Abstract 910

In the mammalian embryo, hematopoietic stem cells (HSC) emerge from vascular beds and colonize the fetal liver. The first HSC are found in the murine fetal liver by embryonic day 12.5 (E12.5). However blood function is required before HSC have formed, and two earlier waves of hematopoietic potential arise to sustain the embryo. The first wave of hematopoietic progenitors are formed in the yolk sac between E7.25 and E8.5 and are termed “primitive” because, along with megakaryocyte and macrophage potential, they differentiate into primitive erythroid cells that mature in the circulation and express embryonic globins. A second lineage of hematopoietic potential has been characterized in the murine and human yolk sac as well as in zebrafish. These cells have been termed “EMP”, erythro-myeloid progenitors, that generate definitive erythroid and myeloid lineages, including granulocytes. We find that EMP emerging in the mouse embryo express many of the markers associated with HSC emergence from hemogenic endothelium. At early stages of their emergence (E8.5), EMP express not only kit and CD41, but also VE-cadherin, CD31 and CD34. A day later, EMP constitute a robust population of over 1,000 cells in the yolk sac and display diminished expression of VE-cadherin and increased expression of CD45. However, unlike HSC, EMP do not express Sca1, but do express the myeloid progenitor marker CD16/32 (low affinity FCgamma receptor II/III). Like CD45, CD16/32 is expressed on a subset of the CD41+/kit+ cells at E8.5. Colony forming assays confirm that EMP potential is found in both CD16/32 positive and negative CD41+/kit+ cells. By E9.5, over 90% of CD41+/kit+ cells also express CD16/32 and all of the hematopoietic colony-forming potential at E9.5 is found in this triple-positive population. The transition from endothelial-associated to hematopoietic-associated genes suggests that EMP may emerge from hemogenic endothelial intermediates. Between E9.5 and E11.5, cells with the EMP immunophenotype are found in the bloodstream and become concentrated in the liver, where evidence of very robust erythro-myeloid differentiation precedes HSC colonization. In culture, EMP rapidly expand, dividing twice daily, and within 6 days generate predominately erythroid cells as well as smaller numbers of megakaryocyte, macrophage and mast cells, but only rare granulocytes. This is in contrast to lin-/kit+ ScaI- bone marrow progenitors grown in the same culture conditions that generate predominately myeloid cells, particularly granulocytes, and rarely mast cells. In agreement with data in the zebrafish, we also do not see evidence of lymphoid potential or RAG2 expression in EMP cultures or expression of the lymphoid markers Flt3 and IL7 receptor on the cell surface of EMP. In order to better understand the lineage potential of EMP, we examined the expression of known transcriptional regulators of bone marrow hematopoiesis. In the adult, relative levels of GATA1 versus Pu.1 are proposed to determine fate between megakaryocyte/erythroid progenitors (MEP- GATA1 hi) and granulocyte/macrophage progenitors (GMP-Pu.1 hi). Consistent with their erythro-myeloid potential, EMP expressed both GATA1 and Pu.1 at intermediate levels compared to adult marrow-derived MEP and GMP. In the adult, Gfi-1 and C/EBPalpha are both proposed to upregulate granulocyte versus macrophage differentiation. We found lower levels of these regulators in EMP compared to GMP, consistent with EMP cultures generating small numbers of granulocytes versus macrophages. In addition, the GATA1 expression within the Pu.1+ GMP is found to increase mast cell potential and, thus, the high GATA1 and Pu.1 expression in EMP may account for their high mast cell potential. Taken together, these data suggest that, like HSC, EMP emerge from hemogenic endothelium and their erythro-myeloid potential is governed by the action of shared regulatory networks. However, the transcription factors and markers are present in the EMP in unique combinations consistent with their specific role in providing a transient initial wave of definitive hematopoiesis in the embryo.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution