The human ESCC cell lines TE1, TE8, TE9, TE10, TE11 and ECGI were obtained from the Riken Bioresource Center Cell Bank (Tsukuba, Japan). HCE4 and HCE7 were obtained from the Korean Cell Line Bank (Seoul, Korea). Jurkat human T cell (TIB-152) were purchased from American Type Culture Collection (Manassas, VA, USA).

All cell lines were cultured in DMEM (Welgene, Inc., Gyeongsan, Korea; cat. no. 001-05) supplemented with 10% FBS and 1% antibiotics (100 U/ml penicillin and 100 µg/ml streptomycin; Thermo Fisher Scientific, Inc., Waltham, MA, USA) at 37°C in an incubator containing 5% CO₂. For siRNA transfection, the cells were transfected using Lipofectamine RNAiMAX 2000 (Invitrogen; Thermo Fisher Scientific, Inc.; cat. no. 11668-019) in OPTI-MEM (Welgene, Inc.; cat. no. 004-01) according to the manufacturer's protocols and incubated for 16 h. The Lipofectamine-containing media were replaced with standard growth media and incubated for 48 h. Human IL7R siRNA #1 (5′-GUAAGAAACCCAGACUGAA3′) and IL7R siRNA #2 (5′-GAGUGAAUGGAGUCCAAGU-3′) was purchased from Bioneer Corporation (Daejeon, Korea). Control siRNA against GFP (5′-GACUUCCCAAGAAAUGCUU-3′) was purchased from Bioneer Corporation.

For drug screening of a small panel of epigenetic drugs, cells were treated with the drugs at 37°C in an incubator containing 5% CO₂ for 24 h. Apicidin was prepared as described in a previous study (16). DZNep was purchased from Cayman, (cat. no. 13828), BIX was from Sigma-Aldrich; EMD Millipore, Billerica, MA, USA (cat. no. SML1073), Vitamin C was from Sigma-Aldrich; EMD Millipore (cat. no. A4403) and C646 was from EMD Millipore (cat. no. 382113).

Total RNA was extracted using TRIzol reagent. The RNA samples were adjusted to the same concentration and reverse transcribed using a High-Capacity cDNA reverse transcription kit (Applied Biosystems; Thermo Fisher Scientific, Inc.; cat. no. 4368813). Synthesized cDNA was performed using a Cycler^® 480 II with LightCycler^® 480 SYBR Green I master mix (Roche Diagnostics, Basel, Switzerland; cat. no. 04887352001), according to the manufacturer's protocols. The RT-qPCR primer sequences used were as follows: IL7R, forward, 5′-GGAGAAAGTGGCTATGCTCAAAA-3′ and reverse, 5′-CCATTCACTTCCAACTGGCTAT-3′; P21, forward, 5′-GAGGCCGGGATGAGTTGGGAG GAG-3′ and reverse, 5′-CAGCCGGCGTTTGGAGTGGTAGAA-3′; GAPDH, forward, 5′-GAGTCAACGGATTTGGTCGT-3′ and reverse, 5′-TGGAAGATGGTGATGGGATT-3′. The RT-qPCR analysis was performed with an initial denaturation step of 5 min at 95°C, followed by 45 cycles at 95°C for 10 sec, 60°C for 10 sec, and 72°C for 10 sec. Quantification of RNA values was determined automatically using the LightCycler^® 480 program (Roche Diagnostics).

The cells were lysed in RIPA buffer containing protease and phosphatase inhibitors. The lysates were centrifuged at 15,000 × g for 15 min at 4°C and the supernatants were transferred to new tubes. Proteins were quantified using the Bradford protein assay, and lysates were diluted as appropriate to ensure the equal loading of total protein. Depending on the experiment, a volume containing 40–80 µg cell lysates from each sample was boiled in sample buffer [60 mM Tris-HCl (pH 6.8), 25% glycerol, 14.4 mM 2-mercaptoethanol and 1% bromophenol blue]. The lysates were then resolved by SDS-PAGE on 12% gels and transferred onto nitrocellulose membranes. The membranes were incubated with 5% skim milk for 1 h at room temperature and incubated with anti-Ac-FOXO1 (Santa Cruz Biotechnology, Inc., Dallas, TX, USA; cat no. sc49437, 1:100), anti-phosphorylated histone H3 serine 10 (pH3ser10; EMD Millipore; cat. no. 06-570, 1:1,000), H3 (Abcam, Cambridge, UK; cat. no. ab1791, 1:2,000), or β-actin (Cell Signaling Technology, Inc., Danvers, MA, USA; cat. no. 4967, 1:2,000) overnight at 4°C. The membranes were washed with TBS-T and probed with an HRP-conjugated secondary antibody (Cell signaling Technology, Inc.; cat. no. 7074, 1:5,000) in 5% skim milk for 1 h at room temperature. Following washing with Tris-buffered saline containing 0.5% Tween, the membranes were visualized using an enhanced chemiluminescence detection kit (LF-QC0103 AB Frotier) following exposure on an LAS-3000 image detection system (Fujifilm, Tokyo, Japan).

The cells were seeded on poly-L-lysine-coated glass coverslips in standard growth media for 16 h. The cells were washed with cold PBS and fixed with 4% paraformaldehyde for 30 min at room temperature. The fixed cells were permeabilized with 0.1% Triton X-100 for 5 min and blocked in 3% goat serum (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA; cat. no. 005-000-121) for 30 min. The cells were stained with Ac-FOXO1 (Santa Cruz Biotechnology, Inc.; cat. no. sc49437, 1:50) and Alexa Fluor^® anti-rabbit 488 secondary antibodies (Jackson ImmunoResearch, Inc.; cat. no. 111-545-003, 1:500) for 30 min at room temperature. The nuclei were counterstained with DAPI. The fluorescent signals were detected, and images were captured using a fluorescence microscope.

The cells transfected with IL7R siRNA were seeded into 6-well plates and incubated in 5% CO₂ at 37°C for 16 h. When the cells reached confluence, a wound was scratched using a pipette tip, fresh culture media was added, and the cells were incubated in 5% CO₂ at 37°C. Images of the scratched areas were captured under a microscope (Olympus IX71; Olympus Corporation, Tokyo, Japan) at 0, 12 and 24 h post-wounding.

Cell cycle assays were performed using the Cycletest Plus DNA reagent kit (BD Biosciences, Franklin Lakes, NJ, USA; cat. no. 340242), according to the manufacturer's protocol. The cell cycle profiles were analyzed using a FACScan flow cytometer (BD Biosciences).

The cells were washed twice with cold PBS, centrifuged at 1,000 × g for 5 min at 4°C, resuspended in FACS buffer containing IL7R antibodies (eBioscience, Inc., San Diego, CA, USA; cat. no. 17-1278-42) and IgGκ (eBioscience, Inc.; cat. no. 17-4714-42), and incubated for 1 h at 4°C. Following washing, fixation buffer was added, and the cells were incubated for 20 min at 4°C. The cells were washed with cold PBS and resuspended in 4% paraformaldehyde. The cells were analyzed using a FACScan flow cytometer (BD Biosciences).

ChIP assays were performed according to the manufacturer's protocol (Upstate Biotechnology, Inc., Lake Placid, NY, USA). DNA was sheared by sonication to achieve fragments between 200 and 500 bp. The sheared DNA was pre-cleared with protein A magnetic beads (EMD Millipore; cat. no. 92590), and 50 µg DNA was then immunoprecipitated by Ac-FOXO1 (Santa Cruz Biotechnology, Inc.; cat. no. sc49437) or AcH3 (EMD Millipore; cat. no. 06-866). Following IP, the recovered chromatin fragments were subjected to PCR analysis. IgG control experiments were performed for all ChIPs and incorporated into the IP/Input (1%) by presenting the results as (IP-IgG)/(Input-IgG). The PCR primer sequences used were as follows: IL7R proximal promoter (pp)1, forward, 5′-CAGGGAATATCCAGGAGGAA-3′ and reverse, 5′-AGGGAGGGAGGAAGACAGAG-3′; IL7R pp2, forward, 5′-TTGGAAGTTGGAGGTAAGCTATTT-3′ and reverse, 5′-TGCTAGGACTACAGGCATGA-3′; P21pp, forward, 5-GACTGGGGGAGGAGGGAAG-3′ and reverse, 5′-CTTCGGCAGCTGCTCACAC-3′.

The human ESCC cohorts datasets (GSE23400 and GSE29001) from the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database (www.ncbi.nlm.nih.gov/geo/) were used, and the GSE80005 mouse ESCC experimental dataset was used.

The results are expressed as the mean ± standard error of the mean. The majority of statistical comparisons were calculated using one-way analysis of variance followed by Bonferroni's post hoc test using GraphPad Prism software version 5 (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered to indicate a statistically significant difference.

IL7R has been reported to be genetically amplified in ESCC, however, the expression patterns and roles of IL7R in ESCC have not been reported (9). In the present study, the expression of IL7R was first compared in publicly available datasets of human ESCC cohorts from the NCBI GEO database (GSE23400 and GSE29001). IL7R was overexpressed in ESCC tissue samples compared with normal tissue samples in cohorts (Fig. 1A). In the animal model data, a significantly higher expression of IL7R was observed between the normal esophagus and ESCC mice (Fig. 1B). Subsequently, the expression of IL7R was examined by RT-qPCR analysis to screen for the mRNA expression of IL7R in various ESCC cell lines compared with the Jurkat human T lymphocyte cell line, which is known to express IL7R (19). The majority of the ESCC cell lines showed comparable expression of IL7R to the Jurkat cells (Fig. 1C). The HCE7 cells showed distinct expression of IL7R among the ESCC cell lines screened (Fig. 1C); therefore, subsequent experiments were performed mainly using the HCE7 cell line.

To investigate whether the expression of IL7R induces oncogenic effects in ESCC, siRNAs were used to knockdown the expression of IL7R in HCE7 and HCE4 cells. The IL7R-specific siRNAs effectively repressed the expression of IL7R at the mRNA and protein levels (Fig. 1D and E). IL7R knockdown induced mild G2 cell-cycle arrest in the HCE7 and HCE4 cells, and increased the level of P21 (Fig. 1F and G). Furthermore, it was observed that the knockdown of IL7R in HCE7 and HCE4 cells reduced the wound-healing capacity (Fig. 1H). Taken together, the results demonstrated that IL7R was overexpressed and may have an oncogenic role in ESCC.

In addition to ESCC, it was observed that IL7R was overexpressed in EACs and the high expression of IL7R was correlated with poor prognosis in terms of disease-free survival rate (P=0.042, data not shown).

To determine a strategy for IL7R regulation, a small panel of drugs that modulate histone acetylation or methylation were screened. Among the small molecules, apicidin (14,20), an HDAC inhibitor, induced marked downregulation of the expression of IL7R in HCE7 cells compared with other drugs (Fig. 2A). To investigate whether IL7R is directly targeted by apicidin, HCE7 and HCE4 cells were exposed to 0.1, 0.5, 1 or 5 µM apicidin for 24 h. The mRNA levels of IL7R were markedly decreased in the two cell lines (Fig. 2B). Apicidin has been reported to induce G1 cell-cycle arrest (14,16), but in the present study, apicidin induced G2 cell-cycle arrest in a dose-dependent manner in HCE7 and HCE4 cells, similar to the loss of IL7R (Fig. 2C). The downregulated the expression of IL7R by apicidin was observed at an early time-point, from 3 h, and >95% protein was depleted at 12 h in the HCE7 cells (Fig. 2D and E). These time-points corresponded with cell-cycle arrest and the downregulation of the mitotic marker pH3Ser10 (21,22) (Fig. 2F and G). Following treatment with apicidin for 24 h, the mRNA and protein levels of IL7R were also downregulated in Jurkat cells (Fig. 2H and I), suggesting that the effects of apicidin on IL7R are not specific to ESCC.

The present study also investigated whether the downregulation in the expression of IL7R by apicidin required de novo protein synthesis. As shown in Fig. 3, pretreatment of the HCE7 cells with the protein synthesis inhibitor cycloheximide did not rescue the effects of apicidin on the mRNA expression of IL7R. These data suggested that the HDAC inhibitor apicidin suppressed IL7R independent of de novo protein synthesis.

To understand the mechanisms underlying the downregulation of IL7R by apicidin, the present study investigated the upstream regulators of IL7R. To date, zinc finger E-box binding homeobox 2 (ZEB2), GATA binding protein 3 (GATA3) and FOXO1 have been reported to regulate the transcription of IL7R (23–27). The present study investigated the putative binding sites of these TFs based on a consensus sequence of the IL7R upstream region covering -5 kb from the transcription start site (TSS). There are several binding motifs for the three TFs (Fig. 4A). Subsequently, a literature search was performed for the post-translational acetylation of the TFs. Whereas FOXO1 has eight sites for acetylation, ZEB2 does not have a putative site and GATA3 has only one site. Of note, FOXO1 has been known to translocate between the cytoplasm and nucleus dependent on post-translational modifications, including phosphorylation and acetylation (28). Therefore, the present study focused on FOXO1 and hypothesized that apicidin induces the acetylation of FOXO1, and that acetylated FOXO1 translocates to the nucleus and affects the transcription of IL7R. To test this hypothesis, western blot and immunocytochemical analyses of acetylated FOXO1 were performed, using acetylated FOXO1 antibody for the detection of three residues following apicidin treatment. An increase in acetylated FOXO1 was observed as early as 15 min and accumulated in the nucleus up to 6 h (Fig. 4B and C). Subsequently, a ChIP assay was used to examine whether there is a change in the recruitment of acetylated FOXO1 to the IL7R DNA regulatory regions. As shown in Fig. 4A, primers were designed for two IL7R promoters, one bearing the FOXO1 binding motif (pp1) and the other covering the TSS (pp2). The binding of acetylated FOXO1 was significantly increased at both promoters accompanied by a decrease in acetylated histone H3 (Fig. 4D), suggesting that acetylated FOXO1 may act as a repressor of the transcription of IL7R. As a control, a ChIP assay was performed for the P21pp. There was a decrease in acetylated FOXO1-binding, however, the increase in acetylated histone H3 was statistically significant at the P21 promoter at 6 h not 3 h, respectively (Fig. 4E). It is possible that the increased level of acetylated FOXO1 by apicidin results in increased binding at the IL7R promoter, however, based on the decrease of acetylated FOXO1 at the P21 promoter, the increased enrichment of acetylated FOXO1 at the IL7R promoter is specific, rather than simply an increased level of acetylated FOXO1. Taken together, the results indicated that apicidin treatment induces acetylated FOXO1 to translocate from the cytoplasm to the nucleus where it bound to the IL7R promoter and repressed transcription.