Microtubule organization in polarized monolayers of MCF-10Aeco cells. (A and B) Schematic diagram of live cell imaging in a polarized epithelial monolayer. The cells were maintained on uncoated inert glass substrates. (A) Fluorescent cells are surrounded by nonfluorescent cells. (B) Three-dimensional images of whole cells were collected by fast z scanning along the apicobasal axis using a Revolution XD system. (C) Three-dimensional distribution of EB1-GFP comets in polarized monolayers. See also Videos 6 and 7. (D) Representative images of EB1-GFP at the apical, middle, and basal cell planes in control and LL5 knockdown cells. The contrast of the images is inverted. (E) Analysis of the EB1-GFP comet distribution along the apicobasal axis under the indicated conditions. The z sections were divided into five equal parts as shown in the left of C, and the number of EB1-GFP comets in each part was counted. For each condition, >53 cells were analyzed, whereas n = 30 for LL5β(ΔM). In LL5-depleted cells, basal EB1-GFP comets are specifically reduced. (F) Velocity of EB1-GFP comet movement at the basal cortex. The conditions showing statistically significant results with P < 0.01 or P < 0.05 versus the mock control are indicated in the table. (G) The three-dimensional distributions of the fluorescence intensities of microtubules (left) and CLASP1 (right) in polarized epithelial monolayers were analyzed using conventional confocal microscopy. Five equally spaced confocal sections of 1-µm thickness were collected from the apical to basal sides of polarized MCF-10Aeco cells under the indicated conditions. For quantitative analysis of the fluorescent signals in each confocal section, ROIs were selected by avoiding centrosomes and Golgi regions, where microtubules are concentrated independent of LL5s. For microtubule and CLASP1 staining, n > 43 and n > 32, respectively. (E and G) The results are presented as means ± SEM (*, P < 0.01 vs. the mock control). Bars, 5 µm.