Abstract P1-08-09: Aurora kinase-A protein stability is negatively regulated by eEF1α2 and PTEN in breast cancer: Prognostic and therapeutic implications

Abstract

Abstract Background: The AURKA gene encoding Aurora kinase-A (Aurora-A) protein is localized on chromosome 20q13 that is frequently amplified and overexpressed across multiple cancer types correlating with patient prognosis. Aurora-A plays a pivotal role in faithful segregation of chromosomes and normal progression of mitosis, peaking at G2/M followed by degradation at the end of mitosis by APC/C (Cdh1). However, regulation of Aurora-A protein stability in human cancer cells is not well elucidated. Here, we show that Aurora-A is targeted for ubiquitination and degradation by SCF complex involving eEF1α2 and PTEN in human breast cancer cells. Methods: Using a panel of breast cancer cell lines, as in vitro models, the eEF1α2 was knocked down or ectopically expressed to test the stability of Aurora-A protein. For in vivo models, tissue micro arrays of human breast cancer were immunostained for Aurora-A and eEF1α2 expression and categorized values were statistically tested by Chi-squared test. In addition, the public breast cancer dataset (Transbig) was used to predict breast cancer prognosis by Kaplan-Meier survival analysis. Results: In breast cancer cell lines and patient samples, an eEF1α2 non-expressing group showed a trend of higher Aurora-A expression than eEF1α2 expressing group, whose trend was significant in patient samples (P<0.05). Knocking down of eEF1α2 enhanced Aurora-A protein stability. In contrast, ectopic expression of eEF1α2 dramatically decreased Aurora-A protein stability. Inhibition of proteasome activity by MG132 could restore the Aurora-A protein in eEF1α2 expressing cells. Biochemical assays showed the direct binding between eEF1α2 and Aurora-A, and eEF1α2 dependent ubiquitination of Aurora-A. Inverse correlation of the expression levels of the two proteins was also observed throug the cell cycle, with eEF1α2 levels being high from G1 through G2 phases while Aurora-A expression peaked from G2/M phase through cytokinesis. Taken together, these findings highlight eEF1α2 as a novel negative regulator destabilizing Aurora-A through ubiquitin-proteasome proteolytic pathway. Further, mechanistic studies revealed that eEF1α2 enhanced the interaction of SCF E3 ubiquitin ligase complex protein; FBXW7 and Cul1, with Aurora-A. In line with this scenario, knocking down of Cul1 increased Aurora-A level. Since PTEN loss was reported to stabilize Aurora-A through inhibiting SCF complex, we tested the significance of PTEN loss in our model. Knocking down of PTEN further stabilized Aurora-A suggesting an independent role of PTEN from eEF1α2 in destabilizing Aurora-A. When eEF1α2 expressing cells were treated with AKT inhibitor, Aurora-A was destabilized with enhanced bindings between Aurora-A and FBXW7/Cul1. Lastly, low PTEN expression correlated with poor prognosis of Aurora-A over expressing breast cancer patients (P<0.01). Conclusions: Aurora-A overexpression in human breast cancer cells may be associated with loss of eEF1α2 and PTEN due to reduced interaction of SCF with Aurora-A. Findings indicate significant prognostic and therapeutic implications of altered expression of eEF1α2/PTEN/Aurora-A pathways among Aurora-A subset of breast cancer patients. Citation Format: Treekitkarnmongkol W, Kai K, Katayama H, Tian W, Rodriguez-Canales J, Sahin AA, Sen S. Aurora kinase-A protein stability is negatively regulated by eEF1α2 and PTEN in breast cancer: Prognostic and therapeutic implications [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-08-09.