NEDD9 depletion destabilizes Aurora A kinase and heightens the efficacy of Aurora A inhibitors: implications for treatment of metastatic solid tumors

Abstract

Aurora A kinase (AURKA) is overexpressed in 96% of human cancers and is considered an independent marker of poor prognosis. While the majority of tumors have elevated levels of AURKA protein, few have AURKA gene amplification, implying that posttranscriptional mechanisms regulating AURKA protein levels are significant. Here, we show that NEDD9, a known activator of AURKA, is directly involved in AURKA stability. Analysis of a comprehensive breast cancer tissue microarray revealed a tight correlation between the expression of both proteins, significantly corresponding with increased prognostic value. A decrease in AURKA, concomitant with increased ubiquitination and proteasome-dependent degradation, occurs due to depletion or knockout of NEDD9. Reexpression of wild-type NEDD9 was sufficient to rescue the observed phenomenon. Binding of NEDD9 to AURKA is critical for AURKA stabilization, as mutation of S296E was sufficient to disrupt binding and led to reduced AURKA protein levels. NEDD9 confers AURKA stability by limiting the binding of the cdh1-substrate recognition subunit of APC/C ubiquitin ligase to AURKA. Depletion of NEDD9 in tumor cells increases sensitivity to AURKA inhibitors. Combination therapy with NEDD9 short hairpin RNAs and AURKA inhibitors impairs tumor growth and distant metastasis in mice harboring xenografts of breast tumors. Collectively, our findings provide rationale for the use of AURKA inhibitors in treatment of metastatic tumors and predict the sensitivity of the patients to AURKA inhibitors based on NEDD9 expression.

Figure 1. Increased NEDD9 expression correlates with expression of AURKA protein in invasive ductal breast adenocarcinomas

(A). Representative images of immunohistochemical staining of tissue microarray with anti-AURKA (top panel) and anti-NEDD9 (bottom panel) antibodies; Normal breast tissue, intraductal carcinoma (IC), invasive ductal carcinoma (IDC), metastatic invasive ductal carcinoma, lymph nodes (MIDC-LN). Scale bar is 200µm. Insets represent x200 enlarged areas indicated in the main panel. (B). Quantification of NEDD9 and AURKA in RUP-relative units of positivity (percentage of cells stained positively in the same core stained by anti-AURKA or -NEDD9 antibodies). Most fitted lines were plotted on the graphs. (C). positivity data for both proteins were analyzed using Spearman’s Rho correlation. (D). positivity data for both proteins were analyzed by Random Forest statistical software.

Figure 2. Depletion of NEDD9 leads to dramatic decrease of AURKA protein

(A). Western Blot (WB) analysis of NEDD9 and AURKA in breast cancer cell lines and (B) in mouse embryonic fibroblasts derived from wild type (WT) and NEDD9 knock out (KO) animals. (C). Representative IHC of anti-AURKA staining in kidney tissues, WT and NEDD9-KO mice. Staining with non-specific IgG (Con.). Scale bar is 25µm. (D). WB analysis and quantification of AURKA, NEDD9, GAPDH upon treatment with anti-NEDD9 (shN1, shN2) or non-specific (shCon) shRNAs, n=3, percentage of AURKA expression to shCon, standard error of mean (+/−S.E.M). Student’s one-tailed t-test **p=0.0006, **p=0.0024 (shCon/shN1 or /shN2), and no significant difference (ns) (shN1/shN2) in MCF7. **p=0.003 and **p=0.0094 (shCon/shN1 or /shN2), and ns (shN1/shN2) in MDA-MB-231. (E). WB analysis and quantification of AURKA and NEDD9 expression in shNEDD9 and shCon cells transfected with WT-NEDD9 or control-RFP (red fluorescent protein) cDNA; n=3, percentage of AURKA expression to shCon, normalized by GAPDH. Student’s one-tailed t-test *p=0.0186, **p=0.0012 (shCon/shN1 or /shN2), ns (shN1/shN2 or shN1/NEDD9 and shN2/NEDD9 re-expression).

Figure 3. NEDD9 regulates stability of AURKA through inhibition of proteasome-dependent degradation

(A). WB analysis of AURKA and NEDD9 expression in shCon-, shNEDD9-MDA-MB-231LN cells treated with cyclohexamide. Expression was normalized by α-tubulin. (B). Quantification of AURKA as in (A), n=3, fold of change of AURKA expression to time point 0h, +/−S.E.M, one-way ANOVA ***p<0.0001, **p=0.005 (shCon/shN1 or /shN2) at each time point (except 0h). (C). WB analysis of AURKA, NEDD9 in shCon, shNEDD9 cells treated with MG132 or vehicle. (D). Quantification of AURKA expression as in (C, +MG132), n=3, fold of change to shCon, +/− S.E.M. Student’s t-test **p=0.0086, **p=0.0084 (shCon/shN1 or /shN2). (E). WB analysis of AURKA ubiquitination in WCL and IP from shCon or shNEDD9 cells transfected with pcDNA3-Myc-Ubiquitin, with MG132. (F). Quantification of AURKA ubiquitination as in (E, IP), n=3, fold of change to shCon, +/− S.E.M. Student’s t-test, **p=0.0098, *p=0.05 (shCon/shN1 or /shN2). (G). WB with anti-AURKA, anti-NEDD9 and anti-cdh1 antibodies. (H). Quantification of AURKA using the following formula (AURKA-IP/AURKA-total) normalized to NEDD9 pull-down as in (G), n=3, plotted as AURKA ratio +/− S.E.M, *p<0.05.

Figure 4. NEDD9 binding to AURKA is necessary for protein stabilization

(A). WB analysis of AURKA and NEDD9 expression in WCL (left) or IP-GFP (right) from 293T cells transfected with pAcGFP-NEDD9-WT, AA, EE A, E, mutants or GFP-control. (B). Quantification of AURKA-coIP as in (A), n=3, plotted as percentage +/− S.E.M. AURKA-coIP with wtNEDD9 assigned 100%. Student’s t-test **p=0.0013, *p=0.010, **p=0.00209, ns (WT/AA, or /EE, or /A or /E respectively). (C). WB analysis of AURKA and NEDD9 expression in WCL of MDA-MB-231 cells expressing doxycycline-inducible NEDD9-AA, -EE or empty vector control. (D). Quantification of AURKA expression as in (C) n=3, relative intensity units (RIU) +/− S.E.M, normalized to actin. Two-way ANOVA, p<0.0001. Student’s t-test, **p=0.0063, 0.0061 (AA /EE (+Dox, 24, 48h)).

Figure 5. NEDD9 binding to AURKA decreases the efficacy of AURKA inhibitors in vitro

(A). WB analysis of AURKA, phT288-AURKA, NEDD9 in shCon or shNEDD9-MDA-MB-231LN cells treated with PHA-680632. (B). Quantification of phT288-AURKA expression as in (A) n=3, normalized to total AURKA and plotted as relative intensity units (RIU) +/− S.E.M. Two-way ANOVA, p<0.0001, (shCon/shN1 or shN2), p=0.0013 (shN1/shN2). Quantification of phS10-Histone H3 (IF), n=3, 1000 cells/treatment, p=0.006, p=0.0067 (shCon/shN1 or shN2), ns (shN1/shN2) (C). WB analysis of total AURKA, phT288-AURKA, NEDD9 expression in cells treated with increasing concentrations of MLN8054. (D). Quantification as in B for the western blots shown in (C), p<0.0001, (shCon/shN1 or shN2), p=0.0041 (shN1/shN2). Quantification of IF shown in Supplementary Fig.2E, n=3, 1000 cells/treatment, p=0.0053, P=0.0086 (shCon/shN1 or shN2), ns (shN1/shN2).

Figure 6. Depletion of NEDD9 combined with AURKA inhibitors leads to decrease in a tumor burden and lung metastasis in breast cancer xenograft model

(A). Representative images of bioluminescence (BLI) of mice orthotopically injected with MDA-MB-231LN parental, shCon or shNEDD9 (shN2) cells. (B). Quantification of BLI data as in (A) plotted as mean log photon flux +/−S.E.M. Linear regression analysis, p=0.0021, p=0.0006 (shCon/shN2), weeks 3 and 4; ns (shCon/parental). (C). Quantification of BLI data, mice orthotopically injected with shNEDD9 or shCon cells and treated with MLN8237 (MLN) or vehicle, 3 independent experiments, n=6 in each group; mean photon flux, +/−S.E.M. Two-way ANOVA, p=0.0061 (shCon-MLN/shN2-MLN, days 7–11). (D). Representative images of BLI, lungs dissected from 4 groups. (E). Quantification of BLI data as in (D), 3 independent experiments, n=6/group; mean photon flux +/−S.E.M. Two-way ANOVA, **p=0.0071 (shCon-Vehicle/shN2-Vehicle); **p=0.045 (shCon-MLN/shN2-MLN).

Figure 7. Combination of AURKA Inhibitors with depletion of NEDD9 reduces colonization potential of human circulating tumor cells (CTCs) in lungs

(A). Representative images of BLI; mice intravenously injected with MDA-MB-231LN cells expressing shCon or shNEDD9(shN2) and treated with MLN8237 (M) or vehicle (V). (B). Quantification of BLI data as in (A), 3 independent experiments, n=6/group, plotted as fold growth of BLI mean radiance, +/−S.E.M. Two-way ANOVA p<0.0001 (shCon-V/shCon-M), (shN2-V/shN2-M), (shCon-V/shN2-V); p=0.0145 (shCon-M/shN2-M). (C). Quantification of BLI as in (A), dissected lungs; 3 independent experiments, n=6/group, mean photon flux +/−S.E.M. One-tailed Student’s t-test p<0.0001 (shCon-V/shCon-M), (shN2-V/shN2-M), p=0.009 (shCon-V/shN2-V), p=0.0335 (shCon-M/shN-M). (D). Model of NEDD9 action on AURKA stability and inhibitors competitive binding: I/III. Under normal physiological conditions with low NEDD9 expression, AURKA is not solely bound to NEDD9 and gets targeted by cdh1 and AURKA inhibitors; II/IV. Overexpression of NEDD9 leads to sequestering of AURKA by NEDD9 and limits cdh1 binding thus stabilizing AURKA and hampers the ability AURKA inhibitors to access ATP packet.