Fig 1.
A) G2/M arrest in HCT116 colon cancer cells. HCT116 cells were treated with the indicated concentrations of AK301, colchicine or vincristine (microtubule inhibitors), or BI2536 (a PLK1 inhibitor) for 16 hours. Cells were then fixed and stained with propidium iodide (PI), and analyzed by flow cytometry. All four drugs induced high levels of G2/M arrest at both concentrations (P < 0.0001) with no significant differences between the compounds. B) HCT116 cells were treated with 500 nM of each of the indicated compounds for 16 h. Cell lysates were prepared and tested for caspase-3 activity using DEVD-AMC fluorogenic assay. AK301 (**P < 0.0001) and BI2536 (*P < 0.05) induced significantly higher levels of caspase-3 activation relative to control cells. C) Apoptosis in AK301 treated cells released from arrest. The left-most panel shows the cell cycle distribution of HCT116 cells under normal growth conditions. In the remaining panels, HCT116 cells were treated with 500 nM of AK301 for 16 hours. AK301 was then removed and cells were allowed to grow in fresh medium for 3, 6, 12, and 24 hours (bottom panel) or returned to AK301-containing medium (top panel). Cells were harvested at the indicated times following the medium change. Cells maintained in AK301 show a relatively stable G2/M arrest, whereas those switched to new medium showed increasing levels of sub-G1 apoptotic cells.
Fig 2.
AK301 withdrawal induces more apoptosis than colchicine withdrawal.
HCT116 cells were treated with 500 nM of AK301 or colchicine for 16 hours, as indicated. Cells were then switched to fresh growth medium for the indicated lengths of time. Flow cytometric analysis of DNA content showed that both AK301 and colchicine arrested HCT116 cells in G2/M phase of the cell cycle. However, upon drug withdrawal, cells arrested with AK301 showed the formation of more sub-diploid cells than those released from colchicine arrest.
Fig 3.
A) Apoptosis of cells exiting mitotic arrest. HCT116 cells treated with 500 nM of each of the indicated compounds for 16 hours. Cells where then either switched to drug-free medium for 8 hours, or treated with fresh drug-containing media. Flow cytometeric analysis of the DNA content showed that AK301 treated cells showed a significantly higher levels of apoptosis after release than cells treated with the other compounds (*P < 0.0001). B) Wild type and p53-null HCT116 cells were treated with 500 nM of AK301 for 16 hours. Cell lysates were prepared and tested for caspase-3 activity using DEVD-AMC fluorogenic assay. The p53-normal HCT116 cells showed more caspase-3 activation than the null cells (*P < 0.001). C) Wild type and p53-null HCT116 cells were treated and released with AK301 as described in 3A. Cells were then processed for flow cytometric analysis. Apoptosis was significantly higher in p53-normal HCT116 cells (*P<0.001). D) ATM activation and p53 stabilization following AK301 treatment. HCT116 cells were treated with 500 nM AK301 for 16 hours, followed by transfer into fresh medium for 0, 4, or 6 hours. Protein was then extracted for analysis. Immunoblot analysis shows phosphorylation of ATM at Ser1981 and phosphorylation and stabilization of p53 (p-p53 Ser15) in treated and released cells. β-actin was used as a loading control. E) Activation of p53 target genes by AK301. HCT116 cells (in biological duplicates) were treated with AK301 as in 3D, with and without a 4 hour release. Expression of Bax, Bak, p21 and Mdm2 was then determined by western blotting. β-actin was used as a loading control.
Fig 4.
A) γH2AX levels in response to treatment with mitotic arrest agents. HCT116 cells were treated for 16 hours with AK301, colchicine, vincristine, or BI2536 at 500 nM. Treated cells were analyzed for γH2AX immunofluorescent staining (Y-axis) and DNA content/PI staining (X-axis) by flow cytometry. B) Quantification of γH2AX staining in mitotically arrested cells. Using the gates indicated in 4A, the percentage of cells entering quadrant 2 (Q2) was calculated and compared for the arrest agents shown. AK301-treated HCT116 cells showed a significantly greater proportion of cells with γH2AX activation (*P < 0.0001).
Fig 5.
A) γH2AX levels in p53-normal and p53-null HCT116 cells treated with AK301. Cells were treated with 500 nM AK301 for 16 hours. Cells were then analyzed for γH2AX immunofluorescent staining (Y-axis) and DNA content/PI staining (X-axis). B) Quantification of γH2AX staining in p53-normal and mutant cells mitotically arrested with AK301. Using the gates indicated in 5A, the percentage of cells entering quadrant 2 (Q2) was calculated and compared. AK301-treated HCT116 cells showed a significantly greater proportion of cells with γH2AX activation (*P < 0.0001) but no significant differences between the wild type and null cells. C) Aurora B inhibitor-induced reduction in γH2AX. HCT116 cells were arrested with AK301 and then treated with the indicated concentrations of the Aurora B inhibitor AZD1152-HQPA for 1 hour. Immunofluorescent images of γH2AX and phospho-histone H3 staining were then captured and quantified. The Aurora B inhibitor induced a reduction in both phospho-histone H3 Ser 28 (a direct Aurora B target) and γH2AX (*P<0.01).
Fig 6.
HCT116 cells were examined by immunofluorescence confocal microscopy.
Cells were treated with 500 nM AK301 for 16 hours, and then processed for γH2AX and γ-tubulin staining (A) or Aurora B and β-tubulin staining (B). The color key and 20 μm bars are shown. The arrow and the arrowheads indicate structures referred to in the text. C) TUNEL staining shows DNA breakage in AK301-treated cells. HCT116 cells were treated with 500 nM AK301 for 16 hours, and then processed for TUNEL staining. Images of representative field is shown with a 20 μm bar. End-labeled DNA is shown in red and DAPI-stained DNA is blue.
Fig 7.
A) γH2AX levels in HT29 colon cancer cells following AK301 treatment. HT29 cells were treated with 500 nM AK301 for 16 hours. Treated cells were then analyzed for γH2AX (red) and phospho-histone H3 Ser28 (green) by immunofluorescent staining and confocal microcopy. DAPI-stained DNA is in blue. Two representative images from both control and AK301 treated cultures are shown. A 20 μm bar is shown in the left panel. B) Cell cycle analysis of HT29 cells treated with AK301 in the presence or absence of TNF. HT29 cells were treated with AK301 (500 nM) and TNF (50 ng/ml) as indicated for 24 hours. Cells were then fixed and stained with PI for cell cycle analysis by flow cytometry.
Fig 8.
Enhanced TNF sensitivity of AK301-treated HT29 cells after release from mitotic arrest.
HT29 cells were arrested in mitosis with AK301 or colchicine. Arresting agents were removed and cells were released from arrest by treatment with the MPS1 inhibitor SP600125 for 2 hours as indicated. Cells were then treated with TNF (as indicated) for 4 additional hours. Under these conditions, AK301-treated cells released from mitotic arrest are the most sensitive to TNF-induced apoptosis as determined by sub-diploid formation.
Fig 9.
Influence of APC mutation on AK301 sensitivity.
A) Mouse colonocyte cell lines that are APC-normal (YAMC) or APC heterozygous with a Min mutation (IMCE cells) were treated with 500 nM AK301 for 16 hours. Cells were then released from arrest by medium replacement and analyzed by flow cytometry at the indicated time points. APC-mutant IMCE cells underwent apoptosis more readily than wild type cells following release from arrest (*P>0.0001). B) Titration of AK301 on APC-normal and heterozygous mutant colonocytes. YAMC and IMCE cells were treated with the indicated concentrations of AK301 for 16 hours and then assessed for sub-diploid formation by flow cytometry and for mitotic arrest by the phospho-histone H3 staining. Significantly higher levels of apoptosis were observed for IMCE cells at AK301 concentrations from 75–125 nM. C) Comparison of Aurora A localization in YAMC and IMCE cells treated with 100 nM AK301. Cells were immunostained for Aurora A (green) with nuclei counterstained with DAPI (blue). Untreated YAMCs and IMCE cells show a normal bipolar localization of Aurora A to the centrosome and spindle in mitotic cells. AK301 treatment of YAMCs restricted Aurora A association with the centrosome, whereas treatment of IMCE cells induced the formation of multiple diffuse Aurora A foci.