A Novel Time-Dependent CENP-E Inhibitor with Potent Antitumor Activity

Abstract

Centromere-associated protein E (CENP-E) regulates both chromosome congression and the spindle assembly checkpoint (SAC) during mitosis. The loss of CENP-E function causes chromosome misalignment, leading to SAC activation and apoptosis during prolonged mitotic arrest. Here, we describe the biological and antiproliferative activities of a novel small-molecule inhibitor of CENP-E, Compound-A (Cmpd-A). Cmpd-A inhibits the ATPase activity of the CENP-E motor domain, acting as a time-dependent inhibitor with an ATP-competitive-like behavior. Cmpd-A causes chromosome misalignment on the metaphase plate, leading to prolonged mitotic arrest. Treatment with Cmpd-A induces antiproliferation in multiple cancer cell lines. Furthermore, Cmpd-A exhibits antitumor activity in a nude mouse xenograft model, and this antitumor activity is accompanied by the elevation of phosphohistone H3 levels in tumors. These findings demonstrate the potency of the CENP-E inhibitor Cmpd-A and its potential as an anticancer therapeutic agent.

Fig 1. CENP-E inhibitor Cmpd-A induces chromosome misalignment during mitosis.

(A) Cmpd-A is a time-dependent inhibitor with an ATP competitive-like behavior. Red and black lines indicate the dose-dependent activity of Cmpd-A in the presence of low (1.25 μM) and high (500 μM) concentrations of ATP, respectively. The blue line indicates the activity of Cmpd-A with a high concentration of ATP, following 1 h of preincubation with CENP-E. The X-axis and Y-axis indicate the concentration of Cmpd-A and % inhibition of CENP-E ATPase activity, respectively. (B) Representative mitotic HeLa cells treated with Cmpd-A (200 nM) or DMSO. Arrows indicate misaligned chromosomes. Blue, green, and red signals indicate 4′,6-diamidino-2-phenylindole (DAPI)-stained DNA, α-tubulin, and CENP-B (kinetochores), respectively. (C) Quantitative analysis of mitotic morphology in the DMSO- or Cmpd-A-treated HeLa cells. The cells were treated for 3 h with 200 nM Cmpd-A or DMSO after dT block release. The DMSO- and Cmpd-A-treated mitotic cells (105 and 106 cells, respectively) were then counted. (D) Inter-kinetochore distance of aligned and misaligned chromosomes in HeLa cells treated with Cmpd-A or DMSO. Prometaphase (left) and metaphase (middle) cells were used as controls for misaligned and aligned chromosomes, respectively. The inter-kinetochore distance was measured between the outer kinetochore markers (HEC1) of individual chromosomes. Statistical analysis was performed using Student’s t-test. Differences were considered significant at P ≤ 0.05 (*) and P ≤ 0.01 (**).

Fig 2. Cmpd-A induces prolonged mitotic arrest accompanied by SAC activation.

(A) Cell cycle histogram of synchronous HeLa cells treated with Cmpd-A (200 nM) or DMSO. Cmpd-A was added at the G2 phase (7 h after dT block release), and the cells were collected at the indicated time points for FACS analysis. (B) pHH3 in synchronous HeLa cells treated with Cmpd-A (200 nM) or DMSO. Cmpd-A was added at the G2 phase (7 h after dT block release), and the cells were collected 12 h after dT block release. Representative results are shown. (C) pHH3 elevation in synchronous HeLa cells treated with Cmpd-A (200 nM) or DMSO. Cmpd-A was added at the G2 phase (7 h after dT block release), and the cells were collected at the indicated time points for FACS analysis of pHH3 staining. The graph indicates quantified pHH3-positive cells (mean ± standard deviation; n = 3). Red and blue lines indicate Cmpd-A- and DMSO-treated HeLa cells, respectively. (D) Immunoblotting of mitosis markers in synchronous HeLa cells treated with Cmpd-A or DMSO. The cells were treated with Cmpd-A or DMSO as described in Fig 2A and collected 12 h after dT block release for immunoblotting.

Fig 3. Cmpd-A exhibits potent antiproliferative activity in HeLa cells accompanied by the accumulation of pHH3.

(A) Correlation between antiproliferative activity and pHH3 elevation in Cmpd-A-treated HeLa cells. The line graph and red bars indicate the relative ATP amount and pHH3-positive cells (%), respectively. Data are presented as mean ± standard deviation (SD) (n = 3). HeLa cells were treated with Cmpd-A at the indicated concentrations. The cells were collected at 72 h and 24 h after treatment for cell viability and FACS analysis, respectively. The relative ATP amount was calculated based on chemiluminescence compared with the 0 nM chemiluminescence value (control). (B) Representative immunofluorescence of BubR1 in mitotic HeLa cells treated with Cmpd-A (200 nM). Arrows indicate misaligned chromosomes. Green, red, and blue signals indicate BubR1, CENP-B, and DAPI-stained DNA, respectively. (C) BubR1-dependent caspase-3/7 activation in response to Cmpd-A treatment. Twenty-four hours after siNS or siBubR1 treatment, the cells were treated with Cmpd-A at the indicated concentration for 24 h. Relative caspase-3/7 activities were calculated based on chemiluminescence compared with the 0 nM chemiluminescence value in each siRNA treatment. Statistical analysis was performed using Student’s t-test. Differences were considered significant at P < 0.05 (*) and P < 0.01 (**). The line plots represent mean ± SD. (n = 4). (D) BubR1-dependent antiproliferation in HeLa cells with Cmpd-A treatment. Twenty-four hours after siNS or siBubR1 treatment, the cells were treated with Cmpd-A at the indicated concentration for 24 h. Relative ATP amounts were calculated based on chemiluminescence compared with the 0 nM chemiluminescence value in each siRNA treatment. Statistical analysis was performed using Student’s t-test. Differences were considered significant at P < 0.05 (*) and P < 0.01 (**). The line plots represent mean ± SD (n = 4).

Fig 4. Time-dependent antiproliferative activity of Cmpd-A in HeLa cells.

(A) Experimental schemes to asses time-dependent antiproliferative activity of Cmpd-A. HeLa cells were treated with Cmpd-A at the indicated concentrations for 4, 8, 24, 48, and 72 h (red arrows) and then the cells were cultured in Cmpd-A-free medium for 72 h (black arrows). Cells were collected 72 h after treatment for cell viability analysis. (B) Time-dependent antiproliferative activity of Cmpd-A in HeLa cells. The relative ATP concentration was calculated based on chemiluminescence compared with the 0 nM chemiluminescence value (control). Data are presented as mean ± standard deviation (n = 8). (C) Quantitative RT-PCR analysis of CENP-E in cancer cell lines and human skin fibroblasts (MRC5). CENP-E expression ratios were quantified using GAPDH expression in each cell line as a control. Data are presented as mean ± standard deviation (n = 3).

Fig 5. Effect of Cmpd-A on mitotic arrest and antiproliferation in untransformed skin fibroblast MRC5 cells.

(A) Effect of Cmpd-A on mitotic arrest in siNS and siBubR1-transfected MRC5 cells. Twenty-four hours after siRNA transfection, MRC5 cells were treated with or without Cmpd-A (200 nM). Cells were collected 24 h after drug treatment for FACS assay. (B) Antiproliferative activity of Cmpd-A in siNS- (blue) and siBubR1-transfected (red) MRC5 cells. Twenty-four hours after siRNA transfection, MRC5 cells were treated with Cmpd-A at the indicated concentrations. Cells were collected 72 and 96 h after drug treatment for ATP assay. Relative ATP levels were calculated based on the chemiluminescence compared with the chemiluminescence value of 0 nM treatment for each. The line plots represent mean ± standard deviation (n = 3).

Fig 6. Cmpd-A exhibits potent antiproliferative activity in multiple cancer cell lines.

(A) Antiproliferative activity of Cmpd-A in multiple cancer cell lines. DU145, COLO205, NIH-OVCAR3, RKO, ES2, SK-OV3, PC-3, SW620, and CAPAN-2 cell lines were treated with Cmpd-A for 3 days at the indicated concentrations. The relative ATP concentration was calculated based on the chemiluminescence compared with the 0 nM chemiluminescence value (control). Data are presented as mean ± standard deviation (n = 3). (B) Correlation between the antiproliferative activity of Cmpd-A and CENP-E mRNA expression in cancer cell lines. The X and Y axes indicate the relative ATP level at 300 nM Cmpd-A treatment and CENP-E mRNA levels in the 14 indicated cancer cell lines, respectively. The relative ATP concentration was calculated based on chemiluminescence compared with the 0 nM chemiluminescence value (control) in each cell line. The raw data of CENP-E mRNA expression was downloaded from the Cancer Cell Line Encyclopedia (http://www.broadinstitute.org/ccle/data/browseData) and processed with MAS 5.0 algorithm.

Fig 7. PK/PD and antitumor efficacy of Cmpd-A in a COLO205 xenograft nude mouse model.

(A) Expression of pHH3 in COLO205 xenografts at the indicated time points after the intraperitoneal administration of Cmpd-A at a dose of 100 mg/kg. (B) Time-dependent PK and PD of Cmpd-A. The green, blue, and red lines indicate plasma concentration, tumor concentration, and tumor pHH3 intensity, respectively. The pHH3 intensity was quantified using the results shown in (A). (C) Immunohistochemistry of pHH3 in the tumor sections from COLO205 xenograft nude mice 24 h after the intraperitoneal administration of Cmpd-A at 100 mg/kg. Black arrows indicate misaligned chromosomes in sections from COLO205 xenografts treated with Cmpd-A. (D) Antitumor efficacy of Cmpd-A in the COLO205 xenograft nude mouse model. COLO205 xenografted nude mice intraperitoneally injected with Cmpd-A at 100 mg/kg or vehicle three times (at 0, 8, and 24 h) on the first day of the study. Representative tumors 8 days after the administration of vehicle or Cmpd-A are shown. (E) Efficacy data plotted as the mean tumor volume (mm³ ± standard error of the mean; n = 5) in COLO205 xenograft nude mice treated with Cmpd-A (red) or vehicle (black). Statistical analysis was performed using Student’s t-test. Differences were considered significant at P ≤ 0.05 (*) and P ≤ 0.01 (**). (F) Bodyweight comparison of COLO205 xenograft nude mice 8 days after the administration of Cmpd-A (red) or vehicle (black). Data are presented as mean ± standard deviation (n = 5). Statistical analysis was performed using Student’s t-test. Differences were considered significant at P ≤ 0.05 (*) and P ≤ 0.01 (**).