Obesity suppresses neutrophil-dependent itaconate signaling promoting ferroptosis in acute pancreatitis Authors: Néstor Jiménez-Cañete, Alicia Aliena-Valero, Caroline A. Bressan, Salvador Pérez, Isabel Torres-Cuevas, Laura Benkenstein, Sandra Pinto, Antonio Cuadrado, Juan Sastre, Sergio Rius-Pérez Published: 2026 ABSTRACT Obesity is a well-established risk factor for increased severity and mortality in acute pancreatitis. However, the mechanisms by which obesity alters pancreatic immune regulation and favors the progression of acute pancreatitis are not elucidated yet. Here, we identify a neutrophil-driven immune–metabolic pathway that controls ferroptosis during pancreatic inflammation. We show that infiltrating myeloid cells represent the principal source of the immunometabolite itaconate during acute pancreatitis. Through paracrine transfer via the SLC13A3 transporter, myeloid-derived itaconate protects pancreatic acinar cells from ferroptosis by sustaining NRF2-dependent antioxidant responses. Obesity disrupts this protective axis by suppressing ACOD1 expression in infiltrating neutrophils. Proteomic profiling of pancreatic neutrophils from obese mice confirmed reduced ACOD1 abundance and decreased expression of enzymes linked to the tricarboxylic acid cycle and pyruvate metabolism. This metabolic reprogramming limits itaconate production and weakens NRF2-driven redox defenses, leading to downregulation of the xCT–GPX4 ferroptosis-protective pathway and increased lipid peroxidation in the pancreas of obese mice with pancreatitis. Pharmacological restoration of itaconate signaling with the cell-permeable derivative 4-octyl itaconate reactivates NRF2 signaling, the xCT–GPX4 antioxidant axis, and the trans-sulfuration pathway, mitigating pancreatic injury. Together, these findings identify neutrophil-derived itaconate as a key modulator of ferroptosis susceptibility and reveal immune cell metabolism as a critical determinant of obesity-associated severity in acute pancreatitis. Introduction Itaconate has recently emerged as a central immunometabolite linking cellular metabolism to inflammatory responses [1]. It is produced from the tricarboxylic acid (TCA) cycle intermediate cis-aconitate by the mitochondrial enzyme aconitate decarboxylase 1 (ACOD1) and accumulates during acute inflammation causing profound metabolic changes in immune cells [2,3]. This metabolic reprogramming not only supports energetic and biosynthetic demands but also generates metabolites that function as signaling mediators during inflammation [4]. In this context, ACOD1-derived itaconate is predominantly induced in activated myeloid cells, particularly macrophages, in response to a broad range of inflammatory and stress-related stimuli, including microbial products, cytokines, and damage-associated molecular patterns (DAMPs) [1,[5], [6], [7], [8]]. Although myeloid cells represent the main source of itaconate, inducible ACOD1 expression has also been reported in other non-myeloid cell types —such as neurons, hepatocytes, and cancer cells— under specific pathological conditions [[9], [10], [11]]. By targeting key metabolic and signaling pathways, ACOD1-derived itaconate acts as a central modulator of inflammatory and redox responses, thereby limiting excessive immune activation [1,12,13]. Itaconate restrains proinflammatory signaling through NF-κB downregulation and suppression of inflammasome activation, while also rewiring cellular metabolism via succinate dehydrogenase inhibition [[14], [15], [16]]. In addition to these anti-inflammatory effects, a major component of itaconate action involves the activation of the NRF2 pathway [17]. Through its electrophilic properties, itaconate enables covalent modification of cysteine residues in kelch-like ECH-associated protein 1 (KEAP1), the substrate adaptor responsible for NRF2 degradation [17]. Alkylation of redox-sensitive KEAP1 cysteines by itaconate —most prominently Cys151, but also additional regulatory residues— prevents NRF2 ubiquitination, allowing NRF2 stabilization and transcriptional activation [17,18]. NRF2 is a master regulator of the antioxidant response and consequently of ferroptosis resistance, a programmed form of cell death characterized by the iron-dependent accumulation of lipid hydroperoxides [19]. Thus, NRF2 promotes transcription of SLC7A11, encoding the cystine–glutamate antiporter system Xc− (xCT), particularly under conditions of oxidative stress [[20], [21], [22]]. Regulation of this transporter is central to control ferroptosis as xCT sustains intracellular cysteine availability for glutathione (GSH) synthesis [19]. GSH serves as an essential cofactor for glutathione peroxidase 4 (GPX4), a selenoenzyme that detoxifies lipid peroxides and prevents iron-dependent lipid peroxidation [23,24]. Therefore, disruption of this NRF2–xCT–GSH–GPX4 axis compromises antioxidant capacity and promotes ferroptotic cell death [19,22,25]. Ferroptosis is accompanied not only by lipid peroxidation but also by broader stress-responsive transcriptional programs, including induction of cytokines and metabolic regulators associated with oxidative injury and cellular adaptation [[26], [27], [28]]. However, the identification of reliable redox regulatory nodes and molecular markers remains challenging [29]. Here, we investigated the role of ACOD1-dependent itaconate metabolism in the regulation of redox homeostasis and ferroptosis in acute pancreatitis, and also its modulation by obesity. Pancreatitis is an inflammatory pancreatic disorder associated with oxidative stress and disrupted cysteine redox metabolism [[30], [31], [32]]. Obesity is a well-established risk factor for increased severity and mortality in acute pancreatitis [[33], [34], [35]]. Nevertheless, the mechanisms by which obesity alters pancreatic immune regulation and favors the progression of acute pancreatitis have not been elucidated yet. In this work, we show that ferroptosis is triggered by obesity-associated impairment of ACOD1-dependent metabolic regulation in infiltrating pancreatic neutrophils in acute pancreatitis. Materials and methods Animals and in vivo experimental procedures Wild-type mice (C57BL/6J background), Acod1-deficient mice C57BL/6NJ-Acod1em1(IMPC)J/J (ACOD1 KO) and Nrf2-deficient mice (NRF2 KO) were used in this study. C57BL/6NJ-Acod1em1(IMPC)J/J were purchased from Jackson Laboratory. Colonies of NRF2-KO mice and NRF2-WT littermates were established from founders kindly provided by Dr. Masayuki Yamamoto (Tohoku University Graduate School of Medicine, Sendai, Japan). Experimental animals were maintained under standard conditions, with food and water ad libitum. Lean mice were fed standard diet (MD.08485, Envigo RMS Division, Indianapolis, IN, USA), whereas obese mice were fed a high-fat diet containing 42% from fat (MD. 88137, Envigo RMS Division, Indianapolis, IN, USA) for 12 weeks. Induction of acute pancreatitis was performed in 12-18 week-old mice by seven intraperitoneal (i.p) injections of cerulein (Sigma-Aldrich, St. Louis, MO, USA) (50 μg/kg of body weight) at 1h intervals. For control groups, mice were treated with physiological saline solution (0.9% NaCl). Mice were sacrificed 1 h after the last injection of cerulein under anhestesia with isofluorane 2-5% by cervical dislocation. For immune cell depletion, mice were treated by a single i.p. injection of clodronate liposomes (50 mg/kg; Liposoma BV, Amsterdam, The Netherlands) two days prior to the cerulein injections. Control animals received an equivalent i.p. dose of PBS-containing liposomes (Liposoma BV, Amsterdam, The Netherlands) following the same experimental schedule. To suppress ferroptosis, mice were treated with liproxstatin-1 (10 mg/kg, Sigma-Aldrich, St. Louis, MO, USA) administered via i.p. injection 1 h before the induction of pancreatitis. To investigate the effects of exogenous itaconate, mice were treated with 4-octylitaconate (4-OI; 25 mg/kg; Sigma-Aldrich, St. Louis, MO, USA) via i.p. injection. The treatment was administered 10 min prior to the first and fourth cerulein injections. All procedures were conducted in compliance with the legislation on protection of animals used for scientific purposes in Spain (RD 53/2013) and the EU (2010/63/EU). The study protocol was reviewed and approved by the Ethical Committee for Animal Experimentation of the University of Valencia (Valencia, Spain) (Protocol Codes: 2020-VSC-PEA-0030 Type 2 and 2023-VSC-PEA-0286 Type 2) Reverse transcription-quantitative polymerase chain reaction (RT-qPCR) A small fraction of the pancreas (around 20 mg) was excised and immediately immersed in RNA-later solution (Ambion, Thermo Fisher Scientific, Waltham, MA, USA) to preserve RNA integrity. Total RNA was extracted using TRIzol™ Reagent (Thermo Fisher Scientific, Whaltham, MA, USA) followed by chloroform phase separation, isopropanol precipitation and 75% (v/v) ethanol washes. The resulting RNA pellet was resuspended in nuclease-free water, and RNA concentration and purity were subsequently quantified using a NanoDrop™ spectrophotometer (Thermo Fisher Scientific, Whaltham, MA, USA). Complementary DNA (cDNA) was obtained by reverse transcription using PrimeScript™ RT Reagent Kit (Takara, Shiga, Japan) according to the manufacturer's instructions in a GeneAmp PCR System 9700 thermal cycler (Applied Biosystems, Whaltham, MA, USA). Gene expression analysis was performed by RT-qPCR using the StepOnePlus™ Real-Time PCR System (Applied Biosystems, Whaltham, MA, USA). Reactions were carried out using either commercial TaqMan® probes (Applied Biosystems, Whaltham, MA, USA) (Supplementary Table 1) with TaqMan Master Mix 2X (Takara, Kusatsu, Shiga, Japan), or specifically designed primers (Supplementary Table 2) with SYBR Green PCR Master Mix (Takara, Kusatsu, Shiga, Japan). Threshold cycle (CT) values were determined for all samples, and relative gene expression (fold change) was calculated using the 2−Δ(ΔCT) method. Briefly, target gene expression was first normalized to the housekeeping gene (ΔCT = CT target - CT housekeeping). Subsequently, ΔΔCT was calculated by comparing the treated group to the control group (ΔΔCT = ΔCTtreated – ΔCTcontrol). Tbp was used as housekeeping gene. Determination of lipid peroxidation Lipid peroxidation was quantified in pancreatic tissue samples using the Lipid Peroxidation Assay Kit (#ab118970, Abcam, Cambridge, United Kingdom), which is based on the reaction of thiobarbituric acid (TBA) with lipid peroxidation-derived products (TBARS). The protocol was carried out following the manufacturer's instructions. The results are expressed as TBARS (nmol/mg protein). Histological analysis A piece of pancreas was extracted and immersed in 4% (w/v) paraformaldehyde (PFA) for fixation. After 24 h of fixation, samples were embedded in paraffin (Sigma-Aldrich, St. Louis, MO, USA), sectioned at 4 μm with microtome, and mounted on glass slides. Following deparaffinization in xylene and rehydration through graded ethanol (100% to 70%, v/v), the sections were stained with hematoxylin and eosin (Sigma-Aldrich, St. Louis, MO, USA). All procedures were conducted at the University of Valencia's Central Service for Experimental Research Support (SCSIE). The extent and severity of necrosis, tissue edema and inflammatory infiltration were evaluated in tissue sections using a 20× objective across 10 non-overlapping fields per sample, in accordance with [36]. Western blot analysis Pancreatic specimens were frozen at −80 °C until homogenization on ice. Proteins were extracted in buffer (100 mg/mL) containing 20 mM Tris–HCl (pH 7.5), 1 mM EDTA, 150 mM NaCl, 0.1% (w/v) SDS, 1% (v/v) Igepal CA-630, 30 mM sodium pyrophosphate, 50 mM sodium fluoride, 50 μM sodium orthovanadate (all from Sigma-Aldrich, St. Louis, MO, USA), protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA) at a concentration of 8 μl/mL and phosphatase inhibitor cocktail (Sigma-Aldrich, St. Louis, MO, USA) in 1:100 dilution. Total protein lysates from cell cultures were harvested by adding 100 μL of protein extraction buffer per well and scraping the cells with a sterile cell scraper. Protein extracts were centrifuged (15,000 × g, 15 min), and protein concentrations were determined by BCA assay. Equal amounts of protein (20–40 μg) were mixed with sample buffer (69.45 mM Tris-HCl, pH 6.8, 10% (v/v) glycerol, 1% (v/v) SDS, 0.005% (w/v) bromophenol blue and 50 mM DTT, all from Sigma-Aldrich, St. Louis, MO, USA) and denatured at 95 °C for 5 min. Samples were then separated by SDS-PAGE under constant voltage (120–150 V), and transferred to nitrocellulose membranes using the Trans-Blot® Turbo™ System (Bio-Rad, Laboratories, Hercules, CA, USA). Membranes were blocked for 1 h in blocking buffer (5% (w/v) bovine serum albumin in TBS-T: 20 mM Tris, 137 mM NaCl, 0.1% (v/v) Tween-20, pH 7.6) to prevent non-specific binding. Incubation with specific primary antibodies was performed overnight at 4 °C (Supplementary Table 3). Following three washes with TBS-T (5 min at room temperature), membranes were incubated with HRP-conjugated secondary antibodies (Supplementary Table 4) for 1 h. Protein bands were developed using Pierce™ ECL Substrate (Thermo Fisher Scientific, Rockford, IL, USA) and visualized using a ChemiDoc™ MP imaging system (Bio-Rad, Laboratories, Hercules, CA, USA). Cell culture The murine acinar cell line 266-6 (CRL-2151; ATCC, Manassas, VA, USA) was cultured in Dulbecco's Modified Eagle Medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS), 100 U/mL penicillin, 100 μg/mL streptomycin, and 0.01% (w/v) Fungizone (all from Gibco, Thermo Fisher Scientific, Waltham, MA, USA). Cells were maintained in the logarithmic growth phase at 37 °C in a humidified atmosphere containing 5% CO2 and were routinely tested for mycoplasma contamination. Ferroptosis in vitro assay The effect of itaconate on acinar cell ferroptosis was evaluated in vitro by incubation with 4-octylitaconate (Sigma-Aldrich, St. Louis, MO, USA). The following four groups were studied: control (C), 4-octylitaconate (O), erastin (E) and erastin + 4-octylitaconate (E + O). Around 250.000 cells were seeded in each well and incubated with 2 mL of complete medium. 4-octylitaconate was diluted in DMSO and added in O and E + O groups to reach a final concentration of 250 μM; the same volume of DMSO was added to groups C and E. After 3h of incubation, erastin (Sigma-Aldrich, St. Louis, MO, USA) was incubated for 24h in E and E + O groups with a final concentration of 2.5 μM. Cell viability was evaluated by propidium iodide (PI) exclusion using a FACSVerse cytometer (BD Biosciences, Franklin Lakes, NJ, USA). Briefly, cells were collected, centrifuged for 5 min at 450 × g, resuspended in 500 μl of PBS and stained with PI (10 μg/mL). Since PI penetrates only cells with compromised membranes, PI-positive events were classified as non-viable. Targeted isolation of neutrophils and macrophages from the pancreas After acute pancreatitis induction, pancreatic specimens were collected in 4 ml of Hanks’ Balanced Salt solution (HBSS; from Gibco, Thermo Fisher Scientific, Waltham, MA, USA). Subsequently, samples were mechanically dissociated and subjected to enzymatic digestion by incubation with D collagenase 1 mg/mL (Roche, Basel, Switzerland) in HBSS at 37 °C for 40 min. Digestion was stopped by addition of 20 mL ice-cold HBSS supplemented with 10% (v/v) FBS. Then, samples were filtrated through 100 μm filter to discard non-digested tissue fragments. Cells were collected by centrifugation at 450 × g for 5 min. Neutrophils were enriched using the Ly6G UltraPure MicroBead Kit (Miltenyi Biotec, Bergisch Gladbach, Germany), and macrophages using F4/80 MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany), following the manufacturer's protocols. Briefly, 107 cells were centrifuged at 300 × g for 10 min, resuspended in 90 μL of PEB buffer, and incubated with 10 μL of the corresponding microbeads for 10 min on ice. Cells were then washed with 2 mL of PEB buffer, centrifuged at 300 × g for 10 min, and resuspended in 500 μL of PEB. Automated magnetic separation was performed using the autoMACS® Pro Separator (Miltenyi Biotec, Bergisch Gladbach, Germany). Positive selection was carried out using the Posseld2 program, which increases purity by sequentially passing the sample through two magnetic columns. Single-cell suspensions were stained with CD45-VioGreen (#130-110-803, Miltenyi Biotec), CD11b-PE (#130-113-806, Miltenyi Biotec, Bergisch Gladbach, Germany), Ly-6G-APC (#130-120-803, Miltenyi Biotec, Bergisch Gladbach, Germany), and F4/80-BV421 (#123131, BioLegend, San Diego, CA, USA) for 15 min on ice and protected from light. After incubation, cells were washed, resuspended in PBS, and analyzed on a FACS Aria Fusion (BD Biosciences, Franklin Lakes, NJ, USA) equipped with 405 nm, 488 nm, 561 nm, and 640 nm lasers, with fluorescence collected using 450/50 nm, 530/30 nm, 586/15 nm, and 670/10 nm emission filters. Neutrophils were identified and sorted as CD45+CD11b+Ly6G+ cells, and macrophages as CD45+CD11b+F4/80+ cells. Proteomics analysis Cells were lysed in RIPA buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% (v/v) NP-40, 0.5% (w/v) sodium deoxycholate, and 0.1% (w/v) SDS). For each sample, 10 μg of protein was reduced with 2 mM DTT at 60 °C for 20 min and subsequently alkylated with 5.5 mM iodoacetamide in the dark for 30 min at room temperature. Protein processing was performed using the Single-Pot Solid-Phase-enhanced Sample Preparation (SP3) protocol [37,38]; proteins were bound to SP3 beads at a 1:20 protein-to-bead ratio in 70% acetonitrile, followed by bead-based cleanup and overnight digestion at 37 °C with 500 ng of trypsin in 50 mM ammonium bicarbonate. The resulting peptides were acidified with 10% (v/v) TFA to a final concentration of 1%. Approximately 200 ng of the digest was loaded onto Evotip Pure tips (EvoSep Biosystems, Odense, Denmark) following the manufacturer's instructions, with sample injection order fully randomized using HyStar (Bruker, Billerica, MA, USA). LC–MS/MS analysis was conducted on a timsTOF fleX mass spectrometer coupled to an Evosep One system, utilizing an Endurance analytical column (15 cm × 150 μm, 1.5 μm) and the 30 SPD chromatographic method. Peptides were ionized via a CaptiveSpray source (1600 V, 180 °C) and analyzed in diaPASEF mode with a scan range of 100–1700 m/z and a 1/K0 range of 0.6–1.6 V s/cm2. Raw data were processed using DIA-NN v1.8 via the FragPipe 21.1 pipeline against the SwissProt Mouse database (2025-04-01 release). Quantification results were filtered at a 1% FDR, and differential expression analysis was performed using the FragPipe-Analyst R Shiny web server. Determination of low molecular weight thiols by UPLC–MS/MS Pancreatic tissues were homogenized in PBS containing 10 mmol/L N-ethylmaleimide (NEM). Perchloric acid was added to achieve a final concentration of 4% (v/v). Samples were centrifuged at 15,000 rpm for 15 min at 4 °C and the supernatants were collected and injected into the UPLC–MS/MS system. Chromatographic analyses were performed using a QTRAP 6500 plus mass spectrometer (Sciex, Framingham, MA, USA). The analytical method used has been previously described [39]. Statistical analysis Data are expressed as mean ± standard error of the mean (SEM). Statistical evaluations were performed using GraphPad Prism 8 software (GraphPad Software, San Diego, CA, USA). Significant differences between multiple groups were determined using one-way repeated measures analysis of variance (ANOVA), followed by Sidak or Tukey's post hoc test for multiple comparisons. For comparisons between two independent groups, a two-tailed Student's t-test was employed. In all cases, a p-value <0.05 was considered statistically significant. Results Infiltrating myeloid cells overexpress ACOD1 and represent the primary source of itaconate in pancreas during acute pancreatitis We first analyze the expression of Acod1, the enzyme responsible for itaconate synthesis, in a murine model of cerulein-induced pancreatitis. Pancreatic Acod1 expression was markedly increased in mice with pancreatitis compared to controls, reaching around 50-fold induction (Fig. 1A).Fig. 1(A) RT-qPCR analysis of pancreatic Acod1 mRNA expression in control (Sham) and cerulein-induced acute pancreatitis (Cer) mice (n = 5 mice per group). (B) Schematic representation of the experimental timeline for macrophage/neutrophil depletion using clodronate-loaded liposomes (Cldr) and pancreatitis induction. (C) Effect of clodronate treatment on pancreatic Acod1 mRNA expression in control (Sham), cerulein-induced acute pancreatitis (Cer) mice and clodronate-treated (Cer + Cldr) mice (n = 3-5 mice per group). (D) Relative Slc13a3 mRNA levels expression in control (Sham) and cerulein-induced acute pancreatitis (Cer) mice (n = 5 mice per group). (E) Impact of immune cell depletion on Slc13a3 mRNA expression in control (Sham), cerulein-induced acute pancreatitis (Cer) mice and clodronate-treated (Cer + Cldr) mice (n = 3-5 mice per group). (F) Representative Western blot images and (G) densitometric quantification of ACOD1 protein levels in neutrophils (Neu) and macrophages (MΦ) isolated from pancreas tissue with cerulein-induced acute pancreatitis; α-vinculin served as a loading control (n = 3 isolations per group). For gene expression analysis, data were normalized to TATA-binding protein (Tbp) and are presented as fold change relative to Sham. Bars represent mean ± SEM; ∗p < 0.05. (A) RT-qPCR analysis of pancreatic Acod1 mRNA expression in control (Sham) and cerulein-induced acute pancreatitis (Cer) mice (n = 5 mice per group). (B) Schematic representation of the experimental timeline for macrophage/neutrophil depletion using clodronate-loaded liposomes (Cldr) and pancreatitis induction. (C) Effect of clodronate treatment on pancreatic Acod1 mRNA expression in control (Sham), cerulein-induced acute pancreatitis (Cer) mice and clodronate-treated (Cer + Cldr) mice (n = 3-5 mice per group). (D) Relative Slc13a3 mRNA levels expression in control (Sham) and cerulein-induced acute pancreatitis (Cer) mice (n = 5 mice per group). (E) Impact of immune cell depletion on Slc13a3 mRNA expression in control (Sham), cerulein-induced acute pancreatitis (Cer) mice and clodronate-treated (Cer + Cldr) mice (n = 3-5 mice per group). (F) Representative Western blot images and (G) densitometric quantification of ACOD1 protein levels in neutrophils (Neu) and macrophages (MΦ) isolated from pancreas tissue with cerulein-induced acute pancreatitis; α-vinculin served as a loading control (n = 3 isolations per group). For gene expression analysis, data were normalized to TATA-binding protein (Tbp) and are presented as fold change relative to Sham. Bars represent mean ± SEM; ∗p < 0.05. This remarkable induction of Acod1 prompted us to investigate the cellular origin of itaconate during acute pancreatitis. To this end, mice were treated with clodronate-loaded liposomes (Fig. 1B), a well-established strategy to deplete macrophages that has also been reported to abrogate neutrophil accumulation in inflamed tissues [[40], [41], [42], [43]]. The effectiveness of clodronate treatment was confirmed by RT-qPCR analysis, which revealed a marked reduction in the expression of both macrophage and neutrophil markers F4/80 and Mpo, respectively (Supplementary Fig. S1A and B). Upon clodronate treatment, pancreatic Acod1 expression was markedly reduced too (Fig. 1C), indicating that the strong induction of Acod1 observed during cerulein-induced pancreatitis was largely dependent on these cell populations. We next examined the expression of Slc13a3, a transporter involved in itaconate uptake [[44], [45], [46]]. Slc13a3 expression was significantly upregulated in the pancreas during cerulein-induced pancreatitis (Fig. 1D). In contrast to Acod1, depletion of macrophages and neutrophils did not alter Slc13a3 expression levels (Fig. 1E), indicating that the pancreatitis-associated induction of Slc13a3 is likely to occur in pancreatic acinar cells and may support itaconate uptake by these cells. To further confirm the cellular sources underlying the strong induction of Acod1, pancreatic cells were separated by magnetic-activated cell sorting (MACS) into macrophage (F4/80+), neutrophil (Ly6G+), and negative fractions. Western blot analysis revealed high levels of ACOD1 protein in both the macrophage and neutrophil fractions, whereas ACOD1 expression was negligible in the negative fraction (Fig. 1F). ACOD1 deficiency reprograms metabolic pathways in pancreatic neutrophils and aggravates acute pancreatitis To investigate the functional role of ACOD1 during acute pancreatitis, we induced pancreatitis in ACOD1-deficient mice (ACOD1 KO). Histological analysis revealed that, compared with wild-type animals, ACOD1 KO mice developed more severe pancreatic injury, characterized by increased interstitial edema and a higher proportion of necrotic areas following pancreatitis induction (Fig. 2A and B). Consistent with these findings, plasma amylase levels were further elevated in ACOD1 KO mice compared with wild-type controls (Fig. 2C), indicating exacerbated tissue damage.Fig. 2(A) Representative histological images of hematoxylin-eosin staining and (B) histological score for inflammation, edema and necrosis of pancreas from wild-type (WT) and ACOD1 knockout (ACOD1 KO) Sham and with cerulein-induced acute pancreatitis (Cer). Scale bar: 50 μm. (C) Amylase activity in plasma of wild-type (WT) and ACOD1 knockout (ACOD1 KO) mice under control condition (Sham) and with cerulein-induced acute pancreatitis (Cer) (D) Volcano plot of differentially expressed proteins in isolated neutrophils from ACOD1 knockout (ACOD1 KO) vs. wild-type (WT) mice with acute pancreatitis. Each point represents a protein; significantly upregulated proteins in WT mice are shown in orange (n = 264) and downregulated proteins in blue (n = 217) (p-value <0.05). ACOD1 is highlighted as an internal validation of the KO model. (E) KEGG pathway enrichment analysis. Bars represent the Normalized Enrichment Score (NES) for Acod1 knockout (ACOD1 KO) vs. wild-type (WT) neutrophils. Only pathways with FDR <0.05 and p-value <0.05 are shown. (F) Detailed list of significant proteins involved in selected metabolic pathways. Protein names in red indicate upregulation in wild-type mice, while those in blue indicate downregulation. Data are presented as mean ± SEM; ∗p < 0.05 (n = 4-5 mice per group). (A) Representative histological images of hematoxylin-eosin staining and (B) histological score for inflammation, edema and necrosis of pancreas from wild-type (WT) and ACOD1 knockout (ACOD1 KO) Sham and with cerulein-induced acute pancreatitis (Cer). Scale bar: 50 μm. (C) Amylase activity in plasma of wild-type (WT) and ACOD1 knockout (ACOD1 KO) mice under control condition (Sham) and with cerulein-induced acute pancreatitis (Cer) (D) Volcano plot of differentially expressed proteins in isolated neutrophils from ACOD1 knockout (ACOD1 KO) vs. wild-type (WT) mice with acute pancreatitis. Each point represents a protein; significantly upregulated proteins in WT mice are shown in orange (n = 264) and downregulated proteins in blue (n = 217) (p-value <0.05). ACOD1 is highlighted as an internal validation of the KO model. (E) KEGG pathway enrichment analysis. Bars represent the Normalized Enrichment Score (NES) for Acod1 knockout (ACOD1 KO) vs. wild-type (WT) neutrophils. Only pathways with FDR <0.05 and p-value <0.05 are shown. (F) Detailed list of significant proteins involved in selected metabolic pathways. Protein names in red indicate upregulation in wild-type mice, while those in blue indicate downregulation. Data are presented as mean ± SEM; ∗p < 0.05 (n = 4-5 mice per group). Given the high expression of ACOD1 in pancreatic neutrophils and their established contribution to pancreatitis pathogenesis [31], we next sought to define the molecular programs associated with ACOD1 deficiency in neutrophils. Neutrophils were isolated from the pancreas of ACOD1-deficient mice and subjected to unbiased proteomic profiling followed by KEGG-based functional enrichment analysis to identify pathways altered by loss of ACOD1 (Supplementary Fig. S2A). Proteomic analysis revealed enrichment of certain metabolic pathways in ACOD1-deficient pancreatic neutrophils. Functional annotation demonstrated overrepresentation of pathways related to the tricarboxylic acid (TCA) cycle, oxidative phosphorylation, and glycolysis/gluconeogenesis, suggesting alterations in bioenergetic metabolism. In addition, multiple amino acid metabolic pathways were enriched, including cysteine and methionine metabolism, glycine/serine/threonine metabolism, and branched-chain amino acid degradation (valine, leucine, and isoleucine) (Fig. 2D, E, F). ACOD1 deficiency restrains NRF2 activation promoting ferroptosis during acute pancreatitis To determine whether loss of ACOD1 affects NRF2-dependent antioxidant responses during pancreatitis, we examined NRF2 activation in pancreatitis in mice lacking ACOD1. In accordance with our previous reports [32], cerulein-induced pancreatitis was associated with robust induction of NRF2 in wild-type mice. In contrast, this response was markedly attenuated in in ACOD1-deficient animals (Fig. 3A and Supplementary Fig. S2B).Fig. 3(A) Representative Western blot images of NRF2, xCT and GPX4 protein levels in samples of pancreatic tissue from wild-type (WT) and ACOD1 knockout (KO) mice Sham and with cerulein-induced acute pancreatitis; β-tubulin served as a loading control. n = 4-5 mice per group (B) Lipid peroxidation levels in pancreatic tissue from wild-type (WT) and ACOD1 knockout (ACOD1 KO) mice Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein (n = 4-5 mice per group). (C) Relative Gdf15 mRNA expression levels in wild-type (WT) and ACOD1 knockout (KO) mice Sham and with cerulein-induced acute pancreatitis (Cer). Data were normalized to TATA-binding protein (Tbp) and are presented as fold change relative to WT-Sham (n = 4-5 mice per group). (D) Cell viability assessment. 266-6 murine pancreatic acinar cells were treated with (+) or without (−) the ferroptosis inducer Erastin and the itaconate derivative 4-Octyl itaconate (4-OI), as indicated. Fold change in cell viability (relative to no treated cells) analyzed by flow cytometry using propidium iodide (PI) staining (n = 3 independent experiments per group). (E) Representative Western blot images of NRF2, xCT, and GPX4 protein levels in 266-6 murine pancreatic acinar cells. Cells were treated with (+) or without (−) the ferroptosis inducer Erastin and the itaconate derivative 4-Octyl itaconate (4-OI), as indicated; β-tubulin served as a loading control. n = 3 independent experiments per group. Data are presented as mean ± SEM; ∗p < 0.05. (A) Representative Western blot images of NRF2, xCT and GPX4 protein levels in samples of pancreatic tissue from wild-type (WT) and ACOD1 knockout (KO) mice Sham and with cerulein-induced acute pancreatitis; β-tubulin served as a loading control. n = 4-5 mice per group (B) Lipid peroxidation levels in pancreatic tissue from wild-type (WT) and ACOD1 knockout (ACOD1 KO) mice Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein (n = 4-5 mice per group). (C) Relative Gdf15 mRNA expression levels in wild-type (WT) and ACOD1 knockout (KO) mice Sham and with cerulein-induced acute pancreatitis (Cer). Data were normalized to TATA-binding protein (Tbp) and are presented as fold change relative to WT-Sham (n = 4-5 mice per group). (D) Cell viability assessment. 266-6 murine pancreatic acinar cells were treated with (+) or without (−) the ferroptosis inducer Erastin and the itaconate derivative 4-Octyl itaconate (4-OI), as indicated. Fold change in cell viability (relative to no treated cells) analyzed by flow cytometry using propidium iodide (PI) staining (n = 3 independent experiments per group). (E) Representative Western blot images of NRF2, xCT, and GPX4 protein levels in 266-6 murine pancreatic acinar cells. Cells were treated with (+) or without (−) the ferroptosis inducer Erastin and the itaconate derivative 4-Octyl itaconate (4-OI), as indicated; β-tubulin served as a loading control. n = 3 independent experiments per group. Data are presented as mean ± SEM; ∗p < 0.05. In accordance with the fall in NRF2 levels, pancreatic protein levels of the NRF2 targets and key ferroptosis-protective factors xCT (SLC7A11) and GPX4 were decreased in ACOD1-deficient animals following pancreatitis induction (Fig. 3A and Supplementary Fig. S2C and D). These changes were accompanied by enhanced lipid peroxidation detected in ACOD1 KO mice with pancreatitis (Fig. 3B). Furthermore, Gdf15 expression, a stress-inducible cytokine previously associated with oxidative and ferroptotic stress [26,27], was significantly increased during cerulein-induced pancreatitis and was further elevated in ACOD1-deficient mice (Fig. 3C). To further confirm the relationship between itaconate signaling and ferroptotic cell death, we performed in vitro experiments in pancreatic acinar 266-6 cells. Ferroptosis was induced by treatment with erastin (2.5 mM; 3h), which resulted in a marked reduction in cell viability (Fig. 3D and Supplementary Fig. S3A). Remarkably, co-treatment with 4-octyl itaconate (4-OI) significantly improved cell viability in erastin-treated cells (Fig. 3D and Supplementary Fig. S3A), indicating a protective effect of itaconate against ferroptotic cell death. Furthermore, erastin treatment diminished NRF2 protein levels, whereas 4-OI preserved NRF2 protein expression during erastin-induced ferroptosis (Fig. 3E and Supplementary Fig. S3B). Accordingly, erastin markedly decreased protein levels of xCT and GPX4 (Fig. 3E and Supplementary Fig. S3C and D). However, although 4-OI restored GPX4 expression, xCT levels remained downregulated in this cellular model (Fig. 3E and Supplementary Fig. S3C and D). Taking into account that ACOD1 deficiency was associated with impaired NRF2 activation and reduced expression of its target genes and ferroptosis-protective factors, we addressed whether NRF2 signaling mediates the downstream effects of ACOD1 on ferroptosis-related pathways in vivo. To this end, we analyzed the response to cerulein-induced pancreatitis in NRF2-deficient mice. Similar to ACOD1 KO animals, NRF2 KO mice developed more severe pancreatic injury, as evidenced by elevated plasma amylase levels compared with wild-type controls (Fig. 4A).Fig. 4(A) Amylase activity in plasma of wild-type (WT) and NRF2 knockout (NRF2 KO) mice Sham and with cerulein-induced acute pancreatitis (Cer) (n = 4-5 mice per group). (B) Representative Western blot images of xCT and GPX4 protein levels in samples of pancreas tissue from wild-type (WT) and NRF2 knockout (KO) mice Sham and with cerulein-induced acute pancreatitis (Cerulein); β-tubulin served as a loading control. (n = 4 mice per group) (C) Lipid peroxidation levels in pancreatic tissue. Lipid peroxidation was measured in wild-type (WT) and NRF2 knockout (NRF2 KO) mice Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein. (n = 4-5 mice per group). (D) Reduced glutathione (GSH) levels in pancreatic tissue. GSH content was quantified in wild-type (WT) and NRF2 knockout (NRF2 KO) Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein. (n = 3-7 mice per group). Bars represent mean ± SEM; ∗p < 0.05. (A) Amylase activity in plasma of wild-type (WT) and NRF2 knockout (NRF2 KO) mice Sham and with cerulein-induced acute pancreatitis (Cer) (n = 4-5 mice per group). (B) Representative Western blot images of xCT and GPX4 protein levels in samples of pancreas tissue from wild-type (WT) and NRF2 knockout (KO) mice Sham and with cerulein-induced acute pancreatitis (Cerulein); β-tubulin served as a loading control. (n = 4 mice per group) (C) Lipid peroxidation levels in pancreatic tissue. Lipid peroxidation was measured in wild-type (WT) and NRF2 knockout (NRF2 KO) mice Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein. (n = 4-5 mice per group). (D) Reduced glutathione (GSH) levels in pancreatic tissue. GSH content was quantified in wild-type (WT) and NRF2 knockout (NRF2 KO) Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein. (n = 3-7 mice per group). Bars represent mean ± SEM; ∗p < 0.05. In mice lacking NRF2, pancreatic xCT levels were markedly reduced under basal conditions and became nearly undetectable following pancreatitis induction (Fig. 4B and Supplementary Fig. S4A). GPX4 levels were also decreased in NRF2 KO mice with acute pancreatitis (Fig. 4B and Supplementary Fig. S4B). Furthermore, NRF2-deficient mice exhibited increased lipid peroxidation together with a dramatic depletion of pancreatic reduced glutathione (GSH) upon pancreatitis induction (Fig. 4C and D). Obesity suppresses ACOD1 upregulation, promoting ferroptosis in pancreas with acute pancreatitis We next characterized the impact of obesity on ACOD1 expression in mice with acute pancreatitis. Obese mice exhibited markedly reduced pancreatic Acod1 mRNA levels compared with lean controls (Fig. 5A). To determine whether Acod1 downregulation was evident at the protein level and within specific immune populations, we performed proteomic profiling of pancreatic neutrophils isolated from obese mice. This analysis confirmed decreased ACOD1 abundance in neutrophils from obese animals (Fig. 5B). Remarkably, KEGG pathway enrichment analysis revealed that this decrease occurred within a broader metabolic context characterized by coordinated downregulation of mitochondrial pathways, including the tricarboxylic acid (TCA) cycle and pyruvate metabolism (Fig. 5C and D). These findings indicate that obesity is associated with suppression of ACOD1 expression within an overall dampened oxidative metabolic program in pancreatic neutrophils.Fig. 5(A) Relative Acod1 mRNA expression levels in lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer). (B) Volcano plot of differentially expressed proteins in pancreatic isolated neutrophils from obese vs lean mice with acute pancreatitis. Each point represents a protein; significantly upregulated proteins in lean mice are shown in orange (n = 88) and downregulated proteins in blue (n = 389) (p-value <0.05). ACOD1 is highlighted. (C) KEGG pathway enrichment analysis. Bars represent the Normalized Enrichment Score (NES) for obese vs. lean pancreatic neutrophils. Only pathways with FDR <0.05 and p-value <0.05 are shown. (D) Detailed list of significant proteins involved in selected metabolic pathways. Protein names in blue indicate downregulation in obese mice. (E) Representative Western blot images of NRF2, xCT and GPX4 protein levels in samples of pancreas tissue from lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer); β-tubulin served as loading control (F) Lipid peroxidation levels measured in the pancreas from lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein. (G) Relative Gdf15 mRNA expression levels in lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer). (H) Representative histological images of hematoxylin-eosin staining of pancreas from lean and obese Sham and with cerulein-induced acute pancreatitis (Cer). Scale bar: 50 μm. For gene expression analysis, data were normalized to TATA-binding protein (Tbp) and are presented as fold change relative to lean sham mice. Data are presented as mean ± SEM; ∗p < 0.05 (n = 5-7 mice per group). (A) Relative Acod1 mRNA expression levels in lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer). (B) Volcano plot of differentially expressed proteins in pancreatic isolated neutrophils from obese vs lean mice with acute pancreatitis. Each point represents a protein; significantly upregulated proteins in lean mice are shown in orange (n = 88) and downregulated proteins in blue (n = 389) (p-value <0.05). ACOD1 is highlighted. (C) KEGG pathway enrichment analysis. Bars represent the Normalized Enrichment Score (NES) for obese vs. lean pancreatic neutrophils. Only pathways with FDR <0.05 and p-value <0.05 are shown. (D) Detailed list of significant proteins involved in selected metabolic pathways. Protein names in blue indicate downregulation in obese mice. (E) Representative Western blot images of NRF2, xCT and GPX4 protein levels in samples of pancreas tissue from lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer); β-tubulin served as loading control (F) Lipid peroxidation levels measured in the pancreas from lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer). Data are expressed as nmol/mg protein. (G) Relative Gdf15 mRNA expression levels in lean and obese mice Sham and with cerulein-induced acute pancreatitis (Cer). (H) Representative histological images of hematoxylin-eosin staining of pancreas from lean and obese Sham and with cerulein-induced acute pancreatitis (Cer). Scale bar: 50 μm. For gene expression analysis, data were normalized to TATA-binding protein (Tbp) and are presented as fold change relative to lean sham mice. Data are presented as mean ± SEM; ∗p < 0.05 (n = 5-7 mice per group). Importantly, in contrast to lean animals, obese mice failed to activate NRF2 in response to cerulein-induced pancreatitis (Fig. 5E, and Supplementary Fig. S5A). Consistent with impaired NRF2 activation, protein levels of the NRF2 targets and ferroptosis-protective factors xCT and GPX4, which were already diminished under obese conditions, remained suppressed following pancreatitis induction (Fig. 5E, and Supplementary Fig. S5B and C). These alterations paralleled the increased levels of pancreatic lipid peroxidation in these mice (Fig. 5F). Furthermore, pancreatic Gdf15 expression was elevated in obese mice compared with lean controls (Fig. 5G), further supporting the presence of ferroptosis. Histological analysis confirmed enhanced inflammatory infiltration, pronounced interstitial edema, and expanded areas of necrosis in obese mice following pancreatitis induction (Supplementary Fig. S5D). These histopathological changes were accompanied by a further increase in pancreatic amylase levels in obese animals compared with lean controls (Supplementary Fig. S5E). To determine whether ferroptosis directly contributes to the exacerbated pancreatic injury observed in obesity, obese mice were treated with liproxstatin-1, a specific inhibitor of ferroptotic cell death. Liproxstatin-1 treatment significantly attenuated pancreatitis severity in obese mice, as evidenced by reduced necrotic areas, as well as decreased plasma amylase levels (Supplementary Figs. S6A, B, C, D, E, F). Exogenous 4-octyl itaconate ameliorates pancreatic injury, restores redox homeostasis and prevents ferroptosis in acute pancreatitis in obese mice Our final approach was to investigate whether exogenous itaconate could mitigate pancreatic injury during cerulein-induced pancreatitis in lean and obese mice. To this end, we administered 4-octyl itaconate (4-OI), a cell-permeable esterified derivative of itaconate that effectively enters cells and mimics the metabolic functions of endogenous itaconate [17,47,48]. In lean mice with cerulein-induced pancreatitis, 4-OI treatment resulted in marked histological improvement compared with untreated pancreatitis controls, with reduced inflammatory cell infiltration and decreased interstitial edema (Fig. 6A and Supplementary Fig. S7A). These histopathological improvements paralleled the reductions in plasma amylase levels (Fig. 6B). 4-OI treatment also attenuated the pancreatitis-associated upregulation of pro-inflammatory cytokines Tnf-α and Il-6 (Fig. 6C and D). Furthermore, expression of the NRF2 target genes Gr, Nqo1, Sod1, and Ho1 was induced during acute pancreatitis and further enhanced following 4-OI treatment, supporting NRF2 activation in the pancreas by exogenous itaconate (Supplementary Fig. S7B).Fig. 6(A) Representative histological images of hematoxylin-eosin staining of pancreas from lean mice under control condition (Sham), with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). Scale bar: 50 μm. (B) Amylase activity in plasma of lean mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (C) Relative Tnfα and (D) Il-6 mRNA levels expression in lean mice Sham, cerulein-induced acute pancreatitis (Cer) mice and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (E) Relative levels in lean mice of methionine (Met), homocysteine (HCys), cysteine (Cys), cystine (CySS), γ-glutamylcysteine (γ-GC), and glutathione (GSH) were determined by UPLC-MS/MS. Experimental groups include basal condition (Sham), acute pancreatitis cerulein-induced (Cer), and cerulein-induced acute pancreatitis and 4-octylitaconate (Cer + 4OI). Asterisks (∗) denote metabolites with statistically significant differences between Cer and Cer+4OI groups. (F) Representative histological images of hematoxylin-eosin staining of pancreas from obese mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). Scale bar: 50 μm. (G) Amylase activity in plasma of obese mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (H) Relative Tnfα and (I) Il-6 mRNA levels expression in obese mice Sham, cerulein-induced acute pancreatitis (Cer) mice and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (J) Relative levels in obese mice of methionine (Met), homocysteine (HCys), cysteine (Cys), cystine (CySS), γ-glutamylcysteine (γ-GC), and glutathione (GSH) were determined by UPLC-MS/MS. Experimental groups include basal condition (Sham), acute pancreatitis cerulein-induced (Cer), and cerulein-induced acute pancreatitis and 4-octylitaconate (Cer + 4OI). Asterisks (∗) denote metabolites with statistically significant differences between Cer and Cer+4OI groups. (K) Representative Western blot images of NRF2, xCT and GPX4 protein levels in samples of pancreas tissue from obese mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI).; β-tubulin served as a loading control. Data are presented as mean ± SEM; ∗p < 0.05 (n = 4-5 mice per group). (A) Representative histological images of hematoxylin-eosin staining of pancreas from lean mice under control condition (Sham), with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). Scale bar: 50 μm. (B) Amylase activity in plasma of lean mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (C) Relative Tnfα and (D) Il-6 mRNA levels expression in lean mice Sham, cerulein-induced acute pancreatitis (Cer) mice and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (E) Relative levels in lean mice of methionine (Met), homocysteine (HCys), cysteine (Cys), cystine (CySS), γ-glutamylcysteine (γ-GC), and glutathione (GSH) were determined by UPLC-MS/MS. Experimental groups include basal condition (Sham), acute pancreatitis cerulein-induced (Cer), and cerulein-induced acute pancreatitis and 4-octylitaconate (Cer + 4OI). Asterisks (∗) denote metabolites with statistically significant differences between Cer and Cer+4OI groups. (F) Representative histological images of hematoxylin-eosin staining of pancreas from obese mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). Scale bar: 50 μm. (G) Amylase activity in plasma of obese mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (H) Relative Tnfα and (I) Il-6 mRNA levels expression in obese mice Sham, cerulein-induced acute pancreatitis (Cer) mice and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI). (J) Relative levels in obese mice of methionine (Met), homocysteine (HCys), cysteine (Cys), cystine (CySS), γ-glutamylcysteine (γ-GC), and glutathione (GSH) were determined by UPLC-MS/MS. Experimental groups include basal condition (Sham), acute pancreatitis cerulein-induced (Cer), and cerulein-induced acute pancreatitis and 4-octylitaconate (Cer + 4OI). Asterisks (∗) denote metabolites with statistically significant differences between Cer and Cer+4OI groups. (K) Representative Western blot images of NRF2, xCT and GPX4 protein levels in samples of pancreas tissue from obese mice Sham, with cerulein-induced acute pancreatitis (Cer) and with cerulein-induced acute pancreatitis and 4-octylitaconate (Cer+4OI).; β-tubulin served as a loading control. Data are presented as mean ± SEM; ∗p < 0.05 (n = 4-5 mice per group). To explore whether itaconate supplementation modulates redox metabolism and could potentially affect the regulation of ferroptosis during acute pancreatitis, we measured reduced glutathione (GSH) levels as well as metabolites of the trans-sulfuration pathway, a key metabolic route linking sulfur amino acid metabolism to glutathione synthesis. We have previously shown that this pathway is blocked during pancreatitis, leading to homocysteine accumulation and reduced flux toward cysteine and glutathione synthesis [32]. Interestingly, we found that 4-OI treatment reversed this blockade, avoiding homocysteine accumulation and favoring restoration of cysteine availability and physiological GSH levels (Fig. 6E). In addition, 4-OI treatment prevented disulfide stress [49], i.e. the oxidation of cysteine to cystine, further supporting preservation of intracellular thiol redox balance (Fig. 6E). To assess the impact of itaconate supplementation on ferroptotic cell death, we next examined the effects of 4-OI in obese mice, in which ferroptosis contributes to pancreatitis severity. 4-OI treatment markedly attenuated pancreatic injury, as evidenced by reduced inflammatory infiltration and decreased necrotic areas compared with untreated obese controls (Fig. 6F and Supplementary Fig. S7C). These effects were remarkably accompanied by much lower plasma amylase levels and reduced pancreatic expression of Tnf-α and Il-6 (Fig. 6G, H, I). In contrast to the accumulation of homocysteine observed in lean mice with pancreatitis, obese mice exhibited reduced pancreatic homocysteine levels together with marked glutathione depletion (Fig. 6J). Administration of 4-OI restored the metabolic flux through the trans-sulfuration pathway and partially prevented glutathione depletion, (Fig. 6J). In addition, 4-OI treatment limited the oxidation of cysteine to cysteine and hence prevented disulphide stress, thus suggesting preserved intracellular thiol redox homeostasis (Fig. 6J). Consistent with this restored redox balance, 4-OI administration also recovered protein levels of the NRF2 targets and ferroptosis-protective factors xCT and GPX4, which were suppressed in obese mice with pancreatitis (Fig. 6K and Supplementary Fig. S7D and E). Discussion Neutrophil infiltration into the pancreatic parenchyma is a well-known feature of acute pancreatitis that plays a central role in amplifying pancreatic inflammation, tissue injury, and systemic complications [31]. Beyond their established functions in cytokine release, oxidative burst, and protease-mediated damage, emerging evidence suggests that metabolic reprogramming of neutrophils may critically drive inflammatory outcomes [[50], [51], [52], [53]]. In this study, we identify ferroptosis as a pathophysiological feature of acute pancreatitis in obese mice and demonstrate that it is due to the suppression of ACOD1 activity in infiltrating pancreatic neutrophils. Although ACOD1 was initially described in macrophages [54], it is also robustly expressed in neutrophils [[55], [56], [57]]. In our study, we detected high ACOD1 expression in both infiltrating neutrophils and macrophages within the inflamed pancreas of lean mice. Accordingly, depletion of infiltrating neutrophils and macrophages resulted in a near-complete loss of pancreatic ACOD1 expression, demonstrating that myeloid cells represent the primary source of itaconate during acute pancreatitis. In contrast, expression of the SLC13A3 transporter, recently implicated in itaconate uptake [[44], [45], [46]], was preserved following myeloid depletion, indicating that parenchymal cells exhibit the capacity to import extracellular itaconate. Hence, we propose that itaconate is produced locally by infiltrating myeloid cells and subsequently taken up by pancreatic acinar cells, which are highly susceptible to ferroptotic stress. In contrast to lean animals, obese mice exhibited suppressed ACOD1 expression within neutrophils in the inflamed pancreas, leading to defective NRF2 activation, sustained downregulation of the xCT–GPX4 antioxidant axis, and enhanced lipid peroxidation in the pancreatic tissue. Therefore, obesity suppresses ACOD1 in infiltrating immune cells thus reducing paracrine itaconate availability, thereby weakening NRF2-dependent antioxidant defences in acinar cells and increasing their vulnerability to ferroptosis. Ferroptosis is increasingly considered a key contributor to pancreatic injury in acute pancreatitis, particularly in severe forms of the disease as well as in certain metabolic contexts such as hypertriglyceridemia or iron overload [[58], [59], [60], [61]]. Experimental evidence has shown that suppression of key ferroptosis regulators, including SLC7A11/xCT and GPX4, exacerbates lipid peroxidation and tissue damage, while pharmacological inhibition of ferroptosis ameliorates pancreatic injury [[61], [62], [63]]. These observations support that the disruption of the cystine–GSH–GPX4 antioxidant axis is a central determinant of disease severity. We further complement these findings by identifying obesity as a key metabolic condition that predisposes the pancreas to ferroptotic cell death, also demonstrating that ferroptosis is also a prominent feature of obesity-associated acute pancreatitis. Previous studies on acute pancreatitis have focused primarily on acinar cell–intrinsic stress responses or systemic metabolic alterations that contribute to ferroptosis susceptibility [[62], [63], [64]]. Nevertheless, the upstream immune-metabolic mechanisms predisposing to ferroptosis in obesity remained poorly defined yet. ACOD1 activity in neutrophils has previously been shown to regulate neutrophil inflammatory responses, promoting reverse migration and facilitating the resolution of local inflammation [55,57]. Here, we identify neutrophil ACOD1-dependent itaconate metabolism as a key immune-metabolic regulator of ferroptosis susceptibility in the pancreas whose protection is lost in obesity. Our analyses focused primarily on neutrophils due to their prominent contribution to acute pancreatitis pathophysiology [31]. However, ACOD1 expression was also detected in pancreatic macrophages and therefore, a cooperative role for macrophage-derived itaconate in maintaining redox balance and protecting against ferroptosis cannot be excluded. One limitation of the present study is that direct measurement of pancreatic itaconate levels was not performed, and therefore the contribution of neutrophil-derived itaconate was inferred from ACOD1 expression and functional analyses. Although our results strongly support a major role for neutrophil ACOD1 activity in acute pancreatitis, the use of neutrophil-specific ACOD1 knockout mice would further strengthen the causal relationship between neutrophil-derived itaconate and ferroptosis protection during acute pancreatitis. It is noteworthy that proteomic profiling of pancreatic neutrophils from obese mice revealed reduced expression of enzymes involved in the tricarboxylic acid (TCA) cycle and pyruvate metabolism, occurring in parallel with decreased ACOD1 abundance. This reduction in TCA- and pyruvate-associated enzymes is intriguing and may reflect a reprogramming of mitochondrial metabolism and signalling in neutrophils. In this context, diminished TCA cycle activity is compatible with limited substrate availability for itaconate synthesis, further supporting the concept that obesity-associated metabolic remodelling of infiltrating neutrophils restricts itaconate production, weakens NRF2-driven antioxidant responses, and promotes ferroptosis during acute pancreatic inflammation. Functionally, we also demonstrate that restoration of itaconate signalling with exogenous 4-octyl itaconate largely attenuates pancreatic injury, reactivates the NRF2–xCT–GPX4 axis, and suppresses ferroptosis in obese mice with pancreatitis. Mechanistically, ACOD1-derived itaconate is known to activate NRF2 through electrophilic modification of KEAP1, thereby stabilizing NRF2 and promoting antioxidant gene expression [17,18]. Accordingly, we found that 4-OI restored NRF2 signalling and increased GPX4 and xCT levels in the pancreas of obese mice with pancreatitis. Interestingly, we observed that 4-OI modulates the trans-sulphuration pathway in acute pancreatitis. We previously reported the blockade of this metabolic pathway in lean mice with pancreatitis due to nitration-mediated inactivation of cystathionine β-synthase (CBS) [32]. 4-OI reversed this blockade, preventing homocysteine accumulation and favouring glutathione synthesis. In obese mice, itaconate supplementation partially prevented glutathione depletion also in these mice, which is consistent with the reduced susceptibility to ferroptosis. In conclusion, our findings show that the immune–metabolic axis driven by ACOD1-dependent itaconate production from infiltrating neutrophils critically regulates ferroptosis susceptibility during acute pancreatitis. Obesity disrupts this pathway by suppressing ACOD1 expression and restraining NRF2 activation, thereby promoting lipid peroxidation and pancreatic injury. Restoration of itaconate signalling through exogenous 4-octyl itaconate re-establishes the antioxidant defence, upregulates the xCT–GPX4 axis, and mitigates ferroptotic pancreatic damage. Taken together, these results identify neutrophil-derived itaconate as a key modulator of ferroptosis susceptibility and reveal immune cell metabolism as a critical determinant of obesity-associated disease severity in acute pancreatitis. CRediT authorship contribution statement Néstor Jiménez-Cañete: Data curation, Formal analysis. Alicia Aliena-Valero: Data curation, Formal analysis. Caroline A. Bressan: Data curation, Formal analysis. Salvador Pérez: Data curation, Formal analysis. Isabel Torres-Cuevas: Data curation, Formal analysis. Laura Benkenstein: Data curation, Formal analysis. Sandra Pinto: Data curation, Formal analysis. Antonio Cuadrado: Formal analysis, Writing – review & editing. Juan Sastre: Conceptualization, Funding acquisition, Supervision, Writing – original draft, Writing – review & editing. Sergio Rius-Pérez: Conceptualization, Funding acquisition, Supervision, Writing – original draft, Writing – review & editing. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.