Plk1 colocalizes with PBIP1 at the interphase and mitotic centromeres. (A), Structure of mammalian Plk1. Numbers indicate amino acid positions. Critical amino acid residues are shown. (B-C), HeLa cells were costained with anti-Plk1 and anti-α-tubulin antibodies (B) or anti-PBIP1 and mouse anti-CREST antibodies (C). Images were collected with a Zeiss LSM510 confocal microscope. Arrows, centrosome-localized Plk1 signals. (D), Model illustrating the Plk1-PBIP1 interaction at the centromeres. PBIP1 functions as a centromeric scaffold to recruit Plk1. Once localized at the centromeres, Plk1 phosphorylates either PBIP1 or other centromeric Plk1 substrates. KD, kinase domain; PBD, polo-box domain.

A model illustrating the self-regulated mechanism of Plk1 localization to the interphase and mitotic centromeres. Early in the cell cycle, PBIP1 accumulates at the interphase kinetochores prior to Plk1 localization (S phase). As Plk1 is expressed in G2, Plk1 interacts with PBIP1 and phosphorylates T78 to create a self-docking site for its PBD (G2 phase). This step is critical to promote its own recruitment to the kinetochores. In early mitosis, a surge in the Plk1 activity induces hyperphosphorylation and delocalization of PBIP1 from the mitotic kinetochores in a manner that is not understood at present (early M phase). As the level of PBIP1 at the kinetochore diminishes, Plk1 liberated from the p-T78 PBIP1 tether binds to other kinetochore components or substrates (marked "X") critical for proper M-phase progression (M phase).

Localization patterns of PBIP1 and Plk1 during the cell cycle. (A), Asynchronously growing HeLa cells were costained with anti-Plk1 and anti-PBIP1 antibodies. Representative images from each stage of the cell cycle are shown. Note that, except G2, inverse correlation between PBIP1 and Plk1 localization is manifest. (B), Schematic diagram illustrating the fluctuation of PBIP1 and Plk1 fluorescence intensities at the centromeres/kinetochores during the cell cycle. During interphase, PBIP1 is tightly localized to centromeres (Red). A rise in the Plk1 level and activity (Blue) during early mitosis induces PBIP1 delocalization from the kinetochores, thus yielding an increased level of non-centromeric PBIP1 population (Green). The free Plk1 population from the p-T78 PBIP1 tether may interact with various PBD-binding proteins or substrates at or near the kinetochores.

Stability of PBIP1 in S and M phases of the cell cycle. To examine the level of PBIP1 stability and to assess the degree of PBIP1 delocalization during the cell cycle, HeLa cells were arrested at the G1/S boundary by double thymidine block, and then treated as described below. Top panel, Cells were released from the G1/S block into fresh medium. Two hours after release (DT 2 h), cells were treated with cycloheximide (20 μM) to inhibit protein synthesis. Where indicated, cells were additionally treated with proteasome inhibitor MG132 (10 μM) to prevent PBIP1 degradation. Bottom panel, Cells were released from the G1/S block into nocodazole (200 ng/ml)-containing medium (DT → Noc) to trap the cells in prometaphase. Ten hours after release, cells were treated as described above. Top and Bottom panels, the same amount of total cellular lysates (50 μg) were treated with either 200 U of λ phosphatase for 1 h at 30°C or left untreated. Samples were mixed with SDS sample buffer, boiled, and separated by 10% SDS-PAGE. Anti-PBIP1 immunoblotting analysis was carried out at the same time for both the top and bottom samples to directly compare all the chemiluminascent signals. The same membrane was stained with Coomassie (CBB) for loading controls. Numbers at the Top and Bottom panels indicate the relative levels of PBIP1 comparison to that of the DT 2 h sample.