SOCS1 coexpression with FLT3-ITD accelerates the development of MPD. (A) Representative bone marrow cytospins from 120-day posttransplantation control and SOCS1 or from FLT3-ITD and SOCS1-T2A-FLT3-ITD moribund mice at the time of death, indicating MPD and ALL, respectively. (B) H&E staining of tissues (spleen, liver, and bone marrow) from control and SOCS1 mice and leukemic FLT3-ITD and SOCS1-T2A-FLT3-ITD mice, indicating MPD and ALL, respectively. Data are representative moribund FLT3-ITD and SOCS1-T2A-FLT3-ITD mice that died of MPD or ALL. (C) Comparison of mean survival (normalized to FLT3-ITD) from 2 independent transplantation experiments. Similar latency in ALL and significantly different latency in the MPD phenotype are shown. (D) Mechanistic model for the cooperation of SOCS1 in FLT3-ITD–mediated leukemogenesis. As we have described recently, constitutively active oncogenic FLT3-ITD directly phosphorylates STAT5 from the endoplasmic reticulum.^21,22  On phosphorylation, STAT5 dimerizes and translocates into nucleus and induces transcription of STAT5 target genes that includes SOCS1. FLT3-ITD–induced SOCS1 expression neither affects FLT3-ITD–mediated proliferative signals nor STAT5 activation, whereas it terminates cytokine-dependent signals. Thereby, SOCS proteins cooperate with FLT3-ITD by “shielding” the cell from cytokine control (eg, IFNα, IFN-γ, IL-3, and IL-6; *P < .05).