Knocking out Atg5 in the eosinophil lineage results in delayed and reduced eosinophil precursor proliferation and maturation. (A) Flow cytometry. Maturity of eosinophils in the bone marrow of control and Atg5^eoΔ mice was assessed by measuring the cell surface expression of chemokine receptor CCR3. Mature eosinophils were defined as Siglec-F⁺CCR3⁺ and immature eosinophils as Siglec-F⁺CCR3^– cell populations (n = 6). Right: representative original flow cytometry data are shown. Each symbol represents a value for an individual mouse throughout. (B) The data of panel A are presented as absolute numbers. (C) Flow cytometry. Eosinophils were differentiated in vitro from bone marrow cells of control and Atg5^eoΔ mice, and their maturity was assessed by measuring the surface expression of CCR3 on the indicated days (n = 3). Right: representative original flow cytometry data are shown. (D) Flow cytometry. Proliferating eosinophil precursors in the S phase of the cell cycle during in vitro differentiation were detected by 5-ethynyl-2'-deoxyuridine (EdU) labeling (n = 3). (E) Flow cytometry. The proliferative status of eosinophil precursors in the G₁, S, G₂, and M phases of the cell cycle during in vitro differentiation was determined by Ki-67 expression measurements (n = 3). (F) [³H]‐thymidine incorporation assay. For each time point, [³H]‐thymidine was added to the cultures of in vitro differentiating eosinophils 24 hours before the measurement. Radioactivity in DNA recovered from the cells was determined with a scintillation counter to determine the extent of cell division (n = 3). (G) CFU assay. Freshly isolated bone marrow cells from control and Atg5^eoΔ mice were plated in methylcellulose media supplemented with recombinant mouse IL-5. After 12 days of incubation, cells were stained with the substrate for EPX, and brown-stained Eo-CFUs of >50 cells were counted in each well (n = 5). Right: representative images of culture wells and single colonies are shown. Scale bars, 10 µm. To confirm the identification of colonies as Eo-CFU, samples were selected and removed from the wells for analysis of Siglec-F expression by flow cytometry as presented in supplemental Figure 1H. (H) Immunoblotting. Protein lysates during in vitro culture of differentiating eosinophils were collected on the indicated days and analyzed for the presence of phosphorylated p38 (Thr180/Tyr182), phosphorylated p44/42 (Thr202/Tyr204), ATG5, and LC3 protein expression. p38, p44/42, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) protein levels were analyzed as loading controls. Representative immunoblots of 3 independent experiments are shown. Values are means ± standard error of the mean, or single data are presented in scatter dot plots in which the medians are indicated as red lines. *P < .05; **P < .01; ***P < .001.