Gfi1 regulates expression of RasGRP1, which is critical for Erk activation by G-CSF. (A) Stable expression of GFP, Gfi1-GFP (left); vector and RasGRP1 (right) in 32D cells assessed by reverse transcription (RT)–PCR with specific primers. Glyceraldehyde phosphate dehydrogenase (GAPDH) expression was used as a control. Representative results from 3 experiments are shown. (B) Expression of Gfi1 in 32D cells specifically promotes expression of endogenous RasGRP1 as measured by RT-PCR. Representative results from 3 experiments are shown. (C) Relative levels of RasGRP1 mRNA expression in parental 32D cells and in 32D cells stably transduced with GFP or GFP-Gfi1 vectors as measured by real-time PCR. Representative results from 3 experiments are shown. (D) Stable expression of Gfi1 in 32D promotes increased Erk1/2 activation by G-CSF (left) but not by IL-3 (right) as assessed by immunoblotting with specific antibodies to phosphorylated Erk1/2. The membrane was reprobed for total Erk1/2. Representative results from 6 experiments are shown. (E) Effect of RasGRP1 silencing in 32D cells stably transduced with RasGRP1. Erk1/2 activation by G-CSF or IL-3 is assessed by immunoblotting. The membranes were reprobed for total Erk1/2. The reduction of RasGRP1 expression in this experiment was 78%, as assessed by real-time PCR. Representative results from 3 experiments are shown. (F) Erk1/2 activation by G-CSF or IL-3 in 32D cells stably expressing RasGRP1 or control cells as assessed by immunoblotting. The membranes were reprobed with the indicated antibodies to assess loading and activation of STAT3 and STAT5. Representative results from 3 experiments are shown. (G) Time-dependent activation of endogenous RasGRP1 in 32D cells cultured with G-CSF as assessed by semiquantitative PCR. Thymus-derived RNA from a Gfi1^+/+ mouse was used as a positive control. Representative results from 5 experiments are shown.