Bem1–Cdc42^GTP binding is not required for GEF localization or cell polarization. (A) Schematic drawing of the proposed Bem1 feedback loop. (B) Bem1 domains and known interacting partners. (C) Bem1 mutations used in this study and their projected effects on Bem1 interactions. (D) Polarization of GFP-Cdc42 in axl2Δ rax1Δ BEM1, axl2Δ rax1Δ bem1Δ, and axl2Δ rax1Δ bem1^N253D cells, as in Fig. 1 B. The plots show means and SEMs from three experiments, with n > 80 cells counted per time point per experiment. (E) Maximum projections of representative polarized cells from D. (F) Maximum projections of representative cells with polarized Cdc24-GFP in axl2Δ rax1Δ bem1^N253D cells after pheromone arrest release in LatA or DMSO. (G) Polarization of Cdc24-GFP in axl2Δ rax1Δ cells and in axl2Δ rax1Δ bem1^N253D cells in LatA, as in Fig. 1 K. For the comparison of peak values at 0°, P < 0.01. (H) Localization of Bem1-GFP or bem1^N253D-GFP in cells with polarized mCherry-Cdc42 after pheromone arrest release in LatA. Z-stack images of representative cells were subjected to a 1 × 1 Gaussian filter before maximum projection. (I) Histogram of Bem1-GFP polarity (maximum pixel intensity/mean pixel intensity per cell) for Bem1-GFP and bem1^N253D-GFP at 50 min after pheromone arrest release in LatA. n = 520 cells were quantified for each genotype. Bars, 5 µm.