Human pancreatic cancer cell lines SW1990 and PANC-1 were purchased from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The human normal pancreatic cell line HPDE6-C7 was obtained from BLUEFBIO biological company (Shanghai, China). All cell lines were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Hyclone, United States), supplemented with 10% fetal bovine serum (FBS, Gibco, United States), 100 U/mL penicillin, and 100 μg/mL streptomycin (Solarbio, China), and incubated at 37°C in a humidified atmosphere with 5% CO₂.

TEOA was obtained from college of pharmacy, Zhejiang University (Zhejiang, China) and dissolved in DMSO. N-acetylcysteine (NAC), Chloroquine (CQ), 3-Methyladenine (3MA) and Rapamycin were purchased from Medchem Express (MCE, United States). DAPI was purchased from Beyotime (Shanghai, China). The following primary antibodies were used: anti-GAPDH (Abcam ab181602) (1:2,000), anti-Catalase (Abcam ab76024) (1:1,000), anti-MnSOD (Abcam ab68155) (1:2,000), anti-LC3 (Sigma L7543) (1:1,000), anti-p-mTOR (Abcam ab109268) (1:2,000), anti-mTOR (Abcam ab134903) (1:1,000), anti-p-p70S6K (Abcam ab131436) (1:1,000), anti-p70S6K (Abcam ab32529) (1:1,000), anti-p-S6 (CST 4858S) (1:2,000), anti-S6 (CST 2217s) (1:1,000), anti-p-4EBP1 (CST 9459S) (1:1,000), anti-4EBP 1 (CST 9644S) (1:1,000), anti-ATG5 (Abcam ab68155) (1:2,000). The related HRP-conjugated secondary antibody was purchased from Beyotime (Shanghai, China).

SW1990, PANC-1 and HPDE6-C7 cells (2 × 10⁴ per well) were plated in 96-well plates (NEST Biotechnology). Cells were treated with TEOA (0–60 μM) for 12 h, 24 h at 37°C. The viability of cells was detected by the Cell Counting Kit 8 (CCK-8) assay. The optical density of each well was measured at 450 nm with a microplate reader (Thermo, United States).

SW1990, PANC1 and HPDE6-C7 cells (2 × 10⁴ per well) were plated in 96-well plates. After being treated with TEOA (0–60 μM) for 12 h at 37°C, LDH activities of TEOA treated cells were measured by LDH cytotoxicity assay kit (Beyotime, Shanghai, China) according to the manufacturer’s instructions.

To further detect the effect of TEOA on normal cells and pancreatic cancer cells, SW1990 and HPDE6-C7 cells (2 × 10⁴ per well) were plated in 96-well plates. After 12 h, cells were treated with TEOA (0–50 μM) for the indicated time at 37°C. Then the living cells and the dead cells were visualized by Calcein/PI cell viability/cytotoxicity assay kit (Beyotime, Shanghai, China) and photographed under EVOS M7000 (Thermo Fisher Scientific, United States).

SW1990 and PANC1 cells were seeded at 6 × 10⁵ cells per well in 6-well plates overnight and then treated with vehicle (DMSO) or TEOA at various concentrations and times. Then the cellular morphology change was observed under the microscope and photographed (Nikon, Japan).

5-ethynyl-2’-deoxyuridine (EdU) staining was performed by a commercial kit (Beyotime, China) according to the manufacturer’s instructions. Briefly, treated cells were incubated with 10 μM EdU for 2 h at 37°C and then fixed with 4% paraformaldehyde for 15 min at 37°C. Subsequently, the cells were permeabilized with 0.3% Triton X for 15 min at room temperature. After washing twice with PBS, the cells were incubated with a click-reaction cocktail in the dark for 30 min. DNA was stained with DAPI for 5 min and visualized under confocal microscopy (Leica, Germany).

For the colony assay, 1 × 10³ cells were seeded in 12-well plates and cultured with TEOA for 7 days. Cells were fixed with methanol for 15 min before staining with 0.5% crystal violet (Beyotime, Shanghai, China). Colonies were counted using ImageJ and then normalized to the number of colonies in the control group.

For the cell migration assay, cells (4 × 10⁴ cells/well) were incubated in the upper transwell chambers (Corning, United States) with the indicated concentration of TEOA dissolved in serum-free DMEM, and DMEM containing 10% FBS was added to the lower chamber as an attractant. After 24 h, the cells on the lower membrane were fixed with 4% paraformaldehyde for 15 min and stained with 0.1% crystal violet. Finally, the cells were photographed under a microscope (Nikon, Japan), and the number of cells was quantified.

To evaluate the effect of TEOA on cell migration in vitro, we performed a scratch assay. Cells were seeded onto a 6-well plate at a density of 8 × 10⁵ cells/well. A sterile 10 μL pipette tip was used to make a straight scratch line on the confluent cell monolayer when the cells had nearly reached confluency. The medium was replaced with 2% FBS medium containing TEOA. Cell scratches were visualized at 0, 24, and 48 h and imaged by inverted microscope (Nikon, Japan).

Cells were seeded in a confocal dish (NEST, China) and treated with TEOA at the designated concentrations, then stained with Mitotracker, Mitosox (Thermo Fisher Scientific, United States), Magic Red cathepsin B (Immunochemistry Technologies) or DCFH-DA (Sigma-Aldrich, St. Louis, MO, United States) for 30 min in the dark. Nuclei were co-stained with DAPI (10 μg/mL). The cells were viewed and imaged under a confocal microscope (Leica, Germany).

Fluorescent probe Tetramethyl Rhodamine Methyl Ester (TMRM, Thermo Fisher Scientific, Waltham, MA, United States) was used to evaluate the MMP. Cells were seeded in 6-well plate (6 × 10⁵ cells per well) overnight and treated with TEOA for 8 h. Then the cells were stained with TMRM (200 nM) for 30 min in the dark, and followed by flow cytometry detection.

To measure the intracellular ATP levels, cells were seeded onto 6-well plates (6 × 10⁵ cells per well) and then treated with TEOA at different concentrations for 8 h. Intracellular ATP levels were measured using ATP assay kits in accordance with the manufacturer’s instructions (Beyotime, China).

The fluorescent probe DCFH-DA (Sigma-Aldrich, St. Louis, MO, United States) was used to measure intracellular ROS production. PANC-1 and SW1990 cells were seeded onto 6-well plates (6 × 10⁵ cells per well) and cultured for 12 h. Subsequently, the cells were directly exposed to TEOA for 8 h at the indicated concentrations and cotreated with or without NAC (0.5 mM). Treated cells were harvested, and intracellular ROS levels were determined by flow cytometry or microplate spectrophotometer (BioTek, United States) according to the manufacturer’s instructions.

Cells were seeded onto a 6-well plate (7 × 10⁵ cells per well) and cultured for 12 h. The cells were treated as designed for 8 h. After being collected and lysed with RIPA buffer containing complete protease and phosphatase inhibitor (Thermo, Waltham, MA, United States) for 10 min, the lysates were centrifuged at 12,000 × g for 15 min at 4°C. BCA Protein Assay Kit (Beyotime, China) was used to quantify the protein concentration in the supernatant. Equal amounts of proteins were separated using polyacrylamide gel electrophoresis and subsequently transferred to a polyvinylidene fluoride membrane (Bio-Rad, United States). The membrane was blocked with 5% skim milk in Tris-buffered saline with 0.1% Tween-20 (TBST) for 2 h and incubated with primary antibodies overnight at 4°C. The next day, the membrane was washed three times with TBST and incubated with the appropriate secondary antibodies for 1 h at room temperature. Finally, the immunoblots were visualized using an enhanced chemiluminescence system (Bio-Rad, United States). ImageJ was used to quantify the band intensities.

We used mRFP-GFP-LC3 adenovirus (Hanbio, Shanghai, China) to further confirm the autophagic flux. Briefly, mRFP-GFP-LC3 adenovirus was transfected into SW1990 cells, in accordance with the manufacturer’s instructions. After 36 h, transfected SW1990 cells were treated as described, followed by staining with DAPI (10 μg/mL) at 37°C for 15 min. Autophagic vesicles were observed and photographed under a confocal fluorescence microscope.

Transmission electron microscope was used to determine the occurrence of autophagy. SW1990 cells were first fixed with 2.5% glutaraldehyde for 4 h and postfixed with 1% OsO₄ in phosphate buffer for 2 h. Then the samples were dehydrated by a graded series of ethanol (30, 50, 70 and 80%) alcohol. Subsequently, the specimens were sectioned in LEICA EM UC7 ultratome and stained by uranyl acetate and alkaline lead citrate for 5 min, respectively. The ultrastructural characteristics of autophagy in different treatment groups were observed by TEM (Hitachi, H-7650, Japan).

For knockdown of ATG5, target shRNA sequences were subcloned into pLVX-shRNA lentivirus vector (Takara, Japan). The shRNA sequences for ATG5 knockdown were shown below. shATG5 forward sequence: GATCCGCCTGAACAGAATCATCCTTAATTCA AGAGATTAAGGATGATTCTGTTCAGGTTTTTTG, shATG5 reverse sequence: AATTCAAAAAACCTGAACAGAATCA TCCTTAATCTCTTGAATTAAGGATGATTCTGTTCAGGCG. The recombinant lentiviral plasmids were verified by sequencing. The empty vector was used as a control.

Cell apoptosis was detected by Annexin V-FITC/PI apoptosis kit (MultiSciences, China) with a flow cytometry. Cells were seeded in 6-well plates and cultured with TEOA at indicated concentrations for 12 h. The cells were then harvested and stained with Annexin V-FITC/PI for 5 min, then followed by flow cytometry.

Data are exhibited as mean ± standard deviation (mean ± SD) from three independent experiments. IC50s were calculated using non-linear regression by GraphPad Prism. Log dose-response curves were plotted using the non-linear regression (Dose-response Inhibition-log (agonist) vs. response module). The differences between two groups were detected using the Student’s t test. The differences between multiple groups were detected using the analysis of variance (ANOVA). p < 0.05 was defined as statistically significant difference. ^★p < 0.05, ^★★p < 0.01, ^★★★p < 0.001.

To evaluate the selective anti-cancer activity of TEOA in pancreatic cancer cells, the cytotoxic effects of TEOA on two pancreatic cancer cell lines (PANC1, SW1990) and normal pancreatic cell line HPDE6-C7 were detected using the CCK-8 assay and LDH release assay. CCK-8 assay showed that TEOA significantly reduced the viability of cancer cells in a dose-dependent manner (p < 0.01). The IC-50s of TEOA at 12 h in the normal pancreatic cell line and pancreatic cancer cell lines were 47.84 μM (HPDEC-C7), 38.02 μM (PANC1), and 32.71 μM (SW1990), respectively (Figure 1A). At 24 h, TEOA also showed preferential tumor cytotoxicity (Supplementary Figure 1A). Similarly, LDH release from TEOA-treated PANC1 and SW1990 cells were obviously higher than TEOA-treated HPDE6-C7 cells (Figure 1A). In Supplementary Figure 1B, Calcein/PI cell viability/cytotoxicity assay further confirmed that TEOA preferentially targeted pancreatic cancer cells, while sparing low toxicity on non-malignant cells.

In addition, changes in cell morphology were monitored using a light microscope. With the increase in time and drug concentration, severe vacuolation was observed in the cytoplasm and the cancer cells tended to be fragmented (Figure 1B). EdU staining and colony formation assay were also performed to evaluate the inhibitory effects of TEOA on cell proliferation. As shown in Figure 1C, with an increase in TEOA concentration, the number of EdU-positive cells decreased gradually. Consistently, Figure 1D shows a concentration-dependent decrease in colony formation. Taken together, our findings indicated that TEOA exhibited an efficient cytotoxic effect on pancreatic cancer cells.

In addition to proliferation, migration of cancer cells is another important hallmark of cancer progression. Therefore, we performed cell transwell assays and scratch tests to determine the influence of TEOA on cell migration. The results showed that TEOA significantly suppressed the migration of pancreatic cancer cells in a dose-dependent manner. The relative migration rate (%) decreased to 49.34 ± 11.69% (p < 0.01), 38.69 ± 4.70% (p < 0.001), and 12.87 ± 2.89% (p < 0.001) after treatment with 15, 20, and 25 μM TEOA, respectively. In PANC1 cells, the relative invasion rate (%) decreased to 65.18 ± 1.84% (p < 0.01), 36.72 ± 6.50% (p < 0.001), and 22.88 ± 7.44% (p < 0.001), respectively (Figures 2A,B). Consistently, the scratching test also exhibits a similar result (Figures 2C,E). The relative migration rate of SW1990 cells incubated with 25 μM TEOA decreased to 37.54 ± 4.57% compared with that of the control group at 48 h (p < 0.01) (Figure 2D). The relative migration rate of PANC1 cells treated with 25 μM TEOA decreased to 38.67 ± 2.99% (p < 0.01) (Figure 2F). Similar trends were observed at 24 h. On the contrary, TEOA exhibits lesser effects on the migration of HPDE6-C7 cells (Supplementary Figure 2). These results revealed that TEOA suppressed the migration of SW1990 and PANC1 cells.

Mitochondria coordinate a large fraction of metabolic, energetic, and physiological processes, and their functional integrity is crucial for the proliferation, migration, and metabolism of cancer cells (Picard et al., 2018). We verified whether TEOA could impair mitochondrial homeostasis in pancreatic cancer cells. We first focused on mitochondrial morphology, an early event of mitochondrial dysfunction, which was visualized by the MitoTracker Red probe and observed under a confocal fluorescence microscope. Compared with the healthy tubular network structure of mitochondria in the control group, TEOA induced concentration-dependent changes in mitochondrial morphology, characterized by reduced cristae and fragmentation, in SW1990 and PANC1 cells (Figures 3A,B). In addition to morphological changes, MMP, another reliable indicator to evaluate mitochondrial functional damage, was monitored by the fluorescent dye TMRM (Kang et al., 2017). The results showed that the MMP of two pancreatic cancer cells was significantly decreased after treated with different concentrations of TEOA (Figures 3C,D). Because MMP is maintained through mitochondrial oxidative phosphorylation, we detected the ATP turnover. As expected, ATP levels in pancreatic cancer cells treated with TEOA showed a similar downward trend (Figures 3E,F). In summary, our results emphasize that TEOA treatment impairs mitochondrial function in pancreatic cancer cells.

Once mitochondrial dysfunction occurs, large quantities of ROS leak from the electron transport chain, resulting in oxidative stress (Panel et al., 2018). The DCFH-DA probe was then used to detect ROS levels in cells treated with TEOA. Confocal images revealed a significant increase of ROS levels in PANC1 and SW1990 cells treated with ascending dose of TEOA (Figures 4A,B). Notably, we could observe a significant accumulation of mitochondrial ROS after the treatment of TEOA, whereas TEOA did not induce significant oxidative stress in HPDE6-C7 cells (Supplementary Figure 3). Consistent with the results of fluorescent confocal microscopy, flow cytometry analysis of DCF also demonstrated that TEOA facilitated the accumulation of cellular ROS (p < 0.05; Figures 4C,D). Furthermore, the expression of oxidative stress-related proteins after TEOA treatment was determined by western blot. The results indicated that antioxidant molecules, including MnSOD and catalase, were upregulated and activated as an adaptive response to maintain tumor cell survival (Figures 4E–H). Furthermore, we found that co-treatment with the ROS scavenger NAC could not restore the MMP, which confirmed the upstream position of mitochondrial dysfunction (Supplementary Figure 4). These results revealed that TEOA impaired mitochondrial homeostasis and thus led to an oxidative stress response.

Mitochondria plays an indispensable role in maintaining the normal physiological processes in cells (Pustylnikov et al., 2018). ROS accumulation due to mitochondrial dysfunction results in the initiation of cell autophagy (Hatchi et al., 2011; Yang et al., 2019). Hence, we measured the levels of autophagy-related proteins in TEOA-treated pancreatic cells to investigate whether TEOA induces autophagy. LC3 is a classical autophagy-related protein, and the conversion of LC3-I to LC3-II is a reliable marker of autophagy initiation. We found that TEOA significantly induced the conversion of LC3, while did not exhibit significant effects on the ratio of LC3II/I in HPDE6-C7 cells (Figures 5A,B; Supplementary Figure 5). The activation of autophagy is accompanied by the formation of autophagosome and the activation of lysosome. Therefore, we used Ad-GFP-LC3B combined with lysosome staining to verify the occurrence of autophagy. As shown in Figure 5C, GFP-LC3B fluorescence is diffusely distributed in the cytoplasm in the basal conditions. After the treatment of TEOA, GFP-LC3B aggregated on autophagosome membrane and co-stained with cathepsin B to form the autolysosome (Figure 5D). Our TEM results also revealed the occurrence of autophagy in TEOA-treated SW1990 cells (Supplementary Figure 6).

In addition, we further verified the autophagic flux using a tandem-tagged mRFP-GFP-LC3 adenovirus and evaluated the results by confocal microscopy. The appearance of red vesicles after TEOA treatment marked the autolysosomal event that occurred after the fusion of lysosomes with autophagosomes. When the autophagic flux is disrupted during the fusion of lysosomes with autophagosomes, autophagosomes are present as yellow vesicles. Compared with TEOA-treated alone, combination of rapamycin further enhanced autophagic flux (p < 0.01). While, ATG5 knockdown effectively inhibited TEOA-induced autophagy (p < 0.01) (Figures 5E,G). The knockdown efficiency of ATG5 in SW1990 cells was evidenced by the western blot assay (Figure 5F). Therefore, we can conclude that TEOA is capable of promoting autophagic flux.

To further demonstrate the mechanisms of autophagy induced by TEOA, we analyzed the AMPK/mTOR/p70s6k/S6 signaling pathway, which was previously demonstrated to play a role in the initiation of autophagy. As shown in Figure 6, TEOA decreased the phosphorylation of mTOR and its downstream proteins, including p70s6k, S6, and 4EBP1. Furthermore, we observed the increase of p-AMPK in TEOA treated pancreatic cancer cells (Supplementary Figure 7). Collectively, TEOA treatment significantly affecting the AMPK/mTOR/p70s6k/S6 signaling pathway, thus increasing the autophagic flux, which might play a critical role in the cytotoxicity of TEOA.

Autophagy is a catabolic process that plays a central role in maintaining cell metabolism and renewal. In cancer cells, autophagy has a dual role, acting as an adaptive stress response to maintain tumor cell survival, but excessive stimulation of autophagy can be lethal for cancer cells owing to the excess elimination of cytoplasmic content. To confirm the role of TEOA-induced autophagic flux in PDAC cells, we silenced ATG5 using lentivirus-mediated shRNA to interfere with autophagy and evaluated the cytotoxicity of TEOA. The results of western blotting and confocal microscopy confirmed that ATG5 knockdown successfully inhibited TEOA-induced autophagy (Figures 5G, 7C). CCK-8 assay showed that the cytotoxic effect of TEOA was partially abrogated by ATG5 knockdown (Figure 7A). In addition, the autophagosome formation inhibitor (early autophagy inhibitor, 3MA) and autophagosome-lysosome fusion inhibitor (late autophagy inhibitor, CQ) were utilized to confirm these findings. CCK-8 assay showed that both of them partly abrogated the decrease in cell viability following treatment with TEOA (Figure 7B). The autophagy inhibitory efficiency of CQ was also confirmed by western blotting (Figure 7D). We next aimed to determine whether enhanced activation of autophagy exerted a synthetic effect with TEOA and provided different lines of evidence to support the cytotoxicity model. Both administration of rapamycin and EBSS induced starvation sensitized the anticancer activity of TEOA in SW1990 cells (Figures 7E,F). Collectively, these data from both pharmacological and genetic approaches suggest that excessive activation of autophagy acts synergistically with TEOA and renders PDAC cells vulnerable to death, whereas inhibition of autophagy can partly abolish the cytotoxicity of TEOA.

The above mentioned results confirmed that ROS accumulation oriented from mitochondrial dysfunction acts upstream of autophagosome formation, leading to the suppression of mTOR and activating of autophagy. Therefore, it is reasonable to assume that TEOA induced ACD could be mitigated by the inhibition of ROS production. The commonly used ROS scavenger NAC was selected for further study. We found that NAC could notably reverse the oxidative stress induced by TEOA (Figures 8A,B). Furthermore, we examined the changes in morphological features and cell viability following treatment with TEOA in the presence or absence of NAC. As expected, both cell morphology features and cell viability assays validated that coincubation with NAC alleviated cell death induced by TEOA (Figures 8C,D). We further performed western immunoblotting and found that NAC cotreatment restored TEOA-induced mTOR/p70s6k pathway inhibition and alleviated the conversion of LC3 (Figure 8E; Supplementary Figures 8, 9). Collectively, our results confirmed that NAC attenuated excessive activation of autophagy by restoring the mTOR signaling pathway and alleviating ACD induced by TEOA treatment.