The Polycomb group (PcG) proteins form multiprotein complexes that play an important role in the maintenance of transcriptional repression of target genes. Loss-of-function analyses show abnormal hematopoiesis in mice deficient for PcG genes including Bmi-1, Mph-1/Rae28, M33, Mel-18, and Eed, suggesting involvement of PcG complexes in the regulation of hematopoiesis. Among them, Bmi-1 has been implicated in the maintenance of hematopoietic and leukemic stem cells. In this study, detailed RT-PCR analysis of mouse hematopoietic cells revealed that all PcG genes encoding components of the Bmi-1-containing complex, such as Bmi-1, Mph1/Rae28, M33, and Mel-18 were highly expressed in CD34−c-Kit+Sca-1+Lin− (CD34−KSL) hematopoietic stem cells (HSCs) and down-regulated during differentiation in the bone marrow. These expression profiles support the idea of positive regulation of HSC self-renewal by the Bmi-1-containing complex. To better understand the role of each component of the PcG complex in HSC and the impact of forced expression of PcG genes on HSC self-renewal, we performed retroviral transduction of Bmi1, Mph1/Rae28, or M33 in HSCs followed by ex vivo culture. After 14-day culture, Bmi-1-transduced but not Mph1/Rae28-transduced cells contained numerous high proliferative potential-colony forming cells (HPP-CFCs), and presented an 80-fold expansion of colony-forming unit-neutrophil/macrophage/Erythroblast/Megakaryocyte (CFU-nmEM) compared to freshly isolated CD34−KSL cells. This effect of Bmi-1 was comparable to that of HoxB4, a well-known HSC activator. In contrast, forced expression of M33 reduced proliferative activity and caused accelerated differentiation into macrophages, leaving no HPP-CFCs after 14 days of ex vivo culture. To determine the mechanism that leads to the drastic expansion of CFU-nmEM, we employed a paired daughter cell assay to see if overexpression of Bmi-1 promotes symmetric HSC division in vitro. Forced expression of Bmi-1 significantly promoted symmetrical cell division of daughter cells, suggesting that Bmi-1 contributes to CFU-nmEM expansion by promoting self-renewal of HSCs. Furthermore, we performed competitive repopulation assays using transduced HSCs cultured ex vivo for 10 days. After 3 months, Bmi-1-transduced HSCs manifested a 35-fold higher repopulation unit (RU) compared with GFP controls and retained full differentiation capacity along myeloid and lymphoid lineages. As expected from in vitro data, HSCs transduced with M33 did not contribute to repopulation at all. In ex vivo culture, expression of both p16INK4a and p19ARF were up-regulated. p16INK4aand p19ARF are known target genes negatively regulated by Bmi-1, and were completely repressed by transducing HSCs with Bmi-1. Therefore, we next examined the involvement of p19ARF in HSC regulation by Bmi-1 using p19ARF-deficient and Bmi-1 and p19ARF-doubly deficient mice. Although bone marrow repopulating activity of p19ARF-deficient HSCs was comparable to that of wild type HSCs, loss of p19ARF expression partially rescued the defective hematopoietic phenotypes of Bmi-1-deficient mice. In addition, transduction of Bmi-1 into p19ARF-deficient HSCs again enhanced repopulating capacity compared with p19ARF-deficient GFP control cells, indicating the existence of additional targets for Bmi-1 in HSCs. Our findings suggest that the level of Bmi-1 is a critical determinant for self-renewal of HSC and demonstrate that Bmi-1 is a novel target for therapeutic manipulation of HSCs.
Read full abstract