YM155 selectively reduced the expression of survivin in ESCC cells

First, we assessed the effect of YM155 on survivin expression in two ESCC cell lines Eca109 and TE13. Western blot analysis showed that YM155 inhibited survivin expression in a dose dependent manner, but had no significant effect on the abundance of other IAP family members such as XIAP and c-IAP1 (Figure 1). These results suggest that YM155 specifically suppresses survivin at low nanomolar concentrations in ESCC cells.

YM155 enhanced cytotoxicity of irradiation in ESCC cells

Next, we evaluated the viability of ESCC cells after 24- and 48- h incubation with increasing concentration of YM155. At 24 h, the IC50 of YM155 on Eca109 and TE13 cells were 21 and 60 nM, respectively. At 48 h, the IC50 of YM155 on Eca109 and TE13 cell lines were 12 and 50 nM, respectively (Figure 2A). The sub-toxic concentrations of YM155 (5 nM and 10 nM) were adopted to investigate the radiosensitivity of the two cell lines.

YM155 reduced irradiation induced accumulation of G₂/M fraction in ESCC cells

To explore the effect of survivin inhibitor on radiation-induced cell cycle checkpoint, we performed cytometric analysis on ESCC cells exposed to 8 Gy of X rays. The results showed that both two cell lines were arrested in G₂ phase of cell cycle (62.5% for Eca109 and 66.1% for TE13). Radiation-induced G₂/M arrest was abrogated by 10 nM YM155 (34.7% for Eca109 and 36.4% for TE13), with a concomitant rise in G₁ and S phases (Figure 3A and B). Exposure of the cells to YM155 alone caused small decrease in G₂/M fraction and slight accumulation of G₁ population (Figure 3A). In order to confirm that YM155 abrogated G₂ arrest, rather than induced a G₁/S- phase block, mitotic inhibitor nocodazole was used. As shown in Figure 3B, the addition of nocodazole (0.4 μg/mL) successfully prevented irradiated cells exposed to YM155 from cell cycle progression. Therefore, YM155 seems to impact the progression of cells from G₂ to M phase. Detailed data of cell cycle distribution were summarized in Table 2.

YM155 enhanced Cdc2 kinase activation and attenuated cyclin B1 expression in irradiated ESCC cells

Histone H1 kinase activity of Cdc2 was suppressed in irradiated Eca109 cells. However, YM155 resulted in significant activation of Cdc2 kinase in irradiated cells, and this effect was most evident when nocodazole was administered post-irradiation to prevent cells from existing M phase (Figure 4A). Cdc2/cyclin B1 complex is the key enzyme regulating G₂ to M transition and is controlled by the phosphorylation at various sites. DNA damage induced G₂ phase arrest is linked to the accumulation of relatively inactive, hyperphosphorylated Cdc2/cyclin B1 complexes which have inhibitory phosphates on two residues (Thr14/Tyr15) of Cdc2. To determine whether YM155 induced abrogation of G₂ checkpoint is mediated by the suppression of Cdc2 inhibitory phosphorylation, we monitored the phosphorylation level of Cdc2 as well as cyclin B1 expression in Eca109 and TE13 cells. Western blot analysis showed that treatment with YM155 alone slightly suppressed basal level of phospho-Cdc2 and cyclin B1. In both ESCC cell lines, irradiation induced inhibitory phosphorylation of Cdc2 on T-14 and Y-15 and accumulation of cyclin B1, which was most obvious at 36 h for Eca109 and at 48 h for TE13. YM155 evidently abrogated X-ray induced Cdc2 hyperphosphorylation and cyclin B1 accumulation without affecting total Cdc2 and phospho-Cdc25C protein levels (Figure 4B).

YM155 inhibited irradiation induced homologous recombination repair (HRR)

To further explore the mechanisms of YM155 induced abrogation of G₂ arrest, we detected γ-H2AX and RAD51 foci in Eca109 cells. As anticipated, radiation produced a γ-H2AX signal as early as 30 min, which was reduced to basal level 24 h post irradiation (data not shown). However, YM155 treatment resulted in a significant prolongation of γ-H2AX signal at least 24 h post irradiation (47.7 ± 3.0%) compared with radiation alone (7 ± 2.8%; P < 0.001) (Figure 5A). These results suggest that YM155 significantly inhibited the repair of DSBs manifested as the persistence of γ-H2AX foci at 4 to 24 h after radiation. We next assessed RAD51 focus formation, as a hallmark of ongoing DNA homologous recombination repair (HRR). In response to radiation alone, Rad51 formed discrete nuclear foci at 8 h and the percentage of RAD51 foci-positive cells increased up to 36 h after treatment. However, in the presence of YM155, the assembly of Rad51 foci in response to radiation was significantly suppressed, which was in contrast with kinetics of γ-H2AX (Figure 5B). Taken together, these findings support the conclusion that YM155 inhibits HRR, likely through the inhibition of Rad51 focus formation in response to radiation.

YM155 enhanced caspase-associated apoptosis in irradiated ESCC cells

To analyze the possible enhancement of radiation-induced apoptosis in ECSS cells after YM155 exposure, we first performed Annexin V/FITC and propidium iodide dual staining to quantify apoptotic cells after treatment with YM155 (10 nM) and radiation (8 Gy), alone and in combination for 48 h. As shown in Figure 6A, the induction of early apoptotic events (Annexin V positive) was most evident when cells were treated with 8 Gy plus YM155. The difference between groups of combinational and each single treatment was statistically significant (P < 0.01) in both Eca109 and TE13 cells. We next detected the cleavage of caspase-3 and PARP. As shown in Figure 6B, higher levels of cleaved PARP and caspase-3 were observed in Eca109 cells treated with 8 Gy radiation plus 10 nM YM155, compared with the cells treated with YM155 or radiation alone. Similar results were observed in TE13 cells.

YM155 inhibited survivin expression and radiosensitized Eca109 in vivo xenograft tumor

To determine the in vivo radiosensitizing activity of YM155 for ESCC, mice bearing Eca109 cells xenograft tumor were utilized. As shown in Figure 7A, YM155 or irradiation alone produced significant tumor volume regression on day 20 (P = 0.006 for irradiation, P = 0.001 for YM155) compared to control group. However, the combination of YM155 and radiation produced more tumor volume regression (P < 0.05 for YM155 plus IR versus single treatment). Notably, the combined treatment significantly prolonged the time required for tumor volume doubling relative to radiation or YM155 alone (11.9 d vs. 2.3 or 3.3 d). No complete regression in either treatment alone was observed. The enhancement factor (EF) was 3.75. The tumor growth delay induced by each modality was summarized in Table 3. All kinds of the treatments were well tolerated by the animals with no evidence of local cutaneous damage or systemic toxicity such as weight loss (Figure 7B).

Immunohistochemical analysis demonstrated that YM155 significantly inhibited survivin expression in either IR treated or untreated tumors (Figure 7C, upper panel). The mean density of survivin staining in tumors from YM155 treated groups (YM155: 0.021 ± 0.0087; IR plus YM155: 0.008 ± 0.0036) was significantly lower than that in YM155 untreated groups (Control, 0.061 ± 0.0025; IR: 0.050 ± 0.0044) (P < 0.01). The TUNEL assay was performed to quantify apoptosis in tumor sections from different groups. As shown in Figure 7C (middle panel), the mean apoptotic index was 2.8% in the control group, 30.6% in the YM155 group, and 22.2% in the IR group; while the combined treatment group had an apoptotic index of 61.2%, significantly higher than either treatment alone (P < 0.01). Furthermore, we performed immunohistochemical staining for cleaved caspase-3. As shown in Figure 7C (lower panel), 57.1% cells were positive for cleaved caspase-3 staining in combined treatment group, compared to 31.2% in YM155 alone group and 10.8% in IR alone group. These data indicated that YM155 radiosensitized ESCC in vivo.

Cell culture and reagents

The human esophageal squamous cell carcinoma cell lines Eca109 and TE13 were obtained from Shanghai Cell Bank (Chinese Academy of Sciences, Shanghai, China). The cells were cultured in RPMI 1640 (Hyclone, Thermo Scientific, MA) supplemented with 10% fetal bovine serum (Hyclone) at 37°C under a humidified atmosphere of 5% CO₂. YM155 monobromide was purchased from SelleckChem (Houston, TX). For the in vitro experiments, YM155 was dissolved in DMSO and diluted in medium to a final DMSO concentration of ≤0.1%. For the in vivo experiments, YM155 was dissolved and diluted in saline immediately before administration. Mitotic inhibitor nocodazole was purchased from Beyotime institute of biotechnology (Shanghai, China).

Irradiation conditions

X-ray radiation was delivered by a 6 MV linear accelerator (Elekta, Stockholm, Sweden) at a dose rate of 250 cGy/min with a source-to-target distance of 100 cm. For tumor irradiation, animals were anesthetized with isoflurane and positioned to place the tumor in the center of 1.0 × 1.0 cm radiation field, with the rest of mice shielded from radiation.

Cell viability assay

Cells were seeded in 96-well plates overnight, then treated with YM155 at various concentrations (0, 1, 2, 5, 8, 10, 25, 50, 100 nmol/L). Twenty-four and forty-eight hours later, 10 microliters of CCK-8 solution (Cell counting kit-8, Dojindo Molecular Technologies, Gaithersburg, MD, USA) was added to each well. Absorbance was determined at 450 nm after 3 hours of incubation. The viability of cells was calculated as following: Viability = (OD_test groupOD_blank group)/(OD_{control group}OD_blank group) × 100 %, and IC₅₀ (half maximal inhibitory concentration) was calculated from the dose–response curves. All experiments were repeated in triplicate.

Western blot analysis

Cells were harvested and homogenized in RIPA lysis buffer and centrifuged at 12,000 rpm for 20 min at 4°C. Protein concentrations of the supernatants were determined using BCA Protein Quantification Kit (Vazyme biotech co., Itd.). Western blot analysis was performed as previously described [38], using rabbit polyclonal antibodies to human survivin and c-IAP1 (R&D systems, MN), XIAP and Cyclin B1 (Cell signaling, MA), phosphor-Cdc25C, Cdc2, phospho-Cdc2, cleaved PARP and cleaved caspase-3 (Santa Cruz, CA).

Clonogenic survival assay

Exponentially growing cells were trypsinized as single-cell suspension and diluted serially to appropriate densities and seeded in triplicate in six-well plates. After cell adhesion, they were treated with 0.1% DMSO (control) or YM155 (5 nM or 10 nM) for 24 h, and then subjected to 0, 2, 4, 6 or 8 Gy X-rays irradiation. The cells were then washed with PBS, cultured in drug-free medium for 14 days, fixed with methanol, and stained with Giemsa. Only colonies containing more than 50 cells were scored. SF (surviving fraction) of each irradiation group was corrected by the PE (plating efficiency) of the nonirradiated control. The cell survival curves were fitted according to a multi-target single-hit model and the survival enhancement ratio (SER) was calculated as the ratio of the mean inactivation dose in control cells divided by the mean inactivation dose in YM155-treated cells. The experiment was repeated for three times.

Flow cytometry for cell cycle

Eca109 or TE13 cells were harvested after 24 h or 48 h treatment with YM155 (10 nM) and/or 8 Gy X rays, respectively. After washing with ice-cold PBS, the cells were fixed with ice-cold 70% ethanol and stored at −20°C for 1 h. Before analysis by flow cytometry, the cells were washed with PBS, resuspended in a staining solution containing 20 μl RNase A solution and 400 μl propidium iodide staining solution (Vazyme biotech co., Itd). Next, cell cycle distribution assessment was performed using a fluorescence-activated cell sorter (BD FACS Calibur).

Cdc2-Associated H1 Kinase assay

A total of 4 × 10⁷ cells were lysed in 400 μl lysis buffer (50 mM Hepes/NaOH, pH7.4, 150 mM NaCl, 1% Triton X-100, 1 mM dithiothreitol, protease/phosphatase inhibitor cooktail) on ice for 20 min. The cell lysate was centrifuged at 15,000 rpm for 10 min at 4°C and concentration of protein in the supernatant was determined using a BCA protein quantification kit (Beyotime, Shanghai, China). Cellular extracts were cleared by incubation with protein A or protein G-agarose (Cell Signaling, MA) for 30 min at 4°C. After centrifuging at 2,500 rpm for 5 min at 4°C, the supernatant was incubated with 4 μg Cdc2 antibody (Cell Signaling, MA) and 20 μl protein A/G agarose beads for 2 h at 4°C. The protein-agarose beads and bound immune complexes were then pelleted by centrifugation and immunoprecipitates were washed twice with lysis buffer and twice with wash buffer (50 mM Hepes/NaOH, pH 7.4, 10 mM MgCl₂, 1 mM dithio-threitol) and then subjected to a Cdc2 kinase assay using histone H1 peptide (Merck Millipore, Darmstadt) as a substrate. The immunocomplex was incubated with 40 μg wash buffer containing 10 μg of histone H1, 10 mM adenosine triphosphate (ATP) for 10 min at 30°C, and then subjected to immunoblotting using phosphoserine/threonine/tyrosine antibody (Abcam, Cambridge, MA).

Immunofluorescence

Immunfluorescence detection of phospho- H2AX and RAD51 foci was performed to monitor DNA double-strand breaks (DSBs) formation and homologous recombination repair (HRR). Cells cultured on coverslips were treated with 10 nM YM155 and irradiated with a dose of 4 Gy to assure a discrimination of individual nuclear foci in immunofluorescence staining. At indicated time points, the cells were fixed by 4% paraformaldehyde for 20 min at room temperature and permeabilized with 0.1% Triton X-100 for 10 min at 4°C. After blocking with Immunol Staining Blocking Buffer (Beyotime, Shanghai, China) for 1 h at room temperature, cells were incubated with antibody for phospho-H2AX (Ser139) (Millipore/Upstate, Temecula, CA) and RAD51 (EMD Millipore, Billerica, MA) at 4°C overnight, followed by staining with Fluorescein (FITC)-conjugated goat anti-mouse IgG (Jackson Immunoresearch, PA) and Rhodamine (TRITC)-conjugated goat anti-rabbit IgG (Jackson Immunoresearch, PA) for 1.5 h at room temperature. Finally, the samples were counterstained with 2 μg/ml DAPI and mounted in 3 μl of mounting medium (Beyotime). Three random fields each containing 50 cells were examined at a magnification of ×100 under a Zeiss LSM5 confocal laser-scanning microscope (Carl Zeiss, Jena, Germany). Nuclei containing ≥10 immunoreactive foci were scored as positive for γ-H2AX, and ≥5 foci as positive for RAD51.

Apoptosis analysis

Annexin V-FITC/PI dual staining was performed by using apoptosis detection kit from Keygen Biotech (Nanjing, China) according to the manufacturer’s instruction. Flow-cytometric analysis for induction of apoptosis was performed as previously described [38].

Xenograft tumor radiosensitivity studies

Animal experiments were approved by Ethics Committee of Nanjing Medical University. Five to six week-old male BALB/C nude mice were provided by Nanjing Medical University Animal Center. A suspension of 1 × 10⁶ Eca109 cells in 0.1 mL PBS was injected s.c. into one site of the right leg of nude mice. Tumors were allowed to grow for 7 days before treatment. Twenty-four nude mice with established tumors (all ~250 mm³) were divided into four groups and treated with (a) vehicle (PBS) alone; (b) a single dose of 8 Gy IR; (c) YM155 alone (5 mg/kg as a 7-day continuous infusion); or (d) YM155 plus IR (a single fraction of 8 Gy IR delivered on day 3 of drug treatment). Body weight and tumor diameter were measured three times per week, and tumor volume was determined according to the formula: (length[L] × width[W]²)/2. The efficacy of each treatment was evaluated by the volume change during the treatment period. Growth delay time (GD) was calculated as the time for treated tumors to reach double in volume minus the time for control tumors to reach double in volume. The enhancement factor (EF) was then determined as follows: EF = (GD_{IR + YM155}–GD_YM155)/GD_IR. The first day of treatment was designated as day 0, and observation continued until the day 20. At the end of observation, mice were sacrificed and tumors were fixed in 10% formalin, embedded in paraffin, and cut into 5 μm-thick sections for immnohistochemistry (IHC) and terminal deoxynucleotidyltransferase (dUTP) nick-end labeling (TUNEL) assay.

Immunohistochemistry

Paraffin-embedded tumor tissue sections were deparaffinized in xylene, rehydrated in graded ethanol, and rinsed twice with PBS. Endogenous peroxidase activity was blocked by incubating sections with 3% hydrogen peroxide in the dark for 15 min. The sections were then incubated overnight at 4°C with polyclonal antibody to survivin (1:50; Abcam, Ltd., Cambridge, United Kingdom) or cleaved caspase-3 (1:100; Cell Signaling, Beverly, MA, USA). After washes, slides were incubated with horseradish peroxidase (HRP)-conjugated anti-rabbit secondary antibody for 1 h at room temperature. Finally, the slides were visualized by incubation with 3, 3′-diaminobenzidine (DAB) (Dako, Hamburg, Germany) and counterstained with hematoxylin (37%). The sections were analyzed under a Zeiss Axiovert A1 light microscope. The intensity for survivin immunoreactitity in tumor cells was evaluated using Image-Pro Plus 6.0 software. Mean intensity was calculated by dividing the integral optical density (IOD) by area. The cleaved caspase-3 labeling index was calculated by dividing the number of activated caspase-3-positive cells by the total number of nuclei (per 1000 tumor cells from four microscopic visual fields).

TUNEL assay

Paraffin-embedded tumor tissue were deparaffinized, rehydrated and blocked in 3% hydrogen peroxide as above. The slides were permeabilized with TritonX-100 in 0.1% sodium citrate for 10 min. TUNEL staining was performed using the in situ cell death detection Kit-POD (Roche Molecular Biology, Mannheim, Germany) according to the manufacture’s protocol. Peroxidase activity was visualized using the liquid DAB substrate chromogen system (Dako), followed by a haematoxylin counterstaining. The percentage of apoptosis was calculated by dividing the number of TUNEL-positive cells by the total number of nuclei (per 1000 tumor cells from four microscopic visual fields).

Statistical analysis

All data were expressed as the mean ± standard deviation (SD). Statistical differences between groups were determined by the unpaired Student’s t test or one-factor ANOVA. All statistical analysis was performed using SPSS statistics 17.0 software (SPSS, Inc., Chicago, IL, USA). A P value < 0.05 was considered statistically significant.