Anaphase onset does not require the microtubule-dependent depletion of kinetochore and centromere-binding proteins.

Abstract

Spindle checkpoint proteins, such as Mad2 and BubR1, and the motors dynein/dynactin and CENP-E usually leave kinetochores prior to anaphase onset by microtubule-dependent mechanisms. Likewise, 'chromosome passenger proteins' including INCENP are depleted from the centromeres after anaphase onset and then move to the midzone complex, an event that is essential for cytokinesis. Here we test whether the cell cycle changes that occur at anaphase onset require or contribute to the depletion of kinetochore and centromere proteins independent of microtubules. This required the development of a novel non-antibody method to induce precocious anaphase onset in vivo by using a bacterially expressed fragment of the spindle checkpoint protein Mad1 capable of activating the APC/C, called GST-Mad1F10. By injecting PtK1 cells in nocodazole with GST-Mad1F10 and processing the cells for immunofluorescence microscopy after anaphase sister chromatid separation in nocodazole we found that Mad2, BubR1, cytoplasmic dynein, CENP-E and the 3F3/2 phosphoepitope remain on kinetochores. Thus depletion of these proteins (or phosphoepitope) at kinetochores is not required for anaphase onset and anaphase onset does not produce their depletion independent of microtubules. In contrast, both microtubules and anaphase onset are required for depletion of the 'chromosome passenger' protein INCENP from centromeres, as INCENP does not leave the chromosomes prior to anaphase onset in the presence or absence of microtubules, but does leave the centromeres after anaphase onset in the presence of microtubules.