Targeting aurora kinase with MK-0457 inhibits ovarian cancer growth

Abstract

Purpose: The Aurora kinase family plays pivotal roles in mitotic integrity and cell cycle. We sought to determine the effects of inhibiting Aurora kinase on ovarian cancer growth in an orthotopic mouse model using a small molecule pan-Aurora kinase inhibitor, MK-0457.

Experimental design: We examined cell cycle regulatory effects and ascertained the therapeutic efficacy of Aurora kinase inhibition both alone and combined with docetaxel using both in vitro and in vivo ovarian cancer models.

Results: In vitro cytotoxicity assays with HeyA8 and SKOV3ip1 cells revealed >10-fold greater docetaxel cytotoxicity in combination with MK-0457. After in vivo dose kinetics were determined using phospho-histone H3 status, therapy experiments with the chemosensitive HeyA8 and SKOV3ip1 as well as the chemoresistant HeyA8-MDR and A2780-CP20 models showed that Aurora kinase inhibition alone significantly reduced tumor burden compared with controls (P values<0.01). Combination treatment with docetaxel resulted in significantly improved reduction in tumor growth beyond that afforded by docetaxel alone (P <or= 0.03). Proliferating cell nuclear antigen immunohistochemistry revealed that MK-0457 alone and in combination with docetaxel significantly reduced cellular proliferation (P values<0.001). Compared with controls, treatment with MK-0457 alone and in combination with docetaxel also significantly increased tumor cell apoptosis by approximately 3-fold (P<0.01). Remarkably, compared with docetaxel monotherapy, MK-0457 combined with docetaxel resulted in significantly increased tumor cell apoptosis.

Conclusions: Aurora kinase inhibition significantly reduces tumor burden and cell proliferation and increases tumor cell apoptosis in this preclinical orthotopic model of ovarian cancer. The role of Aurora kinase inhibition in ovarian cancer merits further investigation in clinical trials.

Fig. 1

A, Aurora-A kinase-mediated autophosphorylation in HeyA8 and SKOV3ip1ovarian cancer cells is selectively inhibited using MK-0457 (as low as 10 nmol/L) as shown by decreased levels of phospho-Aurora-A (Thr²⁸⁸). Inhibition of endogenous substrates, histone H3 (Ser¹⁰) and Cenp-A (Ser⁷), is also elicited by the Aurora-A kinase inhibitor, MK-0457, using doses as low as 10 nmol/L. B, time of onset of Aurora-A kinase inhibition is examined in a time-kinetic experiment using the lowest dose necessary to block phosphorylation of Aurora-A kinase as determined above. MK-0457 decreases Aurora-A phosphorylation (relative to the total amount of Aurora-A) in HeyA8 cells.

Fig. 2

A, cytotoxicity (MTT) graphs depicting the inhibitory concentration at 50% of MK-0457 for HeyA8 and SKOV3ip1cell lines in vitro. HeyA8 (1,000 cells per well) and SKOV3ip1 (2,000 cells per well) were treated with MK-0457 for 96 h. Cell viability was determined using MTT method. Cell viability assay for HeyA8 and SKOV3ip1cells treated with docetaxel alone and after pretreatment with MK-0457. B, cell cycle analysis by flow cytometry of HeyA8 (solid columns) and SKOV3ip1 (open columns) cells shows profound and sustained G₂-M arrest as soon as12 h after inhibition of Aurora-A kinase. Bars, SD. *, P < 0.02, compared with controls. C, marked increases in ≥4N cells were seen 24 h after MK-0457 exposure in HeyA8 cells. D, inhibition of Aurora-A kinase in HeyA8 and SKOV3ip1cells with MK-0457 treatment also results in marked increases in apoptosis relative to untreated controls as indicated by the sub-G₁fraction on flow cytometry. Bars, SD. *, P = 0.02, compared with controls.

Fig. 3

A, Aurora-A kinase inhibition with MK-0457 decreases phospho-histone H3 (P-HH3) in formed ovarian tumors (HeyA8) in a time- and dose-dependent manner. Compared with treatment with vehicle alone, MK-0457 treatment (25 or 50 mg/kg) decreased the number of cells with phospho-histone H3 by 40% to 52% in xenograft ovarian tumors collected from athymic mice as determined by immunohistochemical analysis (original magnification, ×100). *, P < 0.001, versus control within the same time point. B, mean tumor weights and distribution from MK-0457 therapy experiments. Mice inoculated with chemosensitive tumors, HeyA8 and SKOV3ip1, as well as taxane- and platinum-resistant tumors, HeyA8-MDR and A2780-CP20, respectively, were randomized into one of four treatment groups: (a) vehicle alone (control), (b) MK-0457 alone (50 mg/kg), (c) traditional cytotoxic docetaxel (35 μg or 50 μg once weekly, depending on cell line) or cisplatin (160 μg once weekly), or (d) MK-0457 plus docetaxel or cisplatin depending on the cell line. Animals from all groups were sacrificed when control animals were moribund (∼3-4 wk after initiating therapy, depending on the cell line used). All tumors were harvested and the pattern of spread was noted. Tumor weights and the number of nodules were recorded. Columns, mean tumor weights; bars, SE. *, P < 0.01, versus control; †, P < 0.05, versus control; ‡, P ≤ 0.01, versus docetaxel or cisplatin, depending on tumor model.

Fig. 4

A, proliferative index and tumor cell apoptosis of HeyA8 tumors with representative PCNA immunohistochemistry (original magnification, ×100) and terminal deoxynucleotidyl transferase– mediated dUTP nick end labeling (original magnification, ×100) sections, respectively, collected at necropsy. *, P < 0.01, versus control; †, P < 0.05, versus docetaxel. B, real-time reverse transcription-PCR analysis of genes induced by MK-0457 using both human and mouse gene-specific TaqMan probes. Fold induction (in log₁₀) represents the ratio between the means of the expression values of four MK-0457-treated mice and five untreated vehicle controls. Mouse, mouse-specific TaqMan probes; Human, human-specific TaqMan probes; H, human; M, murine.