Intervention of cancer cell mitosis by antitubulin drugs is among the most effective cancer chemotherapies. However, antitubulin drugs have dose-limiting side effects due to important functions of microtubules in resting normal cells and are often rendered ineffective by rapid emergence of resistance. Antimitotic agents with different mechanisms of action and improved safety profiles are needed as new treatment options. Mitosis-specific kinesin Eg5 represents an attractive anticancer target for discovering such new antimitotic agents, because Eg5 is essential only in mitotic progression and has no roles in resting, nondividing cells. Here, we show that a novel selective Eg5 inhibitor, LY2523355, has broad target-mediated anticancer activity in vitro and in vivo. LY2523355 arrests cancer cells at mitosis and causes rapid cell death that requires sustained spindle-assembly checkpoint (SAC) activation with a required threshold concentration. In vivo efficacy of LY2523355 is highly dose/schedule-dependent, achieving complete remission in a number of xenograft tumor models, including patient-derived xenograft (PDX) tumor models. We further establish that histone-H3 phosphorylation of tumor and proliferating skin cells is a promising pharmacodynamic biomarker for in vivo anticancer activity of LY2523355. Mol Cancer Ther; 14(11); 2463–72. ©2015 AACR.

During G2–M progression, duplicated centrosomes are separated by the microtubule-based bipolar mitotic spindle through a complex process involving chromosomes, mitotic kinases, and microtubule-associated proteins (1–3). The bipolar mitotic spindle orchestrates the segregation of replicated chromosomes accurately and equally into two daughter cells to maintain genome stability during cell division. Antimitotic drugs such as antitubulins, including taxanes and vincas, are among the most effective cancer therapies in current clinical use (4). However, these antitubulins also have severe side effects as they disrupt the normal microtubule functions in resting cells, including neurons, which may lead to neuropathic disorders (5–7). Furthermore, emerging resistance to these antimitotic agents renders them virtually ineffective (8).

Eg5, a member of BIMC family of kinesins, is a plus-end directed homotetrameric kinesin (1, 9, 10). It plays a vital role in chromosome segregation and bipolar mitotic spindle organization and elongation by sliding antiparallel microtubules away from each other (11–14). Its function in bipolar spindle formation was found to be evolutionarily conserved (14–16). Eg5 is tightly regulated by mitotic kinases and cell-cycle checkpoint controls (14, 15, 17). Overexpression of Eg5 in transgenic mice caused mitotic defects leading to genome instability and tumorigenesis (18, 19). Unlike microtubules that are present throughout the cell cycle and have important biologic functions in many nondividing cells, Eg5 is expressed and active only during mitosis (20), thus making it an attractive target for discovering new antimitotic drugs with a different mode of action and an anticipated improved safety profile (5, 21). The discovery of small-molecule Eg5 inhibitors spurred strong interest in their development as anticancer agents (6, 22–24).

Inhibition of Eg5 with small molecules disrupts mitotic spindle formation inducing spindle assembly checkpoint (SAC)-mediated mitotic arrest (25) and apoptosis in cancer cells, including taxol-resistant cancer cells (22, 25–27). Recent studies have demonstrated that both activation of SAC and subsequent mitotic slippage are required for apoptosis induction by the Eg5 inhibitors (25, 28). However, how activation of SAC and mitotic slippage are linked to induction of apoptosis is still not well understood.

Eg5 inhibitors generally show excellent anticancer activities in a range of xenograft tumor models (21). However, clinical experiences with several Eg5 inhibitors that entered phase I and II studies have been rather disappointing, resulting in termination of their clinical studies (6). The lack of clinically meaningful activities for these Eg5 inhibitors may be a result of their poor pharmacokinetics or dose selection, which was not optimized on the basis of an understanding of pharmacokinetic and pharmacodynamic relationships. Moreover, early Eg5 inhibitors were developed following a dosing paradigm established for antitubulins. The fact that cell-type variation and the length of time in sustained mitotic arrest contribute to differences in induction of apoptosis by Eg5 inhibitors (25, 29–31) reinforce the need for careful dose and schedule selection.

We describe here the discovery and characterization of LY2523355, a selective allosteric inhibitor of human Eg5, derived through an extensive structure–activity relationship program from a precursor compound identified by a phenotype-based screening for mitotic arrest of HCT116 cells (32). We demonstrate that inhibition of Eg5 in cancer cells by LY2523355 causes rapid apoptosis following SAC-dependent mitotic arrest. We further show that abrogation of SAC by inactivation of cyclin-dependent kinase 1 (CDK1) or knockdown of mitotic arrest deficient 2 (Mad2) causes rapid exit from mitosis and antagonizes induction of apoptosis by LY2523355. LY2523355 shows highly schedule-dependent, broad-spectrum anticancer activity against tumor models, including patient-derived xenograft (PDX) models. We establish that histone H3 phosphorylation of tumor proliferating skin cells is a promising pharmacodynamic marker for LY2523355 activity.

Cancer cell lines, reagents, and treatments

Cancer cell lines (A2780, A549, ARH-77, COLO-205, Calu-6, DU 145, HCT-116, HT-29, Hep 3B2.1-7, LN-18, LN-229, M059K, MDA-MB-231, NAMALWA, NCI-H460, NIH:OVCAR-3, RPMI 8226, SK-BR-3, SK-OV-3, SW620, U-87 MG, HeLa) used for in vitro anticancer and mechanism studies at Eli Lilly and Company were all obtained from ATCC between September 2005 to July 2007, unless otherwise stated. The cell lines were authenticated by short tandem repeat (STR) assay and compared against existing reference standards at RADIL (now IDEXX Bioresearch) and were maintained in media per ATCC instructions and as described previously (33). Sources of the cancer cell lines used for the in vivo xenograft tumor model studies at Kyowa Hakko Kirin as summarized in Table 1 were as follows: HCT-116 cells were obtained in 1998, NCI-H460 cells in 1999, NCI-H929 cells in 2001, and Colo-205, Calu-6, SK-OV-3 cells in 2002 from ATCC; EBC-1 cells were obtained from JCRB Cell Bank in 2002; KPL-4 and KMS-11 cells were obtained from Kawasaki Medical University in 1997 and 2001, respectively; PC-14 and PC-14/CDDP cells were obtained from National Cancer Center Hospital in 2001 and 2002, respectively; A2780 cells were obtained from Japanese Foundation for Cancer Research in 2002; RMG-1 cells were obtained from Ishiwata Hospital in 1999 and MX-1 cells were obtained from Central Institute for Experimental Animals in 1998. These cancer cell lines at Kyowa Hakko Kirin were not authenticated and were maintained as previously described (32).

Table 1.

Summary of the sensitive and insensitive tumors to LY2523355

Number of sensitive tumors (%)Number of insensitive
Type of tumorRegressedaSuppressedbtumorsc (%)
In xenograft tumor models with established cancer cell linesd 
 Colorectal 2/2 0/2 0/2 
 Non–small cell lung 3/5 2/5 0/5 
 Ovarian 1/3 1/3 1/3 
 Breast 1/2 0/2 1/2 
 Multiple myeloma 2/2 0/2 0/2 
 Overall 9/14 (64) 3/14 (21) 2/14 (14) 
In PDX tumor models 
 Non–small cell lung 7/12 4/12 1/12 
 Small cell lung 3/4 1/4 0/4 
 Colorectal 1/4 2/4 1/4 
 Gastric 1/3 2/3 0/3 
 Melanoma 1/5 2/5 2/5 
 Pancreatic 0/1 1/1 0/1 
 Breast 1/1 0/1 0/1 
 Brain 0/1 0/1 1/1 
 Pleuramesothelioma 0/1 1/1 0/1 
 Testis 1/1 0/1 0/1 
 Bladder 1/1 0/1 0/1 
 Ovarian 0/1 0/1 1/1 
 Sarcoma 0/1 0/1 1/1 
 Liver 0/1 1/1 0/1 
 Overall 16/37 (43) 14/37 (38) 7/37 (20) 
Number of sensitive tumors (%)Number of insensitive
Type of tumorRegressedaSuppressedbtumorsc (%)
In xenograft tumor models with established cancer cell linesd 
 Colorectal 2/2 0/2 0/2 
 Non–small cell lung 3/5 2/5 0/5 
 Ovarian 1/3 1/3 1/3 
 Breast 1/2 0/2 1/2 
 Multiple myeloma 2/2 0/2 0/2 
 Overall 9/14 (64) 3/14 (21) 2/14 (14) 
In PDX tumor models 
 Non–small cell lung 7/12 4/12 1/12 
 Small cell lung 3/4 1/4 0/4 
 Colorectal 1/4 2/4 1/4 
 Gastric 1/3 2/3 0/3 
 Melanoma 1/5 2/5 2/5 
 Pancreatic 0/1 1/1 0/1 
 Breast 1/1 0/1 0/1 
 Brain 0/1 0/1 1/1 
 Pleuramesothelioma 0/1 1/1 0/1 
 Testis 1/1 0/1 0/1 
 Bladder 1/1 0/1 0/1 
 Ovarian 0/1 0/1 1/1 
 Sarcoma 0/1 0/1 1/1 
 Liver 0/1 1/1 0/1 
 Overall 16/37 (43) 14/37 (38) 7/37 (20) 

aV/V0 min < 1.

bT/C min < 0.5.

cT/C min = 0.5–1.

dThe cancer cell lines used are: Colo-205, HCT-116, EBC-1, Calu-6, NCI-H460, PC-14, PC-14/CDDP, A2780, SK-OV-3, RMG-1, KPL-4, MX-1, KMS-11, NCI-H929.

Mouse anti-BubR1 antibody was purchased from Becton Dickinson Transduction Labs. Mouse anti-α-tubulin, actin, and anti-Mad2 antibodies were from Sigma-Aldrich. Rabbit anti-pericentrin antibody was from Abcam. PARP1, Mcl1, Bcl2, p38, and anti-cleaved PARP (Asp214) antibodies were from Cell Signaling Technologies. Rabbit anti-phospho-histone H3 (Ser10) and phospho-histone H2A.X (Ser139) and cyclinB1 antibodies were purchased from Millipore. Mouse anti-GAPDH antibody was from Meridian Life Science Inc.. Mad2 ON-TARGETplus SMART pool siRNA and DharmaFECT1 transfection reagent were purchased from Dharmacon. Propidium iodide was purchased from Invitrogen. Paclitaxel, docetaxel, CPT-11, vinorelbine, and cisplatin were from Sigma Aldrich. RO3306 was from Santa Cruz Biotechnology. The Eg5 inhibitor LY2523355 was synthesized internally.

RNAi, compound handling and treatment, immunoblotting, flow cytometry, and immunofluorescence microscopy were performed as previously described (32, 33). Assays for ATPase activity of kinesins and in vitro microtubule polymerization and depolymerization were also done as previously described (32).

Mitotic index and apoptosis measurement

Cancer cells were plated in poly-d-lysine coated 96-well plates (BD cat. no. 356640) and incubated overnight. Cells were then treated with indicated concentrations of LY2523355 for various time periods. Cells were then fixed with 3.7% formaldehyde in PBS for 45 minutes or 1× Prefer fixative solution for 30 minutes, at room temperature, and permeabilized with cold methanol for 10 minutes and then 0.2% Triton X-100 in PBS for 10 minutes. Cells were washed three times with PBS. Cells were then incubated with 100 μg/mL DNase-free RNase and 10 μg/mL of propidium iodide for 1 hour and scanned with Acumen for mitotic index (MI) measurement based on DNA condensation as percentage of cells with condensed DNA. For MI based on histone H3 phosphorylation or apoptosis analysis, cells were incubated overnight at 4°C with anti-phospho-histone H3 antibody or anti-phospho-histone H2AX at 1:1,000 dilution with 5% bovine serum albumin (BSA) in PBS, respectively. Cells were washed three times with 0.2% Triton-X 100 in PBS and incubated with Alexa 488 secondary antibody (1:1,000 in PBS–2% BSA, Invitrogen) for 60 minutes at room temperature. Cells were washed three times again with PBS and stained for 15 minutes in PBS containing 10 μg/mL propidium iodide (Molecular Probes, Invitrogen) and 100 μg/mL RNaseA (Molecular Probes, Invitrogen). The stained cells were scanned with an Acumen Explorer eX3 microplate cytometer (TTP LabTech). The results were expressed as percentage of cells positive for phospho-histone H3-Ser10 or phospho-histone H2AX.

Caspase-3/7 activity after 48-hour treatment was measured using Apo-ONE Homogeneous Caspase-3/7 Assay (cat. no. G7791, Promega) and cell viability after 72-hour treatment was measured using CellTiter-Glo Luminescent Cell Viability Assay (cat. no. G7573, Promega), following the manufacturer's instruction.

Animal models and in vivo pharmacology

Primary human-tumor xenograft models were established and maintained in nude mice as previously described (34). Antitumor efficacy in subcutaneous xenograft tumor-bearing mice with 10 mice per treatment group from either established cancer cell lines or fragments of human tumor explants was evaluated as tumor volume by serial caliper measurements and was calculated as described previously (32, 34). The p388 syngeneic tumor model was developed for high-content imaging analysis using female BDF1 mice, weighing 20 to 23 g, which were purchased from Charles River and acclimated in-house for one week before their use in experiments. p388 murine lymphocytic leukemia cells, obtained from the NCI in 2006, were authenticated by STR assay as described above and maintained in RPMI1640 medium containing 10% FBS. For inoculation, the cells were washed with serum-free medium three times and 1.25 million cells were implanted by intraperitoneal injection into mice. On day 5 after implantation, mice were treated with LY2523355, via either intravenous bolus or intravenous infusion at appropriate doses and durations. Mice were euthanized and the ascitic (intraperitoneal) fluid containing the p388 tumor cells was drawn and analyzed by acumen, flow cytometry, and TUNEL assays for phospho-histone H3, G2–M, and apoptosis. For pharmacokinetic study, we collected the blood samples via cardiac puncture and generated plasma with EDTA and determined LY2523355 exposure in the plasma.

All in vivo studies were reviewed by Ethical Committees of Eli Lilly, Kyowa Hakko, and Oncotest, respectively, and were performed in accordance with the guidelines stated in the International Guiding Principles for Biomedical Research Involving Animals developed by the Council for International Organization of Medical Sciences.

Immunohistochemistry

Xenograft tumors and mouse skin (ears) were fixed in 10% neutral buffered formalin for IHC studies. The sections were trimmed, routinely processed, and embedded in paraffin. Three micrometer sections were immunohistochemically labeled for histone H3-serine 10 phosphorylation using a rabbit polyclonal primary antibody (Upstate, Millipore) without antigen retrieval and with 3,3′-diaminobenzidine as chromagen. Sections were counterstained with hematoxylin.

Threshold concentration effect on cell-cycle–dependent mitotic arrest by LY2523355

Biochemical studies established that LY2523355 is a potent and selective inhibitor of the human Eg5 (Fig. 1A and Supplementary Fig. S1). Cell-cycle analysis showed that LY2523355 caused a dose-dependent increase of cells with 4N DNA content (Supplementary Fig. S2A). Immunofluorescence microscopy further revealed that LY2523355 arrested proliferating normal and cancerous human cells in mitosis with monopolar spindles in vitro (Supplementary Fig. S2B and S2C). Interestingly, inactivation of Eg5 by LY2523355 also inhibited Eg5 localization to the mitotic spindle, suggesting spindle localization of Eg5 is an active process and requires its functional activity (Supplementary Fig. S2C).

Figure 1.

LY2523355 kills cancer cells during mitotic arrest. A, the chemical structure of LY2523355. B, dose-dependent mitotic arrest of HCT116 cells by LY2523355. MI was determined by fluorescent microscopy based on DNA condensation 18 hours after LY2523355 addition. C, mitotic block by LY2523355 is time dependent with a threshold concentration for maximal activity. MI was determined using Acumen high content technology. D, cell-cycle arrest at G2–M and cell killing by LY2523355 were closely correlated from 12 to 24 hours after LY2523355 addition. HCT116 cells were treated with 25 nmol/L LY2523355 for various time intervals and then LY2523355 was removed from the growth medium. Cell-cycle profiles of the treated HCT116 cells were analyzed (top graph) and the ability of the treated cells to grow after LY2523355 removal was measured in real time using Incucyte (bottom graph).

Figure 1.

LY2523355 kills cancer cells during mitotic arrest. A, the chemical structure of LY2523355. B, dose-dependent mitotic arrest of HCT116 cells by LY2523355. MI was determined by fluorescent microscopy based on DNA condensation 18 hours after LY2523355 addition. C, mitotic block by LY2523355 is time dependent with a threshold concentration for maximal activity. MI was determined using Acumen high content technology. D, cell-cycle arrest at G2–M and cell killing by LY2523355 were closely correlated from 12 to 24 hours after LY2523355 addition. HCT116 cells were treated with 25 nmol/L LY2523355 for various time intervals and then LY2523355 was removed from the growth medium. Cell-cycle profiles of the treated HCT116 cells were analyzed (top graph) and the ability of the treated cells to grow after LY2523355 removal was measured in real time using Incucyte (bottom graph).

Close modal

LY2523355-treated HCT116 cells also showed a dose-dependent increase in MI based on phospho-histone H3 or DNA condensation (Fig. 1B). Furthermore, there was a time-dependent increase in MI in LY2523355-treated HCT116 cells reaching a maximum at 12 hours (Fig. 1C). Importantly, LY2523355 at 10 nmol/L was sufficient to cause maximal mitotic arrest at all the time points and concentrations higher than 10 nmol/L had no additional effects either on the timing or the magnitude of mitotic arrest. Mitotic block by LY2523355 was closely correlated with its antiproliferation activity in cell culture as demonstrated by complete inhibition of HeLa cell growth at approximately 10 nmol/L (Supplementary Fig. S2D). Indeed, LY2523355 showed potent, broad-spectrum anticancer activity in vitro at low concentrations (IC50 values ranging from 0.4 nmol/L to14 nmol/L) when tested against a panel of 21 cancer cell lines, which was further confirmed in proliferation assays with the NCI-60 panel (Supplementary Tables S1 and S2).

LY2523355 kills cancer cells as a result of mitotic arrest

LY2523355 at 10 nmol/L or above causes maximal mitotic arrest with monopolar spindle (Fig. 1B and Supplementary Fig. S2B and S2C). A substantial sub-G1 fraction of cells was observed 48 hours after LY2523355 treatment (Supplementary Fig. S2A), indicating that cell death occurred. To establish the causal relationship between mitotic arrest and cell death, real-time measurement of HCT116 cell growth after LY2523355 treatment and subsequent withdrawal was performed. LY2523355 caused maximal mitotic arrest and cell death 12 to 24 hours after addition and cells treated for less than 12 hours were able to resume growth after withdrawal of LY2523355 (Fig. 1D). The data thus strongly suggest that cell killing by LY2523355 is linked to cell-cycle arrest at mitosis. Furthermore, it was observed that LY2523355-induced mitotic arrest and apoptosis were closely associated with each other in a time- and dose-dependent manner (Fig. 1C and Supplementary Fig. S2E).

A role for SAC activation in cancer cell killing by LY2523355

DNA damage and cell death in LY2523355-treated HeLa cells, as indicated by histone H2AX phosphorylation and cleaved PARP (Fig. 2A), occurred in mitotic cells but not in interphase cells (adherent). This confirmed that cell killing by LY2523355 actually occurred during mitotic arrest. Mitotic arrest was indicated by phosphorylation of histone H3 and the spindle checkpoint kinase, budding uninhibited by benomyl Related-1 (BubR1). Immunofluorescence microscopy of mitotic cells further revealed that mitotic cells had condensed DNA, phosphorylated histone H3, and also phosphorylated histone H2AX (Supplementary Fig. S3A).

Figure 2.

Sustained activation of SAC is required for rapid cell death by LY2523355. A, Western blot analysis of mitotic and apoptotic markers in total cell lysates of HeLa cells treated with 25 nmol/L LY2523355 for 24 hours. Mitotic cells were collected after gentle shaking and cells at interphase that remained attached to culture flasks were then collected separately as the adherent fraction. B, time course of 25 nmol/L LY2523355 treatment in HeLa cells. Various mitotic and apoptotic markers were analyzed from total cell lysates by immunoblotting. The p38 MAPK protein was used here as a loading control. C, abrogation of SAC caused rapid exit from mitosis and inhibited cell death by LY2523355. HeLa cells were treated with 25 nmol/L LY2523355 for 12 hours before addition of RO3306, a selective CDK1 inhibitor to abrogate SAC. Total cell lysates were analyzed by immunoblotting.

Figure 2.

Sustained activation of SAC is required for rapid cell death by LY2523355. A, Western blot analysis of mitotic and apoptotic markers in total cell lysates of HeLa cells treated with 25 nmol/L LY2523355 for 24 hours. Mitotic cells were collected after gentle shaking and cells at interphase that remained attached to culture flasks were then collected separately as the adherent fraction. B, time course of 25 nmol/L LY2523355 treatment in HeLa cells. Various mitotic and apoptotic markers were analyzed from total cell lysates by immunoblotting. The p38 MAPK protein was used here as a loading control. C, abrogation of SAC caused rapid exit from mitosis and inhibited cell death by LY2523355. HeLa cells were treated with 25 nmol/L LY2523355 for 12 hours before addition of RO3306, a selective CDK1 inhibitor to abrogate SAC. Total cell lysates were analyzed by immunoblotting.

Close modal

A time course study of multiple mitotic and apoptotic cellular biomarkers showed that mitotic arrest clearly preceded apoptosis (Fig. 2B). LY2523355 treatment caused a dramatic increase in phosphorylation of histone H3, B-cell lymphoma 2 (Bcl2), and BubR1, and accumulation of cyclin B from 4 to 16 hours (Fig. 2B). After peaking at 16 hours, phosphorylation of histone H3, Bcl2, and BubR1, and protein level of cyclin B began to decrease substantially and no cyclin B was detectable 48 hours after LY2523355 treatment. Interestingly, the level of BubR1 protein also markedly decreased 24 hours after LY2523355 treatment. In contrast, apoptosis, indicated by PARP1 cleavage and a decrease in the level of antiapoptotic proteins Bcl2 and Mcl1, occurred 16 hours after LY2523355 treatment and continued up to 48 hours (Fig. 2B). Mcl1 appeared to be most sensitive to LY2523355 treatment; its protein level decreased appreciably at 8 hours and completely disappeared 24 hours after treatment (Fig. 2B). The marked decrease in multiple mitotic markers may indicate a mitotic slippage after a sustained LY2523355-mediated mitotic arrest.

CDK1 kinase activity is required to maintain activation of SAC during mitosis (35). Therefore, to investigate a potential role of SAC in apoptotic cell death caused by LY2523355, we used a selective CDK1 inhibitor, RO3306, to abrogate SAC activation and force cells to exit mitosis rapidly. Upon addition of RO3306 to mitotic HeLa cells 12 hours after pretreatment with LY2523355, cells resumed flat interphase morphology, reattached to the plate quickly, and exited from mitosis. Consistent with rapid mitotic exit, dephosphorylation of histone H3 occurred rapidly upon addition of RO3306 (Fig. 2C). Addition of RO3306 strongly inhibited increase in both PARP1 cleavage and phospho-histone H2AX normally associated with LY2523355 treatment (Fig. 2C). Interestingly, while rapid dephosphorylation of BubR1 (mobility shift) indicates SAC abrogation by RO3306 as expected, BubR1 degradation was also prevented. Taken together, the data support an important role of activated SAC in cancer cell killing by Eg5 inhibitors.

To independently confirm the role of SAC in cell killing by LY2523355, we silenced SAC protein Mad2 by siRNA. As compared with the parental cells, cells with silenced Mad2 failed to be arrested at mitosis and consequently had no activated caspase activity, indicating that SAC has an essential role in mitotic arrest and subsequent cell killing by LY2523355 (Supplementary Fig. S3B–S3D).

Schedule-dependent antitumor activity in vivo by LY2523355

As shown in Fig. 3A, the in vivo antitumor activity of LY2523355 was highly schedule-dependent, although all treatment groups received the same total amount of the compound. Strong antitumor efficacy was observed when LY2523355 was administered with a time interval of <4 days between dosing cycles and the best antitumor activity was observed with daily dosing for 3 consecutive days. However, animals with 3-consecutive day dosing also had significant body weight loss and some mortality. In contrast, LY2523355, when administered once every 7 days, had no significant antitumor activity. It was noted that LY2523355 administered once every 4 days not only showed excellent antitumor activity but also had minimal toxicity (no significant body weight loss; Fig. 3A). Therefore, this dosing schedule was chosen to explore antitumor activity by LY2523355 against a variety of mouse xenograft tumor models.

Figure 3.

Anticancer efficacy of LY2523355 in xenograft tumor models. A, demonstration of schedule-dependent efficacy by LY2523355 in Colo205 tumor models. When tumors reached an average size of 150 mm3, LY2523355 was administered at 9.6 mg/kg to the tumor-bearing mice with 5 mice per treatment group by 1-hour intravenous infusion with various dosing schedules for a total of 3 dosing cycles ranging from daily dosing (q1d) to once every 7 days (q7d). Tumor size and mouse body weight were measured every few days for more than a month after the start of dosing. LY2523355, when dosed once every 4 days, caused marked tumor shrinkage with minimal toxicity seen as body-weight loss. B, comparison of anticancer efficacy by LY2523355 with other chemotherapeutic agents in preclinical xenograft tumor models. Mice bearing non–small cell lung (ECB-1; 160 mm3) or ovarian (A2780) xenograft tumors of an average size of 220 mm3 were treated with the anticancer agents at their maximum tolerated doses and schedules indicated. Tumor size of ECB1 and A2780 xenografts was measured every few days for up to 24 and 28 days, respectively, after the start of dosing.

Figure 3.

Anticancer efficacy of LY2523355 in xenograft tumor models. A, demonstration of schedule-dependent efficacy by LY2523355 in Colo205 tumor models. When tumors reached an average size of 150 mm3, LY2523355 was administered at 9.6 mg/kg to the tumor-bearing mice with 5 mice per treatment group by 1-hour intravenous infusion with various dosing schedules for a total of 3 dosing cycles ranging from daily dosing (q1d) to once every 7 days (q7d). Tumor size and mouse body weight were measured every few days for more than a month after the start of dosing. LY2523355, when dosed once every 4 days, caused marked tumor shrinkage with minimal toxicity seen as body-weight loss. B, comparison of anticancer efficacy by LY2523355 with other chemotherapeutic agents in preclinical xenograft tumor models. Mice bearing non–small cell lung (ECB-1; 160 mm3) or ovarian (A2780) xenograft tumors of an average size of 220 mm3 were treated with the anticancer agents at their maximum tolerated doses and schedules indicated. Tumor size of ECB1 and A2780 xenografts was measured every few days for up to 24 and 28 days, respectively, after the start of dosing.

Close modal

We compared antitumor activity of LY2523355 with other anticancer agents, including paclitaxel, docetaxel, and vinorelbine, all given at their maximum tolerated dose (MTD), in a lung and an ovarian xenograft tumor model. As shown in Fig. 3B, LY2523355 showed better or at least comparable efficacy to other anticancer agents.

LY2523355 has broad-spectrum antitumor activity in vivo

As summarized in Table 1, LY2523355 was studied in 14 subcutaneous xenograft tumor models with established cancer cell lines representing major human tumor histology of colorectal, non–small cell lung, ovarian, breast, and multiple myeloma, including drug-resistant tumors. LY2523355 showed marked antitumor activity in 86% of the xenograft tumor models, with 9 of 14 tumor models having pronounced tumor regression after treatment. Only two tumor models showed minimal sensitivity to LY2523355.

We further tested LY2523355 in 37 subcutaneous PDX tumor mouse models that are shown to largely retain their original tissue histology and cellular morphology. PDX models are hypothesized to closely represent the response to anticancer agents in human subjects (34). We found that LY2523355 had a range of antitumor activities: some tumors were exquisitely sensitive to LY2523355 and achieved total remission (e.g., small cell lung cancer), whereas some tumors were refractory to LY2523355 treatment. The in vivo efficacy data in PDX models are summarized in Table 1. Together, the data show that LY2523355 had broad-spectrum antitumor activity.

In vivo antitumor activity of LY2523355 is associated with mitotic arrest and death of cancer cells

To directly link LY2523355 antitumor activity to inhibition of Eg5 in vivo, we carried out immunohistochemical (IHC) analyses in the Colo205 tumor model. The animals were dosed at 1.1, 3.3, 10, and 30 mg/kg on Days 0, 4, and 8. As expected, LY2523355 showed dose-dependent antitumor activity (Fig. 4A). Immunohistochemistry of tumors one day after each dosing showed that LY2523355 caused a dramatic increase in cancer cells immuno-positive for histone H3 phosphorylation and that had a cellular phenotype consistent with monopolar spindles, indicating mitotic arrest of cancer cells (Fig. 4B). Quantitative analysis of histone H3 phosphorylation showed a correlation with antitumor activity in a dose-dependent manner (Fig. 4C). It is worth noting that at efficacious doses, the increase in histone H3 phosphorylation was followed by cell death on Day 2 after dosing (Fig. 4B). Furthermore, increased cell death was observed after each successive dosing and by Day 8, very few abnormal tumor cells remained, providing a cellular mechanism for marked tumor shrinkage by LY2523355.

Figure 4.

Anticancer activity of LY2523355 is associated with mitotic arrest and subsequent cell death in Colo205 xenograft tumors. LY2523355 was administered by intravenous bolus at the doses indicated for 3 cycles (red arrow) of every 4 days and anticancer activity was assessed by measuring tumor volume (A), and phosphorylation of histone H3 was analyzed by immunostaining (B). Insert is an enlargement of a phospho-histone H3-positive cell to show a phenotype consistent with monopolar spindle and arrows point to dead cells on day 2 after the first dose. C, tumor tissues from 2 mice were harvested, fixed, and processed for phospho-histone H3 immunostaining and positive cells were quantified at various times after each dosing.

Figure 4.

Anticancer activity of LY2523355 is associated with mitotic arrest and subsequent cell death in Colo205 xenograft tumors. LY2523355 was administered by intravenous bolus at the doses indicated for 3 cycles (red arrow) of every 4 days and anticancer activity was assessed by measuring tumor volume (A), and phosphorylation of histone H3 was analyzed by immunostaining (B). Insert is an enlargement of a phospho-histone H3-positive cell to show a phenotype consistent with monopolar spindle and arrows point to dead cells on day 2 after the first dose. C, tumor tissues from 2 mice were harvested, fixed, and processed for phospho-histone H3 immunostaining and positive cells were quantified at various times after each dosing.

Close modal

To further correlate cancer-cell mitotic arrest with antitumor activity in vivo, we examined the effect of LY2523355 in two resistant and two sensitive PDX tumor models (Supplementary Fig. S4A). Immunohistochemistry showed that each tumor had a distinct tissue structure and a low level of phospho-histone H3-positive cells (Supplementary Fig. S4B, a). The number of cells positive for phospho-histone H3 and cell death increased dramatically in the two sensitive tumor models, whereas in the two resistant tumor models there was only a slight increase, if any (Supplementary Fig. S4B, b and c, and 4C). Moreover, IHC analysis of skin tissues observed a marked increase in phospho-histone H3-positive proliferating skin cells in all treated animals (Supplementary Fig. S4D). Taken together, the data demonstrated that the antitumor activity of LY2523355 is linked to Eg5 inhibition in cancer cells and mitotic arrest of proliferating skin cells may serve as a promising surrogate pharmacodynamic biomarker for dose and schedule assessments.

Therapeutic threshold concentration and efficacy by LY2523355

To understand the relationship between therapeutic threshold concentration, duration for which the threshold must be maintained for cell-cycle–dependent mitotic arrest, and subsequent cell killing, we used the p388 ascites tumor model which, being a liquid tumor, is amenable to high-throughput quantitative cell-cycle analysis as compared with immunohistochemistry in solid tumors. Cell-cycle analysis showed a clear threshold dose between 5 to 10 mg/kg for maximal G2–M arrest of cancer cells (Fig. 5A). Indeed, the 5-mg/kg dose markedly extended survival of mice bearing p388 tumors and higher doses did not proportionately increase survival benefit (Fig. 5B).

Figure 5.

Efficacy and therapeutic threshold concentration of LY2523355 in p388 ascites tumor model. A, cell-cycle analysis by flow cytometry of p388 tumor cells from ascites fluid 18 hours after dosing by intravenous bolus. B, survival extension by LY2523355. Note that LY2523355 achieved maximal cell-cycle arrest and significant survival extension benefit at 5 mg/kg. C, changes in phospho-histone H3-positive epidermal cells in the ear skin of vehicle- and LY2523355-treated mice. The top panel shows no phospho-histone H3-positive keratinocytes in a 1 mm length of epidermis from the ear of a mouse that received vehicle. In the bottom panel, multiple phospho-histone H3-positive mitotic cells are present in the same length of epidermis in a mouse that received LY2523355 (bar, 100 μm).

Figure 5.

Efficacy and therapeutic threshold concentration of LY2523355 in p388 ascites tumor model. A, cell-cycle analysis by flow cytometry of p388 tumor cells from ascites fluid 18 hours after dosing by intravenous bolus. B, survival extension by LY2523355. Note that LY2523355 achieved maximal cell-cycle arrest and significant survival extension benefit at 5 mg/kg. C, changes in phospho-histone H3-positive epidermal cells in the ear skin of vehicle- and LY2523355-treated mice. The top panel shows no phospho-histone H3-positive keratinocytes in a 1 mm length of epidermis from the ear of a mouse that received vehicle. In the bottom panel, multiple phospho-histone H3-positive mitotic cells are present in the same length of epidermis in a mouse that received LY2523355 (bar, 100 μm).

Close modal

To further demonstrate the relationship between threshold exposure and cell-cycle arrest, we performed a series of infusion studies with a p388 tumor model. Following a 72-hour infusion, we determined steady-state exposure in the plasma and quantified phospho-histone H3-positive cancer cells in the ascites fluid by high-content imaging and also in skin biopsies by immunohistochemistry. As shown in Table 2, Fig. 5C and Supplementary Fig. S5A and S5B, LY2523355 at the threshold exposure level of about 10 ng/mL, as established in vitro, achieved maximal mitotic block in both cancer cells and proliferating skin cells. TUNEL staining showed that many treated p388 tumor cells were also apoptotic (Supplementary Fig. S5A).

Table 2.

Cell-cycle arrest and steady-state exposure in p388 tumor model after LY2523355 infusion

Total dose (mg/kg)Infusion rate (mg/kg·h)% pHH3Cells positive for pHH3 per unit length of epidermisPlasma (ng/mL)
Vehicle Vehicle 5.07 0.18  
30 0.416 28.37 7.87 94.68 
15 0.208 28.73 5.18 56.16 
10 0.139 20.38 7.75 39.67 
7.5 0.104 20.14 7.86 15.22 
5a 0.069 17.24 11.16 19.60 
3.74a 0.052 22.28 6.09 7.79 
1.22a 0.017 15.81 3.86 4.37 
0.65 0.009 17.70 0.34 1.52 
0.29 0.004 5.70 0.50 1.77 
0.14 0.002 3.98 0.19 1.13 
Total dose (mg/kg)Infusion rate (mg/kg·h)% pHH3Cells positive for pHH3 per unit length of epidermisPlasma (ng/mL)
Vehicle Vehicle 5.07 0.18  
30 0.416 28.37 7.87 94.68 
15 0.208 28.73 5.18 56.16 
10 0.139 20.38 7.75 39.67 
7.5 0.104 20.14 7.86 15.22 
5a 0.069 17.24 11.16 19.60 
3.74a 0.052 22.28 6.09 7.79 
1.22a 0.017 15.81 3.86 4.37 
0.65 0.009 17.70 0.34 1.52 
0.29 0.004 5.70 0.50 1.77 
0.14 0.002 3.98 0.19 1.13 

aDose within the threshold efficacious dose range.

Taken together, both in vitro and in vivo data clearly indicate that efficacy of LY2523355 is highly dosing schedule-dependent and that histone H3 phosphorylation of proliferating skin cells is a promising surrogate pharmacodynamic biomarker for dose and schedule evaluation of LY2523355 to study its anticancer activity.

This article describes in vitro and in vivo studies to characterize the anticancer activity of a novel selective Eg5 inhibitor LY2523355. LY2523355 showed broad-spectrum antiproliferative activity against cancer cell lines. In preclinical xenograft tumor models, LY2523355 also showed a broad-spectrum anticancer activity with complete remission in some. We further demonstrated that in vivo efficacy of LY2523355 is highly dependent on threshold therapeutic concentration and dosing schedule. We also showed that increase in histone-H3 phosphorylation of proliferating skin cells is quantitatively correlated with anticancer activity of LY2523355, thus offering a promising surrogate pharmacodynamic biomarker for the evaluation of LY2523355 anticancer activity.

Our extensive cellular studies showed that the anticancer activity by LY2523355 is mediated through cell cycle–dependent mitotic arrest and requires sustained activation of SAC. LY2523355 activates SAC as a result of monopolar spindle formation, leading to mitotic arrest. Spindle defects such as monopolar spindles activates SAC to delay exit from mitosis by preventing the proteolysis of cyclin B, which is mediated by anaphase-promoting complex (APC), until spindle defects are corrected (35). However, Cancer cells often escape from SAC-mediated mitotic arrest via a poorly defined process known as mitotic slippage. Mitotic slippage has been shown to be caused by gradual degradation of cyclin B in the presence of a functional SAC (36, 37). The time course study with HeLa cells showed that cells that were arrested at mitosis began to escape from the mitotic arrest 16 hours after LY2523355 treatment, as indicated by the decrease of multiple mitotic biochemical markers. Interestingly, we observed that, in addition to cyclin B degradation, mitotic slippage is also associated with BubR1 degradation. Perhaps degradation of BubR1, an essential regulatory component of SAC, is part of the mechanism responsible for mitotic slippage. Analysis of multiple mitotic and apoptotic markers show that induction of apoptosis by LY2523355 clearly followed mitotic slippage. Our data are consistent with the previous report that both activation of SAC and mitotic slippage are required for rapid cancer cell killing (25). This is further supported by the compound wash-out experiment that showed that killing of cancer cells by LY2523355 requires 12 hours or more and coincides with sustained mitotic arrest. Furthermore, phosphorylation of Bcl2 and rapid degradation of Mcl1 observed in cells mitotically arrested by LY2523355 may also contribute to induction of apoptosis. Indeed, recent studies that shows CDK1-mediated phosphorylation of Bcl-xL/Bcl2 to be a functional link between mitotic arrest and apoptosis (38), and that suppression of Mcl1 sensitizes cancer cells to the treatment of anticancer agents, including an Eg5 inhibitor (ARRY-520), in human multiple myeloma (39–41).

Consistent with Eg5 biology and mode of action of LY2523355 as discussed above, we demonstrated that anticancer activity of LY2523355 is highly schedule-dependent and has an optimal threshold exposure level and duration to achieve significant anticancer activity. Furthermore, the in vivo anticancer activity in sensitive xenograft tumor models but not resistant tumor models is associated with first mitotic arrest and then apoptosis in cancer cells very similarly as observed in vitro, indicating that LY2523355 has the same mechanism of anticancer activity in vivo as established in vitro in cell-based studies.

The salient finding in this study on characterization of LY2523355 anticancer activity in vitro and in vivo is that duration of threshold exposure is extremely important to anticancer activity by selective Eg5 inhibitors. This finding has significant implications for a successful effort in the discovery and development efforts not only on selective inhibitors of Eg5, but also on selective inhibitors of Plk1 and aurora kinases as novel new antimitotic cancer drugs. Our data on LY2523355 convincingly showed that compounds targeting these mitotic regulators must have good pharmacokinetic properties that give rise to sustained duration of threshold exposure through careful dose/dosing schedule selection to achieve robust anticancer efficacy. Thus, dosing schedules need to be optimized for each compound in accordance with their potency and pharmacokinetic properties to differentially kill sensitive cancer cells while sparing or minimizing toxicity to, proliferating normal cells. This also provides a good biologic explanation why clinical studies on early generation Eg5 inhibitors, such as SB-715992, that followed the development paradigm of antitubulin drugs showed little clinical activity (6).

Microtubules as target for antitubulins are always present throughout the cell cycle of dividing cancer cells. In contrast, Eg5 is only expressed briefly during mitosis. Furthermore, cancer cell–cycle progression is asynchronous and mitosis is completed very rapidly. Consequently, mitotic cells in tumor tissues are always only a small fraction of the total cancer cell population at any given time. Therefore, Eg5 inhibitors must be available at or above the threshold concentration relative to their potency for an adequate duration to achieve sustained mitotic arrest of most dividing cancer cells as they asynchronously progress into mitosis. The length of this required duration of threshold exposure would also depend on the cell-cycle time of proliferating cancer cells in individual tumors. The importance of sustained duration of threshold exposure of drugs that target cell-cycle regulators now has been borne out by promising clinical results of a second-generation Plk1 inhibitor volasertib (BI6727), which showed very encouraging anticancer activity (42). The reason for clinical anticancer activity of volasertib, in contrast to the first-generation Plk1 inhibitor BI2536, which has similar potency on Plk1, is likely due to the fact that volasertib has a large volume of distribution and a long half-life in humans of up to 110 hours. Such sustained exposure conceivably is able to cause mitotic arrest of proliferating cancer cells as they progress into mitosis and also sustained SAC activation for induction of apoptosis (42, 43).

Eg5 mitotic kinesin, although overexpressed in various cancers, is not a selective cancer target, per se, and plays an essential role in cell proliferation of both cancer and normal cells. It is thus anticipated that the primary dose-limiting toxicity of an Eg5 inhibitor is likely to be myelosupression and gastrointestinal toxicity. As discussed above, unlike traditional antimitotic chemotherapies, treatment at a MTD may not be the best option for application of targeted Eg5 inhibitors. Rather, what is important is the determination of a dose and dosing schedule that achieves the desired duration of exposure to cause sustained mitotic arrest and subsequent mitotic slippage of cancer cells while keeping target-mediated toxicity manageable. Histone H3 phosphorylation (pHH3), which closely correlates with its antitumor activity, is a promising pharmacodynamic biomarker that has the potential to guide dose and schedule selection for optimal cancer treatment of Eg5 inhibitors with improved safety profiles and new anticancer mechanisms as compared with antitubulins.

T. Yin is a scientist Eli Lilly and Company. G.P. Donoho has ownership interest (including patents) in Eli Lilly and Co. A. Aggarwal has ownership interest (including patents) in Eli Lilly. E.H. Westin has ownership interest (including patents) in Eli Lilly. No potential conflicts of interest were disclosed by the other authors.

Conception and design: X.S. Ye, R. Nakai, Y. Ohta, Y. Yamashita, G.P. Donoho, K. Blanchard, E.H. Westin

Development of methodology: L. Fan, R. Nakai, C. Murakata, Y. Yamashita, T. Yin, H. Li, A. Aggarwal

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. Fan, R.D.V. Horn, R. Nakai, T. Yin, K.M. Credille, F.F. Merzoug, H. Li, A. Aggarwal

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): R. Nakai, K.M. Credille, F.F. Merzoug, A. Aggarwal, K. Blanchard, E.H. Westin

Writing, review, and/or revision of the manuscript: X.S. Ye, L. Fan, R. Nakai, T. Yin, K.M. Credille, G.P. Donoho, F.F. Merzoug, E.H. Westin

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): R. Nakai, Y. Ohta, C. Murakata

Study supervision: X.S. Ye, R. Nakai, S. Akinaga, C. Murakata, Y. Yamashita, G.P. Donoho, F.F. Merzoug, E.H. Westin

The authors thank Dr. Richard Gaynor for his interest and support, Dr. Christoph Reinhard for having LY2523355 tested against the gastric cancer cell line panel, Mrs. Michele Dowless for performing Cellomic assays for p388 tumor cells using the ascites tumor model, the scientists at Oncotest, Germany, for testing LY2523355 in PDX models, and Dr. Chunping Yu for carefully reading the manuscript.

This work was supported by Kyowa Hakko Karin and Eli Lilly and Company.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Sawin
KE
,
LeGuellec
K
,
Philippe
M
,
Mitchison
TJ
. 
Mitotic spindle organization by a plus-end-directed microtubule motor
.
Nature
1992
;
359
:
540
3
.
2.
Mishima
M
,
Pavicic
V
,
Grüneberg
U
,
Nigg
EA
,
Glotzer
M
. 
Cell cycle regulation of central spindle assembly
.
Nature
2004
;
430
:
908
13
.
3.
Tanenbaum
ME
,
Medema
RH
. 
Mechanisms of centrosome separation and bipolar mitotic spindle assembly
.
Dev Cell
2010
;
19
:
797
806
.
4.
Dumontet
C
,
Jordan
MA
. 
Microtubule-binding agents: a dynamic field of cancer therapeutics
.
Nat Rev Drug Discov
2010
;
9
:
790
803
.
5.
Huszar
D
,
Theoclitou
ME
,
Skolnik
J
,
Herbst
R
. 
Kinesin motor proteins as targets for cancer therapy
.
Cancer Metastasis Rev
2009
;
28
:
197
208
.
6.
Jackson
JR
,
Patrick
DR
,
Dar
MM
,
Huang
PS
. 
Targeted anti-mitotic therapies: can we improve on tubulin agents?
Nat Rev Cancer
2007
;
7
:
107
17
.
7.
Lobert
S
. 
Neurotoxicity in cancer chemotherapy: vinca alkaloids
.
Crit Care Nurse
1997
;
17
:
71
9
.
8.
Orr
GA
,
Verdier-Pinard
P
,
McDaid
H
,
Horwitz
SB
. 
Mechanisms of Taxol resistance related to microtubules
.
Oncogene
2003
;
22
:
7280
95
.
9.
Kashina
AS
,
Rogers
GC
,
Scholey
JM
. 
The bimC family of kinesins: essential bipolar mitotic motors driving centrosome separation
.
Biochim Biophys Acta
1997
;
1357
:
257
71
.
10.
Kim
AJ
,
Endow
SA
. 
A kinesin family tree
.
J Cell Sci
2000
;
113
:
3681
2
.
11.
Kapitein
LC
,
Peterman
EJ
,
Kwok
BH
,
Kim
JH
,
Kapoor
TM
,
Schmidt
CF
. 
The bipolar mitotic kinesin Eg5 moves on both microtubules that it crosslinks
.
Nature
2005
;
435
:
114
8
.
12.
Valentine
MT
,
Fordyce
PM
,
Krzysiak
TC
,
Gilbert
SP
,
Block
SM
. 
Individual dimmers of the mitotic kinesin motor Eg5 step processively and support substantial loads in vitro
.
Nat Cell Biol
2006
;
8
:
470
6
.
13.
Kashina
AS
,
Baskin
RJ
,
Cole
DG
,
Wedaman
KP
,
Saxton
WM
,
Scholey
JM
. 
A bipolar kinesin
.
Nature
1996
;
379
:
270
2
.
14.
Sawin
KE
,
Mitchison
TJ
. 
Mutations in the kinesin-like protein Eg5 disrupting localization to the mitotic spindle
.
Proc Natl Acad Sci U S A
1995
;
92
:
4289
93
.
15.
Blangy
A
,
Lane
HA
,
d'Hérin
P
,
Harper
M
,
Kress
M
,
Nigg
EA
. 
Phosphorylation by p34cdc2 regulates spindle association of human Eg5, a kinesin-related motor essential for bipolar spindle formation in vivo
.
Cell
1995
;
83
:
1159
69
.
16.
Dagenbach
EM
,
Endow
SA
. 
A new kinesin tree
.
J Cell Sci
2004
;
117
:
3
7
.
17.
Cahu
J
,
Olichon
A
,
Hentrich
C
,
Schek
H
,
Drinjakovic
J
,
Zhang
C
, et al
Phosphorylation by Cdk1 increases the binding of Eg5 to microtubules in vitro and in Xenopus egg extract spindles
.
PLoS One
2008
;
3
:
e3936
.
18.
Liu
M
,
Wang
X
,
Yang
Y
,
Li
D
,
Ren
H
,
Zhu
Q
, et al
Ectopic expression of the microtubule-dependent motor protein Eg5 promotes pancreatic tumourigenesis
.
J Pathol
2010
;
221
:
221
8
.
19.
Castillo
A
,
Morse
HC
 3rd
,
Godfrey
VL
,
Naeem
R
,
Justice
MJ
. 
Overexpression of Eg5 causes genomic instability and tumor formation in mice
.
Cancer Res
2007
;
67
:
10138
47
.
20.
Uzbekov
R
,
Prigent
C
,
Arlot-Bonnemains
Y
. 
Cell cycle analysis and synchronization of the Xenopus laevis XL2 cell line: study of the kinesin related protein XlEg5
.
Microsc Res Tech
1999
;
45
:
31
42
.
21.
Sakowicz
R
,
Finer
JT
,
Beraud
C
,
Crompton
A
,
Lewis
E
,
Fritsch
A
, et al
Antitumor activity of a kinesin inhibitor
.
Cancer Res
2004
;
64
:
3276
80
.
22.
Mayer
TU
,
Kapoor
TM
,
Haggarty
SJ
,
King
RW
,
Schreiber
SL
,
Mitchison
TJ
. 
Small molecule inhibitor of mitotic spindle bipolarity identified in a phenotype-based screen
.
Science
1999
;
286
:
971
4
.
23.
Liu
M
,
Yu
H
,
Huo
L
,
Liu
J
,
Li
M
,
Zhou
J
. 
Validating the mitotic kinesin Eg5 as a therapeutic target in pancreatic cancer cells and tumor xenografts using a specific inhibitor
.
Biochem Pharmacol
2008
;
76
:
169
78
.
24.
Yan
Y
,
Sardana
V
,
Xu
B
,
Homnick
C
,
Halczenko
W
,
Buser
CA
, et al
Inhibition of a mitotic motor protein: where, how, and conformational consequences
.
J Mol Biol
2004
;
335
:
547
54
.
25.
Tao
W
,
South
VJ
,
Zhang
Y
,
Davide
JP
,
Farrell
L
,
Kohl
NE
, et al
Induction of apoptosis by an inhibitor of the mitotic kinesin KSP requires both activation of the spindle assembly checkpoint and mitotic slippage
.
Cancer Cell
2005
;
8
:
49
59
.
26.
Marcus
AI
,
Peters
U
,
Thomas
SL
,
Garrett
S
,
Zelnak
A
,
Kapoor
TM
, et al
Mitotic kinesin inhibitors induce mitotic arrest and cell death in Taxol-resistant and -sensitive cancer cells
.
J Biol Chem
2005
;
280
:
11569
77
.
27.
Kapoor
TM
,
Mayer
TU
,
Coughlin
ML
,
Mitchison
TJ
. 
Probing spindle assembly mechanisms with monastrol, a small molecule inhibitor of the mitotic kinesin, Eg5
.
J Cell Biol
2000
;
150
:
975
88
.
28.
Riffell
JL
,
Zimmerman
C
,
Khong
A
,
McHardy
LM
,
Roberge
M
. 
Effects of chemical manipulation of mitotic arrest and slippage on cancer cell survival and proliferation
.
Cell Cycle
2009
;
8
:
3025
38
.
29.
Bekier
ME
,
Fischbach
R
,
Lee
J
,
Taylor
WR
. 
Length of mitotic arrest induced by microtubule-stabilizing drugs determines cell death after mitotic exit
.
Mol Cancer Ther
2009
;
8
:
1646
54
.
30.
Dalton
WB
,
Yang
VW
. 
Role of prolonged mitotic checkpoint activation in the formation and treatment of cancer
.
Future Oncol
2009
;
5
:
1363
70
.
31.
Shi
J
,
Orth
JD
,
Mitchison
T
. 
Cell type variation in responses to antimitotic drugs that target microtubules and kinesin-5
.
Cancer Res
2008
;
68
:
3269
76
.
32.
Nakai
R
,
Iida
S
,
Takahashi
T
,
Tsujita
T
,
Okamoto
S
,
Takada
C
, et al
K858, a novel inhibitor of mitotic kinesin Eg5 and antitumor agent, induces cell death in cancer cells
.
Cancer Res
2009
;
69
:
3901
9
.
33.
Van Horn
RD
,
Chu
S
,
Fan
L
,
Yin
T
,
Du
J
,
Beckmann
R
, et al
Cdk1 activity is required for mitotic activation of aurora A during G2/M transition of human cells
.
J Biol Chem
2010
;
285
:
21849
57
.
34.
Fiebig
HH
,
Maier
A
,
Burger
AM
. 
Clonogenic assay with established human tumor xenografts: correlation of in vitro to in vivo activity as a basis for anticancer drug discovery
.
Eur J Cancer
2004
;
40
:
802
20
.
35.
Yu
H
. 
Regulation of APC-Cdc20 by the spindle checkpoint
.
Curr Opin Cell Biol
2002
;
14
:
706
14
.
36.
Brito
DA
,
Rieder
CL
. 
Mitotic checkpoint slippage in humans occurs via cyclin B destruction in the presence of an active checkpoint
.
Curr Biol
2006
;
16
:
1194
200
.
37.
Brito
DA
,
Yang
Z
,
Rieder
CL
. 
Microtubules do not promote mitotic slippage when the spindle assembly checkpoint cannot be satisfied
.
J Cell Biol
2008
;
182
:
623
9
.
38.
Terrano
DT
,
Upreti
M
,
Chambers
TC
. 
Cyclin-dependent kinase 1-mediated Bcl-xL/Bcl-2 phosphorylation acts as a functional link coupling mitotic arrest and apoptosis
.
Mol Cell Biol
2010
;
30
:
640
56
.
39.
Lestini
BJ
,
Goldsmith
KC
,
Fluchel
MN
,
Liu
X
,
Chen
NL
,
Goyal
B
, et al
Mcl1 downregulation sensitizes neuroblastoma to cytotoxic chemotherapy and small molecule Bcl2-family antagonists
.
Cancer Biol Ther
2009
;
8
:
1587
95
.
40.
Tunquist
BJ
,
Woessner
RD
,
Walker
DH
. 
Mcl-1 stability determines mitotic cell fate of human multiple myeloma tumor cells treated with the kinesin spindle protein inhibitor ARRY-520
.
Mol Cancer Ther
2010
;
9
:
2046
56
41.
Wertz
IE
,
Kusam
S
,
Lam
C
,
Okamoto
T
,
Sandoval
W
,
Anderson
DJ
, et al
Sensitivity to antitubulin chemotherapeutics is regulated by MCL1 and FBW7
.
Nature
2011
;
471
:
110
4
.
42.
Gil
T
,
Schoffski
P
,
Awada
A
,
Bartholomus
S
,
Selleslach
J
,
Taton
M
, et al
Final analysis of phase I single dose-escalation study of the novel polo-like kinase inhibitor BI 6727 in patients with advanced solid tumors
.
J Clin Oncol
28
:
15s
. 
2010
[suppl; abstr 3061]
43.
Rudolph
D
,
Steegmaier
M
,
Hoffman
M
,
Grauert
M
,
Baum
A
,
Quant
C
, et al
BI 6727, a Polo-like kinase inhibitor with improved pharmacolokinetic profile and broad anticancer activity
.
Clin Cancer Res
2009
;
15
:
3094
102
.