New insights into the physiological functions and anti-cancer drug potential of Aurora kinases

Aurora kinases are essential cell cycle regulators for normal cell division. All three members of the Aurora kinases are overexpressed in cancers and therefore a number of Aurora kinase inhibitors are developed that are undergoing various stages of clinical trials. Three projects related to the biology of Aurora kinases and the potential of Aurora kinases as targets for anticancer drugs are studied in this thesis. First, the effects of the tumor suppressor p53 on the sensitivity to Aurora kinase inhibitors were investigated. Using isogenic cell lines that differ in p53 status and RNA interference, I found that p53 was induced and activated by the Aurora inhibitors. Moreover the absence of p53 significantly sensitized cells to Aurora kinase inhibitors including Alisertib and Barasertib. The Aurora kinases in p53-deficient cells were more robustly inactivated by the Aurora kinase inhibitors, triggering more cells to undergo mitotic slippage, resulting in tetraploid daughter cells. A related study focused on the potential of Aurora B as a ploidy-specific lethality gene. Using models of isogenic diploid versus tetraploid cell lines, investigated their sensitivity towards the Aurora kinase inhibitors ZM447439 and Alisertib. I found that tetraploid cells were sensitized by inhibition of Aurora B but were not affected by inhibition of Aurora A. Aurora B inhibition induced a stronger apoptotic response in tetraploid cells than in their diploid counterparts. Interestingly, the underlying mechanism was due to mitotic slippage and the subsequent excessive genome reduplication, followed by apoptosis. Collectively, my results indicate that Aurora kinase inhibition effectively promotes the formation of polyploid cells by triggering mitotic slippage in p53 deficient cells. The resulting tetraploid daughter cells have a higher sensitivity to extended inhibition of Aurora B or cytokinesis. A detailed knowledge of the functional and regulatory networks of the Aurora kinases is important to understand the effects of the Aurora inhibitors. I used large scale purification arid proteomics, to identify novel binding proteins for Aurora A that included Protein phosphatase 6 and Nucleophosmin. The interaction of Protein phosphatase 6 and Nudeophosmin with Aurora A was verified in vivo and in vitro. I also found that the interaction with Nudeophosmin relied on the activity of Aurora A as well as on the N-terminal oligomerization domain of Nucleophosmin. Nucleophosmin was a substrate of Aurora A in vitro and could inhibit the kinase activity of Aurora A. The tumor suppressor gene p53 is one of the most frequently mutated genes in cancers. Half of the people that developed cancer lack active p53 and p53 deficiency is coupled to poor prognosis. Therefore it is crucial to identify reagents that target p53 deficient cells. The presented results suggest that Aurora kinase inhibitors are particular effective for treating p53- negative cancer cells and might be promising for future therapy approaches.