Cancellation of c-MYC Silencing in Human Induced Pluripotent Stem Cells Contributes to the Efficient in Vitro Production of Platelets with the Ability of Hemostasis In Vivo.

Abstract 1488

Poster Board I-511

Human induced pluripotent stem cells (hiPSCs) generated from somatic cells by introduction of OCT3/4, SOX2, KLF4 and c-MYC represent a potential source of hematopoietic cells for transfusion without the risk of immune rejection. We recently established an in vitro culture system with which hiPSCs could be differentiated into the unique structure of an “in vitro hematopoietic niche” containing hematopoietic progenitors. Upon further cultivation under appropriate conditions, these hematopoietic progenitor cells differentiated into megakaryocytes, which could then generate platelets with morphologies indistinguishable from peripheral blood platelets regardless of either 4-factor iPSCs (n=8, 4-factor hiPSC clones generated from adult dermal fibroblasts through induction with c-MYC) or 3-factor iPSC clones (without c-MYC, n=3). It is well known that iPSC differentiation yields a heterogeneous population of clones. To select the best hiPSC clone for platelet production, we quantified thrombopoiesis with 11 independent hiPSC clones by comparison with human embryonic stem cells (hESCs) evaluated previously (Takayama et al., Blood, 2008) as a reference. Particularly noteworthy is our finding that 4-factor iPSCs have an advantage over 3-factor iPSCs or hESCs (P<0.01) that is mediated through cancellation of c-MYC silencing (re-activation) over the course of differentiation evidenced by RT-PCR studies. Indeed, ectopic expression of c-MYC, but not OCT3/4, SOX2 or KLF4, using a retroviral vector in hESC-derived progenitors accelerated both megakaryopoiesis and thrombopoiesis. By contrast, the platelet activation statuses (i.e., PAC-1 ligation with activated integrin αIIbβ3 following agonist stimulation) were comparable for platelets obtained from 4-factor hiPSCs and hESCs, though levels of c-MYC clearly differed, indicating that at least integrin activation is independent of c-MYC. To further estimate the in vivo functionality of iPSC-derived platelets, we developed a mouse model for transfusion. Irradiation (2.0 Gy, 9 days beforehand) induced thrombocytopenia in NOG (nod-scid/IL-2 γc-null) mice. Subsequent flow cytometry showed that 2 hrs after transfusion (1.0∼1.2×107 platelets per a mouse) of NOG mice via the tail vein, the circulating levels of selected 4-factor iPSC-derived platelets were similar to those of human adult platelets (platelet chimerism of human CD41/mouse CD41; 4∼10%). Moreover, by using our recently established in vivo imaging system, which enables observation of single platelet behavior, we observed that 4-factor iPSC-derived platelets circulate in NOG mice and contribute to the development thrombi within their vessels, suggesting the in vivo functionality of iPSC-derived platelets is intact. A number of studies have suggested that c-MYC can have deleterious effects leading to in vivo oncogeneity after transplantation in vivo. By contrast, our data strongly indicate the importance of c-MYC for platelet generation from hiPSCs and hESCs. Given that anucleate platelets are routinely irradiated before transfusion in clinical settings, use of c-MYC for hiPSCs generation may contribute to the efficient production of HLA-matched platelet concentrates for those requiring repeated transfusion.