IFNAR1-dependent metabolic circuit drives pathogenic neutrophil activation in chronic mucosal inflammation Authors: Cecilia Pessoa Rodrigues, Raquel R Calçada, Mafalda Arnaud, Lukas Kaltenbach, Mika Manser, Anja Scheidegger, Leon Sidney Strauss, Alina Gavrilov, Sharang Kulkarni, Ece Yildiz, Daniela Liberati, Mairene Coto-Llerena, Tamara Hoening, Markus Martin Kramberg, Julian Behr, Anna Mechling, Marvin Hering, Kara G Lassen, Oliver Soehnlein, Emma Doran, Daniel Regan-Komito Published: 2026 ABSTRACT AbstractNeutrophils are double-edged effectors of the innate immune system: while essential for host defense, they can drive significant tissue damage in the context of chronic inflammatory diseases. Identifying strategies to selectively dampen their pathogenic functions without compromising antimicrobial immunity remains a major clinical challenge. Here, we define a conserved IFNAR1-driven circuit mediated by iron homeostasis that sustains neutrophil pathogenicity across distinct mucosal tissues. Using single-cell transcriptomics and a CRISPR-based functional screen, we identified the mitochondrial iron transporter Mitoferrin-1 (SLC25A37) as a key regulator of neutrophil inflammatory programming. Mechanistically, our data indicate that Mitoferrin-1 mediates a feedforward loop in which NET-derived histones activate toll-like receptor 9 (TLR9), which in turn sustains IFN-α/IFNAR1 signaling, rewires mitochondrial metabolism, and ultimately drives tissue damage through markedly increased of NETosis, reactive oxygen species production, and degranulation. Notably, this IFNAR1-driven inflammatory circuit operates entirely independently of antimicrobial function, as pharmacological IFNAR1 blockade fully preserves phagocytic activity against bacterial pathogens in vivo. Targeting this pathway selectively attenuates neutrophil-driven inflammation in both murine models of colitis and acute lung injury, as well as in primary human intestinal organoid systems. Together, these findings identify a targetable metabolic node for selectively disarming pathogenic neutrophil responses in the setting of chronic inflammatory disease.Subject terms: Inflammation, Immunology Introduction Neutrophils, the most abundant white blood cells in circulation, are essential for innate immune defense due to their rapid response to infection and injury. They neutralize pathogens through several key mechanisms, including phagocytosis, degranulation, neutrophil extracellular trap (NET) release, and reactive oxygen species (ROS) production.1–3 Under physiological conditions, these functions are self-limiting2,4 and essential for mucosal homeostasis in the gastrointestinal and respiratory tracts.5 To maintain this delicate balance, neutrophils undergo tightly regulated apoptosis followed by macrophage clearance, preventing the leakage of toxic intracellular contents into healthy surroundings. However, in chronic inflammation, this resolution phase fails, leading to the excessive accumulation and sustained activation of neutrophils that compromise epithelial barrier integrity and drive persistent immune signaling. When these primary cells are continuously recruited to mucosal surfaces without undergoing proper clearance, their prolonged lifespan allows them to transition from protective scavengers into key drivers of chronic tissue destruction, altering local cellular networks. This maladaptive response exacerbates tissue damage through the release of serine proteases, ROS, and proinflammatory cytokines that recruit and activate additional immune cells.6,7 Furthermore, increased NETosis correlates with disease activity in chronic mucosal inflammatory conditions, including inflammatory bowel disease (IBD) and chronic obstructive pulmonary disease (COPD).8–11 In the gut, persistent NET degradation releases cytotoxic histones that disrupt the delicate epithelial monolayer, elevating vascular permeability and fueling a continuous cycle of autoinflammation. In COPD, neutrophil infiltration promotes airway remodeling, mucus hypersecretion, and progressive alveolar destruction (emphysema),6 underscoring their central role in disease progression. Another prominent clinical problem in COPD that is driven and fundamentally caused by neutrophils is the progressive destruction of respiratory architecture. In these patients, the overproduction of matrix metalloproteinases and elastase systematically breaks down the lung parenchyma, severely impairing gas exchange over time.6 Understanding the mechanisms that govern these pathogenic states is critical for developing strategies to hinder neutrophil-mediated injury while preserving essential antimicrobial functions such as phagocytosis. Consequently, identifying cell-intrinsic pathways that selectively decouple destructive degranulation or NETosis from protective pathogen clearance represents the holy grail of modern anti-inflammatory drug discovery. Unfortunately, conventional therapeutic strategies aimed at suppressing these cells often result in systemic immunosuppression, leaving the host highly vulnerable to secondary, opportunistic infections and presenting a major hurdle for clinical translation. To resolve this long-standing clinical dilemma, our approach utilized an unbiased combination of single-cell transcriptomics alongside a targeted, CRISPR-based functional screening platform to profile the heterogeneous activation states of neutrophils isolated directly from inflamed mucosal tissues. By mapping the specific intracellular and transcriptomic changes that occur during chronic mucosal inflammation, we aimed to uncover hidden metabolic checkpoints unique to the pathogenic, hyper-inflammatory state of the cell, effectively looking for pathways that could be targeted to uncouple tissue destruction from protective host immunity. Here, we identified a central role for the mitochondrial iron transporter Mitoferrin-1 (SLC25A37) in regulating neutrophil pathogenicity in IBD and COPD. We demonstrate that mitochondrial iron uptake functions as a metabolic rheostat; increased iron transport promotes an increase in the mitochondrial membrane potential and oxygen consumption through enhanced mitochondrial fusion. This morphological remodeling directly alters metabolic flux, generating localized mitochondrial ROS (mtROS) that act as critical secondary messengers. We found that this mechanism was particularly important for the TLR9 response and type I interferon (IFN) production by neutrophils. Notably, secreted IFN-ɑ then signals through the IFN-ɑ receptor (IFNAR1) in an autocrine/paracrine manner, reinforcing a positive feedforward loop that sustains chronic inflammation. Using murine models of IBD and COPD, alongside human 3D organoid cocultures that partially recapitulated IBD and COPD in vitro, we show that IFNAR1 blockade inhibits neutrophil-driven inflammation and tissue damage. Crucially, this therapeutic inhibition disrupted the pathological feedforward loop, successfully preserving the structural integrity of mucosal barriers and preventing severe epithelial apoptosis. This study is entirely novel because it demonstrates for the first time that this tissue-destructive metabolic circuit operates completely independently of classical antimicrobial machinery, leaving baseline phagocytic activity against bacterial pathogens entirely intact in vivo. By pinpointing Mitoferrin-1 as the central executioner of this hyper-inflammatory pathway, we show that the energetic adaptations driving tissue destruction can be therapeutically separated from the mechanical machinery required for pathogen clearance. These findings identify a targetable metabolic signaling axis to selectively modulate neutrophil activity in chronic mucosal diseases without compromising host defense, providing a conceptual framework for narrow-spectrum immunotherapies that disarm inflammation while preserving patient immunity. Results Infiltrating neutrophils shared transcription signature in IBD and COPD To identify altered infiltrating neutrophil pathways in IBD and COPD, we analyzed 13 public scRNA-seq datasets (5 IBD/intestine, 8 COPD/lung) from patients and healthy controls. To address tissue neutrophil sparsity and annotation discrepancies, we harmonized leukocyte annotations using a machine learning approach.12 We first isolated leukocytes based on PTPRC (CD45) expression, then applied CellTypist to annotate the immune cells (Fig. 1a). Cells with >80% similarity to neutrophils were subsetted, and pseudobulk analysis was performed to compare patient and healthy cohorts. Transcriptional characterization of upregulated genes and pathways in IBD and COPD-associated infiltrating neutrophils. a Schematic overview of the single-cell RNA sequencing (scRNA-seq) integration strategy used to define neutrophil infiltration signatures across inflammatory diseases. Publicly available datasets were integrated for inflammatory bowel disease (IBD: GSE214695, GSE109142, GSE121380, GSE134809, GSE162335) and chronic obstructive pulmonary disease (COPD: GSE136831, GSE173896, GSE178360, GSE183974, GSE145926, GSM3660650, EGAS00001004369). For the pseudo-bulk analysis 850 (IBD) and 170 (healthy) neutrophils were considered for the intestine, while 4100 (COPD) and 1805 (healthy) were considered for the lung. b Venn diagram showing disease-specific and shared upregulated genes in infiltrating neutrophils from IBD and COPD samples. Differentially expressed genes (DEGs) were identified using pseudo-bulk analysis with a significance threshold of p < 0.05. c, d Pathway enrichment analysis of shared upregulated genes. c KEGG pathways and d MSigDB Hallmark gene sets associated with common transcriptional programs in IBD and COPD neutrophils. e Functional classification of shared upregulated genes based on co-expression and network association using GeneMANIA. f Transcription factor (TF) network analysis (TF-TRED) identifying regulatory circuits associated with shared upregulated genes. Nodes represent genes, and colors indicate associated transcription factor pathways. Genes regulated by multiple TFs are connected by edges reflecting shared regulatory interactions This analysis confirmed increased neutrophil infiltration in both diseases (Supplementary Fig. 1a–e). Despite the multifaceted nature of neutrophil roles in IBD, we identified 172 upregulated genes shared between infiltrating neutrophils in IBD and COPD (Fig. 1b, and Supplementary Table 1). These shared genes were enriched for key neutrophil functions, including NET formation, leukocyte migration, degranulation, and interleukin signaling (Fig. 1c, d), and exhibited strong co-expression (Fig. 1e). Transcription factor analysis further revealed activation of inflammatory regulators, including RELA, NFKB1, SPI1, STAT1, and ETS1 (Fig. 1f). Pathway analysis of the shared upregulated genes in disease-infiltrating neutrophils showed an enrichment for interferon alpha/beta signaling (Fig. 1d). We then investigated whether downstream type I IFN signaling was engaged in the neutrophils; we found strong enrichment of type I IFN-related genes (Supplementary Fig. 1b–e). These results agree with elevated ISG15 levels observed during COPD exacerbations.9 While type I IFN-related genes have been previously implicated in IBD patient samples and linked to genetic variants like the IFNAR1-associated SNP rs2284553.10 These results point to a common neutrophil-driven inflammatory program across IBD and COPD, particularly involving type I IFN signaling. CD34-Adult derived neutrophils displayed a similar phenotype to fresh isolated neutrophils To investigate the function of shared upregulated genes found in infiltrating IBD and COPD neutrophils, we developed a CRISPR-compatible in vitro neutrophil differentiation model from human CD34+ hematopoietic stem and progenitor cells (HSPCs) sourced from adult (AD) or cord blood (CB). We opted for this approach due to previously reported technical hurdles using primary human neutrophils for CRISPR screens.11 The three-stage protocol involved: (1) expansion of the progenitor cells,13 (2) myeloblast lineage specification, and (3) neutrophil differentiation using a modified cytokine mix.14 Rosiglitazone, a PPARɣ agonist, was added to enhance neutrophil differentiation.15 After 17 days, ~60–80% of cells displayed neutrophil-like morphology (Supplementary Fig. 2a), with increased CD66b+ and decreased CD34+ populations (Supplementary Fig. 2b), indicating granulocyte lineage differentiation commitment. In vitro derived neutrophils were isolated via negative selection or by negative selection followed by positive selection of CD10 positive cells, which did not trigger activation (Supplementary Fig. 2c). Up to 60% mature neutrophils (CD66b+CD10+) were confirmed by flow cytometry (Supplementary Fig. 2d), with no contamination from other mature immune cells (Supplementary Fig. 2e). However, some immature cells remained in the culture (Supplementary Fig. 2e). High-dimensional flow cytometry revealed ~70% similarity between AD and CB in vitro derived and fresh isolated (FI) neutrophils, versus ~10% similarity with monocytes (Supplementary Fig. 2f). As no significant differences were observed between negative selection alone and negative selection followed by CD10⁺ enrichment, proteomic analyses were performed on freshly isolated neutrophils, negatively selected AD- and CB-derived neutrophils, and the HL-60 cell line (±PMA, 1 μg/mL), which is commonly used as a primary neutrophil surrogate. Principal component analysis (PCA) revealed that HL-60 cells differed from FI-neutrophils (~40% along PC1 and ~23% along PC2), whereas AD- and CB-derived neutrophils showed no separation along PC1 and only ~23% variation along PC2, with AD clustering closer to FI-neutrophils (Supplementary Fig. 2g). AD-derived neutrophils also showed the highest correlation with bone marrow neutrophils at the proteomic level (Supplementary Fig. 2h, i). Overall, these results indicate that AD-derived neutrophils recapitulate the proteomic profile of primary neutrophils, while HL-60 cells and CB-derived neutrophils show lower similarity. To validate our proteomics observations, we investigated the transcriptome profile of the AD neutrophils across the differentiation timeline. We confirmed the transition from progenitor to mature neutrophil signatures by day 17 (Supplementary Fig. 2j, k). We then evaluated the functionality of AD-derived neutrophils, in which we tested phagocytosis (Supplementary Fig. 2l), NETosis (Supplementary Fig. 2m), degranulation (Supplementary Fig. 2n), and migration capacity of these cells (Supplementary Fig. 2o). AD-derived neutrophils showed similar functional capacity compared to fresh isolated neutrophils for the four tested functions. Collectively, confirming that AD-derived neutrophils resemble fresh isolated neutrophils at the proteomic, transcription and function level. CRISPR-Cas9 screen revealed Mitoferrin-1 as a key regulator of neutrophil functions After confirming that AD-neutrophils molecularly and functionally resemble FI-neutrophils, we performed a targeted CRISPR-Cas9 screen. We prioritized genes significantly upregulated in both IBD and COPD patients with confirmed neutrophil protein expression in the Human Protein Atlas (Fig. 2a, and Supplementary Tables 2 and 3). Transfection did not alter AD-neutrophil viability and achieved ~70% efficiency (Supplementary Fig. 3a, b). To validate readout specificity, we targeted canonical genes: NCF2 or SOD2 depletion reduced LPS-induced ROS production, whereas ACTN1 depletion had no effect (Supplementary Fig. 3c). NCF1 depletion selectively impaired both NETosis (Supplementary Fig. 3d) and phagocytosis (Supplementary Fig. 3e). Similarly, NCF1 depletion impaired degranulation, while the migration gene CXCR4 did not (Supplementary Fig. 3e). Mirroring this specificity, VDAC3 depletion selectively impaired ROS production without altering phagocytosis (Supplementary Fig. 3e). Collectively, these results confirm that transfection efficiency and readout specificity were suitable for a larger screen. Functional CRISPR-Cas9 screening in neutrophils identifies SLC25A37 as a regulator of mitochondrial responses and effector functions. a Annotation of the 95 candidate targets identified by single-cell RNA sequencing (scRNA-seq), including validation of protein expression in neutrophils using the Human Protein Atlas. b Schematic overview of the CRISPR-Cas9 screening workflow. Guide RNAs (gRNAs) were transfected at day 17 of in vitro neutrophil differentiation. Functional readouts included phagocytosis, reactive oxygen species (ROS) production, degranulation, NETosis, and motility. Details on gRNA sequences and transfection efficiency are provided in Supplementary Tables 2 and 3. c Venn diagram showing the overlap among 31 functional readouts and the number of targets significantly affecting each function (p < 0.05). d Quantification of histone H3 citrullination in culture supernatants as a measure of NETosis. Cells stimulated with LPS (1 µg/mL) were used as a positive control. The full list of targets significantly affecting NETosis is provided in Supplementary Fig. 3. e, f Functional validation of SLC25A37 depletion. Bar plots show phagocytosis, ROS production, and degranulation in e CRISPR-Cas9–edited in vitro–derived neutrophils (n = 3 independent differentiation experiments) and f freshly isolated neutrophils. g Representative immunofluorescence images of mitochondrial membrane potential based on Mitotracker CMXROS under steady-state conditions (PBS) or following stimulation with extracellular (TLR4 agonist, LPS, 1000 ng/mL) or intracellular (TLR9 agonist, ODN2006, 3 µM) stimuli. Nuclei were stained with DAPI. h Quantification of mitochondrial membrane potential following stimulation with TLR agonists: TLR2 (Pam3CSK4, 1000 ng/mL), TLR3 (Poly I:C, high- and low-molecular-weight; 50 ug/mL), TLR4 (LPS, 1000 ng/mL), and TLR9 (ODN2006, 3 µM). Each data point represents an individual donor (n = 3–10, as indicated). i, j Live-cell imaging of mitochondrial membrane potential (Supplementary Video 1-2). i Representative time-lapse images under steady-state (PBS) or TLR9 stimulation. j Quantification over 1 h (n = 3 independent experiments). k Representative JC-1 staining indicating mitochondrial polarization (red) and depolarization (green). Quantification is shown in Supplementary Fig. 4. l, m Seahorse metabolic analysis of freshly isolated neutrophils (n = 5 donors). l Oxygen consumption rate (OCR) and m extracellular acidification rate (ECAR) under steady-state conditions or following stimulation with TLR4 (LPS) or TLR9 (ODN2006). n SLC25A37 (Mitoferrin-1) protein levels. Left: representative images from three independent donors under steady-state or TLR9 stimulation. Right: quantification of cellular median fluorescence intensity. Data are presented as mean ± SEM. Statistical analyses were performed as follows: Kruskal–Wallis test followed by Dunnett’s multiple comparison test for multi-group comparisons (d–f); one-way ANOVA with Dunnett’s post hoc test (h); two-way ANOVA (j, l, m); and Student’s t-test following assessment of normality using the Shapiro–Wilk test (n). Significance levels are indicated as *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The CRISPR-Cas9 screen was performed in two independent neutrophil differentiation experiments After confirmation of transfection strategies and readouts, we performed the targeted CRISPR-Cas9 screen against the IBD and COPD genes (Supplementary Table 3). Neutrophils were transfected only after complete neutrophil maturation (Fig. 2b). As readouts, we evaluated 5 different functions, namely phagocytosis, ROS production, degranulation, two-dimensional (2D) motility (“distance”), and NET formation (NETosis, based on imaging and H3 citrullination by ELISA) (Fig. 2c, and Supplementary Fig. 3f, g). To elicit robust and reproducible phenotypes across multiple readouts, we used standard LPS stimulation. Despite the lack of specificity in triggering more than TLR4, standard LPS induces a broad spectrum of neutrophil effector responses, including ROS production, degranulation and NET formation, which are key readouts in our multiplexed functional screen. Our screen revealed that among the tested genes, only one candidate gene impaired all five explored functions: SLC25A37 (Mitoferrin-1) (Fig. 2c–e). We confirmed that Mitoferrin-1 impaired neutrophil active using FI-neutrophils (Fig. 2f), in which we observed ~30% mutation efficiency (Supplementary Fig. 3h) that caused a significant reduction at the protein level (Supplementary Fig. 3i). We also did not observe alterations in fresh primary isolated neutrophils viability after transfection (Supplementary Fig. 3j). Collectively, our results revealed that Mitoferrin-1may mediates key neutrophil functions. Mitoferrin-1 regulates neutrophil responses to TLR9 Mitoferrin-1 is a mitochondrial protein that controls iron balance between the cytoplasm and mitochondria16 and is required for proper mitochondrial oxidative phosphorylation and dynamics.16 Mitochondrial dynamics, encompassing both fission and fusion processes, plays a crucial role in mediating innate immune responses.17 Fission facilitates mitophagy and redistribution, while fusion maintains mitochondrial integrity and optimizes energy production. Disrupting this delicate balance impacts immune activation, metabolism, and inflammation, thereby linking mitochondrial dysfunction to autoimmune and chronic inflammatory diseases.18 Given the importance of toll-like receptors (TLRs) in mediating chronic inflammatory diseases19 and the role of Mitoferrin-1 in maintaining mitochondrial homeostasis, we investigated mitochondrial dynamics in freshly isolated neutrophils in response to various TLR agonists (Fig. 2g). We utilized MitoTracker CMXRos, which accumulates in mitochondria in a membrane potential-dependent manner (ΔΨm) (Fig. 2g, h). Through live-cell imaging, we observed that TLR2 (Pam3CSK4) and TLR4 (ultrapure LPS, specific to TLR4), both cell surface receptors, promoted mitochondrial fission-like behavior (Fig. 2g, and Supplementary Fig. 3k), while the intracellular receptors: TLR3 (Poly I:C, high and low molecular weight) and TLR9 (ODN2006) promoted fusion-like behavior (Fig. 2g, and Supplementary Fig. 3k). These findings suggest that extracellular and intracellular pathogen-associated molecular patterns may elicit distinct mitochondrial programs in neutrophils. Among the TLRs examined, TLR9 produced the fastest response time, induced higher mitochondrial membrane potential (Fig. 2h–j, Supplementary Fig. 4a, Videos S1 and S2) and mitochondrial polarization (Fig. 2k, and Supplementary Fig. 4b). Mitochondrial polarization following TLR9 stimulation has mechanistic implications extending beyond a simple activation marker. As TLR9 stimulation promoted cellular ROS production (Supplementary Fig. 4c), phagocytosis (Supplementary Fig. 4d), NETosis (Supplementary Fig. 4e), and enhanced neutrophil migration capacity (Supplementary Fig. 4f). To determine whether the impact of Mitoferrin-1 depletion on mitochondrial activity resulted from mitochondrial cristae morphology collapse or disrupted iron uptake mediated by Mitoferrin-1, we treated cells with the iron chelator deferasirox (DFX).20 DFX treatment blocked mitochondrial polarization upon TLR9 agonist challenge (Supplementary Fig. 4g) and decreased neutrophil ROS production upon TLR9 stimulation (Supplementary Fig. 4h), suggesting that iron uptake is crucial for neutrophil responses to TLR9. We next investigated the global metabolic consequences of TLR9 stimulation by performing real-time metabolic rate analysis using extracellular flux analysis. We found that oxygen consumption rates (OCR) were significantly higher in TLR9-stimulated neutrophils (Fig. 2l), whereas TLR4 stimulation led to increased extracellular acidification rates (ECAR), which primarily reflects lactic acid production from glycolysis (Fig. 2m). To validate that the observed mitochondrial phenotype was independent of mitochondrial quantity, we performed MitoTracker-GreenFM staining. This dye labels mitochondria regardless of their activation state and can be used to evaluate overall mitochondrial presence and morphology. We confirmed that both TLR4- and TLR9-stimulated neutrophils contained mitochondria (Supplementary Fig. 4i) but observed that mitochondrial morphology changed upon TLR4 or TLR9 stimulation (Supplementary Fig. 4i). Together, these findings demonstrate that TLR9 stimulation promotes a distinct metabolic program in neutrophils characterized by enhanced mitochondrial respiration and fusion remodeling, independent of mitochondrial abundance. Since we identified Mitoferrin-1 as a key gene regulating neutrophil function, we investigated whether TLR9 stimulation promotes Mitoferrin-1 expression. We observed that TLR9 agonism indeed enhanced Mitoferrin-1 levels (Fig. 2n). We then examined the mitochondrial profile of Mitoferrin-1-depleted neutrophils with and without TLR9 agonist stimulation. While TLR9 facilitated mitochondrial fusion in control cells, this effect was absent in Mitoferrin-1-depleted neutrophils (Supplementary Fig. 4i). These results suggest that the observed changes in response to TLR9 stimulation are driven by mitochondrial dynamics that promote oxidative phosphorylation upon intracellular insults. Mitochondrial membrane potential regulates type I Interferon production in neutrophils To further understand the role of Mitoferrin-1 in regulating TLR9 responses, we depleted SLC25A37 and performed a real-time metabolic assay. We found that SLC25A37-depleted neutrophils displayed lower OCR (Fig. 3a). Moreover, we observed that TLR9 stimulation increased mitochondrial membrane potential (Fig. 3b) and mtROS (Fig. 3c, d), while these responses were impaired in Mitoferrin-1-depleted neutrophils (Fig. 3b–d). Further supporting the mitochondrial phenotype, we observed increased expression of Optic Atrophy 1 (OPA1) and decreased expression of Dynamin 1 Like (DRP1) upon TLR9 stimulation (Fig. 3e), corroborating that mitochondrial dynamics shifted toward the fusion network when neutrophils were treated with TLR9 agonist. Notably, the upregulation of OPA1 was impaired in Mitoferrin-1-depleted neutrophils (Fig. 3e), thereby supporting the hypothesis that TLR9 stimulation promotes mitochondrial fusion. Mitochondria activation is required for IFN-ɑ production. a Real-time cell metabolic analysis (Seahorse) showing the oxygen consumption rate (OCR) of control neutrophils (transfected with a scramble gRNA, grey, n = 3) and SLC25A37Mmitoferrin-1 depleted neutrophils (100 pmol, gRNA) (green, n = 3). b Quantification of the mitochondria polarization based on JC-1 mito probe red versus green ratio. Green color represents depolarized mitochondria and red represents polarized mitochondria (n = 3). Control neutrophils (transfected with a scramble gRNA, black) and SLC25A37 depleted neutrophils (green). c Representative imaging showing the mtROS levels under PBS or TLR9 stimulated neutrophils. We compared control versus SLC25A37/Mitoferrin-1 depleted neutrophils. d Quantification of the mtROS (n = 6). e Expression of Mitochondrial fusion related gene (OPA1) and mitochondrial fission gene (DRP1) found in control neutrophils or SLC25A37/Mitoferrin-1 depleted neutrophils stimulated with ODN2006 or not. For this analysis we isolated neutrophils from 3 healthy donors. f Representative flow cytometry histogram depicting the mitochondrial membrane potential (left panel) and type I IFN levels. ODN2006, the TLR9 agonist was used as positive control (red line). g Quantification of the type I IFN levels. h Quantification of the mitochondrial membrane potential. For both experiments the cells were treated with PBS (n = 6), ODN2006 (n = 6), LPS (n = 5), SLC25A37 depleted (n = 5, n = 6). The “+” under the bar represents the addition of the compound. i IFN-ɑ quantification upon specific mitochondrial complexes blockers for complex I we used Rotenone, for complex II we used Malonate, for Complex III we used Antimycin A, for Complex IV we used Sodium azide and for Complex V we used Oligomycin (n = 5). j Representative histograms are shown on the left, with corresponding quantification on the right. Human neutrophils were incubated for 2 h at 37 °C with ODN2006 (red) or LPS (teal) in the presence or absence of the iron chelator deferasirox (DFX; 5%, squared pattern). Intracellular labile iron was assessed using a Calcein AM–based assay and analyzed by flow cytometry, in which decreased fluorescence indicates increased iron availability. Importantly, only living cells (based on Zombie dye staining) were included in the Calcein-AM gate. Data are presented as mean ± SEM from n = 4 independent donors. Statistical significance was determined by one-way ANOVA followed by Tukey’s multiple comparisons test. p < 0.01. k ELISA quantification showing the amount of IFN-ɑ present in the supernatant of stimulated neutrophils. The “+” under the bar represents the addition of the compound. DFX: Deferasirox. IgG: Isotype IgG control. Ab: Antibody. l Quantification of IFN-ɑ levels and m percentage of IFN-ɑ positive neutrophils upon PBS, ODN2006 or exogenous IFN-ɑ challenged (500 IU/mL) (n = 5). MFI: Mean Fluorescent Intensity. Statistical analysis was performed by one-way ANOVA followed by Tukey’s multiple comparison. Data variation is presented as ±SEM. n Schematic representation of the proposed positive feedback loop exerted by IFN-ɑ and its receptor IFNAR1. o Bar plot showing the percentage of IFN-ɑ producing neutrophils upon PBS (steady state control, n = 7), ODN2006 (n = 3), Histone 3 citrullination (H3) (n = 10), autologous Mitochondria (n = 3), autologous Mitochondria plus DNAseI (n = 3), and SLC35A25 depleted neutrophils (n = 7). Besides the SLC35A25 depleted group the other groups were transfected with the scramble gRNA control. Statistical analysis was performed by Kruskal–Wallist test followed by Dunn’s multiple comparison. **p < 0.01. p Quantification of type I IFN production upon ODN2006 (ODN) and NETosis protein (Histone 4 (H4), and Histone 3 citrullination (H3Cit), 10 µg/mL). Cells were co-treated with chloroquine in order to test the endosomal formation dependence of the response. q Graphical representation of the colitis model in vivo. r Expression of Mitochondrial fusion related genes (Mfn2 and Opa1) and Slc25a37 found in bone marrow resident neutrophils in mice. For this analysis we isolated neutrophils from the bone marrow using negative neutrophil selection. s Representative imaging showing the mitochondrial polarization based on JC-1, a cationic, lipophilic fluorescent dye that assess mitochondrial membrane potential, forming red-fluorescent aggregates in polarized mitochondria and remaining as green-fluorescent monomers in depolarized mitochondria. t Quantification of the polarized mitochondrial ratio (Red emission/Green emission). Statistical analysis was performed by ANOVA followed by Tukey multiple comparisons (n = 6, healthy donors). *p < 0.05, **p < 0.01 To investigate the immunological relevance of Mitoferrin-1 in regulating TLR9 responses, we assessed type I interferon (IFN) production by neutrophils. TLR9-stimulated neutrophils exhibited robust type I IFN production (Fig. 3f, g), concomitant with enhanced mitochondrial activity (Fig. 3h). Both responses were impaired in Mitoferrin-1-depleted neutrophils (Fig. 3g, h, and Supplementary Fig. 4j). To delineate the contribution of oxidative phosphorylation (OxPhos) to type I IFN production, neutrophils were treated with specific inhibitors of mitochondrial respiratory complexes prior to TLR9 stimulation (Fig. 3i). Disruption of complex I or complex II significantly reduced IFN-α production, while disruption of complex III, IV or V did not alter IFN-α production. Suggesting the dependence of mitochondrial respiration in mediating TLR9-driven signaling. TLR9 activation increased cellular iron uptake (Fig. 3j), which was abrogated by the iron chelator DFX. Confirming the specificity of the TLR9-mediated response, iron uptake was not altered in LPS-stimulated neutrophils (Fig. 3i). Additionally, DFX treatment also damped the type I IFN production by TLR9-stimulated neutrophils (Supplementary Fig. 4k), supporting the existence of an intricate pathway linking iron uptake to type I interferon responses by neutrophils. Collectively, these findings reveal that TLR9 signaling promotes iron-dependent mitochondrial metabolism, linking Mitoferrin-1-mediated iron homeostasis to OxPhos and TLR9 responses in neutrophils. To validate these findings, we measured IFN-α secretion in neutrophil supernatants following TLR9 stimulation (Fig. 3k). We found that TLR9 stimulation promoted increased IFN-α secretion, while DFX treatment significantly impaired IFN-α secretion (Fig. 3k), reinforcing the link between iron availability and interferon signaling. In vitro blocking of IFNAR1 (anti-IFNAR1) led to reduced IFN-α secretion (Fig. 3k) and decreased mitochondrial activity (Supplementary Fig. 4l), while other inflammatory cytokines such as IL-8 remained unchanged (Supplementary Fig. 4m). We hypothesized that secreted IFN-α could contribute to a positive feedback loop, amplifying neutrophil activation. To test this hypothesis, we stimulated neutrophils with exogenous IFN-α in the presence or absence of an IFNAR1-neutralizing antibody (Fig. 3l, m). Prior to the IFN-α challenge, we added monensin, a protein transport inhibitor, to block Golgi transport and cytokine secretion, ensuring that paracrine/autocrine IFN-α was not involved. We observed that exogenous IFN-α stimulation promoted IFN-α production to a comparable degree as ODN2006-treated neutrophils (Fig. 3l, m). These experiments collectively suggest that IFN-α signaling may sustain an autocrine loop in neutrophils in response to TLR9 stimulation. Next, we sought to identify the endogenous ligand that could trigger the IFN-α/IFNAR1 axis in IBD and COPD in situ (Fig. 3n). Considering the DNA nature of ODN2006, we hypothesized that endogenous DNA could act as the trigger, as previously shown in COPD.21 We explored whether autologous mitochondrial DNA or NETosis products (like H3Cit) could promote IFN-α production. H3Cit, but not mitochondrial DNA, facilitated IFN-α production (Fig. 3o, and Supplementary Fig. 4n). In contrast, SLC25A37-depleted neutrophils did not respond to H3Cit (Fig. 3o), indicating a direct link between Mitoferrin-1 and IFN-α production in response to possible in situ triggers. To investigate the molecular mechanisms responsible for IFN-α production, we treated neutrophils with ODN2006 and a pan-TLR inhibitor (MyD88i), which resulted in a significant decrease in IFN-α production (Supplementary Fig. 4o). In contrast, treatment with a cGAS-STING inhibitor did not decrease IFN-α levels (Supplementary Fig. 4o), suggesting that TLR, rather than cGAMP sensors including STING, is the main sensor facilitating IFN-α production. Since MyD88i treatment impaired IFN-α production, we hypothesized that neutrophils sense DNA through TLR9. We therefore examined the neutrophil TLR9 response in recognizing NETosis components (H4 and H3Cit). Chloroquine, which blocks endosomal TLR signaling,22 potently inhibited H4-and H3Cit-induced IFN-α production in a dose-dependent manner (Fig. 3p). This suggests that histones derived from NETosis may trigger TLR9 responses to promote IFN-α production. While these findings highlight a mechanism by which NET-derived H3Cit can promote neutrophil activation, neutrophils in vivo are exposed to multiple signals, including cytokines and type I IFNs from other cellular sources. Thus, H3Cit likely represents one of several ligands that contribute to sustaining the IFN-α/IFNAR1 positive feedback loop in inflamed mucosal tissues. Supporting this hypothesis, a subset of neutrophils identified in infection models have been shown to express interferon-stimulated genes (ISGs), including Isg15.21,22 In the context of mucosal inflammation, we identified SLC25A37 upregulation in neutrophils from COPD and IBD patient datasets (Supplementary Table 1), and observed elevated expression of Slc25a37, Opa1, and Mtfn2 in murine neutrophils in an in vivo DSS-colitis model (Fig. 3q, r). These findings suggest that the metabolic changes in neutrophils are conserved across species. However, given the molecular and functional differences between murine and human neutrophils, we also investigated tissue inflammation using a human system. For IBD modeling, we employed a co-culture of neutrophils with human colon organoids.23 To mimic the inflammatory environment, we pre-treated the gut organoids with a cytokine cocktail (IFN-γ, TNF-α, and IL-1β).23 This treatment increased MCP-1 secretion, epithelial permeability, and neutrophil migration (Supplementary Fig. 4p–s), confirming the establishment of an inflamed state. We then co-cultured these organoids with wild-type or Mitoferrin-1-depleted neutrophils (Fig. 3s). Through live-cell imaging, we observed that under inflammatory conditions, infiltrating neutrophils showed higher mitochondrial potential, while Mitoferrin-1-depleted neutrophils failed to exhibit the same response (Fig. 3s, t). These findings support that in an inflammatory condition, gut-infiltrating neutrophils may exhibit mitochondrial polarization. Therapeutic IFNAR1 blockage ameliorates inflammation in models of IBD and COPD To explore the implications of a positive feedback loop between type I IFN and its receptor on IBD and COPD inflammation at the holistic level, we investigated whether anti-IFNAR1 treatment could ameliorate the inflammatory milieu and disease activity (Fig. 4). Pharmacological inhibition of the IFNAR1 signalling using anti-IFNAR1 attenuates the inflammation in DSS-colitis and CS-COPD models. a Schematic representation of the DSS-induced colitis model treated with anti-IFNAR1 antibody or IgG control. b Colon weight-to-length ratio (g/cm). Statistical analysis was performed using an unpaired Student’s t-test after confirmation of normality. c Total histological score based on leukocyte infiltration, tissue damage, and ulceration. Statistical analysis was performed using an unpaired Student’s t-test after confirmation of normality. d Representative hematoxylin and eosin (H&E) staining of colon sections. Insets show enlarged views highlighting crypt morphology. e Quantification of crypt integrity. f Representative immunofluorescence images showing neutrophil infiltration in the colon, detected by Ly6G staining. Nuclei were counterstained with DAPI. g Quantification of infiltrating neutrophils in the colon. Data passed the Shapiro–Wilk normality test and were analyzed using one-way ANOVA followed by Tukey’s multiple comparison test. h Representative immunofluorescence images showing in situ histone H3 citrullination (H3Cit). i Quantification of H3Cit-positive cells. Data passed the Shapiro–Wilk normality test and were analyzed using one-way ANOVA followed by Tukey’s multiple comparison test. j Schematic representation of the cigarette smoke (CS)–induced COPD model treated with anti-IFNAR1 antibody or IgG control. k Quantification of infiltrating neutrophils in room air (RA) and cigarette smoke (CS)–exposed animals treated with or without anti-IFNAR1 antibody (500 µg/mouse). l Representative histological images of lung sections from IgG- or anti-IFNAR1-treated animals. m Quantification of the mean linear intercept (MLI) in lungs from RA and CS animals treated with anti-IFNAR1 antibody or IgG control. Each dot represents the average of three fields of view from one section. n Cytokine quantification in lung homogenates. o Quantification of mitochondrial membrane polarization in infiltrating neutrophils from RA and CS animals treated with or without anti-IFNAR1. Statistical analysis was performed using one-way ANOVA followed by Tukey’s multiple comparison test. p Representative immunofluorescence images of lung sections from RA or CS animals treated with IgG or anti-IFNAR1. Neutrophils are shown in magenta (Ly6G), nuclei in blue (DAPI), and NETs in cyan (H3Cit). q Quantification of H3Cit median fluorescence intensity normalized to the total lung area (RA IgG n = 8, RA IFNAR1 n = 8, CS IgG n = 7, CS IFNAR1 n = 7; animals from the first experimental batch). Data passed the Shapiro–Wilk normality test and were analyzed using one-way ANOVA followed by Tukey’s multiple comparison test. Data are presented as mean ± SEM We first used an acute DSS-colitis model (Fig. 4a) and observed that IFNAR1 antibody-treated animals displayed lower weight to length colon ratio (Fig. 4b), improved (lower) histology score (based on ulceration, damage and leukocyte infiltration) (Fig. 4c), lower levels of circulating neutrophils (Supplementary Fig. 5a), lower neutrophil activation (Supplementary Fig. 5b, c), lower mitochondrial polarization (Supplementary Fig. 5d) and lower crypt loss (Fig. 4d, e, and Supplementary Fig. 5e). Anti-IFNAR1 treated animals also exhibited a better disease activity index (DAI) at earlier time points of the treatment (Supplementary Fig. 5f) and less body weight loss (Supplementary Fig. 5g). This decrease in neutrophil activity aligns with changes observed in the SLC25A37 in vitro depleted neutrophils. Also, in the colon, we found less neutrophil infiltration (Fig. 4f, g) and NETosis (Fig. 4h, i). The decreased presence of NETosis may reflect the decreased infiltration status. However, by the end of the DSS-colitis experiment we did not observe significant difference in DAI (Supplementary Fig. 5f). This may suggest that DSS-colitis, a model that directly compromises epithelial integrity, may be too acute to observe significant improvements in DAI at later stages due to the damaged epithelium being the initiating insult, rather than neutrophil recruitment only. Collectively, our data suggested that the anti-IFNAR1 treatment decreased the number of circulating neutrophils, in situ, and their activation in the DSS-colitis model. Next, we employed a cigarette smoke (CS)-induced lung inflammation mouse model to investigate the presence of IFNAR1+ neutrophils. We exposed mice to CS twice per day, 5 days per week for 2 weeks (Supplementary Fig. 5h). We observed that CS-treated mice displayed increased numbers of IFNAR1+ neutrophils in their lungs (Supplementary Fig. 5i). These neutrophils were not only activated (IFNAR1+CD62L-) based on the CD62L shedding (Supplementary Fig. 5i) but also displayed higher mitochondrial polarization (Supplementary Fig. 5j), like TLR9-stimulated neutrophils in vitro in terms of mitochondrial phenotype. We then sought to explore if the IFNAR1 blockade could inhibit neutrophilic inflammation and lung pathology in this model. Mice were treated with anti-IFNAR1 or IgG control monoclonal antibodies and exposed to CS or room air (RA) (Fig. 4j). IFNAR1-treated animals displayed less neutrophils in their lungs (Fig. 4k, and Supplementary Fig. 5k). We also observed decreased numbers of monocytes (Supplementary Fig. 5l), which may be related to the fact that neutrophils produce chemokines that attract other innate immune cells. In addition, anti-IFNAR1-treated mice also displayed lower mean linear intercept (MLI) between their alveoli, indicating less tissue damage (Fig. 4l, m). We observed no differences in the bone marrow or blood of these animals, suggesting that the decreased numbers of infiltrating neutrophils in the IFNAR1 antibody-treated mice is due to decreased migration, rather than increased granulopoiesis (Supplementary Fig. 5m, n). Supporting this hypothesis, we observed decreased levels of inflammatory cytokines and chemokines in lung homogenate of IFNAR1 antibody-treated mice (Fig. 4n), including KC, TNFα, MCP-1, IL-6 and IL-12 (Fig. 4n, and Supplementary Fig. 5o), consistent with attenuated pulmonary inflammation. We observed increased levels of IFN-α in the lung homogenate from CS mice, which was dampened in the anti-IFNAR1 treated animals (Supplementary Fig. 5p). Corroborating these findings, when treated with anti-IFNAR1, we observed that the remaining infiltrating neutrophils in the CS model showed lower mitochondrial polarization (Fig. 4o), while the neutrophils within the CS model treated with IgG showed higher mitochondrial polarization compared to RA animals. Moreover, high dimensional flow cytometry analysis revealed the presence of an active neutrophil population present in the CS IgG treated group (Supplementary Fig. 5q). This population exhibited low CD62L and CXCR4, and high CD63 (supplementary Fig. 5r). This population also expressed high levels of IFNAR1 (Supplementary Fig. 5r). Supporting the protective role of anti-IFNAR1 treatment, we did not observe the presence of this active neutrophil population in CS-IFNAR1 treated mice (Supplementary Fig. 5q, r). We also observed decreased NETosis in the smoked animals treated with anti-IFNAR1 (Fig. 4p, q). These data is in line with previous reports indicating that the Ifnr1-/- mice have low myeloperoxidase (MPO) activity in their lungs.24 It is important to note that the anti-IFNAR1 antibody treatment used in these experiments is not neutrophil-specific and may affect multiple cell types in the tissue. Nevertheless, we chose to explore IFNAR1 blockade to understand the holistic consequences of the treatment, thereby mimicking a therapeutic approach. Despite this lack of cellular specificity, our data demonstrate that global anti-IFNAR1 treatment effectively reduces neutrophil activation, mitochondrial polarization, infiltration, and NETosis in both DSS-colitis and cigarette smoke-induced lung inflammation models. Collectively, these findings indicate that systemic IFNAR1 inhibition can dampen inflammation in vivo and correlates with decreased neutrophil infiltration in the affected tissue, supporting the potential of IFNAR1-targeted therapies for modulating excessive neutrophil activity in inflammatory diseases. Disrupting the IFNAR1 feedback loop mitigates inflammation in human organoid models of intestinal and pulmonary inflammation Given the molecular and functional differences between murine and human neutrophils,25 and that the monoclonal IFNAR1 treatment does not specifically target neutrophils, we investigated the molecular cues driving tissue inflammation employing a human system. We first validated that upon TLR9 agonist treatment, IFNAR1-depleted neutrophils displayed lower degranulation and ROS production, while phagocytosis remained unchanged (Fig. 5a). Blocking IFNAR1 with anti-IFNAR1 also only impaired ROS and degranulation, but left phagocytosis unaltered (Fig. 5b). We also observed that neutrophils treated with exogenous IFN-α showed increased ROS production and degradation, both activities were decreased when the neutrophils were treated with anti-IFNAR1 antibody prior to the challenge (Fig. 5b). These results are particularly interesting in the context of our SLC25A37 findings, which demonstrated its role in regulating key neutrophil functions, including phagocytosis. IFNAR1 depletion did not impair phagocytosis, indicating that while inflammatory activation is dampened, essential neutrophil functions necessary for maintaining tissue integrity are preserved (Fig. 5a), underscoring the more selective nature of IFNAR1 modulation. Neutrophil-specific IFNAR1 depletion or Mitoferrin-1 depletion impairs inflammation and neutrophil infiltration in intestinal organoid models. a Validation of IFNAR1 depletion in freshly isolated neutrophils. Left: quantification of IFNAR1 expression. Right: functional characterization of control (scramble gRNA, grey) and IFNAR1-depleted neutrophils (navy) following stimulation with ODN2006. b Functional validation of IFNAR1-dependent neutrophil responses. Neutrophils were treated with ODN2006 in the presence or absence of a MyD88 inhibitor (MyD88i, 30 µM). Each dot represents an individual donor. “+” indicates addition of the specified treatment. c Schematic representation of the colon organoid–neutrophil co-culture system. Human colon organoids derived from healthy donors were maintained under steady-state conditions or exposed to an inflammatory cocktail. After 24 h, organoids were re-embedded and co-cultured with neutrophils transfected with scrambled or IFNAR1-targeting gRNA. A representative image is depicted in detail. Magenta represents neutrophils and nuclear staining shown in cyan. d Representative images showing neutrophil infiltration and IL-6 production in colon organoid–neutrophil co-culture with control (guide SRC), IFNAR-depleted or Mitoferrin-1-depleted neutrophils. Quantification of e neutrophil infiltration and f IL-6 production in colon organoids. Data are derived from 8 neutrophil donors for IFNAR1-deficient neutrophils, from 6 neutrophil donors for Mitoferrin-1-deficient neutrophils and 2 organoid donors. g IL-6 quantification observed in healthy colon organoids treated or not with cytokine cocktail and IL-6 levels in ulcerative colitis organoids. h Representative images of co-cultures using IBD-derived colon organoids, showing neutrophil infiltration, cleaved caspase-3 (cCaspase-3), and IL-6 production. Quantification of i neutrophil infiltration, j IL-6 production, and k cleaved caspase-3 levels in IBD organoids. Data are derived from 6 neutrophil donors and 2 organoid donors. Each dot represents an individual organoid. l Quantification of neutrophil infiltration in colon organoids treated or not with cytokine cocktail; neutrophils were treated prior co-culture with protease inhibitor, NAC or Cl-Amide (PAD4i) (n = 3). Data are presented as mean ± SEM. Statistical analyses are indicated in the corresponding panels To dissect the neutrophil-intrinsic role of IFNAR1 and Mitoferrin-1 in human inflammation, we generated neutrophils with specific IFNAR1 or SLC25A37 knockdowns using CRISPR/Cas9 (gRNA), and subsequently co-cultured them with relevant human organoids. This approach allowed us to selectively interrogate the contribution of IFNAR1-expressing neutrophils and the importance of Mitoferrin-1 in mediating neutrophil inflammation in human-related models. For the IBD modeling, colon organoids were pre-treated with or without a cytokine cocktail (Fig. 5c). Then, we co-cultured these organoids with wild-type or depleted neutrophils (Fig. 5c). First, we confirmed that in the inflammatory milieu, wild-type neutrophils significantly infiltrated the colon organoids (Fig. 5d, e). Contrasting, IFNAR1-depleted or Mitoferrin-1 depleted neutrophils showed reduced migratory capacity (Fig. 5d, e) and IL-6 production compared with wild-type neutrophils even in inflammatory conditions (Fig. 5f). To create a model that closely resembles IBD, we then used ulcerative colitis (UC) patient-derived organoids. Confirming the inflammatory component of IBD, we observed that UC organoids showed higher IL-6 levels at steady state (Fig. 5g), thus, to avoid a cytokine storm phenotype, we opted to keep the UC-neutrophil co-culture without the cytokine cocktail. Using the UC-derived organoids, we observed consistent findings: IFNAR1-deficient or Mitoferrin-1-deficient neutrophils displayed reduced migration (Fig. 5h, i) with lower IL-6 production (Fig. 5j) and fewer apoptotic events (Fig. 5k). In the IFNAR1-deficient co-culture, we also noticed decreased TNFα and IFNγ release (Supplementary Fig. 5a). These results support a role for Mitoferrin-1 and neutrophil-intrinsic IFNAR1 expression in sustaining intestinal inflammation. To explore which neutrophil function associated with the IFNAR1-IFN-α axis mediates the neutrophil infiltration in the gut organoids, we treated wildtype neutrophils with protease inhibitors to hinder the degranulation effect, N-acetyl-L-cysteine (NAC) to hinder ROS, or Cl-amide (PAD4 inhibitor) to hinder the NETosis effect. Surprisingly, there was no difference in the treatment, in which they all impaired neutrophil infiltration (Fig. 5l). This supports that blocking neutrophil function may impair neutrophil infiltration. To extend this to COPD, we generated a 3D alveolar model by exposing alveolar organoids (AO) to 0.1% cigarette smoke extract (CSE) for 5 days (Fig. 6a). CSE exposure enhanced neutrophil migration into the AO (Fig. 6b–e), upregulated CXCR2 expression (Fig. 6f), and increased alveolar cell apoptosis (Fig. 6g), without affecting neutrophil viability (Fig. 6h), thereby recapitulating COPD-associated pathology. We then employed this platform to test IFNAR1-depleted neutrophils (Fig. 6i). We observed that IFNAR1-deficient neutrophils exhibited impaired migration (Fig. 6i, j; Videos S3–S6; Supplementary Fig. 6b), reduced epithelial cell death (Supplementary Fig. 6c), reduced IFNγ, IL-6, and TNFα release (Supplementary Fig. 6d), and CXCR2 expression (Supplementary Fig. 6e, f). We also observed that neutrophils cultured with CSE-treated AO displayed increased IFN-α levels (Supplementary Fig. 6g). In contrast, the IFNAR1-depleted neutrophils failed to produce IFN-α (Supplementary Fig. 6h), thus suggesting the presence of a positive feedback loop mediated by IFN-α that binds to IFNAR1, leading to the maintenance of inflammation in the tissue. Next, we sought to determine the impact of Mitoferrin-1 in mediating the AO neutrophil infiltration. We then employed the AO platform using SLC25A37-depleted neutrophils (Fig. 6k). We found that SLC25A37-deficient neutrophils exhibited impaired migration (Fig. 6l) and reduced IL-6 production (Supplementary Fig. 6i). Neutrophil-specific IFNAR1 depletion or Mitoferrin-1 depletion impairs inflammation and neutrophil infiltration in alveolar organoid models. a Schematic representation of the alveolar organoid (AO) coculture model. Human AOs were cultured under steady-state conditions (DMSO) or exposed to cigarette smoke extract (CSE, 0.1%) for 5 days prior to coculture with neutrophils. b Representative images of AO cocultured with neutrophils, with neutrophils identified by CD66b and MPO staining. c Quantification of neutrophil infiltration into organoids. d Flow cytometry analysis of CD66b and CXCR2 expression in neutrophils (CD66b⁺) following coculture. e Quantification of neutrophil numbers and f CXCR2 expression levels on the infiltrating neutrophils. g Levels of cleaved caspase-3 in AO following coculture with neutrophils. h Viability of neutrophils in AO coculture conditions. i Representative images showing neutrophil infiltration into the AO. j Quantification of infiltrating neutrophils. Each dot represents a single organoid (1 AO donor, 3 neutrophil donors). k Representative images showing neutrophil infiltration into AO exposed or not to 0.1% CSE. l Quantification of infiltrating neutrophils. Each dot represents a single organoid (1 AO donor, 3 neutrophil donors). m Representative image showing neutrophil infiltration (green) and IL6 production (yellow) in airway healthy or COPD-derived organoids (2 airway organoid donors, 4 neutrophil donors). n Quantification of neutrophils infiltrating healthy or COPD-derived organoids. o Quantification of IL-6 production. Data are presented as the mean ± SEM. Statistical analyses are indicated in the corresponding panels Given that COPD-infiltrating neutrophils showed high expression of Mitoferrin-1, we also co-cultured COPD patient-derived airway organoids with control or Mitoferrin-1-depleted neutrophils. We observed increased neutrophil migration and IL-6 production in COPD organoids compared to healthy controls (Fig. 6m–o), both of which were dampened by Mitoferrin-1 depletion (Fig. 6n, o). To identify the neutrophil functions driving tissue damage, we applied targeted pharmacological inhibitors (NAC, PAD4 inhibitors, or protease inhibitors) to the co-cultures. Blocking these pathways reduced apoptotic events in the COPD organoids (Supplementary Fig. 6j). While this approach does not fully replicate the complex in vivo microenvironment, it provides a powerful platform to directly assess neutrophil-driven inflammatory mechanisms, highlighting the potential of targeting neutrophil-specific IFNAR1 signaling to modulate tissue damage in human inflammatory diseases. Collectively, our data demonstrate that mucosal inflammatory cues like H3Cit activate neutrophils to elicit a TLR9 response. This response elevates Mitoferrin-1 levels, driving mitochondrial iron uptake and OxPhos to sustain IFN-α production. Through a positive feedback circuit, this IFN-α maintains tonic activation, perpetuating mucosal inflammation and recruiting additional immune cells. Our findings identify a conserved, neutrophil-intrinsic TLR9–Mitoferrin-1–IFNAR1 loop as a central driver of neutrophil pathogenicity across IBD and COPD, revealing IFNAR1 blockade as a promising strategy to selectively restrain neutrophil-mediated tissue damage in chronic inflammatory diseases. Discussion In this study, we uncovered shared neutrophil-driven inflammatory mechanisms between IBD and COPD by integrating scRNA-seq datasets and performing functional validation using an in vitro human neutrophil model. By employing a functional screen, we identified a central role for mitochondrial metabolism, particularly regulated by Mitoferrin-1, which established an autocrine type I interferon (IFN-α) loop that amplifies neutrophil activation through IFNAR1 signaling. Finally, by employing two different in vivo models that recapitulate key pathological features of IBD and COPD, we found that anti-IFNAR1 treatment hinders inflammation and neutrophil activity. These findings were then confirmed using human 3D organoids spiked with native or IFNAR1-depleted neutrophils, showing that IFNAR1-deficient neutrophils are less active and promote less tissue damage. Recent studies have identified distinct neutrophil subsets characterized by unique phenotypic markers and specialized functions.26 Among the factors influencing neutrophil heterogeneity, a growing body of evidence suggests that neutrophil metabolism may be a key driver of this functional diversity26. Under homeostatic conditions, neutrophils primarily depend on glucose metabolism via glycolysis and the pentose phosphate pathway to meet their energy demands and produce NADPH,27 which is essential for their antimicrobial effector functions.28 However, in inflammatory diseases, this metabolic profile can shift. For instance, unstimulated neutrophils isolated from patients with IBD exhibit increased oxidative metabolism compared to healthy controls.29 Interestingly, in IBD patients, this enhanced oxidative activity negatively correlates with disease severity, suggesting a potential compensatory or regulatory mechanism that diminishes as inflammation progresses.30 Similarly, in COPD, intracellular NAD biosynthesis has been implicated in regulating neutrophil development, activation, and survival, potentially contributing to chronic inflammation.31 Our data indicate that mitochondrial fusion and iron uptake are critical for maintaining elevated oxygen consumption and mitochondrial polarization. Taken together, this data suggests that during chronic inflammation, neutrophils increase the expression of Mitoferrin-1, a mitochondrial iron transporter, to promote mitochondrial fusion and OxPhos. Mitochondrial dynamics are increasingly recognized as key modulators of innate immune cell function, particularly in response to infection stress and chronic inflammation. Mitochondrial fusion, a process orchestrated by mitofusin-1 (MFN1), mitofusin-2 (MFN2), and OPA1,32 has been implicated in the regulation of NET formation. OPA1-dependent ATP production is essential for NETosis and is critical for host defense against Pseudomonas aeruginosa.33 Beyond bacterial infections, mitochondrial function also plays a pivotal role in antiviral responses.34 Mitochondrial antiviral-signaling protein is central to viral sensing,35,36 and recent evidence has demonstrated that mitochondrial fusion is regulated by the SIRT3–OPA1 axis, which is essential for defense against human cytomegalovirus.37 In this study, we demonstrate that neutrophils can sense NET-derived products through TLR9, a process that elicits mitochondrial OxPhos. Specifically, Mitoferrin-1-mediated mitochondrial fusion enhances OxPhos, promoting the production of type I interferons via TLR9 signaling. NETs, which contain citrullinated histone H3 (H3Cit) as a key component, have been implicated in multiple inflammatory diseases. Accordingly, in tuberculosis, it was reported that type I IFN-mediated NET release is associated with granuloma caseation38 and can activate plasmacytoid dendritic cells (pDCs), driving chronic pDC activation and facilitating the onset of autoimmunity in systemic lupus erythematosus (SLE).39 Similarly, in IBD, NETs contribute to mucosal tissue damage and local cytokine release, while in COPD, NETs are abundant in the airways and correlate with disease severity and neutrophil-driven inflammation. These observations indicate that IFN-α may drive neutrophils toward a “pathogenic” phenotype capable of amplifying tissue inflammation. For example, cigarette smoke has been reported to promote neutrophil CXCL8 production in a TLR9 activation-dependent manner,40 and neutrophils isolated41 from SLE patients display a proinflammatory phenotype characterized by type I IFN synthesis. Collectively, these findings identify a novel mechanism by which sterile inflammation may be propagated through mitochondria-fueled NET sensing, with H3Cit-containing NETs serving as a key mediator in both IBD and COPD pathogenesis. Type I IFNs signal through the heterodimeric IFNAR1/IFNAR2 receptor. Activation triggers TYK2, phosphorylating STAT1 and STAT2 to drive nuclear ISG complex formation41. Notably, TYK2 impairment via the rs34536443 and rs12720356 SNPs protects against IBD.42 While global Ifnar1 knockout mice are more susceptible to acute DSS-induced colitis, global loss of IFN-I signaling during the recovery phase favors tissue repair by reducing epithelial apoptosis and monocyte recruitment.43,44 This highlights differing IFNα-IFNAR1 dynamics between acute and chronic inflammation. Because systemic treatments are complicated by IFNAR1 expression across multiple cell types, we utilized organoid co-culture models to achieve neutrophil-specific Ifnar1 deletion. These targeted experiments confirm that neutrophil-intrinsic IFNAR1 modulates inflammatory responses and tissue repair, supporting our proposed differential mechanism. Finally, because systemic Type I IFN blockade risks compromising antiviral defenses, targeting upstream components offers a safer alternative. For instance, a small-molecule TLR9 antagonist could disrupt this pathogenic cascade while preserving antiviral redundancy via TLR7, TLR8, and RIG-I-like receptors. Collectively, our findings highlight a mitochondrial–NET–TLR9–type I IFN axis as a primary driver of mucosal inflammation. While Mitoferrin-1 is an essential component of this circuit, its fundamental roles in neutrophil pathogen sensing may limit its utility as a therapeutic target. Instead, our data suggest that targeting NETosis-TLR9 may represent a novel strategy for selectively disarming neutrophil pathogenicity. By intervening at this specific signaling node, we can effectively hinder the feedforward loop that drives ROS and NETosis, attenuating chronic tissue damage, without compromising essential antimicrobial functions such as phagocytosis. This makes TLR9-IFNa-IFNAR1-directed therapies particularly attractive for chronic mucosal conditions where maintaining host defense is as critical as dampening inflammation. Methods Animals Animal studies were approved by the relevant local regulatory authorities (BS Kantonales V, cantonal approval number #3065 and #3090, national number #35788). All experimental protocols adhered to both federal and cantonal regulations. The experiments were carried out in AAALAC-accredited facilities at F. Hoffmann-La Roche in Basel. Mice were housed under specific-pathogen-free conditions with continuous health surveillance throughout the studies. Sampling of mice Terminal blood collection, lung perfusion followed by lung, spleen and bone collection. Anesthesia: terminal anesthesia, 150 mg/kg body weight pentobarbital diluted in 0.9% NaCl physiol. I.p. CS model A total of 30 female C57BL/6J mice from CRL-F of the age of 9w were included in the study. Having 7 animals for the CS group and 8 animals for the RA group, housed 3–4 animals per cage. Smoking challenge Mice were exposed to whole-body mainstream CS generated from 1R6F research cigarettes (29.1 mg tar/1.896 mg nicotine) using the following standard parameters (ISO 1991): one 60 ml puff of 2 s duration followed by 58 s of fresh air. The cigarette smoke is directed in the exposure chamber (5 liters volume). Flow rate is constant at 0.833 L/min. Animal treatment IFNAR1-inhibiting antibodies (P1AQ1038) or isotype control (P1AQ1040) were dosed twice per week via 200 μL i.p. injection. Therefore, the concentration of the solution was 2.5 mg/mL. Mice were monitored and weighed every day during the smoking exposure. Mice were assessed for difficulty in breathing, activity, posture, general behavior and body weight loss in the appropriate score sheet. Lungs were carefully isolated without the trachea and esophagus and split as follows: (a) Left lobe of lung for flow cytometry. Samples were kept on ice on a 24-well plate in PBS + EDTA. (b) Superior, middle and inferior lobe for homogenization for cytokine and chemokine assay. (c) Post-caval lobe was isolated into a histological cassette and then kept in 10% Formalin for 24 h before transferring to 4 °C PBS. Lung homogenization Lungs used for cytokine evaluation were weighed and snap-frozen in a 7 mL Precellys tube. To each sample 100 µL of lysis buffer (200 mM NaCl, 5 mM EDTA, 10 mM Tris, 10% glycerin and Roche cOmplere protease inhibitors) was added per 10 mg tissue. The samples were then homogenized in the Precelly tissue homogenizer at 5500 rpm, 3 × 20 s and 20 s pause. Afterward, the homogenate was transferred to a 2 mL Eppendorf tube and centrifuged for 15 min at 1500 × g at 4 °C. The supernatant was removed and transferred to a new 2 mL Eppendorf tube and centrifuged at 4 °C at 21,000 × g for 10 min. The supernatant was transferred on a 96-well plate and stored until the Meso Scale Discovery assay was performed (MSD). MSD U-R-Plex To evaluate the inflammatory cytokines IL-12, IL-6, IFN-γ, Rantes and TNF-α we used the MSD U-plex Custom biomarker (Cat. No. K15069M-1; Lot: 504404). Bone marrow collection for smoking model For flow cytometry, mouse bone marrow cells were isolated from pooled femur and tibia by flushing them with cold phosphate-buffered saline (PBS) containing 2% FCS. The fluxed material was filtered in a 40 mm filter and transferred in a 50 mL conical tube. The samples were centrifugated for 5 min at 350 × g at 4 °C. The supernatants were removed and 5 mL of ACK Lysing Buffer (Thermo Fisher Scientific) was added. Cells were then resuspended in FBS-BD-FACS staining buffer and washed twice by 5 min at 350 × g at 4 °C centrifugation. About 1 million cells were incubated with an antibody cocktail containing LyG6 (clone 1A8, Biolegend), Ly6C (clone HK1.4, Biolegend), CD11b (clone M1/70, Biolegend), CD62L (clone MEL14, Biolegend), IFNAR1 (clone MAR1-5A3, Biolegend), CD63 (clone NVG2, Biolegend), CD45 (clone S18009F, Biolgened), Lineage (F4/80 clone BM8, CD3 clone 17A2, B220 clone RA3-6B2) and viability dye. Samples were acquired using an Aurora Cytek cytometer. Dextran sulfate sodium (DSS) colitis model A total of 24 female C57BL/6J mice from CRL-F of the age of 9w were included in the study. The animals were separated into 3 groups (8–7 animals per group). The DSS challenge was performed by adding 2.00% DSS (SKU 0216011090, MP Biomedicals) to the mice drinking water. DSS-colitis treatment IFNAR1-inhibiting antibodies (InVivoMab, Bio X Cell, P1AQ1038) or isotype control (P1AQ1040) were dosed twice per week via 100 µL i.p. Injection (500 µg/100 µL). Daily scoring (Day 4–12) Animals were placed into an empty cage to assess fecal samples, during this time, the body weight was taken and the animals were scored. To assess severity of colitis, the Disease Activity Index (DAI) was calculated as a sum of the scores of body weight loss, stool consistency and blood in stool. Tissue collection For blood collection after euthanasia, we retrieved about 500–700 µl from the right ventricle. The blood was transferred to an EDTA tube. The syringes were also previously coated with PBS-EDTA. For the colon, we dissected the gut and separated the small intestine from the colon. Next, we removed fecal pellets and selected one centimeter of the mid to distal section for histology and kept them in 4% PFA overnight, then transferred them to PBS for at least 24 h until further embedding in paraffin. AI-based histological image analysis for the colon sections Transverse sections of the distal colon were stained with H&E as previously described. Images were acquired at 40× magnification NA missing, producing one digital image per mouse, with 1–4 tissue sections per image. Digital image analysis was performed using the commercial software Visiopharm (Version 2024.07.1.16912 ×64). The workflow was developed to reproduce manual scoring criteria described by Smith et al.45 and Dieleman et al.,45,46 focusing on histological features relevant to colitis, specifically ulceration and crypt damage. The analysis pipeline was internally validated by comparison to manual scoring across multiple DSS-induced acute colitis studies. Each step of the pipeline was optimized through an iterative process: representative fields of view were manually annotated, artificial neural networks were trained, and performance was assessed on randomly selected images. A step was considered validated if visual confirmation of results reached ≥90% accuracy. Once validated, the step was applied to at least 50 images from various studies and disease states. If performance did not meet criteria, further manual annotations were added, networks were retrained, and post-processing steps were adjusted. This iterative process was repeated until consistent alignment with ground truth was achieved. Detailed image analysis workflow Tissue segmentation: Whole transverse tissue sections were detected using a DeepLabv3+ neural network at 1× digital magnification. The model was trained on two classes (background, tissue) with 76,891 iterations and a final loss function value of 0.001. Detected tissues were enumerated to determine tissue count per image. All downstream analyses were performed on individual tissue sections, and final per-mouse values were calculated as the mean of all tissue sections. Tissue segmentation: Whole transverse tissue sections were detected using a DeepLabv3+ neural network at 1× digital magnification. The model was trained on two classes (background, tissue) with 76,891 iterations and a final loss function value of 0.001. Detected tissues were enumerated to determine tissue count per image. All downstream analyses were performed on individual tissue sections, and final per-mouse values were calculated as the mean of all tissue sections. Layer detection: Within each tissue region of interest (ROI), anatomical layers were identified using a DeepLabv3+ network trained with 112,000 iterations and a final loss of 0.017. Images were analyzed at 4× virtual magnification with four classes: background, lumen, mucosa, and gut-associated lymphoid tissue (GALT). GALT regions were excluded from all further analyses. Layer detection: Within each tissue region of interest (ROI), anatomical layers were identified using a DeepLabv3+ network trained with 112,000 iterations and a final loss of 0.017. Images were analyzed at 4× virtual magnification with four classes: background, lumen, mucosa, and gut-associated lymphoid tissue (GALT). GALT regions were excluded from all further analyses. Crypt detection and scoring: Within the mucosa, colon crypts were identified using a DeepLabv3+ network trained with 78,840 iterations and a loss value of 0.022. Images were processed at 10× virtual magnification using three classes: crypt area, crypt loss, and GALT. GALT regions were again excluded to ensure specificity. This step allowed the quantification of intact versus damaged crypt areas, enabling calculation of the percentage of crypt loss within the mucosa. Crypt detection and scoring: Within the mucosa, colon crypts were identified using a DeepLabv3+ network trained with 78,840 iterations and a loss value of 0.022. Images were processed at 10× virtual magnification using three classes: crypt area, crypt loss, and GALT. GALT regions were again excluded to ensure specificity. This step allowed the quantification of intact versus damaged crypt areas, enabling calculation of the percentage of crypt loss within the mucosa. Ulceration scoring: Ulceration was quantified by measuring the total interface length between the lumen and mucosa and comparing it to the length of interface where crypt loss directly abutted the lumen—indicating a loss of epithelial barrier and presence of ulceration. The ratio of ulcerated interface to the total lumen-mucosa interface was used to assign scores: <1%: Score 0 (no relevant ulceration); 1–10%: Score 1 (light ulceration); 10–50%: Score 2 (moderate ulceration); 50%: Score 3 (severe ulceration). Ulceration scoring: Ulceration was quantified by measuring the total interface length between the lumen and mucosa and comparing it to the length of interface where crypt loss directly abutted the lumen—indicating a loss of epithelial barrier and presence of ulceration. The ratio of ulcerated interface to the total lumen-mucosa interface was used to assign scores: <1%: Score 0 (no relevant ulceration); 1–10%: Score 1 (light ulceration); 10–50%: Score 2 (moderate ulceration); 50%: Score 3 (severe ulceration). Immunofluorescence staining Tissue immunofluorescence staining was performed using the Ventana platform. Initially, the slides were pre-warmed to 60 °C and incubated for 8 min. This was followed by deparaffinization, which consisted of three cycles of warming at 60 °C for 8 min each. Subsequently, the Cell Conditioning 1 (CC1) protocol was initiated, during which slides were heated to 95 °C and incubated for 4 min. A gradual temperature ramp was applied, increasing the incubation time by 8-min increments until a total of 88 min was reached. Following CC1, the DISCOVERY inhibitor protocol was applied. Primary antibodies were then manually added: Ly6G (CST, #87048s, 1:50 dilution) and H3Cit (Abcam, #ab5103, 1:50 dilution). The slides were incubated with the primary antibodies at 37 °C for 60 min. Next, the OMap anti-Rabbit HRP protocol (One Drop) was selected and incubated for 16 min. This was followed by the addition of DCC H₂O₂ (One Drop), with a 20-min incubation. The Dual Sequence (DS) Cell Conditioning 2 (CC2) protocol was then carried out. Slides were heated to 100 °C and incubated for 4 min, followed by three 8-min increments for a total incubation of 24 min. Secondary antibodies were then applied. The slides were gradually warmed to 37 °C from low temperature and incubated for 60 min. The DS Multimer HRP method was used next, followed by the OMap anti-Rabbit HRP protocol (One Drop), incubated for 16 min, during which the DS Cy5 detection system was employed. A Cy5 H₂O₂ reagent (One Drop) was then applied and incubated for 20 min. Finally, DAPI was used as a nuclear counterstain. Cell typist analysis To harmonize and annotate the scRNA-seq datasets obtained from IBD and COPD cohorts, we applied CellTypist (v1.5.0) using both pre-trained and custom-built classification models. Prior to annotation, each dataset was processed independently in Scanpy (v1.9.3) following standard quality-control procedures: cells with fewer than 200 detected genes or >15% mitochondrial reads were removed, and genes expressed in fewer than 3 cells were excluded. Gene expression matrices were log-normalized (scanpy.pp.normalize_total, target sum = 1e4; scanpy.pp.log1p) and harmonized using scVI (v1.1.0) to generate integrated latent embeddings for downstream annotation. Custom neutrophil model construction To capture fine-grained neutrophil heterogeneity, we constructed a custom neutrophil-specific CellTypist model using the “custom model” framework. A training dataset was assembled based on the curated neutrophil transcriptional profiles described in reference.12 The training matrix was standardized to match CellTypist formatting, retaining only high-confidence neutrophil subtypes reported in the source publication. Model training was performed using CellTypist’s built-in logistic regression classifier with the following parameters: solver = “lbfgs”; max_iter = 10,000; class balancing enabled (--balanced); feature selection using CellTypist’s default top highly variable genes (2000 HVGs); cross-validation using fivefold internal splits. The resulting model file was saved in the CellTypist model directory and indexed for downstream predictions. Annotation using pre-trained and custom models For broad immune-cell identification, we first ran the Immune Cells v1 pre-trained classifier using the majority-vote prediction mode (celltypist annotate --majority_voting). Predictions were generated directly from the integrated latent representation to avoid batch-driven biases. Subsequently, the custom neutrophil model was applied to the same integrated dataset, restricted to clusters preliminarily identified as granulocyte-enriched based on Leiden clustering (resolution = 0.5). Annotation outputs from both models included class probabilities for each cell. When discrepancies occurred between model predictions for granulocyte clusters, the custom neutrophil classifier was given priority due to its specialized training. Final cell labels integrated:(1) probability thresholds ≥0.6 for high-confidence labels, (2) majority-vote labels for lower-confidence predictions, and (3) manual curation in cases where predictions were ambiguous. Pseudo-bulk Seurat Single-cell RNA-sequencing (scRNA-seq) data from the subset and harmonized neutrophil annotation were processed and analyzed using Seurat in R. After initial quality control, cells were filtered based on mitochondrial RNA content, total UMI counts, and the number of detected features. Counts were normalized using SCTransform (or NormalizeData, depending on what you used—swap as needed), and the resulting object was used for downstream analyses. Following clustering and annotation, pseudobulk profiles were generated to perform differential expression analysis at the sample level. For each biological sample, raw counts were aggregated across all cells belonging to a specific cell type or cluster. This summation yielded a pseudobulk expression matrix in which each column corresponded to a sample-level replicate rather than an individual cell. The pseudobulk matrices were then used as input for differential gene expression (DEG) testing. For each cell type, aggregated counts were compared between experimental conditions using the Seurat wrapper for FindMarkers() configured to operate on pseudobulk data (i.e., specifying the aggregated assay). Genes were considered differentially expressed if they met the thresholds of |log2 fold-change| ≥ X and an adjusted p-value ≤ Y (FDR correction via Benjamini–Hochberg). To ensure that DEGs reflected biological rather than technical differences, only cell types with at least N cells per sample and at least M samples per condition were included in the analysis. Quality checks were performed to confirm that pseudobulk profiles recapitulated expected sample-level structure, including PCA and hierarchical clustering. All analyses were performed using the R environment. In vitro neutrophil differentiation Progenitor cell expansion CD34+cells (Lonza) from adult or cord blood were thawed in X-VIVO-10. The samples were centrifuged at 300 g for 10 min to remove the freezing solution and resuspended in Progenitor cell expansion media (PEM) containing X-VIVO-10 media (Lonza), 20% BIT 9500 Serum Substitute (BIT, Stemcell Technologies), SCF (100 ng/mL), IL-3 (50 ng/mL), FTL3L (50 ng/mL), IL-6 (50 ng/mL), G-CSF (100 ng/mL), TPO (10 ng/mL), insulin-Transferin-Selenium (1×) and DNAseI (1%). The progenitor cells were cultured in 96-U bottom plates in 100 µL PEM. Myeloid polarization medium (MPM) After 3 days in culture, the wells containing the progenitor cells were combined in a 50 mL Falcon tube and centrifuged at 300 × g for 10 min (break off). The pellet was resuspended in the Myeloid polarization media (MPM), containing X-VIVO-10, 20% BIT Serum Replacement, SCF (100 ng/mL), G-CSF (10 ng/mL) and DNAseI (1%). The cells were plated into 96-well plates (10,000 cells/well) with 100 µL MPM. Neutrophil polarization After 4 days in culture with MPM, the wells were combined in each well in a 15 mL Falcon tube and centrifuged at 300 g for 10 min (break-off). The pellet was resuspended in Neutrophil media having G-CSF (100 ng/mL), DNAseI (1%), Glutamine (0.1%), essential amino acids (0.1%) and Rosiglitazone (0.1 ng/mL). All the cytokines were purchased from Peprotech Proteins/Thermo Fisher Scientific. The culture was expanded to 6-well plates containing 2 mL of neutrophil media (0.8 × 106 cells). After 15–17 days, we used the stem cell-negative neutrophil selection following manufacturing recommendations (StemCell technology) and evaluated it based on cell morphology and flow cytometry (CD66b, CD11b and CD34). CRISPR-Cas9 mutagenesis Target selection for CRISPR-Cas9–mediated mutagenesis was performed based on Human GecKO library guides. The selected sgRNA having a GC content <55%, mismatches (MM) self-complementarity = 0, MM0 ≤ 1, MM1 ≤ 1, MM2 ≤ 1, MM3 ≤ 1 were selected. The sgRNA templates were generated using the protocol described by Gagnon et al.47 using the mMESSAGE mMACHINETM SP6 transcription kit (Invitrogen, AM1340). Editing efficiency was further validated by deep sequencing of PCR amplicons. For the transfection of the guides ~100,000 day 17 neutrophils were resuspended in P3 Primary solution (Lonza) containing the sgRNA (20 μM) and Cas9 Protein (TrueCut v2.0, 10 μM) and placed in 16-well stripes cuvettes from Lonza. The cells were then exposed to the DZ100 program and after transfection 100 μL of X-VIVO-10-20% BIT was added to the wells. The cells were kept untouched for 5 min at room temperature. Amplification was performed using Phusion High-Fidelity PCR Master Mix with HF Buffer (New England Biolabs, catalog no. M0530) under the following conditions: 98 °C for 3 min, 98 °C for 10 s, annealing time for 10 s, and 72 °C for 10 s ×35 cycles, and then 72 °C for 5 min. Then, PCR amplicons were purified on PCR purification kit columns before sequencing and sent to MicroSyth for Sanger sequencing. The sequencing results were analysed using the Tide software. NETosis imaging After neutrophil differentiation and isolation using the Stem Cell negative selection, we prepared a neutrophil stock of ~200,000 cells/mL in culture media (X-VIVO-10 Lonza, 10%-BIT). The cells were then cultured in 0.1%-Gelatin precoatted 96-well glass bottom plates at a concentration of 40,000 cells per well. The cells were allowed to settle at room temperature for 15 min. After, the plate was centrifugated at 70 × g for 3 min (no break) and 40 μL of Sytox Green Dye Solution was added to a final concentration of 250 nM, from stock 1:1000 Media (X-VIVO-10 Lonza). As a positive control, we added 10 µL of fMLP (Sigma, 10 μM) or LPS (Sigma). For negative controls, we used 1% DNAse 1 (Roche) or the PADI4 inhibitor (Chloramine, 10 μM). For the live imaging used for the CRISPR-Screen readout, the plate was incubated in the Incucyte (Sartorius), and images were taken for 6 h with an interval of 20 min/30 min. The relative fluorescence of green was used as a proxy for NET formation. For the antibody staining 4 h after challenge, the supernatant was carefully removed and 100 μL of the BD fix buffer was added. The plate was kept in the fridge overnight. Then, the supernatant was removed and the plate was washed with PBS by spinning the plate at 100 × g for 3 min. After the wash, we added an FBS-FACS BD Staining buffer containing anti-Histone H3 (citrulline R2 + R8 + R17) (Abcam) coupled with Zenon (ThermoFisher, Z25402), CD63 (FITC, BD AB_393869) and DAPI. The plate was incubated at room temperature for 2 h. After, the plate was washed twice with PBS and images were taken using the Leica Stellaris 8 LAS X confocal microscope. The functional characterization after the CRISPR-Cas9 perturbation was conducted 12 h after the transfection. Degranulation assay Human neutrophils were isolated from peripheral blood of healthy donors using the StemCell neutrophil negative isolation (#100-0404) or the CD34 in vitro-derived neutrophils after differentiation confirmation by flow cytometry were resuspended in X-VIVO-10 supplemented with 10% BIT 9500 serum substitute. Neutrophils (~106> cells/mL) were stimulated with fMLP (1 μM) or LPS (1 µg/mL) for 60 min at 37 °C, 5% CO2. CD63 detection by live-cell imaging For imaging, high-content imaging plates with a glass bottom (Millipore, #CLS4580-10EA) were coated with 0.1%-gelatin. Neutrophils (~106> cells/mL) were loaded with 100 µL. After 60 min the plates were centrifugated for 2 min at 70 × g. After centrifugation, 75 µL of the wells was removed and 75 uL of FBS-Staining buffer containing fluorescently conjugated anti-CD63 antibodies (clone H5C6, Alexa488). The antibodies were added together with the stimulation and loaded on the Leica X confocal antibody. Images were taken for 100 min having 20 min interval imaging. The microscope is equipped with an environmental chamber setted for 37 °C at 5% CO2. Image analysis was performed using mageJ/Fiji to quantify CD63 surface intensity over time as a proxy for degranulation. Hoechst was used as a counterstaining technique. CD63 detection by flow cytometry For flow cytometry, neutrophils were stimulated as described, then stained with fluorescently conjugated anti-CD63 antibodies (clone H5C6, Alexa488). At room temperature, protected from light for 30 min. Cells were washed with PBS containing 1% FBS and analyzed immediately on a flow cytometer BD Fortessa III. A minimum of 10,000 neutrophils were acquired per sample. Data were analyzed using FlowJo, gating on live neutrophils (based on FSC/SSC), Zombie dye-negative cells, CD66b+ and measuring median fluorescence intensity (MFI) of CD63 to assess degranulation. Phagocytosis assay Phagocytosis assays were performed using pHrodo™ E. coli (A488) for live-cell imaging or pHrodo™ S. aureus (A555) for flow cytometry. Cells were seeded in imaging-grade 96-well plates (neutrophils, 5 × 104 cells/well) and challenged with the appropriate TLR agonist. The pHrodo particles were prepared according to the manufacturer’s instructions and opsonized with 10% human serum for 30 min at 37 °C. Particles were added at a multiplicity of infection (MOI) of 20–50 and plates were incubated at 37 °C, 5% CO₂. For live-cell kinetic measurements, plates were imaged in an IncuCyte S3 system, acquiring phase contrast and green fluorescence channels every 15 min for 24 h. Fluorescence area per field was quantified using IncuCyte analysis software and normalized to cell confluence. For the CRISPR screening, cells in 96-well plates were incubated with pHrodo S. aureus (A555) overnight, and processed for flow cytometry. Cells were washed, stained with viability dye and surface markers as needed, and analyzed on an Aurora Cytotek. Gating excluded debris, doublets and dead cells; pHrodo-positive cells were identified against unstimulated cells. Outcomes reported include % pHrodo+ cells and MFI. Controls included cytochalasin D (5 µM) pre-treatment. All experiments were performed in ≥2 biological replicates. Statistical comparisons were performed using paired t-tests or one-way ANOVA as appropriate. Neutrophil migration assay toward C5a Neutrophil chemotaxis toward the complement component C5a was assessed using the IncuCyte live-cell imaging system. Briefly, AD or CB-derived human neutrophils or freshly isolated neutrophils were resuspended in migration medium (X-VIVO-10 supplemented with 20% BIT Serum replacement) at a density of 1 × 106 cells/mL. Cells were seeded in the upper chamber of an IncuCyte Chemotaxis plate or directly in the imaging well, and a gradient was established by adding recombinant C5a (100 nM) to the lower chamber. Plates were placed in the IncuCyte system, and phase-contrast images were acquired every 20 min for 6 h at 37 °C in a humidified CO2 incubator. Migration was quantified using IncuCyte software by measuring either the number of cells reaching the target region, the average distance traveled, or directional persistence over time. Negative control wells without C5a were included to account for random motility, and all conditions were performed in technical duplicates or triplicates. Real-time cell metabolic analysis For the real-time cell metabolic assay, the Seahorse XFp Cell Mito Stress Test Kit (Agilent Technologies, California, USA) was used following the manufacturer's recommendations. In brief, freshly isolated neutrophils or transfected neutrophils were plated as 60,000 cells/0.1% coated Gelatin wells. After plating, the cells were stimulated with the TLR agonists (indicated in the Figures) or with PMA as a positive control and immediately placed into the prepared Seahorse XF Pro Analyzer. For the analysis, we kept the edge wells containing only 50 µL/well XF media to account for background correction wells. The XFe96 microplate was centrifuged at 70 × g for 1 min with maximum acceleration but no brake at room temperature before incubation in a non-CO2, 37°C incubator for 30 min. 25 µL of each inhibitor/stimulant was added to injection ports A-C of the sensor cartridge (Agilent Technologies, California, USA) as follows, with stated concentrations referring to final well concentrations: Mitochondrial respiration: port A: 2 µg/ml oligomycin, port B: 3 µM FCCP and port C: 2 µM rotenone/antimycin A. A total of 3–5 cycles were settled for the baseline. Mitochondrial membrane potential measurement Mitochondrial membrane potential (Δψm) in neutrophils was assessed using the JC-1 dye (Thermo Fisher Scientific). Cells (106/mL) were incubated with 2 µM JC-1 in a pre-warmed medium for 20 min at 37 °C in the dark. After staining, cells were washed twice with PBS and analyzed by flow cytometry or fluorescence microscopy. Red fluorescence (J-aggregates) indicates polarized mitochondria, while green fluorescence (monomers) indicates depolarized mitochondria. Δψm was quantified as the red/green fluorescence ratio, with a lower ratio reflecting mitochondrial depolarization. CCCP (10 µM, 10 min) was used as a positive control for depolarization, and unstained cells served as negative controls. Mitochondrial ROS measurement Mitochondrial reactive oxygen species (mtROS) were assessed using MitoTracker™ Red CMXRos Far Red (deep red, ThermoFisher, M22426) according to the manufacturer’s instructions. Neutrophils were resuspended at 1 × 106 cells/mL in a pre-warmed culture medium and incubated with 5 µM MitoTracker™ Red CMXRos for 30 min at 37 °C in the dark. Following staining, cells were washed twice with PBS to remove excess dye. Fluorescence was detected using flow cytometry (excitation 640 nm, emission 665 nm) or live-cell imaging. mtROS levels were quantified as mean fluorescence intensity (MFI), and unstained cells were included as negative controls. Simple Western automated Western blot system (WES) Twelve hours after transfection, the human neutrophils were collected into 50 mL conical tubes and spinned at 350 × g for 5 min (no break). After, the pellet was resuspended in RIPA Lysis and extraction buffer (ThermoFisher) and total protein concentration was evaluated using the PierceTM Bicinchoninic Acid (BCA) protein assay (ThermoFisher). Then, the WES assay was conducted following the product recommendation. A total of 10 μg protein was loaded into the WES capillaries. The SLC25A37 (1:10 dilution, Invitrogen ref:PA5-26720) and Vinculin (1:200 dilution, Cell Signaling ref: 13901) were used. Flow cytometry and confocal microscopy Fluorochrome-conjugated monoclonal antibodies were purchased from eBioscience, BD Pharmingen, or BioLegend. Freshly isolated neutrophils, transfected primary neutrophils, or in vitro-derived neutrophils were quantified by direct staining with MitoTracker (Red CMXROS or GreenTM, ThermoFisher), CMxROS (deep red, ThermoFisher, M22426), and mito probe JC-1 staining was performed according to the manufacturer’s instructions. Intracellular cytokine production in the freshly isolated neutrophils was quantified 4 h after stimulation with TLRs and GolgiStop (BD). The cells were fixed and permeabilized using the transcription factor staining buffer set (eBioscience). For the phagocytosis experiments, fresh-isolated neutrophils or in vitro-derived neutrophils were stimulated with fMLP (Sigma), LPS (Sigma), or ODN2006 (Invitrogen) and incubated with the S. Aureus-pH-Rhodo or E. Coli-pH-Rhodo (ThermoFisher) for 4 h. Cells were acquired on Fortressa flow cytometers (BD Biosciences) or Aurora Cytek and analyzed using FlowJo (TreeStar) software. Cells were imaged live on glass bottom dishes coated with 0.1% Gelatin using a Leica Stellaris 8 confocal scanning microscope or with the Incucytes (Sartorius) using the non-adherent protocol and a 20× Nikon objective. Cells were kept in a humidified incubation chamber at 37 °C with 5% CO2 during image collection. Images were deconvolved and analyzed using ImageJ (NIH). Proteomics One million fresh isolated neutrophils, in vitro derived neutrophils, HL-60 and HL-60 stimulated with Phorbol myristate acetate (PMA; 1 μg/mL) were resuspended in 50 μL of LYSE-iST buffer (PreOmics kit, ref: P.O.00027) containing 0.5 μL TCEP 5 mM final (Bond-Breaker, ThermoFisher ref: 77720). The cells were incubated in a heating block at 95 °C, 1000 rpm for 10 min. After, the samples were sheared in the Bioruptor (10 cycles, 30 s ON/OFF). The total protein concentration was quantified with the BCA protein assay (ThermoFisher). Digestion and purification were conducted following the PreOmics kit instructions. Next, we performed the peptide assay following the Peptide Assay (ThermoFisher; ref: 23275) instructions. The samples were resuspended in LC-MS resuspension buffer (2% ACN-0.5% FA). For the MS measurements 7 μL from each 0.2 μg/μL sample was transferred to a 96-well plate. For the MS measurements, the Trap column DNV pepMap (Neo ref:DNV75150PN) and the 25 cm Aurora Series emitter column (25 cm×75 µm ID, 1.7 µm C18) (Ion Optiks; # AUR3-25075C18; Lot#IO2575028066) were used. The samples were acquired in the Vanquish Neo LC system and an Orbitrap Tribrids Ascend MS System. The data were analyzed using the Spectronaut software. Samples were injected randomly, and ~4.881 proteins were identified. RNAseq bulk RNA was extracted using the MiniRNA Isolation Kit (Thermo Fisher Scientific). The quality between was evaluated using the Fragment Analyzer. RNA that passed the quality control (QC) was then used to generate libraries using the NEBNext Library Ultra Low Input Library Preparation. The manufacturer’s recommendations were followed, and the libraries were sequenced on an Illumina NovaSeq6000 sequencer. All sequencing data were performed in three biological replicates, having a total depth of 15 million reads per reads. The reads were conducted at 2 ×150 base pair (bp) of length. Bulk RNA-seq bioinformatic analysis Reads for mouse bulk RNA-seq datasets were mapped following the default Biokit 3.9 (pRED workflow). Briefly, FASTA files were aligned against the human genome version GRCh38. Reads were counted with featureCounts. Differential expression analysis was performed with DESeq2. Multi-FASTA QC statistics indicated data were of high quality, and sequencing depth was sufficient to test for differential expression between conditions. Differentially expressed genes were called with an FDR threshold of 0.05. Correlations between datasets and graphical output were conducted in an R environment and CellTypist. Enzyme-linked immunosorbent assay (ELISA) The human Citrullinated histone H3 (H3Cit) was quantified from the supernatant of the in vitro-derived neutrophils stimulated for 12 h using the H3Cit ELISA kit (Cayman Chemicals, ref: 501620). Secreted IFN-α from 4 h stimulated fresh isolated neutrophils or the mouse tissue homogenate was quantified with the Human IFN-α ELISA Kit (ThermoFisher, ref:BMS6027) or the Mouse IFN-α ELISA kit (ThermoFisher, ref:BMS6027). Tissue source and ethical statement Human tissues, colon and lung, and associated clinical information were obtained from patients undergoing tumor resection at the Humanitas Research Hospital, Milan, Italy. Samples were sourced through a collaborative framework adhering to legal and ethical regulations. The project was reviewed and approved by Comitato Etico Territoriale Lombardia 5 (Ethical approval 3631). Additional tissue samples and annotated data were obtained, and experimental procedures were performed within the framework of the non-profit foundation HTCR, which includes written informed consent from all donors and has been approved by the ethics commission of the Faculty of Medicine in the LMU (number 025-12) and the Bavarian State Medical Association (number 11142) Colon organoids Human sample preparation, crypt isolation and organoid culturing Intestinal tumour-free regions were used for crypt isolation. First, the muscularis, serosa and fat from the basal side were removed with forceps and scissors. Mucus from the luminal side and blood vessels from the apical side were removed with a scalpel. The processed mucosal tissue was then washed five times with PBS supplemented with penicillin-streptomycin (Thermo, #15070063), gentamicin (Thermo, #15750060), and amphotericin (Thermo, J67049-AD). For the isolation of crypts, a 3 cm2 piece was scraped with a coverslip to remove the villi. The remaining tissue was then incubated in cold PBS supplemented with 10 mM EDTA for 30 min on a shaker. The crypts were afterwards scraped off with a cover slip and collected in organoid media (Advanced DMEM/F12 Gibco #12634028, HEPES Gibco #15630080 10 mM, Glutamax Gibco #35050061 1%, Wnt surrogate ImmunoPrecise Antibodies #N001-0.5 mg 0.15 nM, Noggin-CM ImmunoPrecise Antibodies N002 1:50, Rspo3-CM ImmunoPrecise Antibodies #R001-500 ml 1:50, hIGF-1 BioLegend #590908 100 ng/ml, hFGF2/FGF-basic R&D #3718-FB 50 ng/ml, Gastrin I MedChem Express #HY-P1097/CS-0027717 10 nM, A83-01 (ALK4/5/7 inhibitor) R&D #2939 0.5 µM, hEGF PeproTech #AF-100-15-1MG 50 ng/ml, B27 min VitA Thermo #12587010 1:50, N-acetylcysteine Merck/Sigma #A9165-5G 1 mM, Primocin Invivogen #ant-pm1 50 µg/ml). Crypts were centrifuged, the pellet was resuspended in Matrigel (Corning, #356231), and after polymerization (15 min, 37 °C) cultured in organoid media with 10 µM Y-27632 (STEMCELL Technologies, 72305). The organoids were split every week. Therefore, media was removed and after a PBS wash, Gentle Cell Dissociation reagent (STEMCELL Technologies, #07174) was added for 2 min RT. The domes were collected in Falcon tubes, further incubated for 15 min at RT and centrifuged. The pellet was resuspended in organoid-wash media (Advanced DMEM/F12 Gibco #12634028, HEPES Gibco #15630080 10 mM, Glutamax Gibco #35050061 1%, Primocin Invivogen #ant-pm1 50 µg/ml) and resuspended multiple times to break the organoids. This process was repeated before the pellet was resuspended in Matrigel. After passage three, a mycoplasma test was performed (Lonza, #LT07-710) before the organoids were used for experiments. For experimental preparation at day 3–4 after passing, the media was removed and after washing with PBS, the organoids were treated with cell recovery solution (Corning, CLS354253) and incubated for 40 min at 4 °C. The organoids were transferred into a Falcon tube, well resuspended and centrifuged. IBD derived organoids establishment and expansion Mucosal tissues were separated from the muscularis, serosa and fat using forceps and scissors. The processed mucosal tissue was then washed five times with PBS supplemented with penicillin-streptomycin (Thermo, #15070063), gentamicin (Thermo, #15750060), amphotericin (Thermo, J67049-AD). Tissue was minced into small pieces and subsequently enzymatically digested in 1 ml of advanced DMEM/F-12(GIBCO, #12634028) medium supplemented with 20 μ/ml and 20 μg/ml of P/S (ThermoFisher Scientific #15140122), Primocin 0.1 mg/ml (InvivoGen, #ant-pm-1), Hepes 10 mM (GIBCO, #15630056), and 2 nM Glutamax (GIBCO, #35050061), 2.5 mg/ml collagenase IV (Worthington, #LS004189), 0.1 mg/ml DNase IV (Sigma, #D5025), 20 μg/ml hyaluronidase V (Sigma, #H6254), 1% BSA (Sigma, #A3059), and 10 μM LY27632 (Abmole Bioscience, #M1817) for 1 h at 37 °C under slow rotation and vigorous pipetting every 15 min.The tissue lysate was filtered through a 100 μM cell strainer and centrifuged at 300 g for 10 min. The cell pellet was suspended in PBS, and cells were counted using trypan blue (GIBCO, #15250061), Countess™ Cell Counting Chamber Slides (Invitrogen, #C10228) in Countess™ II FL Automated Cell Counter (Invitrogen, #AMQAF1000). Cells were resuspended in growth factor-reduced Matrigel (Corning, #356231) at a concentration of 8000 cells/ul and seeded as 20 μl drops in a tissue-culture dish.After polymerization of the Matrigel, IntestiCult(# 06010) medium was supplemented with 10 μM Rock Inhibitor. Organoid medium was changed every 3 days and, when needed, organoids were passed after dissociation with Gentle Cell Dissociation reagent as previously described for Colon Organoids. Coculture with neutrophils Before co-culture the human colon organoids were gradually exposed to IntestiCult™ Intestinal Organoid Culture Medium (StemCell Technologies, #06010) until fully cultured using this medium. Healthy and patient-derived IBD were maintained in IntestiCult™ Intestinal Organoid Culture Medium (StemCell Technologies, #06010) from thawing. After expansion, organoids were fragmented by mechanical shearing through repeated pipetting and pelleted by centrifugation at 400 × g at 4 °C. The supernatant was removed and organoid fragments were resuspended in an extracellular matrix mixture composed of type I collagen (Ibidi, #5021) and Matrigel (Corning, #356231) at a 1:4 (v/v) ratio. Neutrophils from healthy donor peripheral blood were isolated by negative magnetic selection (StemCell Technologies, #100-0404) and immediately combined with organoid fragments in the matrix, targeting ~30% neutrophils by input cell number per dome, reflecting proportions observed in IBD.48 The cell–matrix suspension was dispensed as 10-µL domes into glass-bottom 96-well plates and allowed to gel at 37 °C according to matrix manufacturer recommendations. Domes were overlaid with prewarmed IntestiCult media and maintained under standard culture conditions until downstream assays. After 24 h, the medium was removed from wells and 4% Fixation solution (BD Cytofix, #554655) was added. The plate was then incubated at 4 °C for 2 h. Afterwards, the fixation buffer was removed and the samples washed twice with stain buffer (FBS) (BD Biosciences, #554556). Then, organoids were blocked using 0.5% BSA in DPBS for 30 min at room temperature. Supernatant was then removed and samples were resuspended in stain buffer supplemented with the primary antibodies CD66b (1:200, FITC, Biolegend, clone: G10F5, #984102), IL-6 (1:100, BD Pharmigen, APC, clone: MQ2-13A5, #561441) and EpCAM/CD326 (1:100, Alexa-700, Biolegend,clone:9C4, #324244) and incubated in staining buffer solution overnight at room temperature. The next day, organoids were washed once in DPBS buffer, and the plate was imaged in a LAS X Leica confocal scanning microscope (Leica) and a 40x Nikon air-objective. Images were deconvolved and analyzed using ImageJ (NIH). Alveolar organoids Culture expansion Adult human pulmonary alveolar epithelial cells (ScienCell, #3240) were cultured within Matrigel (Corning, #356231) domes, in which 24-well plates were used, and one dome was seeded per well. Passaging of organoids was done every 10 days using the Animal Component-Free Cell Dissociation Kit (ACF, StemCell Technologies, #05426) to dissociate the organoids into single cells, as recommended by StemCell Technologies. Cells were centrifuged at 400 × g for 5 min at 4 °C, and replated at a concentration of 1.6 × 105 cells/mL. After 30 min at 37 °C, PneumaCult Alveolar Organoid Expansion (AvOE) seeding medium (StemCell Technologies, #100-0847) was added to the domes, and changed every 3–4 days for normal AvOE medium. Cigarette smoke extract (CSE) treatment and co-culture with neutrophils On day 10, the alveolar organoids were treated with DMSO (Merck, #D138401) or CSE (Fraunhofer Institute for Toxicology and Experimental Medicine ITEM, generated from 1R6F cigarettes) 0.1% for 5 days in PneumaCult AvOE medium. On day 15, organoids were dissociated from the Matrigel into fragments. In brief, gentle cell dissociation reagent (GCDR, StemCell Technologies, #100-0485) was added for 2 min at room temperature. Then, domes were transferred to a 15 mL 1% BSA-coated Falcon and incubated for 10 min at room temperature. After incubation, samples were centrifuged at 400 × g for 5 min at 4 °C. The supernatant was removed and 10 mL of washing buffer (DMEM/F12+glutamax; Gibco, 30% BSA; Merck and HEPES; Gibco) was added. The organoids were then disrupted by mechanical forces through pipetting. Organoids were centrifuged as before, supernatant was removed and organoids were resuspended in Matrigel in a 1:3 dilution and mixed with healthy neutrophils isolated from blood using the negative magnetic beads enrichment (StemCell Technologies, #100-0404). The proportion used was approximately 50% neutrophil to the dome, which corresponds to 600 K neutrophils for 12 domes. The cell mix was plated into 25 µL-domes of matrigel into 24-well plates. After 30 min at 37 °C, 50% of PneumaCult AvOE medium and 50% neutrophil medium supplemented with DMSO/CSE were added to the domes. The plate was then incubated at 37 °C with 5% CO2 for approximately 24 h. For flow cytometry, samples were first dissociated using the ACF kit, as previously described for passaging. Then, cells were centrifuged at 300 × g at 4 °C for 5 min, washed in cold stain buffer (FBS) (BD Biosciences, #554556), and centrifuged as before. Then, the Zombie violet viability dye (Biolegend, #423113) and Human BD Fc Block (BD Biosciences, #564220) were added, and the cells were incubated at room temperature for 15 min protected from light. Followed by washing in a cold stain buffer and centrifugation. Samples were then transferred to a 96-well U-bottom plate and labelled with primary coupled antibodies, EpCAM/CD326 (1:100, Alexa-700, Biolegend,clone:9C4, #324244), CD66b (1:200, FITC, Biolegend, clone: G10F5, #984102), CD62L (1:100, BUV615, BD, clone: SK11, #751364) and CXCR2/CD182 (1:100, RealYellow586, BD, clone:6C6, #753253) were added. The plate was then incubated at 4 °C for 40 min. Afterwards, wells were washed with the stain buffer, centrifuged as before, and the pellet was resuspended in a fresh stain buffer. Samples were filtered before acquisition using Aurora Cytek and analyzed using FlowJo software. Supernatant cytokine quantification Cytokine levels in organoid-neutrophil co-culture supernatants were measured using the Ella automated immunoassay system (ProteinSimple, Bio-Techne) according to the manufacturer’s instructions. Briefly, 50 µL of culture supernatant was collected from each well, centrifuged at 300 × g for 5 min to remove cells and debris, and transferred to the assay plate. Samples were loaded into the Ella microfluidic cartridges pre-coated with capture antibodies for the cytokines of interest (e.g., IL-6, TNFα, IFNγ, MCP-1). The assay was performed following the standard automated protocol, which includes sample incubation, washing, and detection with fluorescently labeled secondary antibodies. Cytokine concentrations were calculated automatically by the Ella software based on standard curves generated using recombinant protein standards. All samples were assayed in technical duplicates, and results were expressed as pg/mL. Airway organoids COPD-derived airway organoids Lung tissue from COPD donors was washed five times with PBS supplemented with penicillin-streptomycin (Thermo, #15070063), gentamicin (Thermo, #15750060), and amphotericin (Thermo, J67049-AD). Clean tissue was minced into small pieces and subsequently enzymatically digested in 1 ml of advanced DMEM/F-12(GIBCO, #12634028) medium supplemented with 20 μ/ml and 20 μg/ml of P/S (ThermoFisher Scientific #15140122), Primocin 0.1 mg/ml (InvivoGen, #ant-pm-1), Hepes 10 mM (GIBCO, #15630056), and 2 nM Glutamax (GIBCO, #35050061), 4 mg/ml collagenase II (Thermo Fisher, 17101015), 0.1 mg/ml DNase IV (Sigma, #D5025), 20 μg/ml hyaluronidase V (Sigma, #H6254), 1% BSA (Sigma, #A3059), and 10 μM LY27632 (Abmole Bioscience, #M1817) for 1 h at 37 °C under slow rotation and vigorous pipetting every 15 min. The tissue lysate was filtered through a 100 μM cell strainer and centrifuged at 300 g for 10 min. The cell pellet was resuspended in 2 ml of advanced DMEM/F-12, and cells were counted. Cells were plated at a concentration of ~1.6 × 105 cells/20ul Matrigel domes (Corning, 356231). After 30 min at 37 °C, medium was added to the domes and changed every 3–4 days. The medium for the organoids was prepared as previously described by Sachs et al.49 Imaging of the co-culture organoids For immunofluorescence, the organoid-neutrophil mix was recovered from Matrigel using cell recovery solution (Corning, #354253). Samples were incubated on ice for 40 min, and then centrifuged at 300 × g at 4 °C for 5 min. Supernatant was removed, and samples were fixed with 4% paraformaldehyde (PFA, MP Biomedicals, #0219998380) for 45 min at room temperature. Samples were centrifuged at 200 × g at room temperature for 5 min and washed with IF Buffer: 0.1% BSA (Sigma-Aldrich, #A9647), 0.2% Triton X-100 (Sigma-Aldrich, #T8787) and 0.05% Tween20 (Sigma-Aldrich, #P9416) in DPBS. Then, organoids were centrifuged and blocked using 5% donkey serum (Sigma-Aldrich, #D9663) and 1% Triton X-100 in DPBS for 1 h at room temperature. Samples were centrifuged and primary antibodies mature surfactant protein C (SP-C, 1:500, SevenHills Bioreagents, #WRAB-76694), CD66b (1:200, FITC, Biolegend, clone: G10F5, #984102), CXCR2/CD182 (1:100, PE/Dazzle594, Biolegend, clone: 5E8, #320722) or IFN-ɑ (BD Pharmigen, PE, clone: 7N4-1, #560097) were incubated in a blocking solution overnight on a rocker at room temperature. On the next day, organoids were washed once in IF buffer and incubated with secondary antibodies anti-Rabbit IgG Alexa Fluor Plus 555 (ThermoFisher Scientific, #A32794) in IF buffer supplemented with 10% donkey serum for 6 h on a rocker at room temperature. DAPI (Sigma-Aldrich, #MD0015) was added to the organoids and then washed first with water and then with DPBS. Afterwards, organoids were resuspended in ProLong Gold Antifade Mountant (ThermoFisher Scientific, #P36930) and mounted on a glass coverslip. Acquisition of the fix co-cultures was done in a Leica Stellaris 8 confocal scanning microscope using a 20× Nikon oil objective or 40× air Nikon objectives. For the JC-1 experiment requiring live cell imaging a Nikon Ti2 Spinning disk was used with a 40× air Nikon objective. Images were deconvolved and analyzed using ImageJ (NIH) and Imaris (Oxford Instruments). Quantification and statistical analysis Data are presented as means ± SEM or SD as indicated in the Figure legend. Each experiment was performed with a minimum of two biological replicates; exact numbers are depicted in associated Figure legends. Except for proteomics and mRNAseq analysis, all statistical analyses were performed using Prism10 software (GraphPad). The statistical tests used and relevant P values are mentioned in the appropriate Figure legends. P values of <0.05 were considered significant. Supplementary information Supplementary Materials Supplementary Video 1 Supplementary Video 2 Supplementary Video 3 Supplementary Video 4 Supplementary Video 5 Supplementary Video 6 Supplementary Fig1 Supplementary Fig2 Supplementary Fig3 Supplementary Fig4 Supplementary Fig5 Supplementary Fig6 Acknowledgements We are grateful to Roche pool blood donors for valuable support in donating blood and the Roche medical team for their support in collecting the samples. The 360o pRED core facilities for Proteomics (especially for Dr. Angelique Augustin, Sabrina Golling and Justine Fidelin), the Flow cytometry core (especially Dr. Telma Lopes, Dr. Irene Calvo and Dr. Sinduya Krishnarajah) and the IVR-Mouse team at Roche for their invaluable help for this project. We acknowledge the support of the non-profit foundation HTCR, which holds human tissue on trust, making it broadly available for research on an ethical and legal basis. We thanked Marius Franscisco Harter and Giacomo Lazzaroni for their help with the colon organoids culture, Dr. Sukanya Guhathakurta for her support with the mitochondrial angle and Dr. Nicolas Mercado for establishing the collaboration with Fraunhofer. We are also grateful to Dr. Nikolche Gjorevski for reviewing our manuscript and providing us with critical comments. We also acknowledged Biorender (biorender.com) for their imaging collection. Finally, we thank all the CMI department members for their critical feedback and the Roche postdoc fellow (RPF) program for their financial support. Author contributions All authors confirm that there are no mistakes, misinterpretations and inconsistency with main text and corresponding legends in the figures and tables. All authors have read and approved of the article. Methodology: C.P.R., R.R.C., M.A., L.K., E.Y., D.L., M.C.L.; Animal experimentation: C.P.R., M.M., A.S., A.G., S.K., T.H., M.M.K., J.B., A.M., M.H.; Project administration: C.P.R., O.S., E.D., D.R.K.; Visualization: C.P.R.; Bioinformatic analysis: C.P.R.; Writing - original draft: C.P.R.; Writing - review & editing: C.P.R., R.R.C., M.H., J.B., M.C.L., K.G.L., O.S., E.D., D.R.D.; Supervision: O.S., E.D., D.R.D.; Conceptualization: E.D. and D.R.K. Data availability All data generated or analyzed during this study are included in this published article and its supplementary information files. The raw counts and sequencing data for RNAseq were deposited at SRA (https://www.ncbi.nlm.nih.gov/sra) with project number SRX29721209 (https://www.ncbi.nlm.nih.gov/sra/SRX29721209). BioProject: PRJNA1293334. Competing interests Hoffmann-La Roche Ltd provided support in the form of salaries to CPR, RRC, MA, MM, AS, AG, SK, EY, DL; MCL, LK, TH, MMK, JB, AM, MH, KGL, ED and DRK but did not have any additional role in the study design, data collection, analysis, decision to publish, or preparation of the manuscript. All the authors declare that they have no known competing financial or personal interests that could have appeared to influence the work reported in this paper. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Emma Doran, Email: emma.doran@roche.com. Daniel Regan-Komito, Email: daniel.regan-komito@roche.com. Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41392-026-02963-3. References Associated Data Supplementary Materials Supplementary Materials Supplementary Video 1 Supplementary Video 2 Supplementary Video 3 Supplementary Video 4 Supplementary Video 5 Supplementary Video 6 Supplementary Fig1 Supplementary Fig2 Supplementary Fig3 Supplementary Fig4 Supplementary Fig5 Supplementary Fig6 Data Availability Statement All data generated or analyzed during this study are included in this published article and its supplementary information files. The raw counts and sequencing data for RNAseq were deposited at SRA (https://www.ncbi.nlm.nih.gov/sra) with project number SRX29721209 (https://www.ncbi.nlm.nih.gov/sra/SRX29721209). BioProject: PRJNA1293334.