Figure 2.
Knockdown of GPS2 blocks human erythroid differentiation. (A) Assays used to test the effect of GPS2 protein on erythropoiesis. CFU-C, CFU in culture. (B-G) CD34+GFP+ cells were selected from human UCB CD34+ cells infected with GPS2 or control (Ctrl) shRNA lentivirus and cultured in the erythroid differentiation medium for the indicated times. Knockdown efficiency of GPS2 was detected by real-time PCR (B) and western blot analysis (C) at day 5. The proportion of CD71+GPA+ cells was analyzed by flow cytometry at days 0, 3, and 5 (D). Hemoglobin production was analyzed by benzidine staining at day 5. Representative images of benzidine-stained cells were shown; scale bars, 20 μm (E). The percentage of benzidine-positive cells was calculated (F). mRNA levels of the indicated erythroid genes in FACS-sorted CD71+GPA+ cells were analyzed by real-time PCR at day 3 (G). (H-I) Clonogenic capacity of CD34+ cells transduced with GPS2 or control shRNA lentivirus. The number (H) and frequency (I) of colonies are represented. All values are mean ± SEM (n = 3 replicates). (J-K) Human CD34+ cells transduced with GPS2 or control shRNA lentivirus were IV injected into sublethally irradiated NSG mice. The percentage of GPA+ cells within the GFP+CD45− population in BM was determined at 3 (J) and 20 (K) weeks after transplantation (3 weeks, n = 6/group; 20 weeks, n = 4/group, mean ± SEM). *P < .05, **P < .01, ***P < .001; 2-tailed unpaired t test.