Local Rho activity pulses are modulated by associated GAP and GEF activities. (a–h) Temporal cross-correlation analysis of the Rho activity sensor mCherry-Rhotekin-GBD and RhoGAP Myo9b localization using a GAP-deficient mutant (a–d; R1695M) or WT EGFP fusion proteins (e–h) in nocodazole-treated U2OS cells (see also Video 6). The cross-correlation analyses between Rho activity and p190RhoGAP are shown in Fig. S4. TIRF images (a and e), and intensity measurements (b and f) within the boxes in panels a and e. Temporal cross-correlation functions (c and g) and frequency distribution (d and h) of cross-correlation maxima derived from panels c and g (n ≥ 21 cells from three experiments; error bars represent 95% confidence interval). (i) Peak width of Rho activity signal (n ≥ 21 cells from three experiments; red lines represent mean). (j) Fourier analysis of Myo9b WT and R1695M mutant signal intensity (n ≥ 21 cells from three experiments). Frame rate is 20/min. Bars, 10 µm. *, P < 0.05; ***, P < 0.001 (d and h, one-sample t test; i, analysis of variance, Dunnett’s post-test).