4-Acetylantroquinonol B (4-AAQB) was kindly provided by Grape King Bio Ltd. Sodium bicarbonate (NaHCO₃), 2′,7′-dichlorofluorescein diacetate (DCFH-DA), penicillin-streptomycin solution (PS), MTT, DMSO, TEMED, SDS, glycine, Tris, isopropanol, Tween-20, bovine serum albumin (BSA), Triton X-100, sodium chloride (NaCl), and GEM were purchased from MilliporeSigma. Propidium iodide (PI) and Annexin V staining kit (cat. no. 559763) was purchased from BD Biosciences. BCA protein assay kit (BC03-500) was purchased from Energenesis Biomedical. β-actin (cat. no. NB600-501), Beclin-1 (cat. no. NB110-87318), vascular endothelial growth factor-A (cat. no. VEGFA, NB100-664), LC3 II (cat. no. NB100-2220), and Atg5 (cat. no. NB110-53818) antibodies were purchased from Novus Biologicals. Bax (cat. no. 5023), Bcl-xL (cat. no. 2762), PI3K (cat. no. 4292), phospho-PI3K (cat. no. 4228), Akt (cat. no. 9272), phospho-Akt (cat. no. 9271), HMGB1, PI3K catalytic subunit type III (Vps34; cat. no. 4263), and MDR1 (cat. no. 6893) antibodies were obtained from Cell Signaling Technology, Inc. The anti-RAGE (cat. no. PA5-24787) antibody was purchased from Invitrogen (Thermo Fisher Scientific, Inc.). Dulbecco's modified eagle medium (DMEM) high glucose medium, fetal bovine serum (FBS), and horse serum were purchased from Gibco (Thermo Fisher Scientific, Inc.).

The Homo sapiens pancreatic cancer cell line MiaPaCa-2 was purchased from the Bioresource Collection and Research Center (Hsinchu, Taiwan). The GEM-resistant MiaPaCa-2^GEMR cell line was established by gradually increasing GEM concentrations to 0.5 µM GEM to induce tolerance as described in a previous report (21). Cells were cultured in DMEM (high glucose) medium supplemented with 10% fetal bovine serum (FBS) plus 2.5% horse serum and 1% PS antibiotic solution (100 U/ml penicillin and 100 µg/ml streptomycin). The two cell lines were maintained in a 37°C incubator with 5% CO₂.

MiaPaCa-2 and MiaPaCa-2^GEMR cells (8×10³ cells/well) were cultured in a 96-well plate overnight. The cells were incubated with various concentrations (0.1–5 µM) of 4-AAQB at 37°C for 48 h. The medium was discarded, and the cells were washed in PBS three times. Cells were incubated with MTT solution (0.5 mg/ml) for 1 h at 37°C. The MTT solution was removed and the formazan crystals were dissolved in 200 µl DMSO. Cell viability was determined using a microplate reader (FLUOstar Omega) at 570 nm and calculated as: Cell viability (%) = Absorbance of the sample/Absorbance of the control) ×100.

MiaPaCa-2 and MiaPaCa-2^GEMR cells (6×10⁴ cells/well) were cultured in a 24-well plate overnight. Cells were treated with 4-AAQB (2 and 5 µM) with or without GEM at 37°C for 48 h. The cells were then incubated with DCFH-DA at 37°C for 30 min. The cellular ROS level was measured using a Fluostar Galaxy reader (BMG Labtechnologies Ltd.) with maximum excitation and emission spectra of 485 and 520 nm, respectively. The level of ROS was calculated as the rate of change in contrast to the level in untreated cells.

MiaPaCa-2 and MiaPaCa-2^GEMR cells (6×10⁴ cells/ml) were cultured in a 6-well plate overnight. Cells were incubated with various concentrations (2 and 5 µM) of 4-AAQB at 37°C for 48 h. The medium was discarded and cells were washed with PBS three times. Cells were harvested by trypsinization and fixed with 70% ethanol at −20°C for 2 h. Cells (1×10⁵) were stained with 0.5 ml PI/RNase staining buffer (cat. no. 550825, BD Biosciences) at room temperature for 30 min. The cell cycle was evaluated using a BD Accuri C6 Plus flow cytometer (BD Biosciences). The cell cycle phase distribution was analyzed by ModFit LT 3.1 software (Verity Software House).

MiaPaCa-2 and MiaPaCa-2^GEMR cells (6×10⁴ cells/ml) were cultured in a 6-well plate overnight. Cells were incubated with 4-AAQB (2 and 5 µM) at 37°C for 48 h, double-stained with Annexin V and PI at room temperature for 15 min and analyzed using a BD Accuri C6 Plus flow cytometer (BD Biosciences). Quadrant 1 (negative for Annexin V and PI) indicates living cells, quadrant 2 (Annexin V-positive, PI-negative) indicates early apoptotic cells, quadrant 3 (positive for Annexin V and PI) indicates the late apoptotic cells, and quadrant 4 (Annexin V-negative, PI-positive) indicates necrotic cells. The data was analyzed by ModFit LT 3.1 software (Verity Software House).

MiaPaCa-2 and MiaPaCa-2^GEMR cells (1×10⁶) were incubated with 4-AAQB (2 and 5 µM) at 37°C for 48 h. The expression of proteins related to apoptosis (Bcl-xL and Bax), autophagy (Atg5, Beclin-1, and LC3 II), and GEM resistance-associated signaling (PI3K, phospho-PI3K, Akt, phospho-Akt, Vps34, RAGE, HMGB1, and MDR1) was determined using western blotting as described in a previous report (21). Briefly, cells were lysed by ice-cold RIPA lysis buffer (Thermo Fisher Scientific, Inc.) containing a cocktail of 1% protease inhibitors (P&C Biotech, Inc.) and 1% phosphatase inhibitors (P&C Biotech, Inc.), and the protein concentration was determined using a BCA protein assay kit (Energenesis Biomedical Co., Ltd.) according to the user guide. A total of 50 µg protein per lane was separated using 10–12% SDS-PAGE, then transferred to PVDF membranes and blocked with 5% non-fat milk in TBS-0.1% Tween-20 (TBST) at room temperature for 1 h. PVDF membranes were incubated primary antibodies (1:1,000) overnight at 4°C. After washing three times with TBST at room temperature for 10 min, the membranes were incubated HRP-conjugated secondary antibodies (1:10,000) at room temperature for 2 h. After antibody incubation, the protein was visualized using the ECL detecting reagent (PerkinElmer, Inc.), and the density of the target protein bands was quantified with the corresponding internal reference proteins by the Biospectrum 810 UVP VisionWorks LS Image Acquisition and Analysis Software (UVP).

MiaPaCa-2 and MiaPaCa-2^GEMR cells (4×10⁴ cells/ml) were cultured in a 96-well plate overnight. The increase in chemosensitivity was evaluated through both cotreatment with 4-AAQB and GEM and 4-AAQB pretreatment methods. For cotreatment, cells were incubated with 4-AAQB (2 and 5 µM) plus GEM at 37°C for 48 h. For pretreatment, cells were pretreated with 4-AAQB (2 and 5 µM) at 37°C for 48 h and then incubated with GEM at 37°C for another 48 h. In the pretreatment method, cells unexposed to 4-AAQB or GEM were incubated in complete medium until the end of the experiment. After treatments, chemosensitivity was determined using the MTT analysis.

Cell migration was evaluated using a gap closure assay. The Ibidi^® culture inserts were placed in a 24-well plate, and then MiaPaCa-2 and MiaPaCa-2^GEMR cells (5×10⁵ cells/ml) were seeded into the inserts and cultured with complete medium overnight. After cell attachment, the culture insert was removed and then the migration distance was recorded after exposure to 4-AAQB (2 and 5 µM) with or without GEM (0.5 µM) at 37°C for 48 h in serum free condition. The cell migration rate was calculated using ImageJ software (version 1.34s; National Institutes of Health). The migration area of the various treatment groups was calculated as a percentage of the untreated group using the following formula: Mean migration area of the experimental group/mean migration area of the control group ×100%.

Cell invasion was measured using a Transwell^® system with an 8-µm pore size (Corning, Inc.) as described previously (22). Briefly, 0.5 mg/ml Matrigel (Thermo Fisher Scientific, Inc.) was coated on the upper chambers of the inserts at 37°C for 1 h. MiaPaCa-2 and MiaPaCa-2^GEMR cells (5×10⁴ cells/well) were cultured with 4-AAQB (2 and 5 µM) under serum-free conditions at 37°C for 48 h on the upper chambers and medium containing 10% FBS served as a chemoattractant in the lower chambers. The invading cells were subsequently fixed in 100% methanol for 10 min at room temperature, stained with 0.5% crystal violet for 10 min at room temperature. The total invasive number of cells was calculated using the Power IX71 Inverted Fluorescence Microscope (Olympus Corporation).

All experiments were repeated independently three times. The data were analyzed using one-way ANOVA followed by Tukey's post hoc test using the SPSS 20 software (IBM, Corp.) and are shown as the mean ± SD. P<0.05 was considered to indicate a statistically significant difference.

MiaPaCa-2^GEMR cells were generated to explore the molecular mechanisms of GEM resistance in pancreatic cancer cells (21). The structure of 4-AAQB is shown in Fig. 1A. The effect of 4-AAQB on cell viability was evaluated in both MiaPaCa-2 and MiaPaCa-2^GEMR cells. Cytotoxicity was significantly increased by 4-AAQB treatment in a dose-dependent manner (0.1–5 µM) for 48 h (Fig. 1B and C). The IC₅₀ values of 4-AAQB in MiaPaCa-2 and MiaPaCa-2^GEMR cells were 1.61 and 3.8 µM, respectively (Fig. 1B and C). Doses of 2 and 5 µM were used in subsequent experiments.

Cell viability is directly affected by cell cycle arrest, apoptosis, and autophagy in cancer cells (18). S-phase cell cycle arrest was observed in MiaPaCa-2 cells subjected to 4-AAQB treatment, whereas the percentage of G0/G1 phase cells was significantly reduced in 4-AAQB-treated MiaPaCa-2 cells compared with untreated MiaPaCa-2 cells (Fig. 1D). In addition, this coordinated change in G2/M cell cycle arrest was observed in 4-AAQB-treated MIAPaCa-2^GEMR cells (Fig. 1E). Previous studies have found that cell cycle regulators not only influence cell division but also induce programmed cell death (23). In the present study, treatment with 4-AAQB led to dose-dependent cell cycle arrest, suggesting that 4-AAQB treatment may inhibit cell cycle progression in pancreatic cancer cells.

Emerging evidence suggests that to increase apoptosis and decrease autophagy are powerful therapeutic strategies for pancreatic cancer progression (24). In the present study, the percentage of late apoptotic cells was significantly induced in a dose-dependent manner following 4-AAQB treatment in both cell lines (Fig. 2). Western blot analysis was used to examine the expression of apoptosis-associated proteins. Consistent with published evidence (24), although the levels of Bcl-xL were not significantly altered by 2 µM 4-AAQB treatment, however, it was significantly reduced following 5 µM 4-AAQB treatment in MiaPaCa-2^GEMR cells compared with the untreated cells (Fig. 3A and B). The levels of Bax significantly increased in MiaPaCa-2^GEMR cells treated with 4-AAQB (Fig. 3A and C). Moreover, the Bax/Bcl-xL ratio significantly increased in 5 µM 4-AAQB-treated MiaPaCa-2 and MiaPaCa-2^GEMR cells (Fig. 3D).

The pro-apoptotic protein Bax directly inhibits autophagy (25). Furthermore, clinical evidence has revealed that autophagy is involved in resistance to gemcitabine/nab-paclitaxel chemotherapy and its cytotoxic response in patients with pancreatic cancer (26). Thus, the expression levels of autophagy-associated proteins (Atg5, Beclin-1 and LC3 II) were assessed using western blotting. The protein expression levels of Atg5 and LC3 II were significantly reduced after 4-AAQB treatment in both cell lines compared with the untreated cells (Fig. 4).

The PI3K/Akt signaling pathway is critical for tumor survival and autophagy initiation (27,28). The phosphorylation of PI3K and Akt significantly decreased in 4-AAQB-treated MiaPaCa-2 and MiaPaCa-2^GEMR cells compared with untreated cells (Fig. 5A-C). Vps34 binds to beclin-1 and subsequently activates autophagy (29). Additionally, a previous study indicated that VPS34 was essential for the autophagy process (30). In the present study, the expression of the Vps34 protein was significantly reduced following the 4-AAQB incubation in both cell lines (Fig. 5D and E).

As shown in our previous study, the PI3K/Akt/MDR1 axis is triggered by HMGB1 and RAGE engagement during GEM chemoresistance (19). Thus, the protein expression levels of HMGB1, RAGE, and MDR1 were examined. The levels of HMGB1 and RAGE were significantly inhibited after 5 µM 4-AAQB treatment in MiaPaCa-2 cells (Fig. 6A-C). Moreover, treatment with both 2 and 5 µM 4-AAQB effectively inhibited HMGB1 and RAGE expression in MiaPaCa-2^GEMR cells (Fig. 6A-C). Accordingly, the level of the MDR1 protein was reduced by 4-AAQB treatment in a dose-dependent manner in both cell lines (Fig. 6D and E).

A previous study showed that suppression of the PI3K/AKT pathway by inducing ROS generation may represent a strategy for cancer treatment (31). Additionally, GEM induces ROS, resulting in an increase in pancreatic cancer cell death, although the opposite effect is observed in GEM-resistant cells (32). In the present study, ROS levels were measured using DCFH-DA staining. The results showed that ROS levels significantly increased in a dose-dependent manner following 4-AAQB treatment in both cell lines (Fig. 7A and B). Notably, ROS accumulation was dramatically increased by 4-AAQB combination with GEM treatment in MiaPaCa-2^GEMR cells compared with 4-AAQB-treated alone or GEM-treated alone condition (Fig. 7B).

Pancreatic cancer is highly malignant with great metastatic capacity (33). An increase in chemosensitivity is important for preventing the local invasion and long distant metastasis of pancreatic cancer (34). The ability of 4-AAQB to increase chemosensitivity was evaluated using 4-AAQB cotreatment with GEM and 4-AAQB pretreatment methods in both cell lines to select an effective drug treatment pattern. Consistent with the results presented in Fig. 1B, cell viability was significantly reduced by GEM and 4-AAQB (2 and 5 µM) treatment in MiaPaCa-2 cells compared with untreated cells (Fig. 8A). Compared with GEM treatment, chemosensitivity was effectively enhanced following 4-AAQB cotreatment with GEM in MiaPaCa-2 cells (Fig. 8A). Cell viability was not altered following 0.5 µM GEM treatment in MiaPaCa-2^GEMR cells (Fig. 8B), indicating that the GEM-resistant pancreatic cancer cell lines was successfully established. Cell viability was significantly reduced both by 4-AAQB treatment alone and by 4-AAQB/GEM co-treatment in cells (Fig. 8B). In addition, the effect of the 4-AAQB pre-treatment on enhancing chemosensitivity was also evaluated. Cells were pretreated with 4-AAQB for 48 h, then incubated with GEM for another 48 h. In MiaPaCa-2 cells, the GEM and 4-AAQB (2 and 5 µM) treatment significantly reduced cell viability (Fig. 8C). Additionally, chemosensitivity was significantly increased following 4-AAQB pre-treatment in MiaPaCa-2 cells compared with GEM-treated cells (Fig. 8C). In addition, cell viability was significantly inhibited following 4-AAQB (2 and 5 µM) pre-treatment with or without GEM incubation in MiaPaCa-2^GEMR cells (Fig. 8D).

4-AAQB treatment effectively repressed VEGFA-mediated cell migration and invasion. VEGFA plays important roles in cell proliferation, angiogenesis, migration, invasion, and cancer metastasis (35). According to a previous study, the upregulation of VEGFA in patients with pancreatic cancer is associated with metastatic disease and shorter overall survival (36). The next experiments were carried out to assess whether 4-AAQB treatment was effective at inhibiting VEGFA production and subsequently preventing cell migration and invasion. Interestingly, VEGFA protein levels significantly decreased in both cell lines subjected to 4-AAQB treatment (Fig. 9A and B). Cell migration was inhibited following 4-AAQB, GEM, and 4-AAQB/GEM co-treatment in MiaPaCa-2 cells compared with untreated cells (Fig. 9C and D). Unexpectedly, the cell migration distance was not substantially reduced by 4-AAQB/GEM co-treatment compared to 4-AAQB treatment alone in MiaPaCa-2 cells (Fig. 9C and D). In addition, 4-AAQB (2 and 5 µM) treatment effectively decreased MiaPaCa-2^GEMR cell migration compared to untreated cells (Fig. 9E and F). Notably, 2 µM 4-AAQB/GEM co-treatment significantly decreased cell migration compared to that of 4-AAQB-treated MiaPaCa-2^GEMR cells (Fig. 9E and F). In addition, the inhibitory effect of 4-AAQB on cell invasion was also evaluated. The cell invasion abilities of MiaPaCa-2 and MiaPaCa-2^GEMR cells were significantly reduced by 4-AAQB (2 and 5 µM) treatment (Fig. 10). Notably, 4-AAQB combined with GEM significantly enhanced the inhibition of cell invasion compared with 4-AAQB-treated MiaPaCa-2 cells l and MiaPaCa-2^GEMR cells (Fig. 10B and D).