gms | German Medical Science

Objective: The mitotic Aurora B kinase is overexpressed in various cancers and several inhibitors for Aurora B are currently under clinical assessments. Yet, the precise mode of the anti-tumoral action of Aurora B inhibitors and the possible impact of p53 on treatment remained unclear. In this study, we investigated the effects of selective Aurora B inhibition using AZD1152-HQPA/Barasertib (AZD) on U87-MG glioma cells, corresponding isogenic p53-deficient cells and a primary GBM cell line. We aimed to mechanistically link the mode of action of Aurora B inhibitors to the p53 status of tumor cells and a postmitotic cell cycle arrest. Furthermore, we sought to elucidate whether AZD-induced mitotic defects and polyploidy defines a point of no return and ultimately causes glioma cell death.

Method: U87-MGwt, U87-MGshp53 and primary HT7606 glioma cells were treated the selective Aurora B inhibitor AZD. Subsequently, cells were subjected to Western blot-, flow cytometry-, immunofluorescence analysis, clonogenic assays, FISH- and SKY-analysis.

Results: Inhibition of Aurora B induced polyploidy and a DNA damage response (DDR) as indicated by 5-fold increased ɣH2AX levels and phosphorylated sensor kinases (ATM, CHK2) (n=3). Accumulating mitotic defects caused a non-apoptotic cell death with complete eradication of all cell lines irrespective of p53 status after continuous AZD-treatment. Noteworthy, inhibition of Aurora B led to poly-merotelic kinetochore-microtubule attachments eventually causing DANN damage. In cells with functional p53 a DDR was accompanied by an inefficient pseudo G1 cell cycle arrest characterized by 5-fold increased p21waf/cip and 10-fold increased Cyclin D1 protein levels (n=3), which could not halt ongoing DANN endoreplication. Strikingly, release of tumor cells from transient AZD-treatment resulted in recovery of viable cells bearing increased aneuploidy.

Conclusions: Our results indicate that p53 status is not a predictive factor for response when applying sufficient exposure times of Aurora B inhibitors. We furthermore provide evidence, that polyploid glioma cells released from AZD-treatment can recover and give rise to aneuploid progenies. This could be of clinical relevance since this indicates that slow growing gliomas might not fully respond to standard protocols of Aurora B inhibition. Vice versa, selective Aurora B inhibition might preferentially target highly proliferative tumors i.e. recurrent GBM.