Supplementary Figure 5: CYRI regulates cell shape and motility and acts as a Meinhardt local inhibitor.

a - Still phase contrast pictures from a random migration assay of control (Vector Ctr) or cyri-b CrispR knockout (#1 and #2) CHL-1 cells plated on collagen-I-coated glass-bottom dishes. Yellow arrowheads denote C-shaped cells. Scale bar = 100 μm (Movie S4). b - Correlation between cell shape and average speed in control (Vector Ctr) or cyri-b CrispR knockout (#1 and #2) CHL-1 cells. Pearson coefficient and R2 value are shown for each condition. (Ctr n = 45 cells, #1 n = 53 cells, #2 n = 42 cells). c - Random migration assay movies were analysed and the percentage of cells presenting a C shape during the time of the experiment is reported. Cochran-Mantel-Haenszel test was performed. ** p ≤ 0.001; *** p ≤ 0.0005. (n = 3 independent assay representing >75 cells). d-e - Western blot analysis of CYRI-B expression in parental (Par.), CrispR control (Vector Ctr) or four independent cyri-b CrispR knockout WM852 melanoma cells (CYRI-B #1-4). Membrane was blotted for CYRI-B and α-Tubulin (d). Bar graph (e) represents quantification of CYRI-B expression relative to the Vector Ctr cell line. (n = 3 independent western blot except for Parental cell line n = 2, see also Supplementary Fig. 7). [Please remove the statistics when n = 2.]. f - Establishment of cyri knockout and CYRI rescued Ax3 D. discoideum cells was confirmed by western blot analysis. Membrane was blotted for D. discoideum CYRI (arrowhead) and MCC1 as a loading control. See also Supplementary Fig. 7. g - Representative phase contrast pictures from an under-agarose chemotaxis assay of Ax3 (WT) and Ax3-derived cell lines migrating towards a folate gradient, as illustrated (Movie S7). Scale bar = 25 μm. h - Cells chemotaxing in a folate under-agarose experiment were tracked automatically using an ImageJ plugin (see material and methods). To monitor the persistence of cell movement, the angle between each step was calculated from x and y positions at each time point and is plotted on the right-hand side graph (Mean and S.D. are shown). One-way ANOVA with Dunn’s post-test was performed. **** p ≤ 0.0001. (WT n = 424 cells, cyri KO n = 581 cells, cyri KO + CYRIWT = 727 cells). i - Pictures extracted from the modelling of CYRI proteins as local inhibitor (Supp. movie 9). In each panel, top graph shows the outline of a simulated cell, with red areas of the periphery having more activator than local inhibitor and blue having more local inhibitor than activator. The x- and y-axes correspond to arbitrary units representing distance. The simulation was run from time t = 0 and panels are taken from times shown (arbitrary units of time). Bottom graph shows the concentrations of the activator (red) and the local inhibitor (blue), where the x-axis represents the scaled arc-length around the perimeter of the cell and the y-axis represents the concentration in arbitrary units. Panel 1 (t = 4) shows the generation of a pseudopod, where the local inhibitor causes the edges of the activated region to sharpen. Panel 2 (t = 8) shows a split in the activator profile (and, consequently, a split in the associated pseudopod), which results from the higher relative concentration of the local inhibitor near the centre of the activated region. Panel 3 (t = 14) shows the cell with two smaller pseudopods, which compete for dominance and determine the direction of migration. Panel 4 (t = 18) shows the winning pseudopod starting the cycle over again and the local inhibitor rising in response to the rise in activator. All data presented are representative of at least 3 biologically independent experiments except if mentioned otherwise. Bar and scatter plots show data points with mean and s.e.m.