AP-1 is required for the peripheral distribution of recycling endosomal domains. (A) IFM analysis of MNT-1 cells treated with control (a–e) or µ1A/γ-adaptin (f–j) siRNAs that had internalized Tf-A488 for 45 min (a and f) and labeled for AP-1 (γ-adaptin; b and g) and the Golgi marker GM130 (c and h). Single labeling (columns 1–3), merged images (column 4), and magnified insets (of boxed areas; column 5) are shown. Note that Tf-A488–containing endosomes were clustered in the peri-Golgi area of AP-1–depleted cells (f–i, arrows). (B) MNT-1 cells treated with control (a–c) or β3A (d–f) siRNAs that had internalized Tf-A488 for 45 min (a and d) were labeled for GM130 (b and e). Single labeling (columns 1 and 2) and merged images (column 3) are shown. (C) Quantitative evaluation of the distribution of Tf-positive endosomes in cells transfected with control, µ1A, or β3A siRNAs. Results are presented as a histogram of the mean distance (from >30 cells each) of Tf-loaded endosomes from the center of the Golgi apparatus labeled by GM130. (D) MNT-1 cells were treated with control (a–c) or µ1A-specific siRNAs (d–l), incubated with 10 µg/ml Tf-A488 for 45 min, fixed, labeled with antibodies to Rab11 (b and e), EEA1 (h), or LAMP-1 (k), and analyzed by IFM. In AP-1–inactivated cells as compared with control cells (a–c), note the peri-Golgi clustering of endosomes labeled with Tf-A488 and Rab11 (d–f, arrows) but not with EEA1 (g–i, arrows) or LAMP-1 (j–l, arrows). (E) Quantification of Tf accumulation in cells treated with control or µ1A/γ-adaptin siRNAs. Mean intensity of total Tf-A488 fluorescence (a) and γ-adaptin labeling (b) was measured and presented as mean ± SD. Bars, 10 µm.