Reagents were purchased from the following companies: ethyl ferulate (purity >99% from BR analysis, Shanghai yuanye Bio-Technology Company, Shanghai, China), dimethyl sulfoxide (DMSO; Tianjin Kemai Chemical Reagent Company, Tianjin, China), Roswell Park Memorial Institute medium 1640 (RPMI1640) and fetal bovine serum (FBS; Biological Industries, Cromwell, CT, USA), Ham’s F12 medium (Lonza, Walkersville, MD, USA), and MEM/EBSS medium (GE Healthcare Life Sciences, Logan, UT, USA). The following antibodies were purchased from Cell Signaling Technology (Beverly, MA, USA) for performing Western blot assays: phosphorylated mTOR (S2481; Cat# 2974), phosphorylated P70S6K (T389; Cat# 9234), phosphorylated AKT (S473; Cat# 4060), phosphorylated GSK3β (S9; Cat# 5558), phosphorylated ERK (T202/Y204; Cat# 4370), total mTOR (Cat# 2972), total P70S6K (Cat# 2708), total AKT (Cat# 4691), total GSK3β (Cat# 12456), total ERK (Cat# 4695), Cyclin D1 (Cat# 2922), p21 (Cat# 2947), and Ki-67 (Cat# 9027). The antibody for β-actin detection was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA), and the horseradish peroxidase (HRP)-conjugated secondary antibodies were purchased from ZSGB-BIO (Beijing, China). Active mTOR recombinant protein used for in vitro kinase assay or in vitro pull-down assay was purchased from Thermo Fisher (Shanghai, China), and inactive p70S6K recombinant protein used for in vitro kinase assay was purchased from SignalChem (Richmond, BC, Canada).

ESCC cell lines (KYSE30, KYSE70, KYSE450, and KYSE510) and JB6 mouse epithelial cells were purchased from the Chinese Academy of Sciences (Shanghai, China) and American Type Culture Collection (ATCC) (Manassas, VA, USA), respectively. SHEE normal esophageal cells were provided by Dr. Yan Zheng (The Affiliated Cancer Hospital, Zhengzhou, China). All cell lines were cytogenetically validated before the experiments. Each cell line was thawed and maintained in culture medium up to 8 weeks. The KYSE30 cells were cultured in a 1:1 mixture of RPMI 1640 medium and Ham’s F12 medium supplemented with 2% FBS. RPMI1640 medium supplemented with 10% FBS was used to culture the other ESCC cells. JB6 mouse epithelial cells were cultured in MEM/EBSS medium supplemented with 5% FBS. All media were supplemented with 1% penicillin/streptomycin. All cells were maintained at 37°C in a humidified incubator under 5% CO₂.

KYSE30 (2.5 × 10³ cells per well), KYSE70, KYSE450, or KYSE510 (2.0 × 10³ cells per well) cells suspended in complete growth medium were seeded in 96-well plates (100 μl/well). After overnight incubation, different concentrations of ethyl ferulate or isometric DMSO were diluted in complete growth medium and added to each well (100 μl/well). After 72 h, 20 μl of the MTT solution (Solarbio, Beijing, China) were added to each well and the cells were incubated for 2 h at 37°C. Next, the cell culture medium was removed and 150 μl of DMSO were added to each well. The formazan crystals were dissolved by gentle agitation, and cell proliferation was measured at 570 nm absorbance using the Thermo Multiskan plate-reader (Thermo Fisher Scientific, Waltham, MA, USA).

KYSE450 or KYSE510 cells (8 × 10³ cells per well) were suspended in 0.3% agar medium (RPMI 1640 containing various concentrations of ethyl ferulate) and then plated on a 0.6% agar containing the same concentrations of ethyl ferulate in 6-well plates. Cells were maintained at 37°C in a humidified incubator under 5% CO₂ for 2 weeks. Colonies were then photographed with an inverted microscope and counted using the Image-Pro Plus software (v.6) program (Media Cybernetics, Rockville, MD, USA).

KYSE450 or KYSE510 cells (1 × 10³ cells per well) were seeded into 6-well plates and incubated overnight before treatment with various concentrations of ethyl ferulate. Cells were maintained at 37°C in a humidified incubator under 5% CO₂ for 10 days. Foci were stained by 0.4% crystal violet staining solution, and the number of foci was counted.

KYSE30, KYSE450, or KYSE510 cells were seeded (9 × 10⁴ cells per 60-mm dish) and incubated for 24 h. The cells were then treated with various concentrations of ethyl ferulate or DMSO for 48 h. The cells were then collected by trypsinization, washed with 1× PBS, and fixed in 70% ethanol at -20°C overnight. After rehydration, cells were incubated in RNase (10 mg/ml, Solarbio) for 1 h and then stained with propidium iodide (PI; 20 μg/ml, Solarbio). The cell cycle was analyzed by flow cytometry.

The kinase assay was performed according to the manufacturer’s instructions (Promega, Madison, WI, USA). Briefly, the active recombinant mTOR (50 ng) protein was mixed with different concentrations of ethyl ferulate in reaction buffer and incubated at room temperature for 15 min. Next, the inactive p70S6K recombinant protein (100 ng) and ATP (Cell Signaling Technology) were added and incubated for 30 min at 30°C. The reaction was stopped by adding 10 μl protein loading buffer. The protein band was detected by Western blot analysis. mTOR activity was assessed using a p70S6K phosphorylation antibody.

Recombinant mTOR (200 ng) protein or ESCC cell lysate from KYSE510 (1 mg) was incubated with ethyl ferulate–Sepharose 4B or Sepharose 4B beads (100 μl, 50% slurry) in reaction buffer. Incubated samples were constantly rotated at 4°C overnight. The following day, the beads were washed 5 times and the protein bands were detected by Western blot analysis.

Small hairpin RNA plasmids against mTOR were constructed as previously described (26). Viral vectors were co-transfected with packaging vectors into Lentix-293T cells using Lipofectamine 2000 (Invitrogen, Grand Island, NY, USA) to generate lentivirus. At 48 h after transfection, the virus-enriched medium was harvested and passed through a 0.45-μm filter. The virus-enriched medium was then supplemented with 10 μg/ml of polybrene and added to KYSE450 and KYSE510 cells grown to ~60% confluence. The cells were incubated overnight. The following day, the cell culture medium was replaced with fresh complete medium. Infected cells were then selected with puromycin (1 μg/ml) for 48 h.

The study was conducted according to guidelines established for the care and use of laboratory mice by the Zhengzhou University Institutional Animal Care and Use Committee (Zhengzhou, China). Severe combined immunodeficiency (SCID) female mice (6–9 weeks old) were used to examine the effect of ethyl ferulate on ESCC patient-derived tumor growth. Human ESCC tumor tissues were excised from Affiliated Cancer Hospital in Zhengzhou University patients who received neither chemotherapy nor radiotherapy treatments before surgery. Tissue histology was confirmed by a pathologist. Informed consent was given by each patient. After tumors grew to an average volume of approximately 200 mm³, the mice were randomly divided into 2 groups containing 20 animals of the LEG110 tissue and 14 animals of the LEG45 tissue as follows: 1) untreated vehicle group (n = 10 or 7, respectively) and 2) 100 mg ethyl ferulate/kg of body weight (n = 10 or 7, respectively). Ethyl ferulate or vehicle (20% DMSO in 20% Tween 80) was administered by oral gavage once per day Monday through Friday. Tumor volumes were measured by Vernier caliper and calculated according to the following formula: tumor volume (mm³) = (length × width × height × 0.52). Mice were monitored until the tumor volume reached approximately 1.5 cm³. The mice were then euthanized, and their tumors, blood, livers, kidneys, and spleens were extracted.

The liver, spleen, kidney, and tumor tissues from mice were prepared for hematoxylin–eosin (H&E) staining or immunohistochemistry (IHC). For H&E staining, the tissue sections were deparaffinized, hydrated, stained with H&E, and then dehydrated. For IHC, tumor tissue sections were baked at 65°C and hydrated. Afterward, the tissue sections were boiled in sodium citrate buffer (10 mM, pH 6.0) for antigen retrieval and then treated with H₂O₂ for 5 min. After incubation with a primary antibody (Ki-67, 1:100) at 4°C in a humidified chamber overnight, the sections were incubated with an HRP-conjugated goat anti-rabbit or mouse IgG antibody (ZSGB-BIO, Beijing, China) for 30 min. Tissue sections were then stained with 3,3′-diaminobenzidine (DAB; ZSGB-BIO, Beijing, China). The IHC staining was observed by microscope and quantified using the Image-Pro Plus software program.

All statistical results were presented as mean values ± SD. The Student’s t test or one-way analysis of variance (ANOVA) was used to compare significant differences. Statistical analyses were performed using Statistical Package for the Social Sciences (SPSS) for Windows (IBM, Inc., Armonk, NY, USA). Statistical significance was considered at P value <0.05.

Ethyl ferulate is a 4-hydroxy-3-methoxycinnamate compound ( Figure 1A ). SHEE normal esophageal cells were used to assess the toxicity of ethyl ferulate. Cell viability was determined by MTT assay after cells were treated with several concentrations of ethyl ferulate for 72 h. Results showed that SHEE normal esophageal cell proliferation was not affected by ethyl ferulate treatment ( Figure 1B ). Therefore, we used ethyl ferulate at concentrations of 20, 40, and 60 μM for the subsequent experiments. To assess the effect of ethyl ferulate on ESCC cell proliferation, we next treated KYSE30, KYSE70, KYSE450, or KYSE510 ESCC cells with ethyl ferulate at different concentrations for 72 h. The MTT assay results revealed that treatment with ethyl ferulate significantly inhibited the growth of ESCC cells in a dose-dependent manner ( Figure 1C ). Furthermore, the focus formation and anchorage-independent ESCC cell growth assays showed that ethyl ferulate suppressed focus number and anchorage-independent growth of KYSE450 and KYSE510 ESCC cells in a dose-dependent manner ( Figures 1D, E ).

Flow cytometry analysis was performed to determine whether treatment with ethyl ferulate affected cell cycle regulation in ESCC cell lines. KYSE30, KYSE450, or KYSE510 ESCC cells were treated with ethyl ferulate at several concentrations for 48 h. The results indicated that ethyl ferulate induced G1 phase cell cycle arrest in a dose-dependent manner ( Figures 2A, B ). Based on the perturbation of the cell cycle induced by ethyl ferulate, we determined the expression of cell cycle marker proteins in KYSE30, KYSE450, and KYSE510 cells after ethyl ferulate treatment. KYSE30, KYSE450, or KYSE510 ESCC cells were treated with ethyl ferulate for 48 h, and cell cycle marker proteins were analyzed by Western blotting. The results of Western blotting indicated that ethyl ferulate increased the expression of p21 and decreased the expression of cyclin D1 ( Figure 2C ).

To determine the underlying molecular mechanism induced upon treatment with ethyl ferulate, we first investigated whether it could affect various EGF-induced signaling molecules in JB6 cells. After 24-h serum starvation, cells were treated with ethyl ferulate for 6 h and then subsequently treated with EGF for 0.5 h. The results indicated that the phosphorylation of mTOR, AKT, GSK3β, and p70S6K was strongly decreased by treatment with ethyl ferulate in a dose-dependent manner, whereas the protein level of phosphorylated ERK1/2 was not affected ( Figure 3A ). Therefore, to further identify the molecular targets of ethyl ferulate, we evaluated the effect of ethyl ferulate on signaling pathways in KYSE30, KYSE450, and KYSE510 cells after treatment with ethyl ferulate for 24 h. Phosphorylated mTOR, AKT, GSK3β, and p70S6K protein levels were strongly reduced after treatment with ethyl ferulate; however, the level of phosphorylated ERK1/2 was unaffected ( Figure 3B ). To determine potential targets of ethyl ferulate, we performed in vitro pull-down assays using ethyl ferulate-conjugated Sepharose 4B beads, Sepharose 4B with a recombinant mTOR protein, or KYSE510 cell lysates. The results showed that ethyl ferulate directly binds with mTOR but not AKT or p70S6K proteins ( Figure 3C ). Furthermore, we performed in vitro kinase assays using an active recombinant mTOR protein and an inactive p70S6K protein to confirm the effect of ethyl ferulate on mTOR activity. The result indicated that ethyl ferulate suppressed the phosphorylation of p70S6K by directly targeting mTOR in a dose-dependent manner ( Figure 3D ). Additionally, to determine whether the activities of other signaling proteins were affected by ethyl ferulate, the activities of a panel of 36 recombinant active kinases and their respective substrates were screened in the presence or absence of ethyl ferulate. The results indicated that mTOR was the sole kinase that was strongly inhibited by ethyl ferulate ( Supplementary Figures S1A, B ).

To determine the influence of mTOR knockdown on ESCC cell growth, we established stable shRNA control or shmTOR cell lines and subsequently used Western blotting to detect the expression level of mTOR protein. The expression of phosphorylated and total mTOR was strongly suppressed in shmTOR #1 and shmTOR #2 cells ( Supplementary Figures S2A, B ). We next examined the effect of mTOR knockdown on ESCC cell growth using an MTT assay. The results indicated that the anchorage-dependent cell growth of KYSE450 and KYSE510 ESCC cells was decreased upon knockdown of mTOR ( Figure 4A ). Furthermore, the focus formation and soft agar assay results showed that focus number and anchorage-independent growth of ESCC cells were significantly reduced upon knockdown of mTOR ( Figures 4B, C ).

We next assessed whether the effect of ethyl ferulate on ESCC cell proliferation is dependent on the expression of mTOR. shmTOR cells were treated with ethyl ferulate for 72 h, and then anchorage-dependent growth was assessed by MTT assay. Results indicated that shmTOR #1 cells were resistant to the growth inhibitory effect of ethyl ferulate compared with shControl cells ( Figure 5A ). Additionally, the focus formation assay and soft agar assay results showed that the inhibitory effect of ethyl ferulate in shmTOR cells is reduced relative to shControl growth ( Figures 5B, C ).

We first assessed the toxicity profile of ethyl ferulate to determine the appropriate concentration for use in our in vivo studies. Mice were orally administered vehicle, ethyl ferulate at 50 mg/kg, or ethyl ferulate at 100 mg/kg once a day Monday through Friday for 2 weeks by the gavage method. Blood samples from mice were collected and used to analyze alanine aminotransferase (ALT), aspartate aminotransferase (AST) activity. The results showed that ALT and AST activities were not significantly altered in mice treated with ethyl ferulate at 50 or 100 mg/kg compared with the vehicle-treated group ( Supplementary Figures S3A, B ; n = 4). Therefore, we used 100 mg/kg ethyl ferulate for the ESCC patient-derived xenograft (PDX) study. To evaluate the antitumor activity of ethyl ferulate in vivo, we used two PDX models: LEG110 and LEG45. LEG110 or LEG45 human ESCC tumor tissues were implanted into the back of the necks of SCID mice, and ethyl ferulate at 100 mg/kg or vehicle was administered by gavage once a day for 32 (LEG110 tissue; n = 10) or 53 (LEG45 tissue; n = 7) days. The results indicated that the treatment of mice with ethyl ferulate significantly suppressed the tumor volume relative to the vehicle-treated group ( Figures 6A, B ). Additionally, mice that were treated with ethyl ferulate exhibited no significant loss of body weight compared to the vehicle-treated group ( Supplementary Figures S4A, B ). Next, we measured the AST and ALT activity from serum to determine the potential toxicity of ethyl ferulate on PDX mice. Results showed that the ALT and AST activities were not significantly altered in mice treated with ethyl ferulate compared with the vehicle-treated group (Supplementary Figure S4C). We next prepared the tumor tissues that were isolated from both PDX cases for IHC staining of Ki-67, a cell proliferation marker protein. Results indicated that the expression of Ki-67 was significantly decreased in LEG110 or LEG45 tissues by ethyl ferulate treatment compared to the vehicle-treated group ( Figures 6C, D ; n = 5). Furthermore, the liver, kidney, and spleen tissues were stained with H&E to evaluate the potential toxicity of ethyl ferulate. The results of the H&E staining showed that there were no obvious morphological changes between treated and untreated groups ( Supplementary Figures S5A, B ; n = 5). Next, Western blotting was performed to investigate whether ethyl ferulate could inhibit protein expression of mTOR and its downstream signaling targets in PDX tumor tissues. The results showed that the phosphorylation of mTOR, AKT, and p70S6K was strongly inhibited in the ethyl ferulate-treated group ( Figures 6E, F ). We could suggest that inhibition of mTOR activity by ethyl ferulate exerts multiple effects on tumor progression, cancer growth, and cell cycle by modulating the mTOR signaling pathway ( Supplementary Figure S6 ).