Mad2 inhibits Cdc20 in vivo. (A) The binding between Cdc20 and Mad2 is stabilized by depleting p31^comet. HeLa cells were treated with siRNA against GAPDH (control), p31^comet, BubR1, or p31^comet and BubR1 for 72 h, arrested at prometaphase with nocodazole + MG132, and harvested by mitotic shake off. Cdc20 was immunoprecipitated and analyzed by quantitative immunoblotting. (B) Quantification of the levels of Mad2 and BubR1 bound to Cdc20 calculated from three independent experiments with the amount of protein in control siRNA set to 1. Mean values are shown at the bottom. Means and SDs were calculated from three independent experiments. (C) Cyclin B1 destruction in cells depleted of Mad2, BubR1, Mad2 + BubR1, or BubR1 + p31^comet. HeLa cells were treated with siRNA against Mad2, BubR1, or Mad2 + BubR1 for 48 h or against BubR1 for 48 h and against p31^comet for 72 h, and the level of ectopically expressed Cyclin B1–Venus was analyzed by time-lapse DIC and fluorescence microscopy at 90-s intervals in the presence of 0.33 µM nocodazole. The means ± SD for all cells from three independent experiments are plotted. n = number of cells analyzed. (D) The rates of Cyclin B1–Venus degradation in C were analyzed as in Fig. 1 G. (E) The timing of Cyclin B1–Venus degradation from NEBD in C was analyzed plotted as a box and whisker chart. (F) Cyclin B1 destruction in the absence of BubR1 is accelerated by reducing Mad2 binding to Cdc20. HeLa cells expressing 3×Flag-Cdc20, wild type, or the R132A mutant were treated with siRNA against BubR1, and Cyclin B1 destruction was assayed as in C. n = number of cells analyzed in three independent experiments. (G) The rate of Cyclin B1–Venus degradation in E was measured and plotted as in D. The center lines are the medians, boxes correspond to the range between 25 and 75% of all the data, and the whiskers correspond to the minimum and maximum of all the data. CTR, control; IP, immunoprecipitation.