Expression of ezrinWT, ezrinY145F, and ezrinY353F in HS2 leukemic proerythroblasts. (A) 606HS2 cells were transfected with pEF-neo EzWT-VSVG, pEF-neo EzY145F-VSVG, pEF-neo EzY353F-VSVG, or pEF-neo empty vector. WCEs were subjected to Western blot using anti-VSVG and anti-ezrin antibodies, and an anti–β-actin antibody as a loading control. The fold increase in ezrin expression is indicated at the bottom of each line. Vertical lines have been inserted to indicate repositioned gel lanes. (B) Proliferation of 606 cells expressing ezrinWT, ezrinY145F, or ezrinY145F. Cells transfected with the pEF-neo vector were used as control. Living cells were plated at 2 × 105 cells/mL following a ficoll gradient centrifugation to remove dead cells. Viable cells were scored daily for 72 hours. Data are means plus or minus SD of 5 independent experiments performed in triplicate. (C) Percentage of death of 606HS2-EzWT, 606HS2-EzY145F, 606HS2-EzY353F, and 606HS2-neo cells. Dead cells were scored daily by Trypan blue exclusion staining. Data are means plus or minus SD of 5 independent experiments performed in triplicate. (D) Processing of caspase-3 in 606HS2-EzWT, 606HS2-EzY145F, 606HS2-EzY353F, and 606HS2-neo cells. After 48 hours of culture, WCEs were subjected to Western blotting using an anti-cleaved caspase-3 antibody (casp p17). β-actin served as a loading control. (E) Colony formation assayed in methylcellulose. A total of 400 606HS2-EzWT, 606HS2-EzY145F, 606HS2-EzY353F, and 606HS2-neo cells were seeded onto 1% methylcellulose-containing medium, and colonies were scored after 6 days. Data represent means plus or minus SD of 6 independent experiments performed in duplicate by 2 investigators. Statistical significant differences were estimated by a 2-tailed Student t test (*P < .001, 606HS2-EzWT and 606HS2-EzY145F cells; **P < .005, 606HS2-EzWT and 606HS2-EzY353F cells). Representative images of colonies after 6 days of culture were acquired with a Nikon Eclipse TE300 microscope (Nikon, Champigny sur Marne, France). Bars correspond to 80 μm. (F) In vivo tumorigenicity. Subcutaneous tumors grown in nude mice injected with 606HS2-EzWT, 606HS2-EzY145F, or 606HS2-neo cells were extracted 3 weeks after injection and weighed. Two independent clones of 606HS2-EzY145F and 606HS2-EzWT or 606HS2-neo cells were each injected into 8 mice. Histograms are the compilation (mean ± SD) of these 2 independent experiments. A bar represents the mean-weight of 16 tumors. **Significant tumor weight reduction in 606HS2-EzY145F compared with 606HS2- EzWT (Student t test; P < .001).