A total of 20, male, 8-week-old, C57BL/6 (20±2 g) mice were purchased from Shanghai Slack Laboratory Animal Center. The mice were housed in an environment with a 12-h light/dark cycle at 24°C and a relative humidity of 50-70%. The mice were given free access to water and food, and the bedding was changed every other day. After 7 days of acclimation, the mice were randomly divided into the following two groups with 10 mice each: Control group and iron overload group. The iron overload mice were injected intraperitoneally with 120 mg/kg body weight of iron dextran (Pharmacosmos A/S) every other week for 12 weeks. Since infused dextran is eliminated 70% by the kidney and 30% by the gastrointestinal tract, the control mice were injected intraperitoneally with saline as reported previously (27). Mice were sacrificed by cervical dislocation at the end of the experiment. The blood and pancreas were collected. Pancreas tissues were rapidly dissected and fixed in 4% paraformaldehyde solution at room temperature for 24 h for histological analysis, or frozen in liquid nitrogen and stored at −80°C until further analysis. All animal experiments were approved by the Committee of Experimental Animal Care of Zhejiang University (Hangzhou, China; approval no. 20077).

Mouse serum was separated from the blood samples by centrifugation at 3,000 × g for 15 min at 4°C. A serum iron assay kit (cat. no. A039-1-1; Nanjing Jiancheng Bioengineering Institute) and total iron binding capacity assay kit (cat. no. A040-1-1; Nanjing Jiancheng Bioengineering Institute) were used to detect the iron level of the mice according to the manufacturer's instructions. For serum iron detection, iron chromogen was added to the serum and incubated in boiling water for 5 min. After cooling and centrifugation at 3,000 × g for 10 min at 4°C, the absorbance at 520 nm was measured by a fluorescence spectrophotometer. For measurement of total iron binding capacity, the serum was mixed with 179.1 mmol/l iron standard so that all transferrin bound iron in the serum. The excess iron in the serum was then adsorbed away using an iron adsorbent, and the iron content was measured by assessing the iron level in the serum. A lipase assay kit (cat. no. A054-2-1) and an α-amylase assay kit (cat. no. C016-1-1) (both Nanjing Jiancheng Bioengineering Institute) were used to detect amylase and lipase levels, respectively, in the mouse serum. α-amylase hydrolyzes the starch in the substrate. In the case of known substrate concentration and excess, the added iodine solution can combine with the unhydrolyzed starch in the substrate to form a blue complex. The absorbance at 660 nm was measured by a fluorescence spectrophotometer to obtain the amount of starch that had been hydrolyzed, and thus the activity of α-amylase was calculated. Latex made of triglyceride and water has a turbid character due to the absorption and scattering of incident light by its micelles. The triglycerides in the micelles are hydrolyzed by the action of lipase, which causes the micelles to split and thus the scattered light or turbidity is reduced. Lipase activity was calculated by measuring the absorbance at 420 nm using fluorescence spectrophotometer and measuring the rate of reduction of scattered light or turbidity.

Paraffin-embedded samples were cut into 5-µm sections. For haematoxylin and eosin (H&E) staining, paraffin sections were deparaffinized with xylene and rehydrated using different concentrations of alcohol, and then stained with haematoxylin for 5 min. The sections were differentiated in aqueous hydrochloric acid for 2 sec and blunted in aqueous ammonia for 15-30 sec at room temperature. The sections were put into eosin staining solution for 5-8 sec at room temperature. The sections were mounted with neutral gum after dehydration using absolute ethanol and xylene. The H&E staining then examined by light microscope (Leica Microsystems GmbH) and image acquisition was performed to analyze the histopathological features of the pancreatic sections.

Prussian blue staining was used to detect iron deposition in the pancreatic tissues. The 5-µm sections of pancreas were deparaffinized with xylene and rehydrated using different concentrations of alcohol. Pancreatic tissue was incubated in a mixture (1:1) of 2% potassium ferrocyanide and 2% hydrochloric acid for 30 min at room temperature. Sections were then rinsed with PBS (cat. no. KGB5001; Nanjing KeyGen Biotech Co., Ltd.) and counterstained with eosin for 20 sec at room temperature. The sections were mounted with neutral gum after dehydration using absolute ethanol and xylene.

Masson staining and Sirius red staining were employed to evaluate the collagen content of the pancreas. For Masson staining, paraffin sections were deparaffinized with xylene and rehydrated using different concentrations of alcohol. The sections were then placed in potassium dichromate standards at room temperature overnight (~17 h). Sections were treated with aqueous phosphomolybdic acid for 1-2 min and counterstained with aniline blue liquid for 5 sec at room temperature. Sections were sequentially treated with 1% glacial acetic acid for 5-10 sec each at room temperature. Sections were dehydrated and mounted using neutral resin glue. For Sirius red staining, pancreatic tissue sections were deparaffinized and stained with Wiegert's iron haematoxylin stain for 15 min at room temperature. Next, pancreatic sections were differentiated with an acidic differentiation solution. After washing the sections in tap water for 10 min, the pancreatic tissue was stained with Sirius red staining droplets for 1 h at room temperature. The sections were gently rinsed with a flowing water stream to remove the surface dye solution. All sections were examined using a DM3000 microscope (Leica Microsystems GmbH). The inflammation, atrophy and fibrosis in the pancreatic tissues were scored from 0 to 3 using a histopathological scoring criteria system, as previously reported (28).

The pancreatic sections were deparaffinized in xylene and then sequentially dehydrated in different concentrations of ethanol solution. Endogenous peroxidase activity was blocked with 3% H₂O₂ for 10 min, and sections were resuspended in antigen retrieval solution (pH 6.0) and boiled in a microwave oven for 5 min. Pancreatic tissue was then blocked with 3% bovine albumin [cat. no. 36101ES25; Yeasen Biotechnology (Shanghai) Co., Ltd.]. for 1 h at room temperature. Sections were incubated overnight at 4°C with anti-CD11b antibody (1:1,000; cat. no. ab133357), anti-F4/80 antibody (1:5,000; cat. no. ab300421) and anti-CD3 antibody (1:150; cat. no. ab16669) antibodies (all Abcam), then washed with Tris-buffered saline containing 0.1% Tween-20 (TBST). The pancreatic tissues were incubated with anti-rabbit IgG H&L (conjugated with horseradish peroxidase; 1:5,000 cat. no. ab205718; Abcam) for 1 h at room temperature, and then stained and developed with horseradish peroxidase for 30 min. Finally, the nuclei were counterstained using haematoxylin solution for 30 sec at room temperature. All sections were examined using a DM3000 microscope (Leica Microsystems GmbH). In this experiment, immunohistochemical quantification was performed using ImageJ software (version 2.0; National Institutes of Health).

The levels of MDA, SOD and GSH-PX in the pancreatic tissues were measured by MDA (cat. no. A003-1-1), SOD (cat. no. A001-1-1) and GSH-PX (cat. no. A005-1-2) assay kits (Nanjing Jiancheng Bioengineering Institute), respectively, according to the manufacturer's instructions. MDA can combine with thiobarbituric acid to form a red product with an absorption maximum at 532 nm using a fluorescence spectrophotometer. SOD generates superoxide anion radical (O₂⁻) through xanthine and the xanthine oxidase reaction system, which oxidizes hydroxylamine to form nitrite, presenting a purple red color under the action of chromogenic agents. Therefore, the activity of SOD can be determined by measuring the absorbance at 550 nm using a fluorescence spectrophotometer. GSH-PX can promote the reaction of hydrogen peroxide (H₂O₂) with reduced GSH to produce H₂O and the oxidized GSH. The activity of GSH-PX can be expressed as the rate of its enzymatic reaction. GSH and dithiodinitrobenzoic acid act to generate 5-thiodinitro-benzoate anions exhibiting a more stable yellow color, and the amount of GSH can be calculated by measuring its absorbance at 412 nm using a fluorescence spectrophotometer.

Total RNA from the pancreas was extracted using Total RNA Extraction Reagent (cat. no. BS259A; Biosharp Life Sciences) and reverse-transcribed into cDNA using Hifair^® III 1st Strand cDNA Synthesis SuperMix for qPCR [cat. no. 11141ES60; Yeasen Biotechnology (Shanghai) Co., Ltd.] according to the manufacturer's protocols. The concentration of RNA was determined using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc.). Real-time qPCR was performed using Hief UNICON^® qPCR SYBR Green Master Mix [cat. no. 11200ES08; Yeasen Biotechnology (Shanghai) Co., Ltd.] and the ABI 7500 Real-Time PCR system (Thermo Fisher Scientific, Inc.). The following thermocycling conditions were used for the qPCR: Initial denaturation at 95°C for 1 min; followed by 40 cycles at 95°C for 15 sec and 60°C for 1 min. The fold difference in gene expression was calculated using the 2^−ΔΔCq method and presented relative to endogenous β-actin mRNA (29). All reactions were performed at least in triplicate. Primer sequences are listed in Table I.

Total pancreatic protein was isolated using Cell Lysis Buffer for Western and IP (cat. no. BL509A; Biosharp Life Sciences). The total protein concentration was measured by BCA Protein Assay kit (cat. no. KGP902; Nanjing KeyGen Biotech Co., Ltd.). Protein samples (25 µg/lane) were separated on 10% gels using SDS-PAGE, and then transferred to PVDF membranes. At room temperature, the membranes were blocked with 5% skimmed milk for 1 h. The membranes were incubated overnight with the following primary anti-bodies at 4°C: β-actin (1:10,000; cat. no. ET1702-52; HuaBio); glutathione peroxidase 4 (GPX4; 1:1,000; cat. no. ER1803-15; HuaBio), SLC7a11 (1:1,000; cat. no. HA600097; HuaBio) and cytochrome c oxidase subunit II (COX2; 1:1,000; cat. no. 4842S; Cell Signaling Technology, Inc.). Next, the membranes were incubated with HRP-linked goat anti-rabbit IgG (1:5,000; cat. no. BL003A; Biosharp Life Sciences) at room temperature for 1 h. The target proteins in the membranes were visualized by enhanced chemiluminescence detection kit (cat. no. BL520A; Biosharp Life Sciences). The band strength was analyzed using ImageJ software (version 2.0; National Institutes of Health) and normalized to β-actin protein intensity.

Results are shown as the mean ± standard error of the mean. Statistical analysis was performed using GraphPad Prism 8.0 software (GraphPad Software, Inc.). Differences between the two groups were compared using an unpaired two-tailed Student's t-test. P<0.05 was considered to indicate a statistically significant difference.

The iron overload mouse model was constructed by intraperitoneal injection of 120 mg/kg body weight of iron dextran every other week for 12 weeks. Serum iron, transferrin saturation and pancreatic tissue iron levels were all significantly (P<0.01) elevated in the mice with iron dextran injection (Fig. 1A-C). Prussian blue staining revealed a large number of Prussian blue-positive spots, hemosiderin, in the pancreas of the mice injected with iron dextran (Fig. 1D). Massive iron deposition was showed in the exocrine rather than the endocrine pancreas. Moreover, the mRNA level of iron storing protein ferritin H (FtH) was significantly (P<0.01) increased in the pancreas of iron-treated mice, while those of iron transporting membrane proteins divalent metal transporter 1 (DMT1), ferroportin 1 (FPN) and transferrin receptor (TfR) were all significantly (P<0.01) decreased (Fig. 1E-H). The protein level of iron storing protein FtH was significantly (P<0.05) increased in the pancreas of iron-treated mice, while those of iron transporting membrane proteins DMT1, FPN, and iron binding receptor TfR were all significantly (P<0.05) decreased (Fig. 1I). The results indicated that the iron overload mouse model was successfully established and that large amounts of iron were deposited in the pancreas.

In view of the obvious pancreatic iron deposition in iron-overloaded mice, the effect of iron overload on the morphology and function of the pancreas was then analyzed. H&E staining showed that the acinar cells of the pancreas in iron-overloaded mice were atrophied and the area of intercellular substance was larger compared with those of control mice (Fig. 2A), which suggested mild pancreatic injury existed in the iron-overloaded mice. Serum amylase and lipase activities were both significantly (P<0.01) increased in the iron-overloaded mice compared with those of the control mice (Fig. 2B and C). Moreover, compared with those of the control group, the mRNA expression levels of pro-inflammatory cytokines, including interleukin (IL)-1β, IL-6 and inducible nitric oxide synthase, were significantly (P<0.01) increased in the pancreas of iron-overloaded animals (Fig. 2D-F). Accordingly, the transcript of anti-inflammatory cytokine IL-10 was significantly (P<0.01) decreased in the pancreas of iron-overloaded mice (Fig. 2G). Furthermore, the mRNA levels of SRY-related high-mobility-group-box gene 9 (a molecular marker of acinar to ductal metaplasia), keratin 19 (a molecular marker of ductal lesions) and vimentin (a molecular marker of stromal response) were all significantly (P<0.01) elevated by iron injection (Fig. 2H-J). The results indicated that iron overload induced mild CP and resulted in pancreatic injury.

Histological evaluation of the pancreatic lesions showed that increased levels of lymphocytes (anti-CD3), neutrophils (anti-CD11b) and macrophages (anti-F4/80) were observed in the pancreas of iron-overloaded mice compared with those of control mice (Fig. 3A). The quantitative positive areas of CD3 (P<0.01), CD11b (P<0.01) and F4/80 (P<0.05) cells showed a significant increase in the pancreas of the iron-overloaded mice compared with those in the control mice (Fig. 3B-D). Furthermore, the mRNA expression of CD11b, CD3e (lymphocyte marker) and mannose receptor C type 1 (macrophage marker) was significantly (P<0.01) increased relative to the control group (Fig. 3E-G). The results indicated that pancreatic accumulation of immunocytes, including lymphocytes, neutrophils and macrophages, was present in iron-overloaded mice, which was characteristic of CP.

Since pancreatitis symptoms were shown in the iron-overloaded mice, Masson's trichrome and Sirius red staining were next performed to assess collagen accumulation in the pancreatic tissue sections to examine the degree of pancreatic fibrosis. The analysis showed that iron overload induced perivascular collagen accumulation (Fig. 4A). Semi-quantitative morphometric analysis demonstrated that the collagen-positive area of tissue sections was significantly (P<0.01) increased in the pancreas of iron-overloaded mice compared with that of the control group (Fig. 4B and C). Additionally, analysis of transcripts of pancreatic fibrosis markers, such as α-smooth muscle actin (α-SMA), collagen type I α1, connective tissue growth factor and fibronectin-1, showed a significant (P<0.01) increase in their levels in the pancreas tissue of iron-overloaded mice (Fig. 4D-G). It was observed that the development of atrophy, inflammatory cell infiltration and fibrosis in the pancreatic tissue of iron-overloaded mice was significantly higher than that of control mice. Statistical comparisons between groups are summarized in Table II.

Oxidative stress is a common pathogenesis of a number of chronic diseases, and it is well known that iron overload affects the redox state. Compared with that of the control group, the MDA level was increased by 1.12-fold (P<0.01) in the pancreas of iron-overloaded mice (Fig. 5A). However, the injection of iron dextran led to a reduction of SOD activity by 52% (P<0.01) and GSH-PX activity by 37% (P<0.01) in the pancreas (Fig. 5B and C). This suggested that excess iron could elevate the pancreatic oxidative stress of pancreatic acinar cells in mice. Moreover, the injection of iron dextran significantly (P<0.05) promoted COX2 (a putative molecular marker of ferroptosis) protein expression, but inhibited GPX4 and SLC7A11 protein expression in the pancreas of iron-overloaded mice compared with the control group (Fig. 5D). The mRNA level of COX2 was also significantly (P<0.01) increased, and GPX4 and SLC7A11 were significantly (P<0.01) decreased in the pancreas of iron-overloaded mice compared with the levels in control mice (Fig. 5E-G). The results indicated that iron overload induced ferroptosis.