DEV032748 Supplementary Material

Files in this Data Supplement:

• Supplemental Figure S1 -

  Fig. S1. Defects in the ALM and its derivatives on loss of gata4, gata5 and/or gata6. (A) gata4 is downregulated in gata5+gata6 morphants at 5 somites, showing that in effect gata5+gata6 morphants are a triple knockdown model. (B) Myelopoiesis is reduced but not abolished in single GATA morphants as seen by a downregulation of l-plastin and mpx in gata4, gata5 and gata6 morphants.

• Supplemental Figure S2 -

  Fig. S2. PLM expression of haemangioblast-associated genes is independent of gata4, gata5 and gata6. The expression of haemangioblast-associated genes was analysed close to the onset of their expression. Expression in the PLM was unchanged in gata5+gata6 morphants. uncx4.1 was used as a somite marker to stage the embryos (white brackets) and the stage of the embryos is shown under each panel. Posterior views. For each gene analysed, n=40-100, the images shown represent the findings for >95% of the embryos.

• Supplemental Figure S3 -

  Fig. S3. gata4, gata5 and gata6 expression is lost in the endoderm but not in the mesoderm of casanova morphants. (A) Expression of gata4, gata5 and gata6 was downregulated in the gut tube of cas morphants (red arrows) at 27 hpf, whilst expression remained in the cardiac regions (black arrowheads). Cardia bidifa was observed in cas morphants (two black arrowheads), whilst wild-type embryos possess a single heart tube (one black arrowhead). Cardia bifida in cas morphants was confirmed by the bilateral expression of cmlc2, which is not seen in control embryos. (B) Expression of gata4, gata5 and gata6 at 5 somites remained normal in the ALM of casanova morphants. Images shown in A are dorsal views, anterior to left; those in B are anterior-dorsal views.

• Supplemental Figure S4 -

  Fig. S4. Pu1 is required for myeloid differentiation but not for anterior haemangioblast formation. (A) Recovery of the head endothelial programme is seen in gata5 and gata6 morphants. fli1 expression appears normal at 22 and 30 hpf. (B,C) To assess the requirement for pu1 in the ALM, embryos were injected with pu1 morpholino (mo) and gene expression was analysed at 5 and 10 somites. (B) The loss of pu1 expression itself in the ALM, and the PLM (data not shown), confirmed pu1 mo efficiency (red arrows). gata2, gata4, gata5 and gata6 were all expressed as in the wild-type embryos in the ALM, as was draculin. Ectopic expression of gata1 was not observed at 5 somites in pu1 morphants. (C) Expression of the haemangioblast-associated genes scl, fli1, lmo2 and hhex was unchanged at 10 somites in pu1 morphants, demonstrating that pu1 is not required for anterior haemangioblast formation. However, cmyb, runx1 and ikaros were specifically lost in the ALM of embryos depleted of pu1 (red arrowheads), confirming a requirement for pu1 in establishing the myeloid programme. Anterior-dorsal views. For each gene analysed, n=43-97, the images shown represent the findings for >95% of the embryos.