Robust Autonomous Generation of Definitive Hematopoietic Stem Cells from Pre-Circulation Murine Embryos

Introduction:

The ontogeny of mammalian hematopoiesis is organized in at least three embryonic hematopoietic waves. The first wave is characterized by the generation of primitive erythrocytes emerging from the yolk sac (YS); a second one produces bipotent erythro-myeloid progenitors also in the YS. Finally, multipotent definitive Hematopoietic Stem and Progenitor Cells (HSPCs) with the ability to reconstitute all mature adult blood lineages and sustain life long hematopoiesis emerge from a subset of endothelial cells named hemogenic endothelium (HE). The embryonic anatomical site of origin of these HSPCs is still under debate in the field. An intra-embryonic origin is well accepted, which primarily encompasses the Aorta-Gonad-Mesonephros region (AGM) and umbilical and vitelline arteries. In contrast, a YS contribution to adult hematopoiesis remains controversial. The same developmental origin ( e.g. endothelial cells with a mesodermal ancestor) makes it difficult to identify specific markers to discern between an AGM- vs. YS-origin using a lineage trace approach. This together with the highly migratory nature of blood cells and the inability of cells isolated from pre-circulatory embryos to robustly engraft in transplantation assays, even after ex vivo culture, has precluded scientists from properly answering these questions. For example, cultured YS and para-aortic splanchnopleura (PSP, the precursor of the AGM) dissected from embryos prior to the onset of circulation resulted in the generation of HSPCs with multilineage engraftment potential only from the embryonic PSP and not the YS (Cumano A. et al., Immunity 2002). However, blood chimerism was low (<5%) and engraftment was achieved only after transplantation into severely immuno-compromised mice ( i.e. Rag2gc^-/-). This low engraftment baseline makes difficult any comparison between intraembryonic tissues (IET) and Yolk Sac because a smaller frequency of YS-derived transplantable HSPCs would be difficult to detect, yielding a false negative result.

Thus, here we sought to interrogate if we could generate robust engraftment from pre-circulation embryos that would allow to rigorously compare the ability to derive definitive HSC in the IET and YS.

Results:

We have established ex vivo culture conditions to allow for the derivation of robust multi-lineage engraftment activity from pre-circulation murine embryos. CD45.2+ embryos at the 2-3 somite pairs (s.p.) stage and 5-7 s.p. embryos were dissected from their YS, cultured as explants in our optimized conditions and transplanted into lethally irradiated CD45.1+/CD45.2+ recipients. Here, we observed multilineage reconstitution of the peripheral blood (PB) for >16 weeks post-transplant in 5/7 and 9/11 recipients, respectively. Moreover these cells were serially transplantable, confirming the generation of self-renewing HSCs. In primary recipients, PB chimerism was >50% in 8/18 total recipients. This robust PB chimerism allowed us to rigorously interrogate the corresponding CD45.2+ YS (cultured under the same conditions). In these transplants, we observed PB engraftment in 0/7 and 0/11 recipients of cells isolated from 2-3 s.p. and 5-7 s.p. cultured YSs, respectively. These data suggest that either YS do not autonomously generate HSPCs or that they can not support their maturation ex vivo . Moreover we found that the engraftment ability from the IET is confined to an emerging CD31⁺CD45⁺c-Kit⁺CD41^- population. Importantly, this population also develops in the YS explants, but lacks engraftment ability. We have repeated these studies with cultured 19-22 s.p. embryos and YSs, a post-circulation time point and still observed a failure of YS-derived cells to engraft irradiated recipients (0/9) while embryo-derived cells (9/9) engraft robustly. These data suggest that YS likely cannot support the maturation of HSPCs that might have migrated from other sites. Only when we transplanted 32-36 s.p. or 46-47 s.p. YS-derived cells we observed robust PB engraftment under our culture conditions (3/7 and 4/6 recipients engrafted, respectively).

Conclusions:

We observed robust generation of HSCs with long-term and serially transplantable multilineage potential from pre-circulation murine embryos. Our study demonstrates that the YS is not able to support the maturation of HSPCs. Further studies would be required to determine if the YS is an actual source of HSPCs.