A novel Aurora-A kinase inhibitor MLN8237 induces cytotoxicity and cell-cycle arrest in multiple myeloma

Abstract

Aurora-A is a mitotic kinase that regulates mitotic spindle formation and segregation. In multiple myeloma (MM), high Aurora-A gene expression has been correlated with centrosome amplification and proliferation; thus, inhibition of Aurora-A in MM may prove to be therapeutically beneficial. Here we assess the in vitro and in vivo anti-MM activity of MLN8237, a small-molecule Aurora-A kinase inhibitor. Treatment of cultured MM cells with MLN8237 results in mitotic spindle abnormalities, mitotic accumulation, as well as inhibition of cell proliferation through apoptosis and senescence. In addition, MLN8237 up-regulates p53 and tumor suppressor genes p21 and p27. Combining MLN8237 with dexamethasone, doxorubicin, or bortezomib induces synergistic/additive anti-MM activity in vitro. In vivo anti-MM activity of MLN8237 was confirmed using a xenograft-murine model of human-MM. Tumor burden was significantly reduced (P = .007) and overall survival was significantly increased (P < .005) in animals treated with 30 mg/kg MLN8237 for 21 days. Induction of apoptosis and cell death by MLN8237 were confirmed in tumor cells excised from treated animals by TdT-mediated dUTP nick end labeling assay. MLN8237 is currently in phase 1 and phase 2 clinical trials in patients with advanced malignancies, and our preclinical results suggest that MLN8237 may be a promising novel targeted therapy in MM.

Figure 1

Inhibition of Aurora-A kinase (Thr288) phosphorylation in MM cells by MLN8237. (A) Expression of Aurora-A kinase protein in a panel of MM cell lines by immunoblotting of equal amounts of total protein (30 μg) with anti–Aurora-A kinase antibody. (B) Immunoblotting with anti–phospho (Thr288)-Aurora-A kinase antibody showed inhibition of Aurora-A kinase autophosphorylation (Thr288) in MM1.S and OPM1, with or without synchronizing with nocodazole. Expression of phospho (Thr288)-Aurora-A kinase protein was shown relative to total expression of Aurora-A kinase protein. Actin was used as a loading control. MM1.S (C) and OPM1 (D) MM cell lines were treated with DMSO or MLN8237 (0.5μM) for 24 hours and then stained with anti–phospho (Thr288)-Aurora-A kinase antibody (red), αtubulin (green), and DNA (blue). Overlapping localization is shown in the merged images. Arrow indicates Aurora-A autophosphorylation on Thr288 in the centrosome (original magnification ×40). Magnified single mitotic cell image is shown in the right panel. The number of mitotic cells with phosphorylated Aurora-A kinase (pT288) relative to total number of mitotic cells are shown. (E) To assess the effect of MLN8237 on Aurora-B kinase and mitosis, MM1.S and OPM1 cells were treated for 1, 24, and 48 hours with DMSO or MLN8237 (0.5μM). Flow cytometric (top panel) and Western immunoblot (bottom panel) analyses in representative MM cell lines show mitotic cells and active Aurora-B with increased DNA, costained with phospho-(Ser10)-histone H3 (Alexa Fluor 488) and PI.

Figure 2

MLN8237 inhibits cell viability and proliferation in MM cells in the presence or absence of BM stroma cells. A panel of MM cell lines and tumor cells obtained from patients with MM were treated with DMSO or increasing doses of MLN8237 (0.0001-4μM) for 48 to 72 hours. (A) Inhibition of cell line viability was determined by MTT assay. Inhibition of cell proliferation was determined by 3[H]thymidine incorporation assay in MM cell lines (B) and in tumor cells obtained from MM patients (C). MM1.S and RPMI8226 cell lines and tumor cells from patients with MM were cocultured with either BM stroma cells from patients with MM or cytokines (IL-6, 10 ng/mL; IGF-1, 25 ng/mL), in the absence or presence of MLN8237 (0.001-4μM) for 24 to 48 hours. (D) MM cell proliferation was measured by 3[H]thymidine incorporation assay in the cocultures with BM stroma cells. (E) MM cell viability was also assessed by MTT assay, in the absence or presence of IL-6 or IGF-1. Data represent mean ± SD of triplicate cultures. Statistical significance indicated (t test, 1-tailed distribution, P < .05).

Figure 2

MLN8237 inhibits cell viability and proliferation in MM cells in the presence or absence of BM stroma cells. A panel of MM cell lines and tumor cells obtained from patients with MM were treated with DMSO or increasing doses of MLN8237 (0.0001-4μM) for 48 to 72 hours. (A) Inhibition of cell line viability was determined by MTT assay. Inhibition of cell proliferation was determined by 3[H]thymidine incorporation assay in MM cell lines (B) and in tumor cells obtained from MM patients (C). MM1.S and RPMI8226 cell lines and tumor cells from patients with MM were cocultured with either BM stroma cells from patients with MM or cytokines (IL-6, 10 ng/mL; IGF-1, 25 ng/mL), in the absence or presence of MLN8237 (0.001-4μM) for 24 to 48 hours. (D) MM cell proliferation was measured by 3[H]thymidine incorporation assay in the cocultures with BM stroma cells. (E) MM cell viability was also assessed by MTT assay, in the absence or presence of IL-6 or IGF-1. Data represent mean ± SD of triplicate cultures. Statistical significance indicated (t test, 1-tailed distribution, P < .05).

Figure 3

MLN8237 induces accumulation of G₂/M phase and apoptosis in MM cells. (A) DNA profiles of MM cell lines, MM1.S and OPM1, as well as tumor cells from MM patient BM treated with DMSO or MLN8237 (0.5μM) for 24 hours were evaluated by flow cytometry. The percentage of cells in G₀G₁, G₂M, and S phases are shown. (B) MM cell lines, RPMI8226 and OPM1, as well as tumor cells from MM patient BM were cultured with DMSO or 0.5 to 1μM of MLN8237 for 24, 48, and 72 hours. Induction of apoptosis and cell death were determined by flow cytometric analysis in MM cells costained with annexin V and PI. (C) Induction of PARP, caspase-9, and caspase-3 cleavage by MLN8237 was measured by immunoblotting in MM cell lines. (D) Representative photomicrograph of β-galactosidase activation in MLN8237 (0.5μM, 48 hours) treated MM1.S cells. Arrows indicate activated β-galactosidase (blue) in cytoplasm. The senescent MM1.S cells (β-galactosidase active, blue stain) were visualized using a light microscope (original magnification ×20).

Figure 4

Molecular changes induced by MLN8237 in cell cycle–regulatory and antitumor pathways. Molecular changes induced by MLN8237 in cell cycle–regulatory and antitumor pathways were analyzed in MM cell lines. MM cell lines were cultured in the absence or presence of DMSO or MLN8237 (0.5-1μM) for various times. Protein level changes were determined by either Western blotting or flow cytometry. (A) Relative expression of negative cell cycle–regulatory molecules PP1α and PP2αA in MM cell lines treated with MLN8237 (0.5μM) for the indicated times. Pixel density of each protein band was measured using ImageJ software (1.37v; National Institutes of Health, http://rsb.info.nih.gov/ij/) and normalized with GAPDH expression. Fold expression per control for each band is shown. *Statistical significance (t test, one-tailed distribution, P < .05; n = 3 independent experiments). (B) Intracytoplasmic expression of phospho(Thr183/Tyr185) SAPK/JNK and phospho(Thr180/Tyr182) MAPK in MM1.S cells after 1-hour exposure to DMSO or MLN8237 (0.5μM). Dotted line indicates isotypic control; blue line, DMSO; and red line, MLN8237 (0.5μM for 1 hour). (C) Molecular changes induced by MLN8237 in p53 pathway were determined by Western blotting in MM cell lines. MLN8237 induced expression of p53, p27, and p21 in MM cell lines. MM cell line OPM1 does not express p21. GAPDH was used as a control to determine total protein expression. (D) Intracytoplasmic expression of cell-cycle checkpoint molecules phospho(Tyr15) Cdc2, phospho(Ser345) Chk1, and phospho(Thr68) Chk2 were analyzed in MM1.S cells after 1-hour exposure to DMSO or MLN8237 (0.5μM). Dotted line indicates isotypic control; blue line, DMSO; and red line, MLN8237 (0.5μM for 1 hour). (E) Molecular changes induced by MLN8237 in α-tubulin activation and formation were analyzed by Western blotting in MM cell lines. MLN8237 induced expression of α-tubulin in MM cell lines at early exposure time, whereas there was decreased expression of α-tubulin after 24 hours of MLN8237 treatment. GAPDH was used as a control to determine total protein expression.

Figure 4

Molecular changes induced by MLN8237 in cell cycle–regulatory and antitumor pathways. Molecular changes induced by MLN8237 in cell cycle–regulatory and antitumor pathways were analyzed in MM cell lines. MM cell lines were cultured in the absence or presence of DMSO or MLN8237 (0.5-1μM) for various times. Protein level changes were determined by either Western blotting or flow cytometry. (A) Relative expression of negative cell cycle–regulatory molecules PP1α and PP2αA in MM cell lines treated with MLN8237 (0.5μM) for the indicated times. Pixel density of each protein band was measured using ImageJ software (1.37v; National Institutes of Health, http://rsb.info.nih.gov/ij/) and normalized with GAPDH expression. Fold expression per control for each band is shown. *Statistical significance (t test, one-tailed distribution, P < .05; n = 3 independent experiments). (B) Intracytoplasmic expression of phospho(Thr183/Tyr185) SAPK/JNK and phospho(Thr180/Tyr182) MAPK in MM1.S cells after 1-hour exposure to DMSO or MLN8237 (0.5μM). Dotted line indicates isotypic control; blue line, DMSO; and red line, MLN8237 (0.5μM for 1 hour). (C) Molecular changes induced by MLN8237 in p53 pathway were determined by Western blotting in MM cell lines. MLN8237 induced expression of p53, p27, and p21 in MM cell lines. MM cell line OPM1 does not express p21. GAPDH was used as a control to determine total protein expression. (D) Intracytoplasmic expression of cell-cycle checkpoint molecules phospho(Tyr15) Cdc2, phospho(Ser345) Chk1, and phospho(Thr68) Chk2 were analyzed in MM1.S cells after 1-hour exposure to DMSO or MLN8237 (0.5μM). Dotted line indicates isotypic control; blue line, DMSO; and red line, MLN8237 (0.5μM for 1 hour). (E) Molecular changes induced by MLN8237 in α-tubulin activation and formation were analyzed by Western blotting in MM cell lines. MLN8237 induced expression of α-tubulin in MM cell lines at early exposure time, whereas there was decreased expression of α-tubulin after 24 hours of MLN8237 treatment. GAPDH was used as a control to determine total protein expression.

Figure 5

MLN8237 induces inhibition of tumor growth in MM xenograft murine model. MM1.S cells were injected subcutaneously into SCID mice. After 2 weeks of tumor engraftment, mice were treated orally with vehicle or MLN8237 (7.5 mg/kg, 15 mg/kg, and 30 mg/kg) for 21 consecutive days. Tumor size, survival, and changes in weight were measured. (A) Tumor growth was significantly inhibited (t test, one-tailed distribution, P < .007). Mean tumor volumes (cubic millimeter [volume]) ± SD (n = 8-10 animals/group) are shown from the beginning of treatment. (B) The overall survival after the injection of tumor cells was significantly increased (log-rank, Mantel-Cox test, P < .03; log-rank test for trend, P < .005) in treated compared with control animals. (C) Treatment with MLN8237 was not associated with significant changes in weight (t test, one-tailed distribution, P = not significant). Mean weight (grams) ± SD (n = 8-10 animals/group) is shown from the beginning of treatment. (D) Induction of apoptosis and cell death in tumors excised from vehicle or MLN8237 (7.5 mg/kg, 15 mg/kg, and 30 mg/kg) treated animals are shown by TUNEL assay. Photomicrographs show apoptotic cells indicated by arrows (brown) with nucleus (blue; original magnification ×20) using light microscopy. The average number of TUNEL-positive cells was determined in 93 mm² area.