Loss of endothelial cell MEKK3 function impairs definitive hematopoiesis in the yolk sac. (A) Visual appearance of E9.5 and E10.5 control (Map3k3^f/+ and Map3k3^f/f) and EC MEKK3^KO (Tie2-Cre; Map3k3^f/-) embryos surrounded by intact yolk sacs. Boxed regions are shown at higher magnification in the image on the right. Note the pale appearance and lack of visible vasculature in the E10.5 EC MEKK3^KO yolk sac. Images are representative of 8 to 10 embryos of each genotype at both timepoints. Scale bars, 500 μm. (B) Hematoxylin and eosin stained sections of E9.5 and E10.5 control and EC MEKK3^KO yolk sacs showing dilated vascular spaces and lack of intravascular round hematopoietic cells in the E10.5 EC MEKK3^KO yolk sac. Images are representative of 8 to 10 yolk sacs per genotype at both timepoints. Scale bars, 20 μm. (C) The number of primitive erythroid colony-forming progenitors (EryP-CFC) in E8.5 control and EC MEKK3^KO yolk sacs. Each data point represents a single yolk sac. N = 10 control and 8 EC MEKK3^KO embryos. Error bars represent mean ± standard deviation (SD) and significance determined by an unpaired, 2-tailed Student t test. (D) Fraction of myeloid, EryP/Mix (erythrocyte progenitors/mixed lineage), and EryP (erythrocyte progenitors) colonies identified from culture of E8.5 control and EC MEKK3^KO yolk sacs. N = 10 control and 8 EC MEKK3^KO yolk sacs. (E) Confocal image of E9.5 and E10.5 control and EC MEKK3^KO yolk sac sections stained for CD31 and RUNX1. White arrows indicate rounded RUNX1⁺CD31⁺ cells; yellow arrowheads indicate flat RUNX1⁺ HECs. Scale bars, 20 μm. (F) Number of RUNX1⁺ cells per square millimeter (mm²) of vascular area in E9.5 and E10.5 control and EC MEKK3^KO yolk sacs. N = 9 E9.5 control, 18 E10.5 control, 17 E9.5 EC MEKK3^KO, and 10 E10.5 EC MEKK3^KO yolk sacs. Each data point represents the mean of >25 vessels analyzed in a single animal’s yolk sac. Error bars represent mean ± SD with significance determined by an unpaired, 2-tailed Student t test. (G) Number of RUNX1⁺ CD31⁺ cells per mm² of vascular area in E9.5 and E10.5 control and EC MEKK3^KO yolk sacs. N = 9 E9.5 control, 18 E10.5 control, 17 E9.5 EC MEKK3^KO, and 10 E10.5 EC MEKK3^KO yolk sacs. Error bars represent mean ± SD with significance determined by an unpaired, 2-tailed Student t test. (H) Method of measuring vascular area and the number of round and flat RUNX1⁺ HECs. Vascular area measurement was determined using ImageJ after defining vascular borders based on CD31 expression (top). Flat endothelial cells were defined as those with a length/width ratio of >2:1, and round cells with a length/width ratio of ≤2:1. (I) Ratio of round/flat RUNX1⁺ cells in E9.5 and E10.5 control and EC MEKK3^KO yolk sacs. N = 9 E9.5 control, 18 E10.5 control, 17 E9.5 EC MEKK3^KO, and 10 E10.5 EC MEKK3^KO yolk sacs. Error bars represent mean ± SD with significance determined by an unpaired, 2-tailed Student t test. (J) The number of EMPs per yolk sac (±SD), enumerated in methylcellulose colony-forming assays. f/+ = Map3k3^f/+; f/- = Map3k3^f-; Cre;f^/+ = Tie2-Cre;Map3k3^f/+; Cre;^f/−= Tie2-Cre;Map3k3^f/−. Cre;^f/+ embryos were used to determine the deletion efficiency of the Map3k3^f allele by Tie2-Cre. (K) Individual EMP colonies analyzed by polymerase chain reaction for deletion of Map3k3^f alleles. f/+ represents colonies from Tie2-Cre; Map3k3^f//+ yolk sacs in which the Map3k3^f allele was not deleted, and Δ/+ represents colonies from Tie2-Cre; Map3k3^f//+ yolk sacs in which the Map3k3^f allele was deleted. f/− and Δ/− represent colonies from Tie2-Cre; Map3k3^f//− yolk sacs in which the Map3k3^f allele was not deleted (f/−) or deleted (Δ/−). Error bars represent mean ± SD of 7 individual Cre; f/+ and Cre; f/− embryos. The significance was determined by an unpaired, 2-tailed Student t test. In all panels **P < .01; ***P < .001; ****P < .0001.