Recruitment of CENP-E to the kinetochore is dependent on TRAMM. (A) HeLa cells were transfected with NS or siRNA against TRAMM, CENP-E, or both TRAMM and CENP-E as indicated. After 24 h, the cells were fixed and stained with antitubulin, ACA, and anti–histone H2B. Shown next to each image is a brightfield image of the cells. The mitotic index is indicated in the brightfield images as a percentage ± SD. The intermediate value obtained for the double depletion was seen in four independent experiments. Bars: (fluorescent images) 5 µm; (brightfield images) 100 µm. A Western analysis of each condition is shown to the right of A. (B) After staining the cells, ACA-positive pairs that were not aligned in the metaphase plate were counted and expressed as 1–5 or >5 pairs. (C) TRAPP subunits in a Gal4-DNA binding domain vector (BD; pGBKT7) were transformed into yeast cells and mated with a yeast strain containing CENP-E (amino acids 2,131–2,701) in a Gal4 activation domain vector (AD; pGADT7). To assess autoactivation, each TRAPP subunit was also transformed into a yeast strain with an empty Gal4 activation domain vector (Ø). Growth on DDO (−Trp/−Leu) plates indicates mating between the two strains, whereas growth on TDO (−Trp/−Leu/−His) plates containing 4 mM 3-aminotriazol (3-AT) indicates a protein–protein interaction. (D) HeLa cells were treated with NS or TRAMM-specific siRNA. After 24 h, the cells were left untreated or treated with nocodazole (Noc) as indicated. The cells were then fixed and stained for ACA and anti–CENP-E. Bars, 5 µm. (E) Fluorescence intensity was measured for CENP-E signals in D that colocalized with ACA. The intensity of NS cells not treated with nocodazole was adjusted to 100%. Values from two independent experiments ± SD are indicated. KD, knockdown.