Prolonged survival and decreased CML in mice reconstituted with BM cells coexpressing P210 and DN STAT5. (A) Schematic depicting a bicistronic retroviral vector used to co-express DN STAT5 and Bcr/Abl in primary murine BM cells. (B) Retroviral supernatants were titered on NIH 3T3 fibroblast cells by anti-Abl and antihemagglutinin (HA) Western immunoblot (recognizing epitope-tagged STAT5 mutants). The positions of the STAT5 mutants were confirmed by anti-STAT5A/5B immunoblot (bottom panel). The carboxy-terminal deletion mutant DN STAT5A/Δ53C (D53c) migrates below endogenous STAT5A/5B (bottom asterisk). P210 bicistronic retroviruses coexpressing neomycin (neo) or a STAT5A point mutant with limited DN STAT5 activity (STAT5A-EE; top asterisk), or sham-transduced cells (–), are shown as expression controls. (C) Survival curves of mice reconstituted with BM cells transduced by DNSTAT5A/Δ53C/P210 (n = 6), or neo/P210 (n = 5). The survival difference between DN STAT5A/Δ53C/P210- and neo/P210-reconstituted animals was statistically significant (P < .001, log-rank test). The results depicted are representative of at least duplicate experiments using equivalent titer bicistronic retroviruses, as described in “Materials and methods.” (D) A bar graph depicting the leukemia phenotype of STAT5A/Δ53C/P210 (n = 10), STAT5A-EE/P210 (n = 6), or control (neo)/P210 (n = 7) mice. (E) A schematic depicting the kinetics of this mouse Bcr/Abl leukemia model, in which the CML-like disease reaches a lethal leukemic burden before the later onset Bcr/Abl ALL and macrophage/monocyte leukemias can develop. By a not-yet fully understood mechanism, STAT5 inhibition has a greater impact on CML progenitors than other BCR/ABL hematopoietic targets, shifting the kinetics and/or spectrum of leukemia by delaying (*), impairing (**), or preventing CML (***).