos are stained in blue for vhnf1 (A, B) and hoxd4a (C, D). Other markers in red are pax2a (A–D) and krox-20 (C, D). Dorsal views show anterior to the top. Treatments were initiated at 4 hpf (sphere stage), and embryos were fixed at 9 hpf (90% epiboly) (A, B) or 10.6 hpf (2 s) (C, D). For vhnf1, 10/10 DEAB-treated embryos showed loss of expression. For hoxd4a, 10/10 DEAB-treated embryos showed loss of hindbrain expression as well as loss of r5 krox-20 expression. r3, rhombomere 3 krox-20 expression. Scale bar: 50 μm. (E) RA synthesis is required for initiation, but not maintenance, of vhnf1 and hoxd4a. Embryos were treated with 5 μM DEAB beginning at stages between 4 hpf and 11 hpf (3 s). Arrows point to normal initiation times of vhnf1 and hoxd4a expression. For vhnf1 analysis, embryos were treated until 10.3 hpf (1 s) then were fixed and stained for vhnf1 expression. For hoxd4a analysis, embryos were treated until 18 hpf (18 s) then were fixed and stained for hoxd4a and krox-20 expression. Levels of hindbrain expression were scored on a 0–4 scale, with 0 representing no expression and 4 representing wild-type levels. n = 10 for each vhnf1 data point. n e 12 for each hoxd4a data point. (F) Specification of r5–r6 and r7 has distinct temporal requirements for RA synthesis. The same series of treatments were used as for panel (E). For r5 analysis, krox-20 expression was scored in 18 hpf embryos. For r6 analysis, val (non-overlapping r5 krox-20) expression was scored in 18 hpf embryos. For r7 analysis, efnb2a expression was scored in embryos fixed after treatment until 11.6 hpf (5 s). Rhombomere length was scored on a 0–3 scale, with 0 representing no expression and 3 representing wild type or longer lengths of expression domains. n e 20 for each r5 data point. n e 10 for each r6 data point. n e 14 for each r7 data point. Error bars represent 95% confidence interval.