Ellagic acid is associated with reduced pancreatic carcinogenesis and modulation of the IL-6/STAT3 pathway

Abstract

Background

The 5-year survival of pancreatic ductal adenocarcinoma (PDAC) remains about 10% despite therapeutic advances, underscoring the need for effective preventive and early intervention strategies. Ellagic acid (EA), a naturally occurring polyphenol, has been associated with demonstrated antitumor activity in several malignancies. However, its potential role in preventing PDAC development remains unclear. In this study, we examined the chemopreventive potential and underlying mechanisms of EA in a hamster model of PDAC induced by a high-fat diet and N-nitrosobis(2-oxopropyl)amine.

Results

Dietary EA (0.1% w/w) was associated with significantly reduction in both the incidence and multiplicity of PDACs compared to controls. The proportion of histologically normal pancreatic ducts increased and the Ki-67 labeling index decreased in pancreatic intraepithelial neoplasia (PanIN) following EA exposure. In vitro, cell proliferation decreased in a dose-dependent manner, G1 arrest was induced, and invasion was diminished after EA treatment. Multiplex Western blotting revealed lower inhibition of the IL-6/STAT3 pathway. The proportion of pSTAT3-positive cells in hamster PanINs and PDACs was significantly lower in the EA-treated groups than in controls. Because a high-fat diet is known to elevate adipocytokines, and resistin (Res) has been implicated in STAT3 regulation, the Res/STAT3 axis was also examined. Res-associated promotion of invasion was observed but there was no proliferation in vitro, and pSTAT3 expression did not increase. Similarly, serum Res levels did not differ significantly across groups, suggesting a limited contribution of the Res/STAT3 pathway in this hamster model.

Conclusions

EA treatment was associated with reduced pancreatic carcinogenesis in vivo. The IL-6/STAT3 pathway appears to be a primary molecular target under our experimental conditions, whereas the contribution of Res is likely minimal. These findings support the potential of EA as a preventive agent against pancreatic cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s41021-026-00353-3.

Introduction

Pancreatic ductal adenocarcinoma (PDAC) remains one of the most lethal cancers, with a 5-year survival rate of approximately 13% (2014–2020) [1]. The dismal prognosis reflects its aggressive nature, challenges in early diagnosis, and the lack of effective treatment options [2]. Therefore, research on novel therapeutic approaches as well as preventive strategies for PDAC is of critical importance.

Ellagic acid (EA) is a naturally occurring polyphenolic compound found in pomegranates, berries, grapes, and walnuts. It exhibits multiple biological activities, including antioxidant, anti-inflammatory, antimicrobial, and anti-diabetic effects. Anti-tumor effects of EA have been reported in vivo in several cancer types, including colon, prostate, liver, lung, oral, breast, urinary bladder, and ovarian cancers, as well as melanoma [3–6]. In a clinical study on hormone-refractory prostate cancer, EA administration did not significantly improve overall survival but was associated with reduced serum prostate-specific antigen levels and decreased chemotherapy-related myelotoxicity [7].

In pancreatic cancer, EA has been shown to inhibit tumor growth in xenograft models using human pancreatic cancer cells, in part by suppressing COX-2 and NF-κB expression in vitro [8]. However, its potential role in preventing pancreatic carcinogenesis has yet to be investigated and the underlying mechanisms of its action in pancreatic cancer also remain to be elucidated.

In this study, we evaluated the chemopreventive potential of EA in a hamster model of pancreatic carcinogenesis and examined the underlying molecular mechanisms.

Materials and methods

Hamster PDAC model

N-nitrosobis(2-oxopropyl)amine (BOP)-induced hamster model of pancreatic carcinogenesis was established, with minor modifications, as described previously [9–12]. Six-week-old female Syrian golden hamsters (Japan SLC, Shizuoka, Japan) received subcutaneous injections of BOP (Toronto Research Chemical Inc., Toronto, Canada) at 10 mg/kg on four alternate days (days 1, 3, 5, and 7). One week after the final injection, the animals were placed on a high-fat diet (The Quick Fat; crude fat 14.53%, crude protein 24.68%, 4.17 kcal/g; CLEA Japan, Tokyo, Japan) for 15 weeks. They were subsequently divided into three groups and provided high-fat diets containing ellagic acid (EA, Tokyo Chemical Industry Co., Ltd., Tokyo, Japan) in the following manner: control (n = 17), EA 0.01% (n = 15), and EA 0.1% (n = 15). A dietary concentration of 0.1% EA corresponds to an approximate dose of 50 mg/kg/day, which has been reported to exert antitumor effects in preclinical studies of prostate, colorectal, and lung cancers [4–6]. When extrapolated to humans, this dose is equivalent to approximately 3 g/day based on body weight. Importantly, in a clinical study of prostate cancer, administration of pomegranate juice at a polyphenol-equivalent dose of 3 g/day demonstrated therapeutic efficacy without significant adverse events, supporting the translational relevance of this dosage [7]. Food and water were available ad libitum; body weight and food intake were recorded weekly. The hamsters were housed with hardwood-chip bedding under controlled conditions (22 ± 2 °C, ~ 50% relative humidity, 12-h light/dark cycle). All protocols used were approved by the Institutional Animal Care and Use Committee of Nagoya City University School of Medical Sciences (IDO19-005).

Histopathological and serum analysis

At the end of the study, all hamsters were euthanized by exsanguination under deep isoflurane anesthesia (4.0% in oxygen), with the blood was collected from the abdominal aorta. The blood was allowed to clot and then centrifuged (3,000 rpm, 10 min, 4 °C) to obtain serum, which was stored at − 20 °C until analysis. Serum resistin (Res) concentrations were measured using a hamster-specific ELISA (MyBioSource, San Diego, CA) according to the manufacturer’s instructions. The pancreas, liver, kidney, and lung were excised and fixed in 10% neutral-buffered formalin, processed for paraffin embedding, and sectioned at 4 μm. For histological evaluation, the pancreas was divided into the gastric, splenic, and duodenal lobes (including the common duct), sliced transversely at ~ 4 mm intervals, embedded routinely, and stained with hematoxylin–eosin (H&E) or subjected to immunohistochemistry (IHC). PDAC and pancreatic intraepithelial neoplasia (PanIN) were independently diagnosed by two reviewers, including a board-certified pathologist, all blinded to group allocation. The PanIN progression score in the duodenal lobe was calculated as a weighted index modified from that of Kato et al. (2021) [10], which weights the lesion thus: (normal = 0, PanIN1 = 1. PanIN2 = 2, PanIN3 = 3, PDAC = 4). For incidental histopathological evaluation, the liver was sampled at one transverse section from each of three lobes, including the gallbladder; the lungs were sampled at one representative section from each lobe (five sections per animal); and the kidneys were sampled at one transverse section from each kidney through the renal papilla. All tissues were processed routinely and examined by H&E staining. For immunohistochemistry, sections were stained on an automated Leica BOND-MAX (Leica Biosystems, Nussloch, Germany) platform. After deparaffinization, heat-induced epitope retrieval was performed in EDTA buffer (pH 9.0), followed by quenching of endogenous peroxidase and blocking of nonspecific binding. The primary antibodies were Ki-67 (clone SP6, Abcam, Cambridge, UK; 1:250) and phospho-Stat3 (Tyr705) (Cell Signaling Technology(CST), Danvers, MA; 1:200). An HRP polymer system with DAB chromogen was used for detection. Whole-slide images were scanned on an Aperio CS2 (Leica Biosystems), and the quantitative analysis was carried out using vendor-supplied software. Within PanIN lesions, Ki-67 and pStat3 labeling indices were determined by counting ≥ 1,000 nuclei per case using standardized thresholds. Hotspot lesions were photographed at 200× magnification for pStat3 in PDAC and ≥ 500 tumor cell nuclei were manually counted to calculate the positivity rate. In addition, specimens containing PDAC lesions composed of more than 100 tumor cells were subjected to TUNEL staining (Takara Bio Inc., Shiga, Japan) according to the manufacturer’s protocol, followed by DAB chromogenic detection. Whole-slide images were acquired using an Aperio CS2 scanner. The number of background cells was quantified, and because TUNEL-positive cells were infrequent, positive cells were counted manually by visual inspection. The percentage of TUNEL-positive cells was then calculated.

Cell culture

Human pancreatic cancer cells (MIAPaCa2 and PANC1) were purchased from the American Type Culture Collection (ATCC, Rockville, MD) and maintained in RPMI1640 (Wako Pure Chemical Industries Co. Ltd., Osaka, Japan) supplemented with 10% fetal bovine serum (FBS). The BOP-induced hamster pancreatic cancer cell line (HPD1NR) was provided by Dr. Masahiro Tsutsumi and has been described previously [10, 13].

Cell proliferation assay

Cell proliferation was quantified using the WST-1 assay (Roche Applied Science, Mannheim, Germany). MIAPaCa2, PANC1, and HPD1NR cells were seeded in 96-well plates at 1.0 × 10⁴ cells in 50 µL per well and incubated overnight at 37 °C in 5% CO₂. The next day, 50 µL of treatment solution was overlaid to reach the final concentrations in a 100-µL well volume: ellagic acid (EA, 0–80 µM; dissolved in DMSO) or recombinant Res (0–400 ng/mL) prepared in complete medium. After 48 h of treatment, WST-1 reagent (per 10 µl) was added to each well and plates were incubated for 2 h at 37 °C. Absorbance was read on a SpectraMax iD3 microplate reader (Molecular Devices, San Jose, CA) at 440 nm with a 600-nm reference; values were calculated as A440 − A600. Proliferation was expressed as a percentage of the 0.1% DMSO vehicle control. Each concentration was tested with n = 8 biological replicates per experiment.

Cell cycle analysis

MIAPaCa2 cells and PANC1were treated with EA (10, 20, or 40 µM, final 0.1% DMSO) and harvested, washed with PBS, and stained with propidium iodide using the Guava^® Cell Cycle Reagent (Merck, Darmstadt, Germany) according to the manufacturer’s instructions. Stained cells were analyzed on a Guava^® easyCyte Single flow cytometer (Merck). Debris and doublets were excluded by scatter/pulse-geometry gating, and ≥ 2,000 events per sample were acquired. Cell-cycle phase distribution (G0/G1, S, G2/M) was calculated with the GuavaSoft™ cell-cycle module.

Migration and invasion assays

Migration assay: MIAPaCa2 and PANC1 cells were grown to confluence in 10-cm dishes. The culture medium was replaced with RPMI 1640 containing EA (0, 20, or 40 µM) or recombinant Res (0, 25, or 50 ng/mL) and cells were pre-incubated for 24 h. Linear wounds were then made in six locations per dish using a sterile 200-µL pipette tip. Phase-contrast images were captured at 0 h and 21 h and the migration area was calculated as the difference in area using ImageJ.

Transwell invasion assay: Cell invasion was assessed using BD BioCoat™ Matrigel™ Invasion Chambers (24-well, 8.0-µm pores; BD Biosciences, Franklin Lakes, NJ). MIAPaCa2, PANC1, and HPD1NR cells were pretreated for 48 h with EA (0, 20, or 40 µM) or Res (0 or 50 ng/mL), after which 1 × 10⁵ cells were seeded into the upper chamber in medium containing the same treatment. The lower chamber contained medium with 10% FBS. After 48 h, non-invaded cells were removed from the upper surface, and the membranes fixed and stained with toluidine blue. Invading cells were counted in six random fields per insert at 100× magnification and the mean value was used as the number of invaded cells (n = 3 inserts per condition).

Western blotting and multiplex western blotting assay

MIAPaCa2 and PANC1 cells were homogenized on ice in T-PER Tissue Protein Extraction Reagent (Thermo Fischer Scientific Inc., Waltham, MA) supplemented with a protease inhibitor cocktail (Roche Diagnostic). Protein concentrations were determined by the Bradford method using protein assay kit (Bio-Rad laboratories, Hercules, CA). Aliquots containing 30 µg of protein samples were mixed with SDS sample buffer, heated for 10 min at 100 °C and then subjected to SDS-PAGE. Proteins separated on 10% or 12% acrylamide gels were transferred onto nitrocellulose blotting membranes (GE Healthcare UK Ltd., Buckinghamshire, UK). The following antibodies were used: phospho-STAT3 (Tyr705) (1:1000, #9145, CST), STAT3 (1:1000, #8768, CST), phospho-NF-κB (Ser 536, 1:1000, #3033, CST), NF-κB (1:1000, #8242, CST), Cyclin D1(1:1000, #2922, CST), CAP1(1:1000, #47055, CST), SNAIL(1:1000, #3879, CST), IL-6 (1:1000, I2123, Merck), phospho-JAK2 (Tyr1007/1008) (1:1000, #3771, CST), JAK2 (1:1000, #3230, CST), Cleaved-Caspase 3 (Asp175) (1:1000, #9661, CST), Caspase 3 (1:1000, #9662, CST), β-Actin (1:5000, A5316, Merck). Band intensity was quantified using Image J.

The protocol for the multiplex Western blotting assay was identical to the conventional Western blotting method up to the point of protein transfer, except that 300 µg of proteins from MIAPaCa2 or PANC1 cells were applied per lane. After transfer, the membrane was inserted into the Mini-PROTEAN Ⅱ Multiscreen Apparatus (Bio-Rad Laboratories Inc.). Antibodies specific for phosphorylated proteins, as listed in Fig. 3B, were obtained from Cell Signaling Technology and Abcam. The protocol for detection was the same as for Western blotting.

Fig. 3.

Suppression of the IL6/STAT3 pathway post EA exposure. (A) Multiplex Western blot of proliferation-related phosphoproteins in MIAPaCa2 and PANC1 cells treated with EA (0, 20, 40 µM) for 48 h, and (B) The list of phospho-antibodies included in the multiplex panel and quantitative analysis of expression changes relative to the Control. (C) Western blots of IL6/STAT3 and NF-κB pathways in MIAPaCa2 and PANC1 cells

Statistical analysis

Differences in quantitative data (mean ± SD) were analyzed using an unpaired two-tailed Student’s t-test for two-group comparisons and one-way ANOVA followed by Dunnett’s post hoc test for multiple groups (vs. the control). All tests were using Graph Pad Prism 8 (GraphPad Software, Inc., La Jolla, CA).

Results

Inhibition of pancreatic carcinogenesis in a hamster model following EA exposure

All hamsters remained healthy with no mortality throughout the experimental period. There were no significant differences in body weight among groups (Fig. 1A). Dietary intake was consistent (8–10 g/day) and stable across groups, with a gradual decrease in estimated EA intake per body weight as the animals grew (Fig. 1B). At sacrifice, final body, liver, and kidney weights also showed no significant group differences (Supplementary Table 1A).

Fig. 1.

Suppression of BOP-induced pancreatic carcinogenesis in hamsters after EA exposure. (A) Body weight (B) and estimated daily EA intake (mg/kg/day) in Control, EA 0.01%, and EA 0.1% groups after BOP and initiation of EA diet. (C) Representative histology of hamster PDAC. Scale bar, 500 μm. (D) Incidence and (E) multiplicity of PDAC. (F) The proportion of normal, PanIN1, PanIN2, PanIN3 and PDAC in all pancreatic ducts (diameter > 200 mm) of duodenal lobes, and (G) the progression score calculated by weighting respective lesions (normal = 0, PanIN1 = 1, PanIN2 = 2, PanIN3 = 3, PDAC = 4). (H) Immunohistochemical findings of Ki-67 and Ki-67 labeling index in PanINs. Data shown as mean ± SD, n = 17 (Control) and 15 (EA 0.01%, EA0.1%). Statistical comparisons were performed by one-way ANOVA with Dunnett’s post hoc test; * P < 0.05, ** P < 0.01, *** P < 0.001 vs. Control

Histologically, the observed PDAC resembled human PDAC, with gland-forming structures on a fibrotic background (Fig. 1C). The incidence and multiplicity of PDACs were significantly reduced in the 0.1% EA group compared with controls, in a dose-dependent manner (Fig. 1D, E). All pancreatic ducts (diameter > 200 μm) in the duodenal lobe H&E slides were classified as normal, PanIN1, PanIN2, PanIN3, or PDAC. While the proportion of normal ducts significantly increased, it decreased for PDACs, in the 0.1% EA group relative to controls (Fig. 1F). Progression scores also significantly reduced following EA treatment (Fig. 1G). As incidental findings, the liver showed biliary cysts, atypical biliary hyperplasia, cholangioma, cholangiocarcinoma, and hepatocellular adenoma (Supplementary Fig. 1A), with no statistically significant differences among the groups (Supplementary Table 1B). In the gallbladder, human biliary intraepithelial neoplasia (BilIN)-like lesions were observed (Supplementary Fig. 1B); although EA treatment tended to increase normal epithelium and decrease high-grade BilIN, these changes did not reach statistical significance (Supplementary Fig. 1C). Pulmonary tumors of various sizes were also identified (Supplementary Fig. 1D), but no significant differences were observed among the groups (Supplementary Table 1C). In the kidneys, no treatment-related histopathological findings were observed in any of the groups. The Ki-67 labeling index, calculated from an average of 2,956 ± 1,776 cells in low-grade PanINs, was significantly suppressed in the 0.1% EA group compared with controls (Fig. 1H). To evaluate apoptosis, TUNEL staining was performed. TUNEL-positive cells were rarely observed in PanIN lesions and were infrequent within PDAC tissues. Quantitative analysis of TUNEL-positive cells in PDAC demonstrated no statistically significant differences among the groups (Supplementary Fig. 2).

Collectively, these results demonstrate that the progression of pancreatic carcinogenesis declined following EA treatment, including PanIN progression and PDAC development, in the BOP-induced hamster PDAC model by inhibiting proliferation.

Inhibition of cell proliferation, migration and invasion in vitro following EA exposure

Since pancreatic carcinogenesis in the hamster model was suppressed following EA treatment, the direct effects of EA on PDAC cells were evaluated. Human PDAC cell lines (MIAPaCa2, PANC1) and a hamster PDAC line (HPD1NR) were exposed to EA (0–80 µM). Cell growth was significantly reduced in EA-treated cells in a dose-dependent manner in all three lines (Fig. 2A). In cell cycle analysis, both MIAPaCa2 and PANC1 cells exhibited a decrease in the G2/M population and an increase in the G0/G1 population, indicating G1 cell cycle arrest (Fig. 2B). In wound-healing assays, cell migration was decreased in MIAPaCa2 and PANC1 cells (Fig. 2C). Invasion assays further demonstrated that the invasive capacity was reduced in MIAPaCa2, PANC1, and HPD1NR following EA treatment (Fig. 2D). Taken together, these findings indicate that PDAC cell proliferation was lower in EA-treated cells, accompanied by G1 arrest and reduced invasiveness in vitro.

Fig. 2.

Inhibition of proliferation, migration, and invasion of PDAC cells after EA exposure. (A) Cell proliferation measured by WST-1 assay in human PDAC (MIAPaCa2 and PANC1) and hamster PDAC (HPD1NR) cell lines (n = 6 per dose). (B) Cell cycle analysis of MIAPaCa2 and PANC1 cells after 48 h EA treatment (n = 3 per dose). (C) Wound-healing assay in MIAPaCa2 and PANC1 cells with quantification of relative migration area (area at 0 h – area at 21 h, n = 6). (D) Matrigel invasion assay using 8-µm-pore chambers; invaded cells counted in MIAPaCa2, PANC1, and HPD1NR cells (n = 3). Data shown as mean ± SD. Statistical comparisons were made using one-way ANOVA with Dunnett’s post hoc test or unpaired two-tailed Student’s t-test; * P < 0.05, ** P < 0.01, *** P < 0.001 vs. the 0.1% DMSO vehicle control

Suppression of STAT3 pathway in vitro in EA-treated cells

To clarify the mechanisms underlying the inhibitory effects of EA treatment on pancreatic carcinogenesis, the phosphorylation status of 18 proteins associated with cell proliferation was examined in MIAPaCa2 and PANC1 cells using multiplex Western blotting (Fig. 3A). As summarized in Fig. 3B, the following phosphorylated proteins were consistently decreased by EA treatment at all tested concentrations in both MIAPaCa2 and PANC1 cells: pMEK (Ser217/221), pERK (Thr202/Tyr204), p-p38 MAPK (Thr180/Tyr185), pSAPK (Thr183/Tyr185), pPDK (Ser241), and pSTAT3 (Tyr705). Among these proteins, pSTAT3 exhibited the most pronounced and concentration-dependent suppression in both MIAPaCa2 and PANC1 cells, and was therefore selected for further focused analysis.

Because previous work has implicated EA in the inhibition of cell proliferation through NF-κB signaling in pancreatic cancer cell lines [14], the expression of proteins within the NF-κB and STAT3 pathways was also evaluated. pSTAT3 expression decreased in a dose-dependent manner in both MIAPaCa2 and PANC1 cells. In contrast, total NF-κB levels were reduced only in MIAPaCa2, without any change in its phosphorylation status, suggesting minimal alteration of the NF-κB axis. In both cell lines, IL-6 expression decreased in a dose-dependent manner in EA-treated cells, consistent with the inhibition of the IL-6/STAT3 pathway as a plausible mechanism. In line with these observations, STAT3 downstream effectors—SNAIL (related to invasion) and Cyclin D1 (related to proliferation)—were likewise reduced in a dose-dependent manner in both cell lines. In contrast, the apoptosis-related marker cleaved caspase-3 was increased only in MIAPaCa2 cells treated with 40 µM EA, whereas no comparable increase was observed in PANC1 cells (Fig. 3C). Although IL-6 is known to regulate pSTAT3 expression through pJAK2, pJAK2 expression was not detected in the present study. In contrast, total JAK2 protein expression was decreased (Fig. 3C). These findings suggest that inhibition of the IL-6/STAT3 pathway is one of the key mechanisms by which cell proliferation and invasiveness in PDAC is suppressed following EA-treatment.

Reduction of pStat3 expression in the hamster model following EA-treatment

Because a reduction was seen in STAT3 pathway activation in pancreatic cancer cell lines following EA treatment, we examined pStat3 in the BOP-induced hamster model. Immunohistochemical pStat3 (Tyr705) positivity was quantified in low-grade PanINs of the duodenal lobe and in PDACs. The pStat3 labeling index was significantly lower in the EA-treated groups than in the control group for both low-grade PanINs and PDACs (Fig. 4A, B).

Fig. 4.

Inhibition of pStat3 expression in PanIN and PDAC of hamsters after EA exposure. Immunohistochemical findings of pStat3 and quantification of nuclear pStat3 labeling index in (A) PanIN lesions (Control, n = 17; EA 0.01%, 0.1%, n = 15) and (B) PDAC (Control, n = 11; EA 0.01%, n = 7; EA 0.1%, n = 7) of hamsters. Statistical comparisons were made using one-way ANOVA with Dunnett’s post hoc test; ** P < 0.01, *** P < 0.001 vs. Control

Resistin-associated enhancement of invasion without affecting proliferation or pSTAT3 expression

In a previous study, EA intake was associated with reduced serum Res levels in KK-Ay mice fed a high-fat diet [15]. Another study reported that pSTAT3 expression was decreased in MIAPaCa2 cells with the knocked-down Res receptor, adenylyl cyclase-associated protein 1 (CAP1) [16]. Based on these findings, we hypothesized that the EA-associated downregulation of serum Res may contribute to the inhibition of pancreatic progression through the pSTAT3 pathway. To evaluate the direct effects of Res on proliferation, migration, and invasion in vitro, we performed WST-1, wound healing, and Matrigel invasion assays using MIAPaCa2 and PANC1 cells treated with Res (25, 50 ng/ml). Res treatment did not affect proliferation in MIAPaCa2 or PANC1 cells (Fig. 5A). In contrast, migration and invasion in both cell lines was promoted significantly (Fig. 5B, C). However, the protein expression of pSTAT3 and CAP1 showed no alteration in these cells (Fig. 5D). Furthermore, serum Res levels showed no significant change post EA intake. Likewise, there was no significant reduction in Res in the pancreas, adipose tissue, or tumor post EA treatment (Fig. 5E). Taken together, these results suggest that while Res may play a role in promoting the migratory and invasive properties of pancreatic cancer cells, it does not appear to substantially contribute to the inhibitory effects associated with EA observed in this study.

Fig. 5.

Res is associated with promotion of migration and invasion in vitro but not the inhibitory effects associated with EA exposure in the hamster model. (A) Cell proliferation measured by WST-1 assay in MIAPaCa2 and PANC1 treated with Res 0–400 ng/ml (n = 6 per dose). (B) Wound-healing assay with quantification of relative migration area (area at 0 h – area at 21 h, n = 6 scratches). (C) Matrigel invasion assay using 8-µm-pore chambers; invaded cells counted (n = 3). (D) Western blots of CAP1/STAT3 pathway in MIAPaCa2 and PANC1 cells treated with Res 25 ng/ml or 50 ng/ml. (E) Res concentrations measured by ELISA in serum, adipose tissue, pancreas tissue, and PDAC tissue. Data are shown as mean ± SD. Statistical comparisons were made using one-way ANOVA with Dunnett’s post hoc test or unpaired two-tailed Student’s t-test; * P < 0.05, *** P < 0.001 vs. Control

Discussion

To our knowledge, this is the first report to demonstrate chemopreventive potential of EA against pancreatic carcinogenesis in an in vivo model, extending prior findings that mainly evaluated EA in vitro or in xenografts. EA has an established safety profile, and oral dosing at 180 mg/day has been investigated clinically in nonalcoholic fatty liver disease, Type 2 diabetes, and prostate disorders [7, 17]. In an eight-week double-blind trial involving patients with Type 2 diabetes, improvements were reported in insulin resistance and oxidative stress, along with reductions in inflammatory cytokines such as IL-6 [17]. Given that diabetes is a recognized risk factor for pancreatic cancer (RR = 1.5) [2], such high-risk populations may represent suitable candidates for future studies evaluating EA as a potential chemopreventive agent.

Mechanistically, many prior studies—both in vitro and in subcutaneous xenograft models—have reported reductions in NF-κB expression following EA exposure [8, 14, 18]. Other reports describe inhibition of the Akt, Sonic hedgehog (Shh), and Notch pathways, as well as suppression of TGF-β signaling and MMP-2/9 expression [19, 20]. Notably, the NF-κB findings in those studies were based on total p65/p50 protein; to our knowledge, phosphorylation at Ser536 was not examined. In our experiments, total NF-κB levels were lower in MIAPaCa2 cells treated with EA, corroborating previous work, whereas no comparable trend was observed in PANC1 cells. Moreover, pNF-κB (Ser536) and the pNF-κB/total NF-κB ratio remained largely unchanged (Fig. 3C). In contrast, the IL-6/pSTAT3 signaling pathway was consistently reduced in both MIAPaCa2 and PANC1, suggesting a common EA-responsive axis in PDAC cells.

Decreased pSTAT3 expression after EA treatment has also been reported in prostate cancer [21], esophageal squamous cell carcinoma [22], and cervical cancer [23], but it has rarely been investigated in pancreatic cancer. In addition, in clinical studies in Type 2 diabetes have shown significant reduction in serum IL-6 levels after EA administration [17]. The IL-6/STAT3 pathway plays critical roles in cancer by regulating cell proliferation, invasion, anti-apoptotic signaling, inflammation, and angiogenesis. In PDAC, STAT3 signaling is particularly important at early stages, contributing to tumor cell proliferation, anti-apoptosis, and invasiveness [24–27]. In the present hamster model, EA treatment significantly suppressed tumor cell proliferation, whereas no significant changes in apoptosis were detected. These findings suggest that the antitumor effects of EA in this model are primarily mediated through suppression of tumor cell proliferation rather than induction of apoptosis. However, TUNEL-positive cells were extremely rare in PDAC tissues, suggesting that the limited detection of apoptosis may reflect insufficient sensitivity of the assay. Because STAT3 regulates multiple downstream programs beyond apoptosis, including inflammatory and angiogenic pathways, further studies will be required to elucidate the full spectrum of EA-mediated STAT3 inhibition in PDACs. Although IL-6 is known to regulate pSTAT3 activation via pJAK2, pJAK2 was not detectable by Western blotting even after prolonged exposure. This may reflect the transient and labile nature of JAK2 phosphorylation, as well as limitations in antibody sensitivity under the experimental conditions used in this study. Alternatively, STAT3 phosphorylation may be regulated by upstream kinases other than JAK2, such as JAK1 or TYK2. Notably, total JAK2 protein expression was reduced, suggesting that EA may decrease JAK2 abundance itself. The precise involvement of pJAK2 in this context remains unclear, and further investigation will be required.

On the other hand, prior studies have reported lower Res secretion in cultured adipocytes exposed to EA [15]. Accumulating evidence indicates that Res modulates the pSTAT3 pathway via the CAP1 receptor [28], and that CAP1-mediated control of pSTAT3 contributes to pancreatic cancer progression [16]. Accordingly, Res/STAT3 signaling was considered a potential upstream regulator in this study. In our high-fat diet–combined hamster model, adipocytokines such as leptin were elevated, and EA was expected to suppress Res secretion [9]. However, in the pancreatic cancer cell experiments, no apparent change in pSTAT3 levels was observed after Res exposure, and there was no significant decrease in Res concentration in serum, adipose tissue, pancreatic tissue, or PDAC lesions in hamsters after EA exposure. These findings indicate that the influence of EA on the Res pathway was limited under the conditions of this study.

Conclusion

In this study, we demonstrated that there was significant suppression of pancreatic carcinogenesis post EA treatment, using the BOP-induced hamster pancreatic cancer model. Our findings suggest that inhibition of the IL-6 /STAT3 signaling pathway may represent a principal mechanism by which EA modulates pancreatic carcinogenesis.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

PDAC

Pancreatic ductal carcinoma

EA

Ellagic acid

BOP

N-nitroso-bis(2-oxopropyl)-amine

PanINs

Pancreatic intraepithelial neoplasias

WST-1

4-[3-(4-iodophenyl)-2-(4-nitrophenyl)-2 H-5-tetrazolio]-1,3-benzene

Res

Resistin

Author contributions

Hiroyuki Kato: Conceptualization, data curation, formal analysis, investigation, methodology, project administration, visualization, writing original draft. Aya Naiki-Ito: Conceptualization, supervision, writing review and editing, investigation. Masayuki Komura: Investigation. Yuko Nagayasu: Investigation. Motonori Sato: Investigation. Xiaochen Kuang: Investigation. Aya Nagano: Investigation. Satoru Takahashi: Supervision, writing review and editing.

Funding

No funding.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Animal study: the Institutional Animal Care and use Committee of Nagoya City University Graduate School of Medical Sciences (No. IDO19-005).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.