Establishment of a Novel Tool to Discover Late Mitotic Protein Functions by the Example of Kinesin Kif18A

The process of cell division requires a profound re-organization of all cellular components, and it is essential for successful replication of organisms that no errors occur during this process. Amongst many other factors, kinesin superfamily members function as active organizers during mitosis. These ATP-hydrolyzing motor proteins use the microtubule cytoskeleton as tracks for moving cargo or directly influence microtubule dynamics at numerous locations in the cell. These functions are essential for a functional mitotic spindle and thus successful chromosome segregation. The kinesin-8 family member Kif18A is a typical example for an important mitotic kinesin. Its absence was shown to result in pronounced chromosome oscillations and failure of chromosome alignment at the cell equator. In some cell types, this leads to mitotic arrest, which is ultimately accompanied by apoptosis or slippage. Mechanistically, the motor protein is known to have both a stabilizing but also depolymerizing effect on microtubules, the details of which remain to be clarified. Upon anaphase onset, the kinesin was shown to accumulate at the central spindle. Besides observations from other species and some small hints from human cell culture experiments, it is unknown whether the kinesin has an important function in late mitosis, and if it does, whether it is similar to the role it plays during early mitosis. A major obstacle to the discovery of a late mitotic function is the fact that many cell types get stuck in a prometaphase-like state upon depletion of Kif18A. As they almost never enter anaphase, and if they do usually in an abnormal manner, it is difficult to observe any phenotypes which could give clues for a potential function. The aim of this work therefore was to establish a tool that specifically can deplete of Kif18A at anaphase onset, thus allowing cells to properly align chromosomes beforehand. Ideally, the tool acts fast and with high reliability in many cells and thus allows a proper investigation on the role of kinesin Kif18A during mitotic exit. After we could confirm the presence of the kinesin-8 motor during mitotic exit and its localization to the central spindle midzone, we successfully established a tool to degrade Kif18A at anaphase onset. This tool, termed CycBmod, consists of the N-terminal 70 amino acids of cyclin B1, but with modifications to ensure efficient degradation without interfering with Kif18A’s function. We could show both in live-cell experiments and by immunoblotting that CycBmod-EGFP-Kif18A is reliably degraded at anaphase onset with rapid kinetics. To generalize the applicability of the tag we attached it to two other proteins with known functions during mitotic exit: the kinesin Kif4A and the kinase Aurora B. We could show that, as previously described, knockdown of Kif4A results in elongated spindles during anaphase. Expression of CycBmod-EGFP-Kif4A in cells depleted of endogenous Kif4A resulted in elongated spindles as well; however, expression of a nonfunctional variant CycBmodND-eGFP-Kif4A, which is not degraded during mitotic exit, rescued the spindle phenotype. In line with this we could show that also CycBmod-eGFP-Aurora B was degraded successfully at anaphase onset and observed phenotypes resembled those observed with knockdown or treatment with an Aurora B inhibitor. We also noticed that degradation efficiency is not identical for every substrate and may depend on subcellular localization, protein abundance or tag accessibility. However, tagging Kif18A was extremely efficient. We therefore set out to observe possible phenotypes of premature Kif18A absence during mitotic exit. We could neither find an effect on chromosome segregation speed, segregation errors, nor nuclear reassembly. Experiments aimed at quantifying central spindle microtubule integrity failed due to problems in proper signal quantifications. In conclusion, we successfully established and characterized a reliable method to study Kif18A in anaphase and could show that it has a promising applicability for other proteins involved in mitotic exit. However, despite many trials, we could not discover a phenotype that occurred in absence of Kif18A during late mitosis.publishe