Neutrophil Irgm1 ameliorates sepsis-induced myocardial dysfunction by promoting Alox15 degradation

TL;DR

This study shows that a protein called Irgm1 in neutrophils helps reduce heart damage caused by sepsis by breaking down another protein called Alox15.

Contribution

The novel finding is that neutrophil Irgm1 prevents sepsis-induced heart dysfunction by promoting Alox15 degradation and inhibiting ferroptosis.

Findings

Irgm1 expression in neutrophils is upregulated in sepsis-induced myocardial dysfunction and correlates inversely with disease severity.

Neutrophil-specific Irgm1 deficiency in mice worsens cardiac dysfunction and inflammation during sepsis.

Irgm1 interacts with RNF213 to degrade Alox15, reducing 15-HETE production and neutrophil ferroptosis, which alleviates SIMD.

Abstract

Sepsis-induced myocardial dysfunction (SIMD), a severe sepsis complication, is characterized by immune dysregulation, with neutrophils playing a central role. While the immunity-related GTPase family M protein (IRGM) in humans and its murine ortholog Irgm1 are key immune regulators, the precise contribution of neutrophil Irgm1 to SIMD pathogenesis remains unclear. This study aims to explore the involvement of neutrophil Irgm1 in SIMD and uncover its mechanisms. This research found that IRGM expression was upregulated in peripheral blood neutrophils from patients with SIMD and inversely correlated with disease severity. In mice, neutrophil-specific Irgm1 deficiency worsened CLP-induced cardiac dysfunction and myocardial inflammation. Mechanistically, Irgm1 interacted with the E3 ubiquitin ligase RING finger protein 213 (RNF213) to facilitate 15-lipoxygenase (Alox15) ubiquitination and degradation, thereby inhibiting neutrophil ferroptosis and suppressing the production of 15-HETE, which alleviates SIMD. In patients with SIMD, the expression levels of Alox15 and the concentrations of 15-HETE were positively correlated with disease severity. Notably, intraperitoneal administration of Alox15-targeting drug PD146176 significantly improved cardiac function in SIMD mice. Collectively, this study highlights the pivotal role of the Irgm1 in attenuating SIMD by restraining neutrophil ferroptosis and 15-HETE production. Irgm1 may serve as a promising prognostic biomarker and a valuable therapeutic target for SIMD.

Image 1

Figures

Full text

Introduction

1

Sepsis, a life-threatening condition characterized by organ dysfunction due to a dysregulated host response to infection, remains one of the leading causes of mortality in intensive care units [1]. Sepsis-induced myocardial dysfunction (SIMD), also known as sepsis-induced cardiomyopathy (SICM) [2], is present in 40% to 70% of cases, as reported by epidemiological studies [3,4]. Myocardial dysfunction in sepsis increases mortality by approximately 40% compared to cases without cardiac dysfunction [5,6]. Current clinical management of SIMD remains primarily supportive and non-specific, aimed at stabilizing hemodynamics and ensuring tissue perfusion rather than directly targeting the specific pathophysiological pathways underlying myocardial injury [2]. Despite advancements in intensive care, clear diagnostic criteria and effective interventions for SIMD remain lacking [2,7]. Moreover, the pathogenesis of SIMD remains incompletely understood [7]. Therefore, understanding the progression and mechanisms of SIMD is critical for developing effective diagnostic and treatment strategies.

As the first line of defense against infection and a key regulator of both innate and adaptive immune responses, neutrophils play a pivotal role in sepsis pathogenesis [8,9]. During sepsis, there is a marked and rapid increase in peripheral blood neutrophil counts [10]. These neutrophils engage in phagocytosis and release inflammatory cytokines, reactive oxygen species, and other antimicrobial substances [10]. In addition to their antimicrobial role, specific neutrophil functional phenotypes correlate with the severity of sepsis [11]. Dysfunctional neutrophils have been implicated in sepsis-related organ injury, and aberrantly activated neutrophils can exacerbate sepsis-induced damage [12,13]. However, the role of neutrophils in SIMD and their regulatory mechanisms remain poorly understood.

The immunity-related GTPase family M protein (IRGM) and its murine ortholog Irgm1, members of the immunity-related GTPase family [14,15], are crucial in regulating immune responses, host defense against infections, immune inflammation, and tumorigenesis [15,16]. While studies have demonstrated the important role of IRGM/Irgm1 in atherosclerosis [17], its involvement in cardiac injury remains unreported. Currently, research on IRGM/Irgm1 in sepsis has primarily focused on sepsis-associated acute lung injury and encephalopathy [18,19], with its role in SIMD has not yet been investigated. Furthermore, most of the current research on Irgm1 in immune cells has focused on macrophages [17,20]. Recently, it has been discovered that Irgm1 also regulates T cell metabolism and function [21]; however, its role in neutrophils, particularly in neutrophils associated with SIMD, remains unexplored.

This study demonstrated that increased neutrophil IRGM expression in patients with SIMD correlates with improved prognosis. Irgm1 promoted the ubiquitination and degradation of Alox15 by directly interacting with RNF213 to alleviate neutrophil ferroptosis and reduce 15-HETE generation, which, in turn, exerted a protective regulatory role in SIMD. Together, these findings provide valuable insights into potential therapeutic strategies for SIMD.

Methods and materials

2

All animal experimental procedures were conducted by the Guide for the Care and Use of Laboratory Animals outlined by the National Institute of Health (NIH) and were approved by the Research Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (SYDW2025-066 and SYDW2025-117). Adult C57BL/6J and transgenic mice of comparable ages were used (According to previous studies in the cardiovascular field, since the interference of protective factors such as estrogen secretion in adult female mice might obscure the effects of the target gene; therefore, this study exclusively used adult male mice [22,23].), and all mice were housed in temperature-controlled rooms and given unrestricted access to food and water. During the data analysis process, researchers maintained blinding regarding the genotypes of mice and experimental grouping information. At 24 h post-CLP, animals were anesthetized with an overdose of inhaled 5% isoflurane/95% oxygen and euthanized by cervical dislocation after tissue collection.

Study population

2.1

Sepsis patients (n = 36) were recruited from the Second Affiliated Hospital of Harbin Medical University (Harbin, China). Patients were diagnosed with sepsis according to the guidelines from The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) [24]. The inclusion criteria for sepsis patients are as follows: 1) Clear indications for infection are found in patients; 2) Patients display secondary organ dysfunction or acute exacerbation of primary organ dysfunction; 3) Organ Failure Assessment (SOFA) score is ≥ 2; 4) Patients do not receive insulin treatment on the day of ICU admission (to exclude the influence of insulin on the metabolic indicators). Patients with sepsis and left ventricular ejection fraction (LVEF) < 50%, cardiac troponin I > 0.02 ng/mL [4,25] were selected for the current study. Healthy controls (n = 10) were recruited from individuals undergoing health checkups at the Second Affiliated Hospital of Harbin Medical University. All patients were included in the study within 18 h of ICU admission and gave written informed consent. This study was approved by the Research Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (KY2025-128) and performed according to the criteria set by the Declaration of Helsinki. All participants provided informed consent. Detailed clinical characteristics and laboratory information are shown in Table S1.

Animals

2.2

C57BL/6J (B6) mice were procured from the Animal Supply Center of Harbin Medical University. The S100a8-Cre (C57BL/6J background) and Irgm1^flox/flox^ (C57BL/6J background) mice were sourced from Cyagen Biosciences Inc., Hefei, China. A crossbreeding strategy was employed to generate neutrophil-specific Irgm1 gene knockout mice, designated as Irgm1-cKO. The genotypic ablation efficiency was ascertained through Western blot analysis and reverse transcription polymerase chain reaction (RT-PCR).

Cecal ligation and puncture model

2.3

The mouse model of sepsis-induced myocardial dysfunction (SIMD) was established by the cecal ligation and puncture (CLP) as previously reported [26,27]. Briefly, male mice aged 9-12 weeks were anesthetized by inhalation of isoflurane (-induction at 5% isoflurane/95% oxygen, maintenance at 1.5% isoflurane/98.5% oxygen). Then, mouse abdomens were disinfected, a small incision was made about 1 cm from the midabdominal line, and the cecum was bluntly separated and freed. The cecum was ligated about 1 cm from the tip of the cecum and then pierced 2 times with a 22-G needle. The cecum was slightly compressed so that it releases a small amount of fecal matter from the hole, then the cecum is returned to the peritoneal cavity and closed behind the abdomen. Then the abdominal wall and skin are sutured separately. For fluid resuscitation, pre-warmed normal saline (37 °C, 50 mL/kg) was administered via intraperitoneal injection. Postoperative analgesia was provided according to established protocols from previous CLP studies [28,29], with buprenorphine administered subcutaneously (0.05 mg/kg body weight) 30 min prior to surgery and repeated every 6–8 h thereafter. Mice were then placed on a thermostatically controlled heating pad until full recovery from anesthesia. Sham-operated mice underwent identical surgical procedures, including laparotomy, without cecal ligation and puncture. Antibiotics were withheld postoperatively to allow assessment of the systemic inflammatory response. During the early postoperative period, general condition, locomotor activity, and body temperature were monitored at regular intervals (every 4 h). Mice exhibiting signs of imminent death or complete immobility were promptly euthanized to prevent prolonged suffering [28].

For pharmacological interventions, PD146176 (Alox15 inhibitor; 10 mg/kg, HY-103157, MedChemExpress) or 15-HETE (Arachidonic acid metabolites; 0.5 mg/kg, sc-200944, Santa Cruz Biotechnology) was administered intraperitoneally 15 min before CLP. Compounds were dissolved in a vehicle solution composed of DMSO:PEG300:Tween80:PBS (4:30:5:61) [30]. Ferrostatin-1 (Fer-1, HY-100579, MedChemExpress), a ferroptosis inhibitor, was first dissolved in dimethyl sulfoxide (DMSO) and then diluted with saline to a final DMSO:saline ratio of 1:9. Mice received an intraperitoneal injection of Fer-1 (1 mg/kg) 15 min before cecal ligation and puncture (CLP), while control mice were administered the vehicle [31]. Animals were randomly assigned to experimental groups, and all subsequent procedures and analyses were performed in a blinded fashion.

Echocardiography analysis

2.4

Cardiac function was measured by echocardiography using a VisualSonics Vevo 3100LT ultrasound system (VisualSonics, Inc.) and a 30 MHz MX400 probe transducer at 24 h post-CLP. Mice were maintained under anesthesia with 1.5% isoflurane/98.5% oxygen to keep their heart rate (HR) at 400-500 beats per minute. Cardiac imaging was performed in parasternal long-axis B-mode, with M-mode measurements taken perpendicular to both the interventricular septum and left ventricular posterior wall at the papillary muscle level.

Histological analysis

2.5

As previously described [5,32]-micron thick heart sections were dewaxed, rehydrated, and washed, and hematoxylin-eosin (H&E) staining was used to assess organ damage and inflammatory cell infiltration. In short, H&E staining is performed according to the manufacturer's instructions(G1120, Stain Kit Solarbio). For each heart sample, five paraffin sections at different levels were prepared, and four fields of view were selected from each section for statistical analysis. Images were captured with the microscope (Olympus, U-RFL-T) at 20 × magnification. The cardiac injury score was referred to the following myocardial tissue structural parameters: myocardial fiber arrangement; nuclear swelling and pyknosis; inflammatory cell infiltration; edema; or interstitial congestion [33,34]. All tissue processing, staining, slide scanning, and histopathological scoring were performed by investigators blinded to the group assignment of the samples.

Immunohistochemical staining was performed according to the manufacturer's instructions (PV-600, ZSGB-BIO). Subsequently, sections were rinsed with preheated citrate buffer and heated in a pressure cooker for 7 min. Endogenous peroxidase activity was quenched by incubating sections with 3% hydrogen peroxide in the dark for 30 min. After PBS washes, non-specific binding was blocked with serum for 30 min, followed by overnight incubation with primary antibody at 4 °C. Sections were then incubated with the corresponding secondary antibody for 1 h at room temperature, developed with DAB (ZLI9018, ZSGB-BIO), and counterstained with hematoxylin. Stained sections were imaged using a slide scanner (Olympus, U-RFL-T) at 20 × magnification.

TUNEL assay

2.6

The TUNEL procedure was performed as previously described [35]. In accordance with the instructions provided in the kit manual (40307ES20, YEASEN), the TUNEL assay solution was applied, followed by incubation of the mixture in the dark at 37 °C for 60 min. Subsequently, the slices were incubated in 5% BSA and 0.5% saponin (Beyotime, P0095) for sealing, and then incubated overnight with anti-α-actin antibody. Each section was incubated with its respective fluorescently labeled secondary antibody at room temperature for 1 h, followed by counterstaining of the nuclei with DAPI (Beyotime, Cat. No. C1005). Stained tissue sections were viewed using a Laser Scanning Confocal Microscopy (Zeiss). The number of TUNEL-positive nuclei in each cardiac sample was quantified in 10 high-power fields (HPFs) across three different sections. Data are expressed as the mean number of TUNEL-positive nuclei per HPF. All tissue section preparation, image acquisition, and subsequent quantitative analysis were performed in a "blinded" manner.

Adoptively transferred neutrophils

2.7

The method of neutrophil adoptive transfer was conducted as previously reported [36]. Specifically, the in vitro purified Irgm1-cKO neutrophils were incubated with vehicle or 10 μM PD146176 at 37 °C for 4 h, and they were washed twice with PBS. Induction of sepsis was conducted in WT mice by CLP. Based on previous studies [[37], [38], [39]],we administered 5 × 10^6^ neutrophils to recipient mice via intravenous injections.

Isolation of mouse neutrophils from bone marrow and blood

2.8

Mouse neutrophils were isolated as previously described [40]. Briefly, the isolation of mouse neutrophils from bone marrow and blood was performed by using a mouse neutrophil separation medium (TBD, Tianjin) according to the manufacturer's protocol. Neutrophils isolated from bone marrow were resuspended in RPMI 1640 (C11875500CP, Thermo Fisher) and then stimulated with lipopolysaccharide (LPS) (BS904, Biosharp) at 5 ng/ml for 12 h prior to subsequent use. For neutrophil purification, we employed the magnetic bead sorting method. In summary, following the protocol of the Mouse Neutrophil Cell Isolation Kit (BeaverBeads, 7097), neutrophils with a purity exceeding 95% can be obtained.

Isolation of human neutrophils from blood

2.9

Human neutrophils were isolated as previously described [40]. Briefly, the isolation of human blood neutrophils was performed by using a human neutrophil separation medium (TBD, Tianjin) according to the manufacturer's protocol. Briefly, a pipette was used to carefully absorb the blood sample, and the sample was added to the liquid surface of the separation solution and centrifuged for 20-30 min at 550-650 g. A pipette was used to carefully absorb the neutrophil layer in the separation solution. The plasma was prepared by centrifugation at 1000g for 15 min at room temperature. These samples were collected and stored in the Biobank of the Key Laboratory of Myocardial Ischemia, Ministry of Education.

Quantitative real-time polymerase chain reaction

2.10

Quantitative real-time PCR (qRT-PCR) was performed as previously described [32]. For tissue samples, specimens were precisely dissected into small fragments using sterile forceps and rapidly ground into fine powder under liquid nitrogen. The powdered tissue was then transferred into pre-cooled Eppendorf tubes for subsequent processing. For cell samples, cells were harvested by centrifugation at 1500 rpm for 5 min. Total RNA was extracted from tissue homogenates or cell pellets using an RNA extraction kit (SM132-01, SEVEV, China) according to the manufacturer's instructions. RNA concentration and purity were assessed using a Shimadzu BioSpec-nano spectrophotometer (Shimadzu Corporation, Japan). Subsequently, 1 μg of total RNA was reverse transcribed into cDNA using the ReverTra Ace™ qPCR RT Kit (FSQ-101, TOYOBO, Japan). Real-time quantitative PCR was conducted on a CFX96 real-time PCR system (Bio-Rad, Hercules, CA, USA). Relative mRNA expression levels were calculated using the 2^−ΔΔCt^ method, with β-actin or GAPDH serving as the internal reference gene. All primer pairs were synthesized by Invitrogen (Thermo Fisher Scientific) and are listed in Supplementary data 2.

ELISA assay

2.11

The ELISA procedure was performed as previously described [27]. Briefly, blood was collected via retro-orbital bleeding prior to euthanasia and transferred into EDTA-treated collection tubes. Samples were then centrifuged at 2500 rpm for 15 min at room temperature. The resulting supernatant (plasma) was carefully separated and stored in tubes at −80 °C for subsequent analysis. Cytokine and chemokine levels were measured in mice and human plasma by using ELISA kits according to the manufacturer's instructions (R&D Systems and Thermo Fisher Scientific).

Reactive Oxygen Species (ROS) assay

2.12

As previously mentioned [41], the determination of ROS levels was conducted using a ROS detection kit (S0035 S; Beyotime). Neutrophils were transferred to a light-protected 96-well plate and treated with CM-H2DCFDA (5 μM) before being incubated in a 37 °C cell culture incubator for 30 min. Unincorporated CM-H2DCFDA was removed by washing with PBS. Subsequently, the absorbance of each well was measured using a Varioskan LUX Multimode Microplate Reader (Thermo Fisher Scientific, Waltham, MA, USA) with an excitation wavelength of 495 nm and an emission wavelength of 530 nm.

Ferrous ion (Fe2+) assay

2.13

The determination of Fe^2+^ levels was carried out using the Cell Ferrous Ion Red Fluorescence Detection Kit (S1070 S; Beyotime). Neutrophils were transferred to light-protected 96-well plates and treated with RhoNox-6 staining solution (100 μl per well). The plates were then incubated in a 37 °C cell culture incubator for 30 min. Subsequently, the absorbance of each well was measured using the Varioskan LUX Multimode Microplate Reader (Thermo Fisher Scientific, Waltham, MA, USA) with an excitation wavelength of 545 nm and an emission wavelength of 585 nm.

Transmission Electron Microscopy (TEM)

2.14

The mouse heart underwent fixation overnight at 4 °C in a 2.5% glutaraldehyde solution, followed by postfixation in 1% osmium tetroxide, acetone dehydration, and embedding in epoxy resin. Semithin sections (1 μm) were toluidine blue-stained for microvascular localization. Ultrathin sections (70-90 nm) were placed on Formvar-coated 75-mesh copper grids (1595 E, Merck), post-stained with uranyl acetate and lead citrate, and analyzed via transmission electron microscopy.

Proximity Ligation Assay (PLA)

2.15

The PLA was performed following the NaveniFlex PLA protocol. After treatment, the neutrophil culture on coverslips was washed with PBS three times, then fixed with 4% paraformaldehyde (PFA) for 15 min and permeabilized using 0.1% Triton X-100 (Beyotime) for 10 min. After washing with filtered PBS, add the blocking solution to each sample for 1 h at 37 °C. Incubate the sample with the primary antibody in NaveniFlex Diluent overnight at 4 °C. Subsequently, the samples were covered with a sufficient volume of Navenibody working solution and incubated in a preheated humidity chamber at 37 °C for 60 min. Add enzyme 1 to configure reaction system 1, cover the coverslips with reaction system 1, and incubate in a preheated humidity chamber for 30 min at 37 °C. Subsequently, configure reaction system 2 (take care to protect from light). Cover the coverslips completely with reaction system 2 and incubate in a preheated humidity chamber for 90 min at 37 °C. After washing, coverslips were counterstained with DAPI (blue) and mounted onto glass slides. Images were acquired using a Laser Scanning Confocal Microscopy (Zeiss).

Proteomics and analysis

2.16

Total protein was extracted from WT and Irgm1-cKO neutrophils from the Peripheral Blood of WT and Irgm1-cKO mice after CLP (n = 3 for each group). Subsequently, add iodoacetamide and perform the reaction in the dark at room temperature for 45 min. Precipitate and wash the proteins with acetone, then add trypsin. After overnight enzyme digestion at 37 °C, acidify the peptides and remove salts using a C18 SPE column. Evaporate the eluted peptides using a vacuum concentrator. Perform conventional reverse-phase gradient separation under alkaline conditions using high-performance liquid chromatography. Dissolve each pre-separated fraction in liquid A and centrifuge at 20,000g for 5 min. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was employed to analyze each component, with the raw data obtained from the instrument being imported into the Proteome Discoverer (v2.3) search software. The secondary mass spectrometry data were then searched against the Mus_musculus database. Subsequently, Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were utilized to classify the significantly differentially expressed proteins in terms of their biological functions.

Cell transfection

2.17

Plasmids (His-IRGM, Flag-RNF213, HA-Alox15, and Myc-Ub) were constructed by Miaolingbio (Wuhan, China). Plasmids (oe-IRGM) and si-RNA (si-Alox15, si-RNF213, and si-IRGM) were constructed by generalbiol (An Hui, China). HL-60 cell line was purchased from the National Collection of Authenticated Cell Cultures (Shanghai, China). HEK293T cells were kindly provided by the oricellbio (Guangzhou, China). The si-Control, si-IRGM, si-RNF213, oe-IRGM, or oe-RNF213 (MiaoLing, Wuhan, China) was transfected into HL-60 cells using Lipofectamine 3000 reagent (L3000015, Invitrogen) according to the instruction manual. The plasmids (His-IRGM, Flag-RNF213, HA-Alox15, and Myc-Ub) and si-RNA (IRGM and RNF213) were transfected into HEK-293T cells using Lipofectamine 3000 reagent (L3000015, Invitrogen) according to the instruction manual (Given that ferroptosis was not significantly induced in IRGM-overexpressing cells following LPS stimulation, whereas it was markedly observed in IRGM-knockout cells under the same conditions, we employed the ferroptosis inducer RSL-3 to trigger ferroptosis in IRGM-overexpressing cells. Conversely, IRGM-knockout cells continued to be stimulated with LPS).

IP-LC-MS/MS analysis

2.18

Neutrophils isolated from the bone marrow of WT mice were treated with LPS or vehicle, after which the neutrophils were lysed using immunoprecipitation lysis buffer. The cell lysate was then subjected to immunoprecipitation with Irgm1 antibody (1:100) or IgG (1 μg/ml). Candidate proteins interacting with Irgm1 were identified through mass spectrometry analysis (conducted by Lu Ming Biotech (Shanghai), China). Ultimately, substrate proteins capable of binding to Irgm1 were identified by screening based on the scores and masses of the detected proteins, in conjunction with the control IgG group.

CO-IP

2.19

Total protein was obtained via IP lysis buffer that included protease inhibitors (Roche Holding AG, Basel, Switzerland). The extracts were incubated with anti-Irgm1, anti-RNF213, or anti-IgG for 24 h at 4 °C. Following the addition of Protein A/G agarose, incubation continued at 4 °C for 3 h. Subsequently, centrifugation was performed at 12,000 rpm for 5 min to collect the precipitate, which was then washed and resuspended in 40 μl SDS lysis buffer. After boiling for 5 min, the final analysis of the precipitate was conducted via immunoblotting using the specified antibodies.

Molecular docking

2.20

The protein models utilized for docking are Irgm1, RNF213, and ALOX15. Protein-protein molecular docking was performed using the ZDOCK SERVER (https://zdock.wenglab.org/). Before docking, protein structures were prepared with PyMol 2.4. The docking poses were ranked based on their docking scores, with the top 10 highest-scoring conformations retained for further analysis. The highest-scoring model was identified as the optimal docking complex. Finally, PyMol 2.4 was used to visualize the results, allowing for direct observation of the ligand-receptor binding interactions.

Western blot

2.21

The Western blot procedure was performed as previously described [32]. Samples were lysed in RIPA buffer containing protease inhibitors (Thermo Fisher Scientific, Cat. #36978), and protein concentration was determined using the BCA kit (Thermo Fisher Scientific, Cat. #23225). Extracted proteins were mixed with 1X loading buffer and boiled at 95 °C for 10 min 20 μg protein per lane were resolved by SDS-PAGE (gel percentage chosen according to target molecular weight) and transferred to PVDF membranes (GE Healthcare Life Sciences, Cat. #10600023). After incubating the membranes at room temperature for 1 h, the primary antibody was added, and incubation was carried out on a shaker at 4 °C. Following washes, membranes were incubated with the secondary antibody at room temperature. Detection was performed using an enhanced chemiluminescence (ECL) kit (Bio-Rad, Cat. #1705061), and images were acquired on a ChemiDoc imaging system (Bio-Rad).

Statistical analysis

2.22

Statistical significance was assessed in GraphPad Prism (Version 9.5) following the testing procedures outlined in the figure legends. The number of replicates for each sample is provided in the respective legends. Results are presented in the form of mean ± SD. Outliers were identified using the Grubb test or the ROUT test (Q = 1), followed by normality testing. To explore the associations between clinical indicators, a simple linear regression analysis was conducted. Spearman or Pearson correlation analysis methods were employed to calculate the correlation coefficient r and the significance level p-value. The unpaired Student's t-test was employed to compare variables that follow a normal distribution across two groups, whereas the Mann-Whitney U test was utilized for the comparison of variables in two groups that are not normally distributed. Additionally, one-way ANOVA with Bonferroni post-hoc analysis was applied for comparisons among multiple groups (>2 groups). The Log-rank (Mantel-Cox) test was used to compare the survival curves. Statistical significance was assigned to results where p < 0.05.

Result

3

IRGM/Irgm1 is upregulated in neutrophils and is associated with the prognosis of patients with SIMD

3.1

Data from the Human Protein Atlas (www.proteinatlas.org) revealed that IRGM exhibits the highest RNA expression in oocytes, followed by granulocytes, with relatively elevated levels in T cells, and lower expression in macrophages, fibroblasts, smooth muscle cells, and cardiomyocytes (Fig. 1A). To this end, we investigated the expression patterns of IRGM in neutrophils of patients with SIMD. Validation experiments confirmed increased IRGM mRNA (Fig. 1B) and protein levels (Fig. 1C and D) in neutrophils isolated from the peripheral blood of patients with SIMD. No significant differences in age, gender, or comorbidities were observed between the sepsis and healthy control groups (Table S1).Fig. 1Upregulation of IRGM was negatively correlated with sepsis severity. (A) Expression levels of IRGM RNA across various human cell types (top 20 expressions, sourced from The Human Protein Atlas). In these cells, the highest expression of IRGM is found in oocytes, followed by granulocytes, with relatively high expression in T cells. (B) Relative IRGM mRNA expression levels of patients with Sepsis-Induced Myocardial Dysfunction (SIMD) or healthy control (Ctr) individuals (n = 5 per group). (C and D) Western-blot analysis (C) and quantification (D) of IRGM expression in neutrophils from patients with sepsis or healthy controls (n = 6 per group). (E) Correlation between neutrophil IRGM mRNA expression and Sequential Organ Failure Assessment (SOFA) scores in patients with SIMD (n = 36 per group). (F-J) Correlations of neutrophil IRGM mRNA expression with PCT (F), IL-6 (G), cTnI (H), LDH (I), and BUN (J) levels were analyzed in patients with SIMD (n = 36 per group). Data are represented as the mean ± SD. Unpaired Student t-test for B through D. Spearman correlation analysis for E through J. ∗∗∗∗p < 0.0001.Fig. 1

Although human IRGM and murine Irgm1 exhibit distinct biochemical properties, their overlapping functions in innate immunity and autophagy are well documented. Subsequent analysis of single-cell transcriptomic data from immune cells in septic mice (GSE249975) [42] revealed significant upregulation of Irgm1 in peritoneal neutrophils from CLP-induced mice compared to sham-operated controls (Fig. S1A and B). These observations were corroborated in vivo using a cecal ligation and puncture (CLP) mouse model of sepsis, which demonstrated elevated Irgm1 mRNA and protein expression in peripheral blood neutrophils (Fig. S2A–C). At 24 h post-CLP, Irgm1-positive cells were detected in myocardial tissue, whereas no expression was observed in sham-operated hearts. These Irgm1-expressing cells exhibited spherical or ovoid morphologies and were distributed within myocardial interstitial spaces (Fig. S2D), suggesting a critical role for Irgm1 in post-sepsis pathology.

To assess the diagnostic relevance of IRGM expression in sepsis, its correlation with laboratory markers of disease severity in patients with SIMD was examined. A negative correlation was found between IRGM expression in peripheral blood neutrophils and the Sequential Organ Failure Assessment (SOFA) score (Fig. 1E), indicating its potential involvement in SIMD pathogenesis. Additionally, IRGM mRNA levels in neutrophils negatively correlated with inflammatory markers PCT and IL-6, as well as organ injury indicators including cTnI, LDH, and BUN (Fig. 1F–J). These findings indicate that reduced IRGM expression is strong association with disease severity in SIMD, highlighting its potential as a prognostic biomarker.

Neutrophil-specific Irgm1 deficiency exacerbates CLP-induced myocardial dysfunction

3.2

The SIMD model was successfully established in C57BL/6 mice via CLP, as evidenced by impaired cardiac function (Fig. S3A–E). Myocardial tissue from the CLP group exhibited extensive structural damage, characterized by disorganized myocardial fibers with indistinct contours, varying degrees of myolysis and nuclear fragmentation, as well as interstitial edema, hyperemia, and infiltration of red blood cells and inflammatory cells, primarily monocytes and lymphocytes (Fig. S3F and G). These pathological alterations were accompanied by upregulated cardiac mRNA expression of atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) (Fig. S3H), elevated mRNA levels of pro-inflammatory cytokines TNF-α, IL-1β, and IL-6 (Fig. S3I), and increased serum concentrations of cardiac injury markers cTnI and LDH (Fig. S3J and K).

To investigate the role of neutrophil-specific Irgm1 in SIMD, neutrophil-specific Irgm1 knockout (Irgm1-cKO) mice were generated, with successful knockout confirmed via Genotyping and qPCR (Fig. S4A–B). In neutrophils, Irgm1 expression was markedly reduced but not completely deleted, whereas the expression of Irgm1 was not altered in monocytes from the cKO mice (Fig. S4C). Furthermore, Irgm1-cKO mice exhibited normal development, body weight, and cardiac function (Fig. S4D–I). Additionally, the blood cell counts in Irgm1-cKO mice were comparable to those in WT mice (Fig. S4J–P).

Interestingly, Irgm1 deficiency significantly reduced survival time (Fig. 2A) and exacerbated cardiac dysfunction compared to WT mice post-CLP (Fig. 2B–F). Septic Irgm1-cKO mice displayed markedly higher cardiac mRNA levels of ANP and BNP (Fig. 2G), as well as elevated plasma concentrations of myocardial injury markers cTnI and LDH (Fig. 2H and I). Furthermore, Loss of Irgm1 further resulted in significantly increased cardiac mRNA levels of pro-inflammatory mediators post-CLP (Fig. 2J). Histopathological analysis revealed pronounced myocardial fiber disorganization, interstitial edema, and inflammatory cell infiltration in Irgm1-deficient mice following CLP (Fig. 2K and L). TUNEL staining confirmed a significant increase in apoptotic cells in the hearts of septic Irgm1-cKO mice compared to septic WT mice (Fig. 2M and N).Fig. 2Neutrophil-specific Irgm1 deficiency exacerbates the cardiac dysfunction following sepsis. (A) Survival analysis of WT (n = 22) and Irgm1-cKO (n = 30) mice after CLP performance was monitored for 24h. (B) Representative M-mode echocardiograms were obtained from WT and Irgm1-cKO mice after CLP. Scale bars = 2 mm (vertical) and 0.1 s (horizontal). (C–F) Echocardiographic measurements of left ventricular ejection fraction (EF, C), fractional shortening (FS, D), left ventricular internal diastolic dimension (LVIDD, E), and left ventricular end-diastolic volume (LVEDV, F) in WT and Irgm1-cKO mice after CLP or sham operation (n = 8 per group). (G) Relative ANP and BNP mRNA expression levels of the cardiac in WT and Irgm1-cKO mice after CLP. (H and I) Graphs showing the levels of cTnI (H) and LDH (I) in the serum of WT and Irgm1-cKO mice after CLP (n = 5 per group). (J) Relative mRNA expression levels of cardiac inflammation-related genes in WT and Irgm1-cKO mice after CLP. (K and L) Representative images of H&E staining (K) and quantitative analysis (L) of myocardial injury (n = 5 per group). Scale bars: 50 μm for high magnification, 500 μm for low magnification. (M and N) TUNEL staining (M) and quantification (N) of apoptotic cardiomyocytes in heart tissues from WT and Irgm1-cKO mice after CLP (n = 5 per group). Scale bar = 20 μm. (O) Representative TEM images of heart sections from WT and Irgm1-cKO mice after CLP. Scale bars: 0.5 μm for high magnification, 1 μm for low magnification. Data are represented as the mean ± SD. Log-rank test for A. Unpaired Student t-test for C through N. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.Fig. 2

Given the essential role of mitochondria in SIMD, mitochondrial morphology was examined via electron microscopy. Irgm1 deficiency markedly exacerbated mitochondrial disarrangement, cristae loss, and membrane disruption in the left ventricular wall of SIMD mice (Fig. 2O). Collectively, these results underscore the critical role of neutrophil Irgm1 in mitigating cardiac inflammation and preserving cardiac function during SIMD.

Irgm1 deficiency exacerbates CLP-induced cardiac dysfunction by promoting neutrophil ferroptosis

3.3

To elucidate the mechanisms underlying Irgm1-cKO-mediated SIMD, proteomic analyses were conducted on peripheral blood neutrophils from CLP-induced Irgm1-cKO mice to identify alterations in protein expression profiles. A total of 455 differentially expressed proteins were detected between Irgm1-cKO and WT neutrophils, with 191 proteins upregulated and 264 downregulated (Fig. 3A and B). Gene Ontology (GO) enrichment analysis revealed significant enrichment in the ferroptosis signaling pathway (Fig. 3C). Given the critical role of neutrophil ferroptosis in disease pathogenesis, the impact of Irgm1 deficiency on neutrophil ferroptosis was first examined in vitro. GPX4, a key marker of ferroptosis, is typically downregulated or inhibited during ferroptosis [43], and our findings demonstrated that Irgm1 deletion significantly reduced GPX4 expression (Fig. S5A and B). Elevated ROS levels, as assessed by DCFH-DA staining, and increased intracellular Fe^2+^ levels observed via FerroOrange staining collectively support the hypothesis that Irgm1 deficiency promotes ferroptosis in neutrophils (Fig. S5C and D). In vivo, ferroptosis in neutrophils was assessed in CLP-induced sepsis cardiomyopathy mice. Compared to CLP-induced WT mice, CLP-induced Irgm1-cKO mice exhibited significantly reduced GPX4 expression in peripheral blood neutrophils (Fig. 3D). In line with the in vitro findings, Irgm1 deficiency significantly increased ROS production (Fig. 3E) and elevated intracellular Fe^2+^ levels (Fig. 3F). TEM observations also revealed that neutrophils from the Irgm1-cKO group exhibited mitochondrial atrophy and a reduction in the number of mitochondrial cristae (Fig. 3G). Collectively, these results indicate that Irgm1 absence promotes ferroptosis in neutrophils.Fig. 3Specific deletion of Irgm1 exacerbates SIMD by promoting neutrophil ferroptosis. (A) PCA analysis of the proteomic profiling from WT neutrophils and Irgm1-cKO neutrophils. Each plot represents one biological replicate. (B) Volcano plots analysis of proteins comparing between WT neutrophils and Irgm1-cKO neutrophils. (C) Signaling pathway classification according to GO analysis. (D) Representative Western blot images and quantitative analysis of indicated proteins (n = 6 per group). (E) CM-H2DCFDA staining of the level of ROS in the WT and Irgm1-cKO neutrophils. (F) RhoNox-6 staining of the level of intracellular Fe2^+^ in the WT and Irgm1-cKO neutrophils. (G) Representative TEM images of neutrophils in WT and Irgm1-cKO mice. Scale bars = 1 μm. (H) Schematic illustration of Ferrostatin-1 (Fer-1) treatment in Irgm1-cKO mice. Echo: Echocardiography. (I) Representative M-mode echocardiograms obtained from Irgm1-cKO mice given Vehicle or Fer-1 after CLP. Scale bars = 2 mm (vertical) and 0.1s (horizontal). (J-M) Echocardiographic measurements of left ventricular ejection fraction (EF, J), fractional shortening (FS, K), left ventricular end-diastolic volume (LVEDV, L), and left ventricular internal diastolic dimension (LVIDD, M) from Irgm1-cKO mice given Vehicle or Fer-1 after CLP (n = 8 per group). (N) Relative ANP and BNP mRNA expression levels of the cardiac in Irgm1-cKO mice given Vehicle or Fer-1 after CLP. (O and P) Graphs showing the levels of cTnI (O) and LDH (P) in the serum of Irgm1-cKO mice given Vehicle or Fer-1 after CLP (n = 5 per group). (Q) Relative mRNA expression levels of cardiac inflammation-related genes in Irgm1-cKO mice given Vehicle or Fer-1 after CLP (n = 5 per group). (R) Representative images of H&E staining and quantitative analysis of myocardial injury. Scale bars: 50 μm for high magnification, 500 μm for low magnification. Unpaired Student t-test for D through R. ∗∗p < 0.01; ∗∗∗∗p < 0.0001.Fig. 3

To investigate whether Irgm1 regulates ferroptosis in neutrophils, HL-60 cells (a human neutrophil-like cell line) were transfected with either an IRGM overexpression plasmid or si-IRGM. Ferroptosis was induced in IRGM-overexpressing cells using RSL-3 (a ferroptosis inducer), while si-IRGM-transfected cells were stimulated with LPS. IRGM knockdown led to a further decrease in GPX4 expression, whereas IRGM overexpression resulted in a partial restoration of GPX4 levels (Fig. S6A and B). Furthermore, IRGM knockdown significantly increased the production of ROS (Fig. S6C) and intracellular Fe^2+^ levels (Fig. S6D), while IRGM overexpression partially reversed these changes (Fig. S6E–H). The efficiency of IRGM overexpression and knockdown is shown in Fig. S6I and S6J, respectively. Collectively, these results indicate that IRGM/Irgm1 in neutrophils serves as a regulator of ferroptosis, with elevated Irgm1 expression effectively mitigating ferroptosis.

To further explore the role of ferroptosis in the promotion of SIMD by Irgm1 deficiency, Irgm1-cKO mice were treated with a ferroptosis inhibitor (Fer-1) (Fig. 3H). Fer-1 administration significantly restored GPX4 expression (Fig. S7A and B). Additionally, Fer-1 treatment effectively reduced ROS and Fe^2+^ levels in peripheral blood neutrophils (Fig. S7C and D). These findings suggest that Fer-1 treatment alleviates ferroptosis in Irgm1-cKO neutrophils.

Fer-1 treatment also significantly improved cardiac function (Fig. 3I–M) in Irgm1-cKO mice. Compared to septic WT mice, Fer-1-treated mice exhibited reduced mRNA expression of ANP and BNP (Fig. 3N), along with decreased plasma levels of cTnI and LDH, key biomarkers of myocardial injury (Fig. 3O and P). Additionally, cardiac mRNA levels of pro-inflammatory mediators were significantly downregulated following Fer-1 treatment post-CLP (Fig. 3Q). Histopathological analysis revealed alleviated interstitial edema and significantly reduced infiltration of inflammatory cells in the hearts of Fer-1-treated mice (Fig. 3R). Collectively, these results highlight that neutrophil Irgm1 functions as a regulatory factor in ferroptosis, capable of ameliorating SIMD by inhibiting neutrophil ferroptosis.

Irgm1 regulates the degradation of Alox15 to inhibit neutrophil ferroptosis

3.4

To further elucidate the mechanism by which Irgm1 exerts its protective effects against SIMD, potential downstream targets mediating this role were investigated. Proteomic analysis revealed a significant upregulation of Alox15 among the ferroptosis-associated proteins (Fig. 4A). Among the top ten upregulated proteins ranked by Sum PEP Score, multiple ferroptosis-associated proteins were identified, with Alox15, a key regulatory factor in ferroptosis, drawing particular attention (Fig. 4B). Neutrophil-specific Irgm1 deficiency following sepsis resulted in a marked increase in Alox15 protein levels (Fig. 4C), while Alox15 mRNA levels remained unchanged between WT and Irgm1-cKO groups (Fig. 4D). No significant difference in Alox15 mRNA levels was observed in LPS-stimulated Irgm1-cKO neutrophils compared to the WT group (Fig. 4E), but Alox15 protein levels were substantially elevated (Fig. 4F). Knockdown of IRGM in HL-60 cells treated with LPS led to upregulation of Alox15, while IRGM overexpression significantly suppressed Alox15 expression (Fig. 4G and H). However, Alox15 mRNA levels were not notably altered by the modulation of IRGM expression (Fig. 4I and J).Fig. 4Irgm1 negatively regulates the Alox15 protein level. (A) Heatmap of the differentially expression proteins (DEPs) in the neutrophil from WT or Irgm1-cKO mice with SIMD. (B) Potential downstream differential target proteins identified by proteomic analyses. (C) Representative Western blot images and quantitative analysis of blood neutrophils from WT and Irgm1-cKO mice after CLP (n = 6 per group). (D) The mRNA expression levels of Alox15 in blood neutrophils from WT and Irgm1-cKO mice after CLP (n = 4 per group). (E) The mRNA expression levels of Alox15 in WT and Irgm1-cKO neutrophils treated with LPS (n = 4 per group). (F) Representative Western blot images and quantitative analysis of neutrophils treated with LPS from WT and Irgm1-cKO mice (n = 6 per group). (G) Representative Western blot images and quantitative analysis of Alox15 expression in HL-60 cells with IRGM knockdown (si-IRGM) (n = 6 per group). (H) Representative Western blot images and quantitative analysis of Alox15 expression in HL-60 cells with IRGM overexpression (n = 4 per group). (I and J) The mRNA expression levels of Alox15 in HL-60 cells with IRGM overexpression (I) or IRGM knockdown (si-IRGM) (J). (K) Representative Western blot images and quantitative analysis of Alox15 expression in HL-60 cells treated with MG132 (n = 4 per group). (L and M) The Ubiquitination level of Alox15. Data are represented as the mean ± SD. One-way ANOVA followed by Bonferroni test for D, E, and K. Unpaired Student t-test for C, F through J. ∗p < 0.05; ∗∗∗∗p < 0.0001. ns indicates not significant.Fig. 4

Given that Irgm1 modulates Alox15 protein expression without affecting its transcription, Irgm1 likely influences the posttranslational modification and stability of Alox15 independently of transcriptional regulation. To assess this, Alox15's half-life was measured using cycloheximide. Irgm1 deficiency significantly prolonged Alox15's half-life, a phenomenon reversed by IRGM overexpression (Fig. S8A and B). Treatment with the proteasome inhibitor MG132 partially restored Alox15 protein expression (Fig. 4K), suggesting that Irgm1 regulates Alox15 degradation via the ubiquitin-proteasome pathway. As expected, Alox15 ubiquitination decreased following Irgm1 knockdown (Fig. 4L), whereas IRGM overexpression increased Alox15 ubiquitination (Fig. 4M). These findings indicate that Irgm1 negatively regulates Alox15 protein levels and promotes its ubiquitination.

Next, the dependency of Irgm1-mediated regulation of neutrophil ferroptosis on Alox15 was explored. Irgm1-cKO neutrophils were treated with the Alox15-specific inhibitor PD146176, and the results demonstrated that inhibition of Alox15 significantly elevated GPX4 expression in LPS-treated Irgm1-cKO neutrophils (Fig. S9A and B). Alox15 inhibition also reduced intracellular Fe^2+^ levels (Fig. S9C) and ROS levels (Fig. S9D) in LPS-treated Irgm1-cKO neutrophils. This indicates that Alox15 plays a critical role in ferroptosis induced by Irgm1 deficiency. Further experiments with transfection of either si-IRGM or si-Alox15 into HL-60 cells to assess the effects of Alox15 knockdown on ferroptosis. Alox15 knockdown significantly attenuated the inhibitory effect of IRGM knockdown on GPX4 expression in LPS-stimulated HL-60 cells (Fig. S9E and F). Additionally, Alox15 knockdown reduced the elevated intracellular Fe2^+^ levels (Fig. S9G) and ROS levels (Fig. S9H) induced by IRGM knockdown in neutrophils. Finally, IRGM or Alox15 plasmids transfected into HL-60 cells indicated that Alox15 overexpression attenuated the promoting effect of IRGM on GPX4 expression in LPS-stimulated HL-60 cells (Fig. S9I and J). Alox15 overexpression also increased intracellular Fe2^+^ levels (Fig. S9K) and ROS levels (Fig. S9L) in neutrophils with IRGM overexpression. The efficiencies of Alox15 overexpression and knockdown are shown in Fig. S9M and S9N, respectively. These findings suggest that Irgm1 deficiency promotes ferroptosis in neutrophils by negatively regulating Alox15 protein levels.

Irgm1 interacted with RNF213 and promoted the RNF213-mediated proteasomal degradation of Alox15

3.5

Since Irgm1 lacks E3 ubiquitin ligase activity, it is unlikely to directly induce substrate ubiquitination. Instead, Irgm1 may regulate proteinase-mediated ubiquitination of Alox15 through interactions with ubiquitin ligases. To explore this, an IP-MS analysis of immunoprecipitated Irgm1 in neutrophils was performed to identify potential Irgm1-interacting proteins involved in regulating Alox15 ubiquitination and degradation (Fig. 5A). A total of 647 Irgm1-interacting proteins were identified, with the E3 ubiquitin ligase RNF213 emerging as a particularly interesting candidate (Fig. 5B). Additionally, mass spectrometry revealed a potential interaction between Alox15 and Irgm1 (Fig. 5B).Fig. 5Irgm1 regulates Alox15 degradation by binding to RNF213. (A) Schematic diagram of the experimental design for IP–MS analysis. (B) Potential Irgm1-interacting proteins identified by MS analysis. (C) Irgm1, RNF213, and Alox15 molecular docking assay. (D) Representative images of immunofluorescence staining of Irgm1 along with Alox15 in neutrophils (left), and analyses of the co-location curve (right). Scale bar = 2 μm. (E) Representative images of immunofluorescence staining of RNF213 along with Alox15 in neutrophils (left), and analyses of the co-location curve (right). Scale bar = 2 μm. (F) Western blot analysis of RNF213 and Alox15 immunoprecipitated with anti-Irgm1 antibody in neutrophils. (G) Western blot analysis of Irgm1 and Alox15 immunoprecipitated with anti-RNF213 antibody in neutrophils. (H) Proximity ligation assay (PLA) reveals the subcellular localization of the interaction between RNF213 and Alox15. Red spots indicate the signals of the RNF213 and Alox15 interaction; blue shows DAPI-stained nuclei. Scale bar = 2 μm. (I) Representative Western blot images and quantitative analysis of Alox15 expression in HL-60 cells with RNF213 knockdown (si-RNF213) (n = 4 per group). (J) Representative Western blot images and quantitative analysis of Alox15 expression in HL-60 cells with RNF213 overexpression (n = 4 per group). (K) RNF213 deficiency might exacerbate the decrease in Alox15 ubiquitination levels resulting from IRGM silencing. (L) IRGM specifically enhanced RNF213-regulated Alox15 ubiquitination. (M) Effect of Irgm1 overexpression on Alox15's binding to RNF213, n = 3 independent experiments. (N) Effects of Irgm1 knockdown on the binding of RNF213 to Alox15 in LPS-treated HL-60 cells by Co-IP, n = 3 independent experiments. Data are represented as the mean ± SD. Unpaired Student t-test for I and J. ∗∗∗∗p < 0.0001.Fig. 5

To further elucidate the regulatory mechanisms of Irgm1, RNF213, and Alox15 during SIMD, molecular docking experiments and immunofluorescence co-localization were conducted. Molecular docking revealed the formation of a tri-protein complex (Fig. 5C). Immunofluorescence analysis demonstrated co-localization of Irgm1 and Alox15 in neutrophils exposed to LPS stimulation, with RNF213 also showing significant co-localization with Alox15 (Fig. 5D and E). Co-immunoprecipitation assays confirmed that the Irgm1 antibody precipitated both RNF213 and Alox15 in LPS-treated neutrophils (Fig. 5F), while the RNF213 antibody specifically precipitated Alox15 and Irgm1 (Fig. 5G). Additionally, the in situ proximity ligation assay revealed an interaction between RNF213 and Alox15 in the cytoplasm of peripheral blood neutrophils from SIMD mice (Fig. 5H).

Knockdown of RNF213 significantly increased Alox15 protein levels (Fig. 5I), while overexpression of RNF213 resulted in a significant reduction in Alox15 protein levels (Fig. 5J). Furthermore, overexpression of RNF213 significantly diminished the upregulation of Alox15 induced by IRGM knockdown (Fig. S10A), while silencing RNF213 reversed the downregulation of Alox15 by IRGM overexpression in HL-60 cells (Fig. S10B). These results suggest that RNF213 mediates Irgm1's regulation of Alox15 protein levels and plays a pivotal role in Irgm1-induced Alox15 ubiquitination.

Given that RNF213 functions as an E3 ubiquitin ligase, this study examined whether the interaction between Irgm1 and RNF213 modulates Alox15 ubiquitination and degradation in HEK-293T cells. Silencing of IRGM significantly reduced Alox15 ubiquitination levels, while RNF213 deficiency further decreased Alox15 ubiquitination (Fig. 5K). In contrast, IRGM overexpression notably increased Alox15 ubiquitination, with this effect further amplified by RNF213 overexpression (Fig. 5L). Immunoprecipitation studies revealed that the precipitation of Alox15 increased with RNF213 antibody following IRGM overexpression, whereas Alox15 precipitation decreased with Irgm1 downregulation (Fig. 5M and N). These results confirm that Irgm1 facilitates the interaction between RNF213 and Alox15, thereby promoting the ubiquitination and degradation of Alox15.

The Alox15/15-HETE axis contributes to SIMD in the context of Irgm1 deficiency

3.6

15-Hydroxyeicosatetraenoic acid (15-HETE), a primary metabolite produced by Alox15 through arachidonic acid metabolism [30,43]. Consistent with Alox15/15-HETE's pivotal role, validation experiments confirmed that the levels of 15-HETE were significantly elevated in the serum of Irgm1-cKO mice as well as in the supernatants of LPS-stimulated Irgm1-cKO neutrophils (Fig. 6A and B). To assess the impact of 15-HETE in SIMD, CLP-induced mice were treated with exogenous 15-HETE (Fig. 6C). This intervention significantly reduced survival time (Fig. 6D) and further deteriorated cardiac function (Fig. 6E–I). Compared to septic WT mice, 15-HETE-treated mice exhibited increased cardiac mRNA levels of ANP and BNP (Fig. 6J), along with elevated plasma concentrations of cTnI and LDH—key markers of myocardial injury (Fig. 6K and L). Additionally, cardiac mRNA levels of pro-inflammatory mediators were significantly upregulated following 15-HETE treatment post-CLP (Fig. 6M). Histopathological analysis revealed exacerbated myocardial fiber disarray, interstitial edema, and inflammatory cell infiltration in the hearts from 15-HETE-treated mice (Fig. 6N). TUNEL staining further confirmed a significant increase in the number of apoptotic cardiomyocytes compared to septic WT mice (Fig. 6O). Electron microscopy analysis confirmed that 15-HETE treatment markedly aggravated mitochondrial damage, including mitochondrial disarrangement and cristae loss (Fig. 6P). Taken together, these findings demonstrate that 15-HETE is a critical mediator in SIMD.Fig. 6The regulatory role of Alox15/15-HETE in CLP-induced SIMD. (A) Graphs showing the levels of 15-HETE in the serum of WT and Irgm1-cKO mice after CLP (n = 5 per group). (B) Graphs showing the levels of 15-HETE in the supernatants of neutrophils from WT and Irgm1-cKO mice treated with LPS (n = 5 per group). (C) Schematic illustration of 15-HETE treatment in WT mice. The mice were intraperitoneally injected with 15-HETE (0.5 mg/kg). (D) Survival analysis of WT mice given Vehicle (n = 20) or 15-HETE (n = 26) after CLP was monitored for 24h. (E) Representative M-mode echocardiograms obtained from WT mice given Vehicle or 15-HETE after CLP. Scale bars = 2 mm (vertical) and 0.1 s (horizontal). (F–I) Echocardiographic measurements of left ventricular ejection fraction (EF, F), fractional shortening (FS, G), left ventricular internal diastolic dimension (LVIDD, H), and left ventricular end-diastolic volume (LVEDV, I) from WT mice given Vehicle or 15-HETE after CLP (n = 8 per group). (J) Relative ANP and BNP mRNA expression levels of the cardiac in WT mice given Vehicle or 15-HETE after CLP. (K and L) Graphs showing the levels of cTnI (K) and LDH (L) in the serum of WT mice given Vehicle or 15-HETE after CLP (n = 5 per group). (M) Relative mRNA expression levels of cardiac inflammation-related genes in WT mice given Vehicle or 15-HETE after CLP (n = 5 per group). (N) Representative images of H&E staining and quantitative analysis of myocardial injury. Scale bars: 50 μm for high magnification, 500 μm for low magnification. (O) TUNEL staining and quantification of apoptotic cardiomyocytes in heart tissues from WT mice given Vehicle or 15-HETE after CLP (n = 5 per group). Scale bar = 20 μm. (P) Representative TEM images of heart sections from WT mice given Vehicle or 15-HETE after CLP. Scale bars: 0.5 μm for high magnification, 1 μm for low magnification. Data are represented as the mean ± SD. Log-rank test for D. Unpaired Student t-test for A, and B, F through O. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.Fig. 6

To evaluate the in vivo role of Alox15, an adoptive transfer experiment was performed (Fig. S11A). Donor Irgm1-cKO neutrophils, pretreated with the Alox15 inhibitor PD146176, were adoptively transferred into recipient mice. This transfer resulted in a significant improvement in cardiac function compared to untreated Irgm1-cKO neutrophils (Fig. S11B–F). Compared to SIMD mice infused with untreated Irgm1-cKO neutrophils, those receiving PD146176-treated Irgm1-cKO neutrophils exhibited significantly lower cardiac mRNA expression of ANP and BNP (Fig. S11G), reduced pro-inflammatory mediator levels (Fig. S11H), and decreased plasma concentrations of myocardial injury markers cTnI and LDH (Fig. S11I and J). Histopathological analysis revealed that myocardial fiber disorganization, interstitial edema, and inflammatory cell infiltration were markedly attenuated in PD146176-treated Irgm1-cKO neutrophil recipients compared to SIMD mice infused with untreated Irgm1-cKO neutrophils (Fig. S11K). TUNEL staining confirmed a significant reduction in apoptotic cardiomyocytes (Fig. S11L), and electron microscopy showed less severe mitochondrial damage (Fig. S11M). These results underscore the crucial regulatory role of the Irgm1/Alox15 axis in SIMD pathogenesis, highlighting its potential as a therapeutic target.

To assess the potential of Alox15 expression in neutrophils as a diagnostic marker for sepsis, Alox15 levels in peripheral blood neutrophils were analyzed, revealing significant upregulation at both the mRNA and protein levels (Fig. S12A and B). To explore the relationship between Alox15 in neutrophils and sepsis progression, its correlation with disease severity and laboratory prognostic markers was examined. The mRNA levels of Alox15 in neutrophils positively correlated with disease severity (Fig. S12C), suggesting a potential role in SIMD pathogenesis. Further analysis demonstrated a significant positive correlation between Alox15 mRNA expression in neutrophils and inflammatory markers such as PCT and IL-6, as well as organ damage indicators like cTnI, LDH, and BUN (Fig. S12D–H). Subsequently, we explored the correlation between 15-HETE and clinical outcomes. Plasma 15-HETE concentrations were significantly elevated in patients with sepsis compared to the healthy control group (Fig. S12I). Additionally, the mRNA levels of IRGM in neutrophils negatively correlated with plasma 15-HETE concentrations (Fig. S12J). Plasma 15-HETE levels were positively correlated with the SOFA score (Fig. S12K) and inflammatory markers PCT and IL-6 (Fig. S12L and M). Furthermore, 15-HETE levels also positively correlated with organ damage indicators such as cTnI, LDH, and BUN (Fig. S12N–P). These findings highlight the strong correlation between Alox15 and its metabolite 15-HETE with sepsis severity, highlighting their potential as biomarkers for SIMD.

Pharmacological Inhibition of Alox15 improves CLP-induced cardiac dysfunction

3.7

Although neutrophil Irgm1 negatively regulates Alox15 protein levels, the lack of specific drugs targeting Irgm1 limits its therapeutic potential. Therefore, the in vivo benefits of targeting Alox15 as a treatment for SIMD were investigated (Fig. S13A). Compared to vehicle-treated WT mice, PD146176 treatment significantly reduced the plasma concentrations of 15-HETE (Fig. S13B). In the SIMD mouse model, administration of the Alox15 inhibitor PD146176 significantly prolonged survival time (Fig. S13C) and improved cardiac function (Fig. S13D–H). Histopathological analysis showed that PD146176 alleviated myocardial fiber disorganization, interstitial edema, and inflammatory cell infiltration in CLP-induced mice (Fig. S13I). PD146176 treatment also reduced plasma levels of myocardial injury markers, including LDH and cTnI (Fig. S13J and K), and decreased ANP and BNP mRNA expression (Fig. S13L). Additionally, mRNA levels of pro-inflammatory mediators were downregulated (Fig. S13M). TUNEL staining revealed a significant reduction in cardiac apoptosis in SIMD mice treated with PD146176 (Fig. S13N), and electron microscopy indicated reduced mitochondrial damage in the myocardium (Fig. S13O).

To assess the specific in vivo mechanisms of Alox15, PD146176 was administered to Irgm1-cKO mice following CLP (Fig. 7A). Treatment with PD146176 significantly alleviated neutrophil ferroptosis (Fig. S14A–C) and a corresponding reduction of plasma 15-HETE concentrations compared to vehicle-treated Irgm1-cKO mice (Fig. 7B). Additionally, PD146176 treatment significantly extended the survival time of Irgm1-cKO mice (Fig. 7C) and improved cardiac function (Fig. 7D–H). H&E staining showed reduced myocardial edema and inflammatory cell infiltration (Fig. 7I), while serum levels of myocardial injury markers LDH and cTnI were significantly decreased (Fig. 7J and K). Cardiac mRNA levels of ANP and BNP were downregulated (Fig. 7L), along with inflammatory gene expression (TNF-α, IL-1β, IL-6) in cardiac tissue (Fig. 7M). TUNEL staining confirmed a reduction in myocardial apoptosis (Fig. 7N), and electron microscopy revealed attenuated mitochondrial damage (Fig. 7O). These results demonstrate that Alox15 inhibition via PD146176 effectively mitigates SIMD, particularly in the context of Irgm1 deficiency.Fig. 7Administration of PD146176 can eliminate the detrimental effects caused by Irgm1 deficiency in vivo. (A) Schematic illustration of PD146176 treatment in Irgm1-cKO mice. The mice were intraperitoneally injected with PD146176 (10 mg/kg). (B) Graphs showing the levels of 15-HETE in the serum of Irgm1-cKO mice given Vehicle or PD146176 after CLP (n = 5 per group). (C) Survival analysis of Irgm1-cKO mice given Vehicle (n = 24) or PD146176 (n = 18) after CLP performance was monitored for 24h. (D) Representative M-mode echocardiograms obtained from Irgm1-cKO mice given Vehicle or PD146176 after CLP. Scale bars = 2 mm (vertical) and 0.1 s (horizontal). (E-H) Echocardiographic measurements of left ventricular ejection fraction (EF, E), fractional shortening (FS, F), left ventricular internal diastolic dimension (LVIDD, G), and left ventricular end-diastolic volume (LVEDV, H) of Irgm1-cKO mice given Vehicle or PD146176 after CLP (n = 8 per group). Scale bars = 2 mm (vertical) and 0.1 s (horizontal). (I) Representative images of H&E staining and quantitative analysis of myocardial injury. Scale bars: 100 μm for high magnification, 500 μm for low magnification. (J and K) Graphs showing the levels of LDH (J) and cTnI (K) in the serum of Irgm1-cKO mice given Vehicle or PD146176 after CLP (n = 5 per group). (L) Relative ANP and BNP mRNA expression levels of the cardiac in Irgm1-cKO mice given Vehicle or PD146176 after CLP (n = 5 per group). (M) Relative mRNA expression levels of cardiac inflammation-related genes in Irgm1-cKO mice given Vehicle or PD146176 after CLP (n = 5 per group). (N) TUNEL staining and quantification of apoptotic cardiomyocytes in heart tissues from Irgm1-cKO mice given Vehicle or PD146176 after CLP (n = 5 per group). Scale bar = 20 μm. (O) Representative TEM images of heart sections from Irgm1-cKO mice given Vehicle or PD146176 after CLP. Scale bars: 0.5 μm for high magnification, 1 μm for low magnification. Data are represented as the mean ± SD. Log-rank test for C. Unpaired Student t-test for B, and E through N. ∗p < 0.05; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.Fig. 7

Discussion

4

This study highlights the critical role of Irgm1 in regulating neutrophil ferroptosis during SIMD. Initially, Irgm1-cKO mice exhibited an exacerbated inflammatory response and more severe cardiac dysfunction following CLP. Proteomic analysis further revealed that the depletion of Irgm1 in neutrophils enriches ferroptosis pathways and upregulates Alox15. Mass spectrometry and co-immunoprecipitation analyses identified that Irgm1 interacts with the E3 ubiquitin ligase RNF213, promoting the ubiquitination and degradation of Alox15, thereby inhibiting neutrophil ferroptosis and 15-HETE production. Ultimately, pharmacological inhibition of Alox15 with PD146176 effectively mitigated SIMD in Irgm1-cKO mice. In summary, these findings highlight that Irgm1 inhibits ferroptosis and 15-HETE production in neutrophils by promoting Alox15 degradation, thereby alleviating SIMD.

Irgm1 (IRGM), is a key regulator of innate immunity and cellular homeostasis [17,20,21,44]. Its functional importance is underscored by studies across different cell types and disease models. For example, myeloid cell-specific deletion of Irgm1 increases susceptibility to bacterial infection [45], and knockout of Irgm1 in CLP-induced acute lung injury (ALI) model mice significantly exacerbates symptoms of sepsis-induced ALI [18]. Furthermore, systemic Irgm1 knockout worsens sepsis-associated encephalopathy [19]. In line with these findings, our study reveals that neutrophil-specific deletion of Irgm1 aggravates cardiac dysfunction and myocardial injury in sepsis, thereby identifying a previously unrecognized, cell-type-specific protective role for Irgm1 in SIMD.

Neutrophils play a pivotal role in regulating iron homeostasis and nutritional immunity during infection. Ferroptosis in neutrophils can be considered a consequence of dysregulated iron homeostasis, where bacteria exploit free iron ions to synthesize essential materials for replication, thereby exacerbating infection [46,47]. Rina et al. [48] demonstrated that neutrophil ferroptosis induces immunosuppression in cancer, promoting tumor progression. Zhao et al. [49] showed that tumor-infiltrating neutrophils (TIN) escape ferroptosis and achieve persistence through an Acod1-dependent immunometabolic switch, enhancing metastasis. Additionally, glutathione peroxidase 4-regulated ferroptosis in neutrophils induces systemic autoimmunity, promoting lupus development, and methods to reduce neutrophil ferroptosis significantly alleviate disease progression in lupus mice [50]. Our findings indicate that neutrophil-specific Irgm1 deficiency promotes neutrophil ferroptosis and exacerbates SIMD. Ferroptosis inhibitor treatment improves SIMD in Irgm1-cKO mice, suggesting that neutrophil ferroptosis could be a potential therapeutic target for SIMD.

Alox15, a key lipoxygenase enzyme, catalyzes the oxidation of polyunsaturated fatty acids (PUFAs) and plays a role in both physiological processes and the pathogenesis of neurodegenerative, inflammatory, and hyperproliferative diseases [51,52]. Elevated Alox15 expression and its metabolites have been implicated in various inflammatory conditions, including arthritis [53], asthma [54], and atherosclerosis [55]. Consistent with previous findings, Chen et al. have discovered that the gene expression of Alox15 was elevated in the peripheral blood of elderly patients with sepsis [56]. This study further reveals that Alox15 expression in neutrophils is upregulated in patients with SIMD and positively correlated with the severity of sepsis. Alox15 primarily induces tissue inflammation and oxidative stress through ferroptosis [43]. Previous studies have demonstrated that upregulation of Alox15 can enhance cellular susceptibility to ferroptosis [57]. In ischemic-injured myocardium, the expression of Alox15 is upregulated, exacerbating myocardial ischemia-reperfusion injury by promoting ferroptosis, whereas Alox15 knockout reduces ferroptosis [58]. Recent studies have demonstrated that targeting Alox15 inhibition can suppress ferroptosis. For instance, Deng et al. showed that Lachnospiraceae-bacterium mitigate ischemia-reperfusion injury in fatty donor livers by inhibiting ferroptosis through the suppression of Alox15 expression [59]. Hua et al. demonstrated that alliin exerts neuroprotective effects by inhibiting Alox15-dependent ferroptosis [60]. In the current study, we observed that Alox15 expression is elevated following sepsis, and its significant upregulation upon Irgm1 deficiency is closely associated with ferroptosis. Inhibiting Alox15 alleviated ferroptosis in Irgm1-cKO neutrophils, and overexpression of IRGM partially mitigated the ferroptosis induced by Alox15 overexpression. These findings indicate that Irgm1 serves as a regulator of neutrophil ferroptosis and offers a therapeutic avenue by inhibiting Alox15 to suppress ferroptosis.

15-HETE is one of the primary metabolites produced from the metabolism of arachidonic acid catalyzed by Alox15 [30,43]. Ischemic conditions induce Alox15 upregulation in cardiac tissue, leading to increased 15-HETE production and contributing to thrombosis [61]. Yang et al. reported that activation of the Alox15/15-HETE axis promotes apoptosis and inflammatory cell infiltration in hepatic ischemia-reperfusion models [30]. Alkayed et al. found that 15-HETE enhances coronary microvascular resistance by increasing intracellular calcium concentration in mVSMCs [62]. This study is the first to demonstrate that 15-HETE exacerbates SIMD and systemic inflammatory response in a CLP model.

Ubiquitination is a crucial posttranslational modification that regulates protein stability and function [63]. RNF213, an E3 ubiquitin ligase, modulates intracellular protein levels by recognizing and ubiquitinating target proteins, thereby influencing various physiological and pathological processes [64,65]. Recent studies have indicated that RBX1 (RING-box protein 1) interacts with Alox15 to exert E3 ubiquitin ligase activity, regulating the stability of Alox15 protein [66]. This study demonstrates that high expression of Irgm1 promotes the interaction between RNF213 and Alox15, ultimately leading to the ubiquitination and degradation of Alox15, thereby elucidating a novel regulatory mechanism of Alox15 in SIMD.

PD146176 is a specific inhibitor of Alox15. Previous studies have demonstrated that PD146176 ameliorates apoptosis and inflammatory cell infiltration in hepatic ischemia-reperfusion injury by targeting Alox15 [30]. Additionally, in a mouse model of unilateral ureteral obstruction, PD146176 attenuated renal fibrosis and inflammation through Alox15 inhibition [67]. The present study demonstrates that the Alox15-specific inhibitor PD146176 effectively reduces neutrophil ferroptosis, thereby alleviating SIMD and systemic inflammation. These findings suggest that targeting Alox15 with PD146176 may improve cardiac function in patients with sepsis and mitigate organ damage, particularly in individuals with IRGM deficiency.

Despite providing novel insights, this study has certain limitations. First, while Irgm1 was shown to promote Alox15 degradation through its interaction with RNF213, the involvement of additional regulatory proteins in this process cannot be excluded. Future studies are warranted to investigate the specificity and ubiquity of this pathway. Secondly, no specific therapeutic agents targeting Irgm1 are currently available for clinical intervention; thus, the development of potential Irgm1 agonists or protein stabilizers, through strategies such as structure-based drug design and high-throughput screening, represents a promising direction for future research. Finally, although previous studies have identified IFN-γ and TLR pathways as upstream regulators of Irgm1 [15,68], the mechanisms underlying Irgm1 upregulation in neutrophils following SIMD remain unclear. Therefore, future investigations could employ transcription factor binding assays and chromatin immunoprecipitation (ChIP) to further elucidate the regulatory mechanisms of Irgm1 induction.

In conclusion, this study establishes neutrophil Irgm1 as a key regulator of SIMD. The interaction between Irgm1 and RNF213 facilitates Alox15 ubiquitination and degradation, thereby suppressing neutrophil ferroptosis and 15-HETE production and improving SIMD outcomes. Inhibition of Alox15 effectively mitigates myocardial damage, and IRGM, Alox15, and 15-HETE may emerge as potential prognostic biomarkers for sepsis. Targeting Alox15 with pharmacological agents such as PD146176 represents a promising therapeutic strategy for mitigating SIMD, particularly in individuals with IRGM deficiency.

Ethics statement

All animal experimental procedures were conducted according to the Guide for the Care and Use of Laboratory Animals outlined by the National Institute of Health (NIH). This study was approved by the Research Ethics Committee of the Second Affiliated Hospital of Harbin Medical University and performed according to the criteria set by the Declaration of Helsinki. All participants provided informed consent.

CRediT authorship contribution statement

Zeng Wang: Funding acquisition, Investigation, Software, Validation, Visualization, Writing – original draft, Writing – review & editing. Jiaxiang Sun: Formal analysis, Methodology, Resources, Software, Validation, Visualization. Mingyang Wang: Investigation, Methodology, Software, Validation, Visualization. Lai Wei: Investigation, Methodology, Software, Validation, Visualization. Fengyi Liu: Methodology, Validation, Visualization. Wenhua Liu: Investigation, Software, Validation. Yige Liu: Methodology, Validation, Visualization. Jiaxin Wang: Software, Validation. Fujian Tan: Methodology, Validation. Bo Yu: Conceptualization, Project administration. Zhiqiang Li: Conceptualization, Project administration. Shaohong Fang: Conceptualization, Formal analysis, Funding acquisition, Project administration, Supervision. Yong Sun: Conceptualization, Data curation, Funding acquisition, Project administration, Resources, Supervision.

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.