Unlimited self renewal capacity and the ability to differentiate into any cell type make human pluripotent stem cells (PSC) a potential source for the ex vivo manufacture of red blood cells (RBC) for safe transfusion. Current methods of RBC differentiation from PSC suffer from low yields of RBCs, most of which contain embryonic and fetal rather than adult hemoglobins. We have previously shown that dimerization of the intracellular component of MPL (the thrombopoietin receptor), induces expansion of myelo-erythroid progenitors (MEP) from human cord blood as well as their terminal differentiation into enucleated RBC through unique, EPO-independent mechanisms (Parekh et al, 2012). Our goal was to investigate the potential of intracellular MPL dimerization to induce erythropoiesis from human PSC and to identify the signaling pathways activated by this strategy.


Human embryonic stem cell (hESC) lines H1 and HES3 were transduced with a lentiviral vector to express the fusion protein F36V-MPL (containing the ligand binding domain F36V and the intracytoplasmic portion of MPL). Dimerization of F36V-MPL was accomplished by addition of the synthetic ligand AP20187 (aka CID) during culture (with or without erythropoietin) on OP9 stroma in the absence of other cytokines. F36V-MPL transduced-hESC that did not receive CID and F36V-transduced hESC cultured with CID served as negative controls. Flow cytometry and Colony Forming Unit (CFU) assays were used to analyze erythroid differentiation. Phosflow and Western Blot were used to analyze cell signaling. MEP generated during hESC differentiation were defined as cells co-expressing GlyA and CD41a/CD42a.


F36V-MPL dimerization induced significantly more Glycophorin A+ cells (P=0.0001; n=5) and 10-fold higher number of erythroid CFU (P=0.0007; n=15) as compared to negative controls. The effect was consistent across different hESC cell lines. This effect was seen in the absence of any hematopoietic cytokines, including erythropoietin (EPO), a critical cytokine for erythropoiesis and an integral component of all ex vivo PSC erythroid differentiation protocols, indicating that MPL dimerization alone is sufficient to induce erythropoiesis from hESCs. Erythroid output was further enhanced in an additive manner in the presence of EPO (P=0.006; n=5). In order to identify the point at which MPL dimerization affects erythropoiesis, CID was added during differentiation directly from hESC or to isolated MEP generated from hESC. CID and EPO increased the number of MEP compared to untreated controls, demonstrating that MPL dimerization induces the generation of early erythroid progenitors. In addition, CID drove erythroid differentiation from MEP more efficiently than EPO, demonstrated by a significantly higher frequency of total erythroid cells (P=0.02; n=3), and 4-fold increase in yield of enucleated RBC. Globin analysis by HPLC demonstrated that although no detectable beta-globin expression was observed with EPO, CID treatment induced the presence of beta-globin and increased the gamma: epsilon globin ratio, suggesting a shift toward definitive erythropoiesis. Signaling studies found that, unlike the full-length MPL receptor, which activates both STAT5/JAK2 and AKT pathways, F36V-MPL dimerization activated AKT but not STAT5 or JAK2 phosphorylation. PI3K/AKT inhibitors (LY294002 and AKT inhibitor IV) effectively inhibited erythroid differentiation of transduced hESC cultured in the presence of CID (P=0.001; n=4) indicating that MPL dimerization induced erythropoiesis is dependent on AKT signaling. Gene expression analysis by qPCR indicated that MPL dimerization upregulates a network of genes (downstream of AKT signaling) associated with the regulation of cell cycle, apoptosis, and erythroid differentiation, including GATA1, CDKN1A, RB1, VEGFA, and BCL-xL with a corresponding reduction in both apoptosis and cell cycle progression.


In summary, we have identified a novel EPO-independent approach that is not only more efficient at erythropoiesis but is also able to augment EPO induced erythropoiesis. The mechanistic insights gained from this study opens up potentially new approaches toward the generation of therapeutically relevant quantities of RBCs for transfusion.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.