The PANC-1 PDAC cell line was purchased from Procell Life Science & Technology Co., Ltd. (Wuhan, China). The cell lines used in this study were authenticated by short tandem repeat (STR) profiling and and tested negative for mycoplasma. Small interfering RNAs (siRNAs) against Beclin 1 were synthesized by GenePharma (Shanghai, China), the sequences are shown in Table 1. We obtained the pCMV3-COL11A1 plasmid from Sino Biological (Cat.No: HG18256-UT; Beijing, China). LY294002 (a PI3K inhibitor) and erastin (a ferroptosis promoter) were obtained from Beyotime (Shanghai, China). Rapamycin (RAPA, autophagy promoter) was obtained from MedChemExpress (Cat.No: HY-10219, Monmouth Junction, NJ, USA).

Cells were seeded in 96-well plates before receiving different treatments. Cell viability was assessed at 24, 48, and 72 h intervals using a Cell Counting Kit-8 (CCK-8) assay in accordance with the manufacturer’s instructions. After the indicated treatments, cells seeded in a 96-well plate were incubated with 10 µL of CCK-8 reagent (Cat.No: CK04, Dojindo, Kumamoto, Japan) per 100 µL of culture medium for 1–4 h at 37 °C in a humidified incubator. The absorbance at 450 nm was then measured.

All cellular extracts were obtained by lysis in radio-immunoprecipitation assay (RIPA) buffer (Solarbio, Beijing, China). Protein concentrations were determined using a bicinchoninic acid (BCA) kit (ThermoFisher Scientific, Waltham, MA, USA). Subsequently, 30 µg protein per sample was separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and electrotransferred onto a polyvinylidene difluoride (PVDF) membrane. After blocking with 5% non-fat milk, the membrane was incubated with primary antibodies at 4 °C overnight, followed by incubation with the corresponding secondary antibody for 1 h. Finally, protein bands were detected using ECL reagent (Millipore, Burlington, MA, USA) and visualized using the LAS500 gel imaging system (GE, Massachusetts, NY, USA). The antibodies used are listed in Table 2.

Total RNA was extracted using the Total RNA Extraction Kit (Promega, Madison, WI, USA) according to the manufacturer’s instructions. Then, it was reverse-transcribed into cDNA using a HiFiScript first-strand cDNA synthesis kit (CWBIO, Beijing, China). Real-time PCR was performed using ChamQTM Universal SYBR qPCR Master Mix (CWBIO, Beijing, China) and detected using an Applied Biosystems 7500 Fast Dx Real-time PCR Instrument (Thermo Fisher Scientific, Rockford, IL, USA). The relative mRNA expression was calculated using the 2^{-Δ⁢Δ⁢Ct} method. The primer sequences used in this study are listed in Table 3.

PANC-1 cells were seeded onto coverslips placed in a 24-well plate. After treatment, the cells on the coverslips were fixed and permeabilized followed by blocking in 10% goat serum for 1 h. Then the coverslips were incubated with LC3 antibody (1:200) at 4 °C overnight, followed by incubation with fluorescently labeled secondary antibody at room temperature for 1 h. Nuclei were stained with DAPI. After washing three times with phosphate buffered saline with Tween-20 (PBST), images were captured using a fluorescence microscope.

FeRhoNox-1 (Cat. No.: MX4558; MaokangBio, Shanghai, China) was used to detect intracellular iron content. Cells were seeded in 24-well plates, cultured overnight, and then treated with rhCOL11A1/pCMV3-COL11A1 and erastin for 48 h. After collection, cells were washed twice with preheated PBS, incubated with 5 µM FeRhoNox-1 at 37 °C for 60 min in the dark, and then washed three times with preheated PBS. Nuclei were stained with DAPI, followed by three additional PBS washes, and images were acquired by fluorescence microscopy. Differences in average fluorescence intensity reflect varying intracellular iron levels.

BODIPY™ 581/591 C11 (Cat.No.: D3861; Thermo Fisher Scientific, USA) was used to detect intracellular lipid peroxidation. Cells were seeded in 24-well plates, cultured overnight, and treated with rhCOL11A1/pCMV3-COL11A1 and erastin for 48 h. After collection, cells were washed twice with preheated PBS, incubated with 10 µM BODIPY™ 581/591 C11 at 37 °C for 30–60 min in the dark, and washed three times with preheated PBS. Nuclei were stained with DAPI, followed by three additional PBS washes. Images were captured via fluorescence microscopy, and differences in average fluorescence intensity reflect varying levels of lipid peroxidation.

PANC-1cells were seeded onto coverslips that had been previously placed in a 24-well plate. Following treatment, cells were washed and incubated with JC-1 working solution at 37 °C for 30 min. After washing three times with pre-warmed buffer to eliminate unbound dye, fluorescence detection was performed immediately, with red fluorescence indicating aggregated JC-1 and green fluorescence representing monomeric JC-1. A Mitochondrial Membrane Potential Detection Kit (JC-1) was purchased from Beyotime (No: C2006).

Cells were seeded into 6-well plates and cultured overnight. After different treatments, each sample was collected and washed twice with PBS. The cell precipitate was resuspended in a reagent with a volume three times that of the precipitate, and the resulting suspension was lysed by ultrasound (200 W, ultrasound 3 s, interval 10 s, repeated 30 times) and centrifuged at 4 °C for 10 min. Then the samples were analyzed using a glutathione (GSH) content detection kit (No: BC1175; Solarbio).

After treatment, the cells were collected and incubated in lysis buffer for 2 h at 4 °C. Then the mixture was centrifuged for 10 min to obtain the supernatant for subsequent analyses. Cell samples were lysed in ice-cold RIPA buffer containing protease inhibitors. After centrifugation, the supernatant was collected. For the reaction, 100 µL of supernatant was mixed with 200 µL of 8.1% SDS, 750 µL of thiobarbituric acid (TBA) working solution (0.8% thiobarbituric acid in 20% acetic acid, pH 3.5), and 750 µL of 20% acetic acid. The mixture was vortexed, incubated at 95 °C for 60 minutes, and then cooled on ice. Absorbance was measured at 532 nm in a 96-well plate. MDA concentration was quantified against a standard curve prepared using tetraethoxypropane and normalized to total protein content determined by the BCA assay. Lipid peroxidation was quantified using an MDA assay kit (No: S0131S; Beyotime).

Cells were seeded in a 24-well plate and treated for 48 h. After discarding the culture medium, the samples were processed for analysis according to the manufacturer’s protocol. Intracellular ROS levels were measured via immunofluorescence assays using an ROS assay kit (No: S0033s; Beyotime).

PANC-1 cells were seeded into 6-well plates at a density of 3 $\times$ 10³ cells per well and cultured under standard conditions (37 °C, 5% CO₂) for 7–14 days while receiving separate treatments. The cells were fixed for 15 min, air-dried, and stained with Giemsa stain for 20 min before being washed with PBS. The colony formation rate was calculated using the following formula: (number of colonies formed $÷$ number of inoculated cells) $\times$ 100%.

The nude mice (BALB/c-nu, weighing 13~16 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. To establish subcutaneous xenograft models, 20 nude mice were randomly divided into two groups. The control group (n = 5) received subcutaneous injection of PANC-1 cells at a dose of 5 $\times$ 10⁶ cells per mouse, while the experimental group (n = 15) was subcutaneously injected with COL11A1-overexpressing PANC-1 cells at the same dosage (5 $\times$ 10⁶ cells/mouse) to evaluate the effect of COL11A1 overexpression on tumor growth.

Treatment began when the implanted tumors reached a volume of approximately 50 mm³, and the mice were divided into four groups. The control group was administered with physiological saline via intraperitoneal injection. A total of 15 nude mice inoculated with the COL11A1-overexpressing cell line were equally divided into three subgroups, with the following treatment regimens applied via intraperitoneal injection respectively: The first subgroup received physiological saline; the second subgroup was treated with erastin at a dose of 30 mg/kg, 2–3 times per week; the third subgroup was given RAPA at a dose of 4 mg/kg, 2–3 times per week. All experimental animals were humanely euthanized by cervical dislocation in accordance with institutional animal care and use guidelines.

Tumor volumes were calculated from formula: TV = 1/2 $\times$ a $\times$ b², where “a” is the tumor length and “b” is the width and both “a” and “b” were measured by a vernier caliper. Post treatment mice were sacrificed on day 22, and tumors were removed for further analysis.

The experimental procedures and the animal use and care protocols were approved by the Committee on Ethics of Tianjin Institute of Medical and Pharmaceutical Sciences.

All statistical analyses and graphical visualizations were conducted with GraphPad Prism v 6.0.1 for Windows (GraphPad Software, Boston, MA, USA). Data from three independent experiments are presented as the mean $\pm$ standard deviation (SD). To compare the experimental groups with the control group, Student’s t-test and analysis of variance were employed. p 0.05 was considered statistically significant.

Kaplan Meier survival analysis showed that the expression of COL11A1 was negatively correlated with the survival rate of patients with pancreatic cancer (Fig. 1a). The results of the CCK-8 assay showed that COL11A1 inhibited PANC-1 cell death, and erastin and RAPA weakened the effects of COL11A1 on PANC-1 cells (Fig. 1b,d). Moreover, cloning experiments confirmed that COL11A1 inhibits PANC-1 cell death, and that its inhibitory mechanism may be linked to autophagy and ferroptosis (Fig. 1c).These results indicate that COL11A1 inhibits pancreatic cancer cell death by regulating autophagy and ferroptosis.

Fig. 1.

COL11A1 inhibited PANC-1 cell death by regulating autophagy and ferroptosis. (a) Kaplan–Meier Analysis of the correlation between COL11A1 expression and survival outcomes in patients with pancreatic cancer. (b) CCK-8 assay was used to detect cell viability under different treatments with COL11A1, erastin (10 µM), or RAPA (100 nM). (c) Colony formation assay was performed in PANC-1 cells with different treatments. (d) CCK-8 assay was performed, and the results are presented as a bar chart showing the viability of PANC-1 cells among different treatment groups. Data are expressed as the mean $\pm$ SD from three independent experiments with triplicate wells per group. n = 3, *p 0.05; **p 0.01; ns, not significant. COL11A1, collagen type XI alpha 1; CCK-8, Cell Counting Kit 8; RAPA, rapamycin.

To further validate that COL11A1 participates in regulating ferroptosis in pancreatic cancer cells, we performed western blotting to assess the expression of key ferroptosis-related proteins under COL11A1 modulation. The data demonstrated that COL11A1 significantly enhanced the protein and mRNA expression levels of GPX4, ferritin heavy chain 1 (FTH1), and solute carrier family 7 member 11 (SLC7A11), supporting its regulatory role in the ferroptosis process (Fig. 2a and Fig. 3c). Iron ion content and lipid peroxidation are key indicators of ferroptosis. Specifically, MDA is an end-product marker of lipid peroxidation, whereas GSH is a key antioxidant molecule inhibiting lipid peroxidation. In PANC-1 cells treated with COL11A1, MDA levels were decreased whereas GSH levels were increased (Fig. 2b). Next, the BODIPY 581/591 C11 probe was used to detect lipid peroxidation levels in erastin-pretreated PANC-1 cells. Cells in the COL11A1 treatment groups exhibited a distinct fluorescence shift from green to red, indicating a significant reduction in lipid peroxidation levels (Fig. 2d). Given that lipid peroxidation primarily occurs in mitochondria, we used the JC-1 probe to detect mitochondrial membrane potential (MMP) changes in PANC-1 cells. The COL11A1 treatment group showed predominant red fluorescent J-aggregates, indicating that COL11A1 attenuated erastin-induced MMP disruption in PANC-1 cells (Fig. 2c). Finally, we performed immunofluorescence staining to assess the intracellular levels of iron and ROS—two core mediators of ferroptosis. The results showed that compared with the control group, iron and ROS levels were significantly downregulated in PANC-1 cells treated with COL11A1 (Fig. 2e,f). Furthermore, no significant differences were observed in the key ferroptosis-related indicators between the rhCOL11A1 treatment group and the pCMV3-COL11A1-transfected group. Taken together, these results demonstrated that COL11A1 exerts an inhibitory effect on ferroptosis in PANC-1 cells.

Fig. 2.

COL11A1 inhibited ferroptosis in pancreatic cancer cells. (a) Western blotting was used to detect the protein expression levels of GPX4, SLC7A11, and FTH1 under treatment with erastin (10 µM) and transfected pCMV3-COL11A1, or rhCOL11A1. (b) GSH/MDA kit was used to detect the expression level of GSH/MDA under treatment with erastin (10 µM) and transfected pCMV3-COL11A1 or rhCOL11A1. (c) The JC-1 probe was employed to assess the impact of COL11A1 on MMP. Scale bar: 20 µm. (d) Representative images of C11-BODIPY^581/591 stained COL11A1 overexpressed PANC-1 cells treated with 10 µM erastin for 12 h. A shift in the green-to-red ratio indicates lipid oxidation. Scale bar: 10 µm. (e) The FeRhoNox-1 probe was employed to determine the iron ion concentration under different treatment conditions. Scale bar: 40 µm. (f) Immunofluorescence was used to detect the expression of ROS by adding DCFH-DA under different treatment. Scale bar: 40 µm. n = 3, *p 0.05. GPX4, glutathione peroxidase 4; SLC7A11, solute carrier family 7 member 11; FTH1, ferritin heavy chain 1; MDA, malondialdehyde; MMP, mitochondrial membrane potential; ROS, reactive oxygen species; GSH, glutathione; JC-1, 5,5’,6,6’-tetrachloro-1,1’,3,3’-tetraethylbenzimidazolcarbocyanine iodide; DCFH-DA, 2’,7’-dichlorodihydrofluorescein diacetate.

Fig. 3.

COL11A1 inhibited autophagy in pancreatic cancer cells. (a) Western blotting was used to detect the expression levels of LC3 and p62 under treatment with RAPA (100 nM) and transfected pCMV3-COL11A1 or rhCOL11A1. (b) Immunofluorescence was used to detect the expression of LC3 under treatment with RAPA (100 nM) and transfected pCMV3-COL11A1 or rhCOL11A1. Scale bar: 40 µm. (c) RT-PCR was used to detect the mRNA expression level of GPX4, SLC7A11, FTH1, p62 and LC3 under treatment with erastin (10 µM) or RAPA (100 nM) and transfected pCMV3-COL11A1, or rhCOL11A1. (d) rhCOL11A1 or pCMV3‑COL11A1 plasmid was co‑transfected with CMV-TurboRFP-EGFP-LC3-PGK-Puro plasmid into PANC‑1 cells, followed by treatment with 100 nM RAPA. The fluorescence spots were detected. Scale bar: 20 µm. n = 3, *p 0.05. LC3, microtubule-associated proteins 1A/1B light chain 3.

To further confirm that COL11A1 is involved in regulating autophagy in pancreatic cancer cells, we analyzed its effect on the expression of autophagy-related proteins. Western blotting analysis showed that COL11A1 reduced the conversion of LC3I to LC3II and increased p62 expression in PANC-1 cells (Fig. 3a). Immunofluorescence staining revealed decreased average fluorescence intensity of LC3 in COL11A1-treated PANC-1 cells (Fig. 3b). RT-PCR results showed that COL11A1 reduced the mRNA expression of LC3 and increased p62 mRNA expression in PANC-1 cells (Fig. 3c). Indeed, PANC-1 cells were transfected with CMV-TurboRFP-EGFP-LC3-PGK-Puro plasmid to detect autophagy flux, and the results showed that in COL11A1-overexpressing PANC-1 cells, there was a significant reduction in the number of both yellow and red-only puncta compared with wild-type cells. To monitor autophagic flux, PANC-1 cells were transfected with the CMV‑TurboRFP‑EGFP‑LC3‑PGK‑Puro plasmid. The results showed that both the pCMV3‑COL11A1 plasmid and rhCOL11A1 significantly blocked autophagic flux (Fig. 3d). Collectively, these results indicated that COL11A1 inhibits autophagy in PANC-1 cells.

Literature reports indicate that autophagy is involved in regulating the ferroptosis process of cells [9, 14]. To further explore the molecular mechanism by which COL11A1 regulates autophagy and ferroptosis in PANC-1 cells, we simultaneously co-treated the cells with 3-methyladenine (3-MA) (an autophagy inhibitor), RAPA (an autophagy inducer), and pCMV3-COL11A1. Changes in ferroptosis-related proteins were detected using western blotting and fluorescence assays. The results showed that 3-MA upregulated the expression of GPX4, FTH1, and SLC7A11, thereby enhancing the inhibitory effect of COL11A1 on ferroptosis in PANC-1 cells. By contrast, RAPA exerted the opposite effect, inhibiting the regulatory role of COL11A1 in ferroptosis (Fig. 4a). In addition, immunofluorescence results demonstrated that the levels of ROS and iron content were increased in the COL11A1+RAPA combined treatment group compared with the COL11A1 monotherapy group, whereas these levels were decreased in the COL11A1+3-MA combined treatment group (Fig. 4b,c). We further examined the effects of RAPA and 3-MA on COL11A1-induced changes in GSH and MDA levels. The results showed that RAPA attenuated the regulatory effect of COL11A1 on GSH and MDA levels, whereas 3-MA enhanced this regulatory effect of COL11A1 (Fig. 4d). Collectively, these findings demonstrated that autophagy inhibition partially alleviated ferroptosis in the context of COL11A1 exposure. In summary, COL11A1 diminishes cellular sensitivity to ferroptosis by inhibiting autophagy.

Fig. 4.

COL11A1 regulated the ferroptosis process of pancreatic cancer cells through autophagy. (a) Western blotting was used to detect the expression levels of GPX4, SLC7A11, and FTH1 under treatment with erastin (10 µM) and transfected pCMV3-COL11A1, pCMV3-COL11A1+RAPA, or pCMV3-COL11A1+3-MA. (b) Immunofluorescence was used to detect the expression of ROS by adding DCFH-DA under different treatments. Scale bar: 40 µm. (c) Immunofluorescence was used to detect the expression level of iron visualized with the FeRhoNox-1 probe under different treatments. Scale bar: 40 µm. (d) The GSH/MDA kit was used to detect the expression level of GSH/MDA under transfected pCMV3-COL11A1, pCMV3-COL11A1+RAPA, or pCMV3-COL11A1+3-MA. n = 3, *p 0.05. 3-MA, 3-methyladenine.

Research has shown that autophagy is intertwined with the AKT/Beclin 1 signaling pathway [31]. To further explore the molecular mechanism by which COL11A1 regulates autophagy, we assessed the effect of COL11A1 on the AKT/Beclin 1 signaling pathway via western blotting, while additionally investigating its impact on phosphorylated mTOR (p-mTOR). The results showed that COL11A1 promoted the phosphorylation of AKT, mTOR, and Beclin 1 (Fig. 5a). Furthermore, western blotting analysis demonstrated that LY294002 attenuated the regulatory effect of COL11A1 on the expression levels of LC3, p62, and p-Beclin 1 (Fig. 5c). To further confirm that COL11A1 regulates autophagy via the AKT/Beclin 1 signaling pathway, we performed immunofluorescence experiments. The results demonstrated that LY294002 and siBeclin 1 attenuated the inhibitory effect of COL11A1 on LC3 expression (Fig. 5b). In addition, we assessed the effects of LY294002 and siBeclin 1 on COL11A1-induced changes in ROS and iron content via fluorescence-based assays. The results revealed that LY294002 and siBeclin 1 partially counteracted the regulatory effect of COL11A1 on ROS and iron content (Fig. 6a,b). Furthermore, the nude mouse tumor-bearing assay demonstrated that implantation of COL11A1-overexpressing PANC-1 cells significantly promoted in vivo tumor growth, whereas tumor volumes were markedly reduced following erastin or RAPA treatment (Fig. 6c). Collectively, these findings indicate that COL11A1 serves as a critical driver of pancreatic cancer progression, and its pro-tumorigenic function is closely associated with the ferroptosis and autophagy pathways. These data suggest that COL11A1 regulates the ferroptosis of pancreatic cancer cells through AKT/Beclin 1-dependent autophagy (Fig. 6d).

Fig. 5.

COL11A1 regulated the autophagy of pancreatic cancer cells through the AKT/Beclin 1 signaling pathway. (a) Western blotting was used to detect the expression levels of AKT, p-AKT, mTOR, p-mTOR, Beclin 1, and p-Beclin 1 under treatment with pCMV3-COL11A1 or rhCOL11A1. (b) Immunofluorescence was used to detect the expression of LC3 under treatment with RAPA (100 nM) and transfected pCMV3-COL11A1, pCMV3-COL11A1+LY294002, or pCMV3-COL11A1+siBeclin 1. Scale bar: 40 µm. (c) Western blotting was used to detect the expression levels of Beclin 1, p-Beclin 1, p62, and LC3II/I under treatment with RAPA (100 nM) and pCMV3-COL11A1 or pCMV3-COL11A1+LY294002. n = 3, *p 0.05. p-mTOR, phosphorylated mTOR.

Fig. 6.

COL11A1 regulated ferroptosis of pancreatic cancer cells through the AKT/Beclin 1 signaling pathway. (a) Immunofluorescence was used to detect the expression of ROS by adding DCFH-DA under treatment with erastin (10 µM) and pCMV3-COL11A1, pCMV3-COL11A1+LY294002, or pCMV3-COL11A1+siBeclin 1. Scale bar: 40 µm. (b) The results of immunofluorescence analysis of iron content in PANC-1 cells following different treatments are shown iron was visualized with FeRhoNox-1, and nuclei were stained with DAPI. Scale bar: 40 µm. (c) Wild-type PANC-1 cells and COL11A1 stably overexpressing PANC-1 cells were subcutaneously implanted into nude mice, respectively. In addition, two other groups were established, in which nude mice bearing COL11A1 stably overexpressing PANC-1 cell xenografts were treated with erastin (30 mg/kg) or RAPA (4 mg/kg), separately. The tumors were obtained at the end of the experiment. (d) Mechanism of COL11A inhibiting ferroptosis in pancreatic cancer by inhibiting AKT/Beclin 1 dependent autophagy. n = 3, *p 0.05.