Lineage tracing studies suggest that the placenta is not a de novo source of hematopoietic stem cells

Definitive hematopoietic stem and progenitor cells (HSPCs) arise from a small number of hemogenic endothelial cells (HECs) within the developing embryo. Understanding the origin and ontogeny of HSPCs is of considerable interest and potential therapeutic value. It has been proposed that the murine placenta contains HECs that differentiate into HSPCs. However, during human gestation HSPCs arise in the aorta considerably earlier than when they can first be detected in the placenta, suggesting that the placenta may primarily serve as a niche. We found that the Runx1 transcription factor, which is required to generate HSPCs from HECs, is not expressed by mouse placental ECs. To definitively determine whether the mouse placenta is a site of HSPC emergence, we performed lineage tracing experiments with a Hoxa13Cre allele that specifically labels ECs in the placenta and umbilical cord (UC), but not in the yolk sac or embryo. Immunostaining revealed Hoxa13Cre lineage-traced HECs and HSPCs in the UC, a known site of HECs, but not the placenta. Consistent with these findings, ECs harvested from the E10.5 aorta and UC, but not the placenta, gave rise to hematopoietic cells ex vivo, while colony forming assays using E14.5 fetal liver revealed only 2% of HSPCs arose from Hoxa13-expressing precursors. In contrast, the pan-EC Cdh5-CreERT2 allele labeled most HSPCs in the mouse placenta. Lastly, we found that RUNX1 and other HEC genes were not expressed in first-trimester human placenta villous ECs, suggesting that human placenta is not hemogenic. Our findings demonstrate that the placenta functions as a site for expansion of HSPCs that arise within the embryo proper and is not a primary site of HSPC emergence.