Emergence of periodic circumferential actin cables from the anisotropic fusion of actin nanoclusters during tubulogenesis

Figures

• 

  Fig. 1. Emergence of actin nanoclusters in the apical cortex during tracheal tube expansion. a Schematic of the tracheal tube constituted by multiple cells. The intercellular periodic circumferential actin cables appear beneath the apical (luminal) membrane. The right image shows a representative image of F-actin pattern at stage 16 visualized by phalloidin staining. Three experiments were repeated independently. Scale bar, 5 µm. b Time course of cortical actin orientation, actin cable elongation, tracheal tube expansion, and tube elongation at stage 15. The time point when the circumferential cables formed was defined as 0* min. c, d Actin pattern visualized by lifeact::GFP in the fixed embryos at stages 15 and 16, respectively. Two representative images for stage 15 are shown in (c). Lower square images indicate a magnified view of the yellow square (left) and its binarized image (right). Scale bars: 2 µm (upper) and 0.5 µm (lower). Eight and seven independent tracheal tubes were observed for stage 15 and stage 16, respectively. e, f Comparison of the cluster-fitted ellipses between stage 15 (n = 356, N = 22) and stage 16 (n = 271, N = 21) with aspect ratio (f), and angle of the major…

• 

  Fig. 2. Nanocluster fusion is more stable in the circumferential direction. a Time-lapse fluorescent images and binarized images of lifeact::GFP expressing tracheal cells with 0.32-s intervals at stage 15. The fusion sites of clusters were labeled with magenta (see Supplementary Fig. 2 and “Fusion stability”). The right-most images are the temporal projection of the six frames, color-coded with Spectrum. Upper: Actin pattern at −40 min. Actin nanoclusters change in shape and size dynamically. Lower: Actin pattern at 0* min. Although the circumferential cables become apparent, the cables are still unstable. Scale bar, 0.5 µm. b Duration of the fusion of the nanoclusters with respect to the orientation. The duration of circumferential fusion became significantly longer than that of the diagonal fusion at 0* min. 0° corresponds to the longitudinal axis, whereas 90° indicates the circumferential axis of the tube. The numbers of fusions (n) are written below. Boxplots represent the median (bar), mean (black diamond) plus minima and maxima with lower and upper quantiles. For statistical analysis, the two-sided Kruskal–Wallis test (shown in parentheses, P = 0.565 in −40 min, P = 0.322 in −30…

• 

  Fig. 3. Revealing nanocluster-associated molecules that regulate cluster fusion and anisotropy response. a Median orientation and average major length of the actin patterns are plotted for the genetically manipulated tracheal cells (“OrientationJ analysis” and “Cluster analysis”). Each circle indicates different genetic conditions: control (red), RNAi lines with mild defects (gray), RNAi lines with strong defects (blue, denoted with the target genes), and mutants found after the RNAi screen (green). See Supplementary Table 2 for details. b Confocal projection images and binarized images of control, RNAi-expressing, or mutant embryos at early stage 16. More than six independent tracheal tubes were observed for each genotype. Scale bars: 2 µm (upper) and 0.5 µm (lower). c A network of protein-protein interaction between genes, whose downregulation displayed strong phenotypes. Solid lines indicate direct interaction and dotted lines indicate indirect interaction with a single interactor. The genes are grouped by representative molecular functions. d Colocalization between the proteins of interest with actin (lifeact::mScarletx2 or lifeact::GFP) at stage 15. Pearson’s correlation coefficient…

• 

  Fig. 4. Self-organized structures reproduced by agent-based model simulation. a Schematics of the simulation model, including actin filaments (blue) and two types of connectors: crosslinkers (yellow) and motors (red). See Supplementary Notes for details. b Steady-state solutions for different conditions. Each column refers to different numbers of cross-linkers: Low (L), moderate (M), and High (H) from left to right. Each of the four rows refers to no anisotropy and no motor (blue circle, “1”); with anisotropy and no motor (orange triangle, “2”); no anisotropy with motors (yellow square, “3”); and with anisotropy and motors (red diamond, “4”). See Supplementary Table 3 for detail information. In each condition, the snapshot (left), the filament density distribution (middle), and the binarized filament density distribution (right) are displayed. The actin pattern that meets the criterion of circumferential cable formation is marked by a red frame (4-M). c Characterization of the clusters. The median orientation and average major length are quantified using the binarized filament density distribution (“OrientationJ analysis” and “Cluster analysis“). Each data point corresponds to the…

• 

  Fig. 5. Circumferential motion of the actin nanoclusters is DAAM- and Zasp52-dependent. a The Motion anisotropy analysis of small actin nanoclusters during circumferential cable formation at stage 15. The x axis corresponds to the longitudinal axis, whereas the y axis corresponds to the circumferential axis of the tracheal tube. The circumferentially biased motion indicated by the ratio (|∆y | /|∆x | ) larger than unity became apparent from −30 min. b Particle image velocimetry (PIV) analysis of the cortical actin dynamics in tracheal cells. The velocity vectors color-coded with the vector length are superposed. The minimum and maximum lengths (nm) of the vectors are indicated below the color box. Scale bar, 0.5 µm. c Transition of mean displacement (nm/s) of the nanoclusters from isotropic phase to anisotropic phase calculated by the velocity vectors of PIV analyses. Line plots represent mean ± SD. The two-sided Student’s t test was performed for the statistical analysis. P = 0.466 in −40 min (n = 4), P = 0.0729 in −30 min (n = 4), P = 0.0657 in −20 min (n = 3), P = 0.0176 in −10 min (n = 4), or P = 0.0038 in 0* min (n = 4). ∗P < 0.05, ∗∗P < 0.01. d The Motion anisotropy analysis of…