Experimental observations of stability of polarization for different geometries.

<p>A: <i>cla4</i>Δ cells were exposed to <i>α</i>-factor for varying amounts of time to observe a series of shapes of projections: spherical (n = 26 cells), short projection (n = 31 cells) and long projection (n = 34 cells). Cells were then characterized as either stable or unstable by comparing the spatial location of the polarization cap immediately after the application of LAT-A and 30 minutes after. If the polarization cap was in the same location after 30 minutes with no actin cables (due to the LAT-A) then the cell was characterized as stable, otherwise it was unstable. The trend of spherical cells and long projection cells being more stable than shorter projection cells is in line with our theoretical prediction. The differences in the percentage of stable polarizations in the short projection cells versus the spherical (p < 0.05) and long projection (p < 0.001) cells were statistically significant by the chi-square test. B: DIC and Fluorescent images of GFP tagged Ste20 right after the application of LAT-A (t = 0min) and 30 min after for a representative cell of each shape. The spherical and long projection cells here show that the polarization cap stays in the same spatial location throughout the experiment while the short projection cell re-polarizes in a different location. We predict that this instability of the shorter projection cells is due to an interaction between the geometry and the dynamics of the Cdc42 polarization network. All scale bars here represent 2 <i>μm</i>. Time-lapse videos can be seen in <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1006241#pcbi.1006241.s012" target="_blank">S1</a>–<a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1006241#pcbi.1006241.s014" target="_blank">S3</a> Movies in the Supporting Information.