Kinetics of ribosome neosynthesis during erythroid differentiation. Human primary erythroblasts were derived from CD34⁺ progenitors cultured for 6 days in the presence of SCF, IL-6, IL-3, and dexamethasone (1 µM); with SCF, IL-6, and IL-3 for 1 day; with SCF and EPO between days 7 and 11; and then with EPO alone. (A) Amplification curve and cell size. Human erythroblast proliferation was expressed as the cumulative number of cells at each indicated time (closed symbols). Cell size was measured using forward scatter (FSC) light expressed as the mean fluorescence intensity (MFI; open symbols). Means ± SEM of 4 experiments. (B) Expression of the SCF receptor CD117 and GlyA by flow cytometry during human erythroblast differentiation. (C) Proportion of progenitor, ProE, baso1, baso2, polyE, and orthoE cells at the indicated days, determined by May-Grünwald-Giemsa–stained cytospins. Mean ± SEM of 4 independent experiments. (D) RNA quantification in ribosome fractions. Quantities of RNA in ribosomes of human erythroblasts purified by ultracentrifugation on sucrose gradient were measured at OD 260 nm per 10⁶ cells. Mean ± SEM of 6 experiments. (E) Quantification (top) of pre-rRNA 45S and β-globin (HBB) transcripts by qRT-PCR. Transcripts amounts were normalized to B2M and UBC transcript amounts, and normalized relative quantities (NRQs) were calculated. Mean NRQ ± SEM of 3 experiments. Fluorescence in situ hybridization (bottom) for pre-rRNA 45S in human erythroblasts at the indicated culture times. Slides were stained using a 5'ETS probe conjugated to Alexa 647. Images were obtained on a Leica DMI6000 inverted microscope with spinning disk and analyzed with ImageJ. *P < .05. (F) Absolute quantification of RPs by MS/MS plotted from publicly available data from Gautier et al.²⁰  Results are expressed as the median protein copy number of 4 independent experiments. **P < .01; ***P < .001, by Student t test. (G) Schematic experimental design of SILAC labeling. Human erythroblasts were metabolically labeled for 24 hours, and ribosomes were purified by differential centrifugation and analyzed by LC-MS/MS. IMDM, Iscove’s modified Dulbecco’s medium. (H) Kinetics of ribosome neosynthesis. In each experiment, the percentage of each neosynthesized RP was calculated as (H/H+L) ×100. Ribosome neosynthesis was defined as the median of neosynthesized RP percentages. Mean ± SEM of 4 experiments. (I) Concomitant decrease in RPL and RPS neosynthesis. The percentage of neosynthesis (calculated as in panel H) of each protein was plotted in 2-dimensional scatterplots to compare the value at each time point with the reference value at day 7. RPL: purple dots; RPS: pink dots. Results are representative of 4 independent experiments. (J) Comparison of the rate of neosynthesis of each RP at different time points of the human erythroid differentiation. The H/L ratio was normalized to the median of the H/L ratios of all RPs of the same subunit and transformed to log2 values. RPs in red are those with the highest rate of neosynthesis and incorporation into the ribosome. The heat maps are representative of 4 independent experiments. (K) Ribosome neosynthesis by pulse SILAC in human erythroblasts at day 10 (proE/baso1). Cells were cytokine-starved for 4 hours before incubation in SILAC medium with 10 UI/mL EPO, 100 ng/mL SCF, or EPO+SCF, in the presence or absence of 2 µM of the c-Kit inhibitor masitinib for 24 hours. For each experiment, the median of neosynthesized RP percentages was determined. Results are shown as means ± SEM of 3 independent experiments. **P < .01, by Student t test. (L) Cytology of sorted human proEs/baso1 (GlyA^low/Band3⁻) and baso2 (GlyA^high/Band3⁺) after May-Grünwald-Giemsa staining. Original magnification ×100. (M) Ribosome neosynthesis by pulse SILAC of sorted human erythroblasts GlyA^low/Band3⁻ and GlyA^high/Band3⁺ incubated in SILAC medium with EPO+SCF for 24 hours. Results representative of 1 experiment and expressed as the median of neosynthesized RP percentages are shown.