Contribution of basophil-like leukemia cells to maintenance of LICs in BM and CML development. (A) MCPT8-DTR donor-derived LICs were subjected to the CML model. DT (400 ng/mouse) was intravenously injected at the indicated time points. As a control, the same dose of DT (Glu52) was injected in a similar manner. The numbers of WBCs in PB (B), SP weights (C), and the numbers of GFP⁺KLS⁺ LICs in the tibial BM (D) were determined. Each symbol represents an individual mouse (n = 6). **P < .01 by the Mann-Whitney U test. (E) DT or DT (Glu52) was injected at the indicated time points. Starting 2 weeks after BM transplantation, the numbers of WBCs were determined weekly. Leukemia-free survival rates were determined (n = 12). Mice that died of leukemia or those in which the WBC count reached 15 000 cells per μL were diagnosed with leukemia. **P < .01 by the log-rank test. (F) At 20 days after transplantation of MCPT8-DTR donor-derived LICs, GFP⁺lineage⁻c-kit⁺ CML progenitor cells were isolated from the BM of primary CML mice treated with DT or DT (Glu52). Forty thousand CML progenitor cells were mixed with 10⁶ total BM cells and intravenously injected into sublethally irradiated secondary recipient mice. From 10 days after transplantation, the numbers of WBCs were determined every 5 days. Leukemia-free survival rates were determined (DT, n = 7; DT (Glu52), n = 10). N.S., no significant difference by the log-rank test.