Functional Hematopoietic Cells Derived From Mouse Embryonic Stem Cells.

Abstract 2304

Despite two decades of effort, deriving long-term repopulating hematopoietic stem/progenitor cells (HSPCs) from embryonic stem cells (ESCs) has proven to be extremely difficult. Both embryoid body (EB)-based and stroma-based methods have been extensively explored. However, robust production of HSPCs from C57BL/6J-derived mouse ESCs (mESCs) has not yet been reported. Furthermore, in vivo engraftment of mES-derived HSCs (from any strain) has been achieved only with forced expression of HoxB4 or related oncogenes, which creates significant limitations for most studies. Here, we describe a stroma-based co-culture method to differentiate HSCs and progenitor populations from C57BL/6J-derived mESCs. After simple co-culture on OP9 stroma cells for one week, C57BL/6J-derived mESC lines differentiate into cells that mark as HSCs, CMPs, GMPs, and MEPs (by immunophenotyping); these cells are capable of giving rise to erythrocytes, monocytes, and mast cells (by morphology and immunophenotyping) after another week of culture in methylcellulose with hematopoietic cytokines (SCF, IL-3, IL-6, and Epo). Similar in vitro hematopoietic differentiation has been achieved in several different C57BL/6J-derived mESCs (B6/Blu, B6-GFP, LK1, and B6 albino), B6/SVJ129 mESCs (R1), various SVJ129-derived mESCs (SWT16, EDJ22, and SCC10), and five independent C57BL/6J mouse embryonic fibroblast (MEF)-derived induced pluripotent stem cell (iPSC) lines. C57BL/6J ESCs derived from CAGGS-GFP transgenic mice (B6-GFP ESCs, which express high levels of GFP in all hematopoietic lineages) were used to determine whether we could obtain long-term engraftment of the OP9 differentiated ESCs. B6-GFP ESCs cultured for 7 days on OP9 cells were sorted by Kit⁺ surface staining. Sorted cells (1×10⁵, 2×10⁵, 4×10⁵) were transferred into immunocompromised NSG mice via retro orbital injection (n=1 mouse per dose). Peripheral blood from the recipients injected with 2×10⁵ and 4×10⁵ cells showed 5% GFP positivity in the peripheral blood at weeks 12 and 16 post-transplant, while recipients injected with 1×10⁵ cells had no detectable GFP+ cells in the periphery. Bone marrow cells and spleens were harvested at week 22. The recipient injected with 4×10⁵ cells showed 5% GFP positivity in the bone marrow and 20% in the spleen. Engraftment was multi-lineage. Myeloid compartments (CD34⁺, CD11b⁺, Kit⁺, and Gr-1⁺) showed similar or less GFP positivity than overall bone marrow and spleen cells. Lymphoid (CD3⁺ and B220⁺) and erythroid (Ter119⁺) compartments also showed similar GFP positivity compared to overall bone marrow cells. However, lymphoid and erythroid compartments contained significantly higher GFP positivity (30–60%) than overall spleen cells. We have now modified the procedure to transfer 1×10⁶ unfractionated B6-GFP ESCs grown for 7 days on OP9 stroma directly into NSG recipients. We have detected short-term engraftment 4 weeks post-injection in the peripheral blood of one recipient and multilineage splenic engraftment 8 weeks post-injection in two recipients, confirming that short-term repopulating cells are indeed generated by this method. Secondary transplants using the GFP⁺ bone marrow cells from the long-term engrafted mouse have been performed. This approach could be a valuable tool for studying the hematopoietic development of a variety of mESC lines, and potentially, iPSC lines as well.