GBC-SD and SGC-996 cells were obtained from The Cell Bank of Type Culture Collection of the Chinese Academy of Sciences (Wuhan, China) and were cultured in RPMI-1640 medium supplemented with 10% FBS, 100 µg/ml streptomycin and 100 U/ml penicillin (all from Gibco; Thermo Fisher Scientific, Inc.). Normal human intrahepatic biliary epithelial cells (HIBECs) were obtained from ScienCell Research Laboratories, Inc. and were incubated using epithelial cell medium (ScienCell Research Laboratories, Inc.). Cells were incubated at 37°C with 5% CO₂. TRAIL was obtained from PeproTech, Inc. (cat. no. 310-04). Rocaglate CR-1-31B was synthesized as previously described (12) and other reagents were obtained from Sigma-Aldrich; Merck KGaA, unless otherwise stated.

Human GBC and paired normal gallbladder tissues were collected from 42 patients during gallbladder resection for GBC (22 males and 20 females) between July 2015 and June 2018 at the Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. The patients did not receive chemotherapy prior to surgery. The use of human GBC tissues was approved by Tongji Hospital Research Ethical Committee, and written informed consent was obtained from all the patients.

GBC-SD and SGC-996 cells were cultured in 96-well plates in complete medium (Gibco; Thermo Fisher Scientific, Inc.) with CR-31 (0–200 nM for 48 h) and CR-31 (100 nM) and/or TRAIL (100 ng/ml) for the indicated time intervals (0–24 h). An MTT assay kit (cat. no. ab211091; Abcam) was used to assess cell viability. Briefly, following removal of the medium, MTT (150 µl; 0.5 mg/ml) was added to each well and then cultured for 2 h at 37°C in an incubator. The medium was removed and DMSO (150 µl) was added to each well and the plates were shaken. Absorbance at 570 nm was analyzed using a microplate reader (BioTek Instruments, Inc.). Results were normalized to DMSO-treated cells.

For the colony formation assays, cells were plated in a 12-well plate and incubated for 12 h at 37°C with 5% CO₂, and then incubated with CR-31 (100 nM) and/or TRAIL (100 ng/ml) for another 12 h at 37°C with 5% CO₂. Cells were washed with PBS, fixed using 4% paraformaldehyde for 15 min, and subsequently stained using crystal violet for 15 min.

Cells were treated with CR-31 (100 nM) or DMSO for 12 h, and incubated with TRAIL (100 ng/ml) for an additional 12 h at 37°C with 5% CO₂. An Annexin V/propidium iodide (PI) kit (cat. no. v13242; Invitrogen; Thermo Fisher Scientific, Inc.) was used to assess cell apoptosis, according to the manufacturer's protocol. Flow cytometry was used to determine the proportion of apoptotic cells. Flow cytometry was performed on a FACSCalibur system (BD Biosciences).

RT-qPCR was performed as described previously (13). Briefly, total RNA was extracted from GBC-SD cells using TRIzol^® reagent (cat. no. 15596018; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. A total of 1 µg total RNA was used as the template for cDNA synthesis using a reverse transcription kit (cat. no. K1691; Thermo Fisher Scientific, Inc.). Equal quantities of cDNA were used for PCR analysis. The sequences of the primers were: c-FLIP forward, 5′-CGCTCAACAAGAACCAGTG-3′and reverse, 5′-AGGGAAGTGAAGGTGTCTC-3′; and β-actin forward, 5′-AGTGTGACGTCGACATCCGC-3′ and reverse, 5′-GACTCGTCGTACTCCTGCTT-3′. PCR primers were purchased from Sangon Biotech Co., Ltd.

Western blot analysis was performed as described previously (13). Briefly, GBC-SD and SGC-996 cells were washed twice with ice-cold PBS and then lysed with ice-cold RIPA lysis buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) supplemented using EDTA-free protease inhibitor cocktail (cat. no. 11836170001; Roche Diagnostics), and followed by protein concentration determination with BCA assay. Lysates were kept on ice for 30 min and then centrifuged at 14,000 × g for 20 min at 4°C. Equal amounts (20 µg) of proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel (SDS-PAGE) and then transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were block with 5% bovine serum albumin (BSA) in Tris-buffered saline Tween-20 for 2 h and then incubated with primary antibodies at 4°C overnight. Immunoreactive proteins were detected with horseradish peroxidase-conjugated secondary antibodies. The primary antibodies used were: Caspase-8 (cat. no. 9746; Cell Signaling Technology, Inc.), caspase-3 (cat. no. 9662; Cell Signaling Technology, Inc.), cleaved-caspase-3 (cat. no. 9664; Cell Signaling Technology, Inc.), eIF4A (cat. no. ab31217; Abcam), c-FLIP (cat. no. ab8421; Abcam) and α-tubulin (cat. no. sc-5286; Santa Cruz Biotechnology, Inc.).

The treated cells were fixed using 4% formaldehyde for 15 min at room temperature, washed and permeabilized using 0.5% Triton X-100 for 20 min at room temperature. The cells were treated for 1 h using Alexa Fluro 647-labeled IgG (H + L) (cat. no. ab150115; Abcam) and an anti-c-FLIP antibody (cat. no. ab8421; Abcam). Nuclei were stained using DAPI. Cells were washed with PBS and visualized using confocal laser scanning microscopy (Olympus Corporation).

Highly pure small interfering (si)RNA non-targeting control (CTRL; 5′-aacuuacgcugaguacuucga-3′) or siRNA eIF4A (Sangon Biotech Co., Ltd.) were transfected into GBC-SD cells using Lipofectamine^® 3000 (cat. no. L3000008; Invitrogen, Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. Briefly, 2.5 µl Lipofectamine 3000 and siRNA (50 nmol/l) were added to the medium (100 µl), and then the mixture was added to the cells, which had been washed. Cells were subsequently cultured for 5 h at 37°C, after which, the medium was removed and replaced with fresh medium for 48 h. Cells were treated using DMSO or TRAIL for 12 h following transfection. After that, the cells were used for subsequent experiments immediately.

GBC-SD cells (~2×10⁶ cells) were suspended in 100 µl 1:1 Matrigel/medium and then injected subcutaneously into the right flank of male BALB/c nude mice (n=20; age, 6-weeks old). The animals were housed and assayed under conditions of control temperature (22±2°C), humidity (45–65%), and artificial light (12-h light-dark cycle) with free access to rodent chow and water. The mice were randomly separated into four equal groups (n=5 per group): i) Control group [receiving vehicle (10% DMSO and 90% olive oil)]; ii) TRAIL group [intraperitoneal injection of TRAIL (80 µg/kg) in 60 µl olive oil]; iii) CR-31 group [intraperitoneal injection of CR-31 (2 mg/kg) in 60 µl olive oil]; and iv) TRAIL/CR-31 group [intraperitoneal injection of TRAIL (80 µg/kg) and CR-31 (2 mg/kg) in 60 µl olive oil]. Treatment was performed once every 2 days for 28 days. Body weight and tumor volume were monitored on a weekly basis. Tumor volumes (mm³) were calculated as follows: Volume=length × S²/2 (where S is the shortest diameter). After 28 days, mice were anesthetized by intraperitoneal injection using 10% chloral hydrate (350 mg/kg) and were sacrificed using cervical dislocation. A comprehensive evaluation of death by respiratory, heartbeat, pupil and nerve reflex of these mice was carried out and recorded. Finally, tumor specimens were collected for immunohistochemistry and TUNEL staining.

The mice were procured from HFK Bioscience Company. Animal experiments were performed according to the Guidelines of Laboratory Animals of Tongji Hospital, which is approved by the National of Health (NIH publication 85–23 revised 1996).

Immunohistochemistry was performed using anti-cleaved-caspase-3 (cat. no. 9664; Cell Signaling Technology, Inc.) and anti-eIF4A (cat. no. ab31217; Abcam) antibodies, as previously described (14). Two independent individuals examined the proportion of positive cells in ≥5 fields of view (magnification, ×400), and tissues were scored as follows: 0, negative; 1, <25% positive cells; 2, 25–50; 3, 50–75; and 4, >75%. Images were visualized and calculated using a Nikon microscope (Nikon Corporation). TUNEL assays was performed on paraffin-embedded tissue sections using a one-step TUNEL apoptosis assay kit (Beyotime Institute of Biotechnology), according to the manufacturer's protocol. Briefly, samples were incubated with TUNEL reaction mixture for 1 h at 37°C in the dark and then washed twice with PBS. Nuclei were counterstained with DAPI (300 nM; cat. no. D8417; Sigma-Aldrich) for 10 min at room temperature and mounting with antifade medium (cat. no. P0126; Beyotime Institute of Biotechnology) and then washed twice with PBS. The condensed or fragmented nuclei of apoptotic cells were observed using fluorescence microscopy (Olympus Corporation) (magnification, ×200) in 20 fields of vision.

All data were analyzed using GraphPad Prism version 8.4.2 (GraphPad Software, Inc.). All data are presented as the mean ± standard deviation. Student's t-test was used to compare differences between two groups, and multiple groups were compared with one-way or two-way ANOVA followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

To assess whether CR-31 could sensitize GBC cells to TRAIL-induced apoptosis, GBC-SD and SGC-996 cells were used, as they exhibit strong chemoresistance to TRAIL (7). The treatment of GBC-SD and SGC-996 cells with CR-31 resulted in dose-dependent growth inhibition, with an IC₅₀ of ~100 nM (Fig. 1A). Furthermore, the efficacy of CR-31 on TRAIL-mediated cytotoxicity was assessed. CR-31 and TRAIL alone were weakly cytotoxic to GBC-SD and SGC-996 cells; however, the treatment of GBC-SD and SGC-996 cells using CR-31 increased the cytotoxicity to TRAIL (Fig. 1B). Moreover, although CR-31 or TRAIL did not reduce the colony numbers, the combined treatment with CR-31 and TRAIL significantly reduced the colony formation of GBC-SD and SGC-996 cells, indicating a synergistic effect (Fig. 1C and D). Similar results were observed based on the Annexin V/PI assay in GBC-SD and SGC-996 cells. TRAIL or CR-31 alone resulted in increased early apoptosis and necrosis in GBC-SD (22.28 and 9.77%, respectively) and SGC-996 cells (23.1 and 10.5%, respectively). However, combined apoptosis and necrosis were increased in both GBC-SD (41.48%) and SGC-996 cells (41.5%) (Fig. 1E and F). Collectively, the results showed that CR-31 may highly sensitize chemo-resistant GBC-SD and SGC-996 cells to TRAIL.

To investigate the clinical significance of eIF4A in human samples of GBC, the abundance of eIF4A was detected by immunohistochemistry in GBC and paired normal gallbladder tissues. Immunohistochemistry analysis suggested that eIF4A abundance was significantly higher in the 42 GBC samples compared with the paired normal control tissues (Fig. 2A). Furthermore, the eIF4A protein was strongly expressed in GBC-SD and SGC-996 cells compared with the HIBECs (Fig. 2B). Increased expression of c-FLIP in GBC cells has been shown to be associated with TRAIL resistance (7), thus the effect of CR-31 on the abundance of c-FLIP was determined. The results showed that eIF4A and c-FLIP were abundantly expressed in GBC cells, and CR-31 treatment significantly decreased the levels of c-FLIP protein in GBC-SD cells (Fig. 2C). Notably, there were no significant changes in c-FLIP mRNA levels (Fig. 2D), suggesting a translational mechanism of regulation of expression. Furthermore, confocal microscopy showed that CR-31 resulted in a decrease in the plasma localization of c-FLIP (Fig. 2E). Collectively, the results showed that eIF4A may be a therapeutic target and CR-31 can downregulate the abundance of c-FLIP at the translational level in GBC.

To assess the effect of knockdown of eIF4A on c-FLIP expression and CR-31 treatment, an eIF4A-knockdown GBC-SD cell line was established. siRNA-eIF4A resulted in a significant decrease in both eIF4A and c-FLIP expression (Fig. 3A). Crystal violet staining showed eIF4A knockout GBC-SD cells exhibited significantly reduced colony formation when treated with TRAIL compared with the control group (Fig. 3B and D). Furthermore, the results of Annexin V-PI apoptosis by flow cytometry showed the same effect; the proportion of apoptotic and necrotic cells following TRAIL treatment was 24.78 and 8.73%, respectively, in the siRNA-eIF4A-transfected cells, and 43.15% in siRNA-eIF4A-transfected cells treated with TRAIL (Fig. 3C and E), showing that knockdown of eIF4A in GBC-SD cells resulted in similar effects to treatment with CR-31.

As TRAIL-mediated apoptosis is mediated through the activation of the caspase cascade (15), whether caspase-8 and caspase-3 expression was increased in GBC-SD cells treated with TRAIL or CR-31 alone was next determined. The results showed that the levels of pro-caspase-8 were weakly decreased when treated with CR-31 or TRAIL alone, along with an increase in cleaved-caspase-8 expression (Fig. 4). Notably, when combined, CR-31 and TRAIL significantly increased the levels of cleaved-caspase-8 (Fig. 4). Notably, TRAIL or CR-31 alone resulted in slightly increased levels of cleaved-caspase-3 in GBC-SD cells (Fig. 4). However, CR-31 and TRAIL combined significantly increased the levels of cleaved-caspase-3 (Fig. 4). Therefore, the results suggest that CR-31 enhanced TRAIL-mediated apoptosis in a caspase-dependent manner.

To evaluate the effects of CR-31 on TRAIL in vivo, a BALB/c nude mice model of GBC was established using GBC-SD cells. The tumor volume and weights of GBC-SD xenografts were measured. Tumor weights after 4 weeks of treatment showed a decreasing trend from control, to TRAIL to CR-31 to combination treatment (Fig. 5A and C). However, tumor weight did not exceed 10% of body weight of mice. Interestingly, there was a notable reduction in tumor volumes treated with CR-31 compared with those treated with TRAIL (Fig. 5A and C), and this may due to an increase in the production of TRAIL from natural killer cells in vivo (16). Notably, CR-31 neither resulted in the notable reduction of body weight nor showed evidence of toxicity (Fig. 5B) and the maximum percentage of weight loss did not exceed 8% of body weigh of mice, indicating that CR-31 was safe in vivo. The apoptotic effect of CR-31 and TRAIL were assessed in vivo using cleaved-caspase-3 and TUNEL staining. Upregulated expression of cleaved-caspase-3 were observed, and there was an increase in apoptosis when treated with the combined treatment compared with either TRAIL or CR-31 alone, which was in agreement with the in vitro results (Fig. 5D-G). Collectively, the results showed that CR-31 enhanced TRAIL-induced apoptosis of GBC xenograft tumors in vivo. Taking these findings together, we conclude that CR-31 can enhance TRAIL-triggered apoptosis by downregulating c-FLIP expression at the translational level and then activating the caspase cascade in TRAIL-resistant GBC cells (Fig. 6).