(A) BL/6 LSCs were cultured in the presence and absence of BM CD8⁺ CTLs from CML-bearing BL/6 overnight at a ratio of 1:1 in triplicate followed by plating in methylcellulose. Colonies were enumerated 7 days later. For secondary platings, cells isolated from primary colony assays were replated in methylcellulose in the absence of T cells; t test (Groups: LSCs, n = 3 mice; LSCs + CML CD8, n = 4). (B) BL/6 LSCs were cultured overnight in the presence and absence of CD8⁺ CTLs derived from the BM of perforin-proficient and -deficient CML mice at a ratio of 1:1 in triplicate followed by plating in methylcellulose. Colonies were enumerated 7 days later; 1-way ANOVA followed by Tukey’s multiple comparison (Groups: LSCs, n = 3 mice; LSCs + BL/6 CML CD8, n = 6; LSCs + Prf^–/– CML CD8, n = 4). (C–L) BCR-ABL1-GFP–transduced LSKs were injected i.v. into nonirradiated BL/6 (n = 7) and Prf^–/– (n = 8) recipient mice. (D–G) Spleen weight; t test (BL/6 CML, n = 7 mice; Prf^–/– CML, n = 8 mice), absolute numbers of L-splenocytes, of L-lin^–, and LSCs in the BM of CML mice; t test (BL/6 CML, n = 7 mice; Prf^–/– CML, n = 8 mice) 15 days after CML induction. (H) Gating strategy to define LSC subpopulations; cells are pregated on lin^–GFP⁺Sca-1⁺c-kit⁺ cells. Representative images from 1 of n = 7 (BL/6) and 1 of n = 8 Prf^–/– CML mice are shown. (I–L) Absolute numbers of LSC subpopulations; t tests (BL/6 CML, n = 7 mice; Prf^–/– CML, n = 8 mice). (M) BM cells from primary BL/6 (n = 7) and Prf^–/– (n = 8) CML mice were injected i.v. into lethally irradiated secondary BL/6 recipients, and survival was monitored; log-rank test. Data are displayed as mean ± SEM. *P < 0.05, **P < 0.01 and ***P < 0.001.