Abstract
Abstract 964
Chronic myelogenous leukemia (CML) is a hematopoietic stem cell (HSC) disease caused by BCR-ABL oncogene, and now newly targeted therapies are warrented for CML due to the minimal effect of BCR-ABL-targeted tyrosine kinase inhibitors toward CML stem cells. As CML stem cells are known to show some similarity with HSC, utilizing HSC-specific factors as a guide for analyzing CML stem cells is of great significance. Since Evi1 is a transcription factor which is highly expressed within normal HSC compartment and it is frequently activated in myeloid malignancies including a blastic phase (blast crisis) of CML (CML-BC), it is supposed that CML stem cells could have a close relation with Evi1. Here in this study, with Evi1-GFP knock-in mice, which we have recently generated, we developed murine models of CML in a chronic phase (CML-CP) and CML-BC for uncovering new properties of CML stem cells. In Evi1-GFP knock-in CML-CP model, we found that Evi1 positive CML cells account for about 0.1–0.5% of total bone marrow (BM) cells and that almost all of them showed no lineage markers. Furthermore, Evi1 is predominantly expressed in the CML stem cell fraction (Lin- Sca-1+ c-kit+ (LSK)), but its expression is sharply downregulated even in myeloid progenitor (Lin- Sca1- c-kit+ (MP)) cells and in more differentiated cells. Even within CML LSK cells, Evi1 expression widely varies and Evi1-high LSK cells show an enhanced colony-forming capacity compared with Evi1-low LSK cells. As for cell cycle status, Evi1-high CML LSK cells are mostly in G0/G1 phase although Evi1-low CML LSK cells or CML myeloid progenitor are more in S/G2/M phase. When CML LSK cells are cocultured with OP-9 stromal cells, only Evi1-high LSK cells could made cobblestone areas. Comparison of Evi1-high cells with Evi1-low cells in normal and CML LSK compartments by gene expression profiles showed that a more quiescent feature and a less differentiated feature in Evi-high CML LSK cells than in Evi1-low CML LSK cells. Moreover, Evi1-high CML LSK cells have a close correlation with TGF-beta signaling and calcium signaling. In addition, Evi1-high normal LSK cells had the most quiescent and the least differentiated profiles, which suggested that Evi1-high CML LSK cells could keep self-renewal capacity with high proliferation capacity. In concert with our data of Evi1-trafficking CML mouse, in CML patients, we have also recently found that CD34+ 38- CML stem cells showed higher EVI1 expression than CD34+ 38+ CML progenitor cells or total CML cells, which implies that EVI1 could mark CML stem cells as well as normal HSCs. These data indicate that in our Evi1 trafficking CML model high Evi1 expression could enrich CML stem cells and that Evi1 could have a crucial role in CML pathogenesis. In Evi1-GFP knock-in CML-BC model, which more differentiated myeloid progenitors are likely to have a high leukemia initiating potential, a sizable fraction of MP leukemic cells show distinct Evi1 expression. Remarkably, in vivo transplantation assay revealed CML-BC stem cells that can recapitulate the disease are exclusively enriched in Evi1-high MP fraction. Evi1-high MP cells showed a replating capacity in colony assay while Evi1-low MP cells could not. Moreover, Evi1-high MP cells are more actively cycling than Evi1-low MP cells. Our data revealed a limited fraction with high Evi1 expression within stem/progenitor cells possesses enhanced proliferative and leukemia-initiating capacities in CML. As opposed to these CML models noted above, in Evi1-GFP knock-in AML model by MLL-ENL, Evi1-high leukemic cells showed no advantage in leukemia initiating potential. Additionally, other Evi1-GFP knock-in AML models by MOZ-TIF2 and TEL-PDGFRb/AML1-ETO never showed Evi1-high fraction both in BM and spleen, which might suggest the high affinity of Evi1 with stem cell disease as CML. The current study provides us with a new tool for dissecting pathogenesis and exploiting novel targeted therapies to eradicate CML stem cells. An establishment of Evi1-related therapy for CML stem cells, which could be applied to EVI1-high malignancies, is currently being explored.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal