Le compartiment stromal du foie fœtal et l'hématopoïèse

During ontogeny, hematopoiesis occurs in sequential waves across distinct anatomical sites. The first wave arises in the blood islands of the yolk sac (YS) around E7.25. Primitive erythroid progenitor cells appear and give rise to erythrocytes, macrophages, and megakaryocytes. The second wave starts around E8.25. Erythro-myeloid progenitors (EMPs) generated in the vascular plexus of the YS through an endothelial-to-hematopoietic transition mechanism, giving rise to definitive erythroid progenitors, bipotent macrophage-granulocyte progenitors, mast cells, and megakaryocytes. Around E10.5, multipotent hematopoietic stem cells (HSCs) appear in the dorsal aorta, vitelline, and umbilical arteries, capable of producing all blood cell types. They subsequently colonize the fetal liver (FL) and undergo expansion and differentiation. Finally, they colonize the bone marrow (BM) where they remain throughout adult life. Unlike the BM, embryonic liver provides a unique microenvironment that supports robust HSC expansion, whereas postnatal BM maintains HSC quiescence. Understanding the FL niche is thus critical for deciphering the mechanisms governing HSC self-renewal versus differentiation—a key challenge in improving HSC transplantation therapies for hematological disorders. Despite the clinical importance of HSCs, their limited availability in cord blood and adult sources restricts therapeutic applications. While most niche studies focus on the BM, elucidating FL stromal components is essential for uncovering the pathogenesis of pediatric hematological disorders and optimizing in vitro HSC expansion. However, the stromal composition of the FL and its role in HSC regulation remain poorly defined. To understand the cues supporting hematopoietic stem and progenitor cell (HSPCs) and regulators mediating HSPC expansion versus differentiation during embryonic development, we identified and characterized the most relevant non-hematopoietic stromal components in the FL at different gestational stages. To emulate in vivo FL microenvironments, we optimized 2D and 3D culture systems of freshly isolated primary FL stromal cells. We developed an in vitro 3D model composed of three stromal subsets (hepatoblasts, mesenchymal, and mesothelial cells) that expand within 4 days in culture, maintaining structural and cytokine profiles comparable to in vivo FL stromal components. Notably, co-culture of the FL stromal populations with HSPCs in the absence of exogenous cytokines induced a 4-fold expansion of multipotent progenitors within four days. In vitro clonal multilineage potential assays showed that cultured HSPCs after expansion in the spheroids have similar potential to proliferate and differentiate into multilineage myeloidlymphoid, myeloid, and lymphoid cell clones relative to freshly isolated FL HSPCs. This 3D model further supported hematopoietic differentiation into different cell lineages, including Gr-1+ myeloid, B lymphoid cells, and Ter119+ erythrocytes. Around 8-fold increase in total hematopoietic cell output was observed within four days. Interestingly, hepatoblasts were shown to be the cell type that produces the most hematopoietic cytokines in the FL environment. They are the only cells that produce interleukin-7, which is essential for lymphoid development, and erythropoietin, which is involved in erythrocyte production. Strikingly, hepatoblasts alone could facilitate HSPCs' expansion and differentiation as effectively as the three stromal compartments. Hepatoblasts were also the main cell type secreting kit ligand, a major hematopoietic cytokine. Inhibition of the SCF/c-kit pathway with the ACK2 antibody in vitro significantly reduced HSPCs' expansion. Therefore, Kit ligand appears to play an important role in the expansion of HSPCs. (...)