Effect of BCR-ABL expression level on IM sensitivity in primary bone marrow cells. (A) BCR-ABL GFP-transduced murine BMMCs were analyzed by FACS for GFP staining intensity and sorted into GFP-high (BM-BA^high) and GFP-low (BM-BA^low) fractions using gates as indicated (left); transduced cells were cultured for 72 hours after sorting, and total protein was extracted. Western blot analysis was performed using anti-ABL antibodies. The membrane was reprobed using an anti-actin antibody as a loading control (right). (B) Left: empty vector GFP (Mig-v)–expressing BMMCs were sorted according to the indicated gates into high and low GFP-expressing populations and separately treated with 3μM IM for 48 hours. Apoptosis was assessed using propidium iodide staining as indicated. A representative flow cytometry plot of 2 independent experiments is shown. SSC indicates side scatter; FSC, forward scatter. Right: propidium iodide staining to measure IM-induced apoptosis by flow cytometry in BM-BA^high– and BM-BA^low–expressing BMMCs. Apoptosis is displayed relative to untreated controls after treatment for 48 hours with IM at 1 or 3μM. Columns and error bars represent means ± SEM (*P < .05; ***P < .0001). (C) We sorted and seeded 1000 BM-BA^high and 1000 BM-BA^low cells, respectively, in 1 mL of semisolid methylcellulose medium with and without 3μM imatinib, and after 14 days colonies were counted. The data were obtained from 3 independent experiments; columns represent means ± SEM (***P < .001 according to Mann–Whitney t test). (D) Sorting strategy of bone marrow cells transduced with BCR-ABL GFP to BA^high/lowlin⁻sca1⁺c-kit⁺ (LSK) and lin⁻c-kit⁺ (LK) populations (i). Apoptosis assessment in BA^high/low LK and LSK after treatment with IM at 3μM for 48 hours using propidium iodide staining and flow cytometry (ii). CFU assay as in panel C (iii) using sorted BA^high/low LK and LSK (***P = .008 according to Mann-Whitney t test; n.s., P > .05).