Fig. 4: CD24^pos YS HE has lymphoid-myeloid potential and LYVE1^pos YS HE has erythroid-myeloid potential.

a, UMAPs depicting the expression of Cd24a (marking the AGM and YS-A clusters) and Mcam (marking cells toward the endothelial end of all three trajectories). b, Correlation between Cd24a and Mcam, transcript expression in clusters mix, a1 and a2 (YS-A trajectory) (top). Correlation between Lyve1 and Mcam, transcript expression in clusters b1, b2 and b3 (YS-B trajectory) (bottom). c, Flow cytometry on extra-embryonic CD45^negCD41^negTER119^neg (Lineage negative) CD31^posKIT^pos cells from Runx1:RFP reporter embryos. MCAM, LYVE1 and CD24 antibodies were used to analyze the proportion of Runx1 (RFP) expressing cells in different subpopulations. Each dot represents cells from a single YS. E9.5 MCAM^posCD24^posLYVE1^neg n = 7, E10.5 MCAM^negCD24^negLYVE1^pos n = 6, all other samples n = 8. Bars represent the average ± s.e.m. d, Heatmap displaying the distribution (as percentage) of different CD45^negCD41^negTER119^negCD31^posKIT^pos FACS-sorted, scRNA-sequenced cell populations across the in silico EHT clusters defined in Fig. 2b. Based on k-nearest-neighbor classifier approach. MCAM, LYVE1, CD24 sorting profiles are depicted on the x axis. Purple boxes indicate the expected/predicted cluster (y axis) for the sorted population (x axis) based on the data presented in Fig. 4b. e, Single-cell hematopoietic assays of YS-A HE (KIT^posCD31^posLIN^negLYVE1^negCD24^posMCAM^neg) and YS-B HE (KIT^posCD31^posLIN^negLYVE1^posCD24^negMCAM^neg) cultured on OP9 feeder cells for 14 days. The percentage of wells with proliferating hematopoietic cells is shown. f, Lineage distribution of the hematopoietic cells shown in e as determined by flow cytometry for myeloid (GR1 and MAC1/CD11b), erythroid (TER119) and lymphoid (CD19) markers. g, Violin plots depicting LMP (top) and EMP (bottom) scores across early progenitor clusters p1 and p2. EMP and LM signatures have been previously published and are listed in Supplementary Table 1. Embedded boxplots indicate the median (horizontal line), the upper and lower hinges represent the 75th and 25th percentile and whiskers extend to 1.5 × interquartile range. Two-sided Wilcoxon rank-sum tests were used (with P values adjusted via the Benjamini–Hochberg procedure to control the FDR) to compare EMP and LMP as well as clusters p1 and p2. h, Violin plots depicting prospective LMP fate (top) and EMP fate (bottom) scores across early progenitor clusters p1 and p2 as well as EMP and LMP populations. EMP fate (8 genes) and LMP fate (14 genes) signatures (Supplementary Table 1) were extracted by intersecting pairwise differential gene expression results (EMP versus LMP and P1 versus P2; Extended Data Fig. 3f). Embedded boxplots indicate the median (horizontal line), the upper and lower hinges represent the 75th and 25th percentile and whiskers extend to 1.5 × interquartile range. Two-sided Wilcoxon rank-sum tests were used (with P values adjusted via the Benjamini–Hochberg procedure to control the FDR) to compare EMPs and LMPs as well as clusters p1 and p2.

Source Data