HCT116 and HCT8 human colorectal cancer cells were purchased from the Type Culture Collection of the Chinese Academy of Sciences (Shanghai, China). Both cell lines were cultured in RPMI-1640 medium supplemented with 10% (v/v) fetal bovine serum (both from Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 100 µg/ml streptomycin and 100 U/ml penicillin at 37°C and 5% CO₂. α-hederin was purchased from Sigma-Aldrich; Merck KGaA (Darmstadt, Germany). ROS inhibitor N-acetyl-L-cysteine (NAC), oxidation-activated fluorescent dye DCHF-DA and Hoechst 33258 were purchased from Beyotime Institute of Biotechnology (Haimen, China). Autophagy inhibitor 3-MA was purchased from Sigma-Aldrich; Merck KGaA.

Anti-light chain (LC)3 (12135-1-AP), anti-caspase-9 (10380-1-AP), anti-caspase-3 (19677-1-AP), anti-P62 (18420-1-AP) and anti-Beclin 1 (11306-1-AP) antibodies were purchased from ProteinTech Group, Inc. (Chicago, IL, USA). Anti-phosphorylated (p)-mTOR (#2971), anti-mTOR (#2972), anti-β-actin (8456), anti-p-Unc-51 like autophagy activating kinase 1 (ULK1; #4634), anti-ULK1 (#4773), anti-p-P70S6K (#9205), anti-P70S6K (#9202), anti-poly(ADP-ribose) polymerase (PARP; #9532), anti-p-AMPK (#2531), anti-AMPK (#2532), anti-B cell lymphoma (Bcl)-2 (#2872), anti-Bcl-2 associated X protein (Bax; #2772), anti-Bcl-xl (#2762), horseradish peroxidase (HRP)-linked anti-rabbit immunoglobulin (Ig)G (#7074) and HRP-linked anti-mouse IgG (#7076) were obtained from Cell Signaling Technology, Inc. (Danvers, MA, USA).

Flow cytometry assays were performed on a FACSCalibur system (Becton-Dickinson and Company, Franklin Lakes, NJ, USA). Apoptosis rate assay was performed using an Annexin V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) apoptosis detection kit (BD Biosciences, San Jose, CA, USA). Cells were digested using trypsin, washed twice with cold PBS and then incubated with Annexin V-FITC and PI at room temperature for 30 min. Subsequently, the apoptosis rate was determined using flow cytometry. Intracellular ROS levels were determined through flow cytometry using DCHF-DA, an oxidation-activated fluorescent dye. Cells in dishes were loaded with PBS containing 10 µM DCHF-DA for 20 min at 37°C. Then, cells were washed thrice with cold PBS, digested with trypsin and washed twice with cold PBS in turn. ROS assay was performed using a FACSCalibur flow cytometer (Becton-Dickinson and Company) and the data were analyzed using FlowJo software, version 7.6 (Tree Star, Inc., Ashland, OR, USA).

Cells were digested, dispersed and plated in a 96-well plate at a density of 1×10⁵ cells/well. Cells were incubated with Hoechst 33258 (0.5 µg/ml; Beyotime Institute of Biotechnology) for 3 min at room temperature. Washed 3 times with PBS, cells were observed and photographed directly under a microscope.

Cells were digested, dispersed and plated in a 96-well plate at a density of 1×10⁵ cells/well. Cell viability was analyzed using Cell Counting Kit-8 (Dojindo Molecular Technologies, Inc., Kumamoto, Japan) according to the manufacturer's protocol. Absorbances of the wells were read at 450 nm using a plate reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Single cells were seeded into a 35-mm cell culture dish at a density of 1,000 cells/well and cultured with α-hederin at concentrations of 0, 10 and 60 µM for 14 days. The medium was replaced every 2 days, a total of six times. Ultimately, cells were treated in turn with 4% paraformaldehyde for 20 min and crystal for 20 min, and were washed with PBS at least twice in room temperature. The colonies were counted, and images were captured. Colonies were determined as those able to form a colony consisting of ≥50 cells.

Cells were lysed using cell lysis buffer (Beyotime Institute of Biotechnology) with Protease Inhibitor Cocktail (100X; Sangon Biotech Co., Ltd., Shanghai, China) and 1 mM phenylmethylsulfonyl fluoride (Sigma-Aldrich; Merck KGaA). Following centrifugation at 4°C and 12,000 × g for 15 min, protein concentration in the supernatant of the lysates was determined using the Bradford Coomassie Blue G-250 method. Proteins (40 µg per lane) were separated using 10% SDS-PAGE, and transferred onto polyvinylidene difluoride (PVDF) membranes (EMD Millipore, Billerica, MA, USA) in sequence. Thereafter, PVDF membranes were blocked with 5% bovine serum albumin (Sigma-Aldrich; Merck KGaA) in TBS containing 0.05% Tween-20 at room temperature for 2 h. Subsequently, PVDF membranes were incubated with primary antibodies at 1:1,000 in TBS containing 0.05% Tween-20 and 5% bovine serum albumin overnight at 4°C for 16 h. Next, PVDF membranes were incubated with corresponding HRP-conjugated secondary antibodies (1:1,000) at room temperature for 2 h. Finally, membranes were visualized using an enhanced chemiluminescence Western blot detection system (EMD Millipore). In the present study, all relative protein concentrations were normalized to that of β-actin.

LC3 lentivirus and mutation lentivirus were purchased from Shanghai GeneChem Co., Ltd. (Shanghai, China). The LC3 lentivirus consisted of the red fluorescent protein Stub-RFP, green fluorescent protein Sens-GFP, and autophagy marker protein LC3. HCT116 and HCT8 cells were infected with these lentiviruses to establish four cells line stably expressing LC3 (HCT116 LC3, HCT8 LC3) or LC3-mutation (HCT116 LC3-mu, HCT8 LC3-mu).

At first, cells were digested, dispersed and plated in a 24-well plate. When cells reached 60% confluence, the medium was replaced with infection mixture. The mixture contained RPMI-1640 medium, lentiviruses (MOI=100) and polybrene (5 µg/ml; Shanghai GeneChem Co., Ltd.). Following incubation for 24 h, the mixture was replaced with normal medium containing FBS. These established cell lines were used to detect the occurrence of autophagy. In the absence of autophagy, the RFP-GFP-LC3 fusion protein diffuses in the cytoplasm. When autophagy is formed, it translocates into the autophagosome membrane. The fusion protein will form bright fluorescent spots, which can be observed under fluorescence microscope. Fluorescent dot aggregation was observed via confocal laser microscopy. One spot is equivalent to an autophagosome, and the activity of autophagy can be evaluated by counting. As the control, RFP-GFP-LC3-mu fusion protein could not participate in the formation of autophagosome membrane.

Small interfering (si)RNA for the human AMPK gene (Gene ID: 94557300) was synthesized by Biomics Biotechnologies Co., Ltd. (Nantong, China). The AMPK siRNA sequence was 5′-GAUAUCAGGGAACAUGAAUdTdT-3′. The control sequence was 5′-UAAGGCUAUGAAGAGAUAC-3′. Cells were transfected with the aforementioned siRNAs at a concentration of 50 nM using Lipofectamine 3000 (Life Technologies; Thermo Fisher Scientific, Inc.) for 48 h.

After being treated with 10 µM α-hederin for 24 h, HCT116 and HCT8 cells were digested with trypsin and washed twice with cold PBS. Subsequently, these cells were fixed overnight using 3% glutaraldehyde in 0.1 M phosphate buffer (pH 7.2) at 4°C and post-fixed in 1% OsO₄ buffer for 1 h at 4°C. After being dehydrated in a graded series of ethanol, the cells were embedded in spur resin at 56°C overnight. The spur resins were cut into ultrathin sections (60 nm) and stained with saturated solutions of uranyl acetate and lead citrate at room temperature for 10 min. Finally, the autophagosomes in these cells were observed using an electron microscope.

A total of 10 male nude mice (BALB/c nu/nu; age, 5 weeks) were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China). The initial body weight of all nude mice was ~18 g. Nude mice were bred in specific pathogen-free (SPF) conditions at the Laboratory Animals Breeding and Research Centre, Shanghai University of Traditional Chinese Medicine (Shanghai, China). The housing conditions for mice were as follows: Room temperature, 20-26°C and 12-h light/dark cycle. The mice received 5 g food and 100 ml water per 100 g body weight per day. Two million HCT116 cells in 0.1 ml PBS were injected into subcutaneous tissues of each mouse. After 2 weeks, mice were weighed and randomly divided into control group and α-hederin group (n=5/group). α-Hederin group mice were injected intraperitoneally with α-hederin (2 mg/kg body weight) and the control group were injected intraperitoneally with the same volume of PBS (control) every 3 days for 3 weeks. Finally, the tumor-bearing mice were sacrificed and the tumors were excised and weighed. The protocol was approved by the Ethics Committee of Shuguang Hospital affiliated with Shanghai University of Traditional Chinese Medicine.

All tumors from tumor xenograft model were formalin-fixed for 24 h at room temperature, embedded in paraffin, sectioned serially to 5-µm thickness and mounted on glass slides. Subsequently, immunohistochemical staining of LC3 was carried out using a standard protocol (30). Anti-LC3 antibody (18725-1-AP; 1:50) for immunohistochemistry was purchased from ProteinTech Group, Inc. HRP-labeled goat anti-human IgG (A0201; 1:50) was purchased from Beyotime Institute of Biotechnology. Anti-LC3 antibody and HRP-labeled goat anti-human IgG were all diluted in PBS containing 5% bovine serum albumin (Sigma-Aldrich; Merck KGaA). Meanwhile, serial sections were stained with hematoxylin and eosin (H&E).

All data were subjected to a single-factor analysis of variance with post hoc Fisher's least significant different test, or Student's t-test using the SPSS 13.0 software package (SPSS, Inc., Chicago, IL, USA). Data are presented as the mean ± standard deviation. P<0.05 was considered to indicate a statistically significant difference.

Previous studies have suggested that α-hederin has anticancer properties against different cancers (24,27,29,31-36), including colorectal cancer (29). To evaluate the effect of α-hederin on the proliferation of colorectal cancer cells, its effect on the viability of two colorectal cancer cell lines (HCT116 and HCT8) was tested. Furthermore, colony formation assay was performed to determine cell proliferation.

In cell viability assay, HCT116 and HCT8 wells were divided into seven groups according to α-hederin concentration (0, 2, 4, 8, 16, 32 and 64 µM). Each group contained six wells. As presented in Fig. 1A, the cell viability of the two cell lines decreased significantly with the increase of α-hederin concentration. The half maximal inhibitory concentration (IC₅₀) of α-hederin in HCT116 was 21.85, 12.47 and 9.32 µM for 24, 48 and 72 h, respectively. Similarly, IC₅₀ values of α-hederin in HCT8 cells treated for 24, 48 and 72 h were 22.76, 12.51 and 9.39 µM, respectively. Therefore, α-hederin decreased the viability of both colorectal cancer cell lines in a dose- and time-dependent manner. Colony formation assay using the adhesive culture system was used to analyze the antiproliferation effects of α-hederin. It was demonstrated that α-hederin treatment resulted in a dose-dependent decrease in the efficiency of colony formation (Fig. 1B). α-Hederin concentrations at 10 µM caused a significant decreased in the number of colonies in both colorectal cancer cell lines (P<0.001). Furthermore, after increasing α-hederin concentrations to 60 µM, no colonies were observed.

Previous studies have demonstrated that α-hederin may induce cell apoptosis (27,37). In the present study, α-hederin-induced apoptosis rates of colorectal cancer cells were determined using flow cytometry with Annexin V/PI (Fig. 2A) and microscopy using Hoechst 33258 staining (Fig. 2B). Flow cytometry results indicated that the α-hederin-induced (0, 10 and 60 µM) apoptosis rates were 5.51±1.18, 8.97±1.69 and 21.75±2.85% in HCT116 cells. Apoptosis rates in HCT8 cells were 3.54±1.3, 7.66±1.89 and 15.66±2.36% (Fig. 2A). Compared with control, 10 µM α-hederin induced higher levels of apoptosis in HCT116 and HCT8 cells (P<0.01), with 60 µM α-hederin inducing even greater levels of apoptosis (P<0.001). In addition, Hoechst 33258 staining exhibited results similar to those for flow cytometry (Fig. 2B).

The mitochondrial signaling pathway is one of the major pathways involved in apoptosis (38). To observe the effect of α-hederin on the mitochondrial signal pathway, the expression of mitochondrial apoptosis-related proteins was measured in HCT116 cells. As presented in Fig. 2C, compared with the control, the expression of activated caspase-9 (cleaved caspase-9), activated caspase-3 (cleaved caspase-3), PARP degradation (cleaved PARP), and Bax were all increased in HCT116 cells treated with α-hederin. Furthermore, α-hederin increased these proteins expression of HCT116 cells in a time-dependent manner. Conversely, α-hederin decreased the expression of Bcl-2 and Bcl-xl in a time-dependent manner.

Analysis of Figs. 1A and 2A indicated that the apoptosis rate did not match the inhibition rate of α-hederin on colorectal cancer cell viability. This suggested that α-hederin-induced cell death may have involved other mechanisms. Autophagy rates in HCT116 and HCT8 cells were determined using LC3 lentivirus (39). As a negative control, mutant LC3 (LC3-mu) was used as it could not bind to autophagosomes and form bright fluorescent spots. As presented in Fig. 3A, α-hederin induced LC3 but not LC3-mu aggregation in HCT116 and HCT8 cells. Compared with controls, 10 µM α-hederin exhibited significantly higher autophagic rate in HCT116 and HCT8 cells (P<0.001). The presence of autophagosomes in HCT116 cells was also evaluated using an electron microscope. As presented in Fig. 3B, autophagosomes were present in α-hederin-treated HCT116 and HCT8 cells. This result further supported that α-hederin induced autophagy in colorectal cancer cells.

In vitro results had demonstrated that α-hederin could induce autophagy in colorectal cancer cells. To investigate the inducing autophagy effect of α-hederin in vivo, a subcutaneous xenograft model of HCT116 cells in nude mice was used. As presented in Fig. 4A, α-hederin significantly inhibited tumor growth compared with the control. According to the results of H&E staining (Fig. 4B), tumors treated with α-hederin exhibited marked necrosis. LC3 puncta was assessed using immunohistochemistry to evaluate the effect of α-hederin on autophagy in vivo. As presented in Fig. 4B, the presence of LC3 puncta was observed in samples treated with α-hederin. In addition, the necrotic area also exhibited highly aggregated LC3 puncta. While, the control exhibited significant diffuse cytoplasmic staining without puncta. These results suggested that α-hederin could inhibit tumorigenicity through promoting autophagy of colorectal cancer cells in vivo.

Given that dephosphorylation of p-mTOR and degradation of LC3 I to LC3 II are the major mechanisms involved in autophagy (40), LC3 II protein levels were used to determine the extent of cell autophagy (41). After treating HCT116 cells with α-hederin for 24 h, cell lysates were used to detect p-mTOR and LC3 II protein levels. As presented in Fig. 5A, an increase in α-hederin concentration resulted in a gradual increase in LC3 II levels but a gradual decrease in p-mTOR protein levels. HCT116 cells were also treated with 10 µM α-hederin for 6, 12 and 24 h. The results demonstrated that, over time, α-hederin caused a gradual decrease in p-mTOR, p-ULK1, p-P70S6K and P62 protein levels but a gradual increase in p-AMPK and beclin-1 protein levels (Fig. 5B).

AMPK/mTOR is a major signaling pathway involved in autophagy (42). In this signaling pathway, AMPK serves as the activator of autophagy. AMPK activation induces dephosphorylation of mTOR, which separates it from the ULK1 complex. The subsequent dephosphorylation of ULK1 then initiates autophagy (43). To verify the role of the AMPK/mTOR pathway in α-hederin-induced autophagy, the expression of autophagy-related signals was detected in HCT116 cells treated with AMPK siRNA. It was demonstrated that AMPK siRNA restored the expression of p-mTOR, p-P70S6K and p-ULK1, which had been decreased by α-hederin (Fig. 5C). Results for p-AMPK indicated that although α-hederin increased LC3 II, AMPK knockdown did not restore LC3 II.

Previous studies have demonstrated that ROS is a major factor in α-hederin-induced apoptosis (37,44). ROS production is also important for AMPK/mTOR pathway activation. To verify the role of ROS in α-hederin-induced AMPK/mTOR pathway activation, oxidation-activated fluorescent dye DCHF-DA was used to detect intracellular ROS using flow cytometry. HCT116 cells were treated with 10 µM each of α-hederin and ROS inhibitor NAC for 24 h at 37°C then incubated in RPMI-1640 medium supplemented with 5 µmol/ml DCHF-DA for 30 min at room temperature. Fig. 6A demonstrated that although α-hederin increased ROS levels, ROS inhibitor NAC reversed this effect. Furthermore, NAC inhibited α-hederin-induced AMPK/mTOR pathway activation (Fig. 6B). Subsequently, the effects of NAC and α-hederin on apoptosis were detected by flow cytometry. As presented in Fig. 6C, NAC alone did not affect the apoptosis rate but NAC reduced the apoptosis rate treated by α-hederin in HCT116 cells (P<0.001). Additional, results from confocal microscopy demonstrated that NAC could inhibit α-hederin-induced autophagy (Fig. 6D).

To clarify the interplay between apoptosis and autophagy induced by α-hederin in colorectal cancer, the effects of autophagy inhibitor 3-MA on apoptosis and autophagy were investigated. HCT116 cells were treated with 2 mM 3-MA for 48 h to evaluate the effect of 3-MA alone on cell viability. The results indicated that 2 mM 3-MA alone had no significant effect on cell viability of HCT116 cells (Fig. 7A). Then, HCT116 cells were treated with α-hederin (2, 4, 8, 16, 32 and 64 µM) or α-hederin plus 2 mM 3-MA for 48 h, after which cell viability was tested. As presented in Fig. 7B, 2 mM 3-MA significantly reduced the effect of α-hederin on cell viability (P<0.05). Furthermore, HCT116 cells were treated with 2 mM 3-MA, 10 mM α-hederin and 2 mM 3-MA with 10 mM α-hederin for 48 h. As presented in Fig. 7C, 3-MA had no significant effect on α-hederin-induced apoptosis. Finally, apoptosis rate was detected by flow cytometry. It was also demonstrated that 3-MA blocked the α-hederin-induced autophagosome formation (Fig. 7D).