Fig. S3

Moesin1 is required for perfusion of the ISVs as a result of defects in tube formation.Tg(fli1:egfp)^y1 embryos injected intravascularly with Rhodamine-dextran (TMRD) in order to assess vascular perfusion. Bright-field (A,D,G,J,M,P), EGFP fluorescence (B,E,H,K,N,Q) and microangiography (c,f,i,l,o,r) images are shown. (A-F) Embryos were injected with 4 ng control MO (a-c) or 4 ng of moesin1 MO (d-f) at the 1- to 2-cell stage and processed for microangiography at 39 hpf. At 39 hpf, the EGFP-expressing ISV endothelial cells finished migration in both control and Moesin1 knockdown embryos. Minor disorganization is observed in some of the ISVs, with endothelial cells failing to make contacts with one another in the dorsal longitudinal anastomotic vessel (arrow in E). The axial vessels, comprising the dorsal aorta (arrowhead in C) and the posterior cardinal vein (arrow in C), show perfusion with TMRD under both conditions. (g-l) At 54 hpf, control MO-injected embryos (g-i) display perfusion in both the axial vessels and ISVs in the trunk. By contrast, embryos injected with 4 ng moesin1 MO (j-l) fail to perfuse TMRD in most of the ISVs and axial circulation is reduced. Examination of the endothelial cells in the ISVs labeled with EGFP showed they were present and in the proper locations. Taken together, these observations suggest that knockdown of Moesin1 leads to defects in endothelial tubulogenesis. (m-r) Defects in perfusion of the ISVs are still observed at 78 hpf after knockdown of Moesin1 (p-r), while the endothelial cells persisted. Normal vascular perfusion is observed after injection of control MO (M-O), indicating that the defects in vascular perfusion are not a result of developmental delay in the embryo.