Eosinophil extracellular traps: heterogeneity of their stimuli, components, and functions

Abstract

Eosinophils participate in immune regulation through their granule proteins and cytokines. Recent studies demonstrate eosinophil functional versatility through the mechanism of eosinophil extracellular traps (EETs). EET formation occurs via suicidal EETosis and vital EET release. EETs contain chromatin- or mitochondrial-derived DNA, granule proteins, nuclear proteins, and cytosolic components that vary depending on the type and intensity of stimuli. Synthetic compounds, pathogenic microorganisms, endogenous molecules, and co-stimulatory factors stimulate EET formation via diverse signaling pathways through receptors that rely on or operate independently of NADPH oxidase-mediated reactive oxygen species (ROS) production, and peptidylarginine deiminase-4 (PAD4)-dependent histone modification. Necroptosis, pyroptosis, and autophagy pathways also contribute to EET biogenesis and subset heterogeneity. Here we summarize EET formation, compositional profile, and functional heterogeneity across disease states and future potential for novel immune intervention.

Underappreciated evaluation of eosinophil extracellular traps and their heterogeneity

Growing evidence indicates that eosinophils have multifaceted roles in maintaining physiological homeostasis and in various pathologies, including allergic responses and infections [1]. As an example, eosinophils play protective roles in acute cardiovascular injuries, such as myocardial infarction [2], abdominal aortic aneurysm [3], cardiac hypertrophy, and heart failure [4], but can be detrimental to chronic cardiovascular injuries, such as atherosclerosis [5,6]. Clinical studies on several types of CVD have supported these observations [7,8].

One primary mechanism by which eosinophils exert their functions is their secretory capacity. Like many other granulocytes, eosinophils undergo degranulation following stimulation, a process involving the release of granule-derived mediators [9]. Emerging evidence has identified extracellular trap (ET) formation as an additional and significant pathway through which eosinophils mediate their biological and pathological effects [10]. The observation of leukocytes releasing extracellular DNA traps was first made in neutrophils as a form of programmed cell death, triggered by microbial stimuli or other activators [11]. Eosinophil extracellular traps (EETs) were subsequently identified as a rapid and non-lytic process in which eosinophils release mitochondrial DNA along with granule proteins major basic protein 1 (MBP-1) and eosinophil cationic protein (ECP) upon sustained stimulation by interleukin-5 (IL-5) or interferon-γ (IFN-γ) [12]. EET formation has been detected in various tissues, including nasal polyps, lungs, bronchoalveolar lavage fluid (BALF), gastrointestinal tract, and blood vessels [13–15]. The presence of EETs in tissues, along with their key components - such as ECP-DNA complexes, Charcot-Leyden crystals (CLC), and double-stranded DNA (dsDNA) - has been closely associated with disease progression and prognosis, thereby serving as potential biomarkers [16,17].

EETs as integrated complexes from eosinophils have shown their activities in promoting type 2 innate immune cell activation and inflammation and antimicrobial and anti-helminthic effects in vitro [18,19]. EET components, including nucleic acids, histones, and eosinophil granule proteins, also serve as effector complexes essential for extracellular signaling. Existing studies have shown that EETsare highly diverse in their formation, components, and functions across different fields. Various stimuli can trigger EET formation, and each type of stimulus often activates different mechanisms and signaling pathways. EET morphology varies across conditions in a stimulus-dependent manner. The molecular composition of EETs also varies depending on the type and strength of the stimulus. EET-derived DNA may originate from either mitochondria or chromatin, while their protein content includes a complex mix of granule proteins, membrane proteins, cytoplasmic proteins, cytoskeletal proteins, and organelle proteins. However, the precise contribution of these components towards the pathophysiological functions of EETs still remain elusive. Finally, the different compositions of EETs can result in a range of effects, which may be beneficial or harmful, depending on the context. Therefore, investigating EETs as homogeneous complexes under varying conditions may oversimplify their pathophysiological roles, overlooking their inherent heterogeneity and complexity. The past five years have marked a notable advancement in our understanding of EET heterogeneity since they were first identified and highlights the challenges in studying eosinophils, EETs and ETs of other granulocytes. [20] (Figure 1).

Figure 1.

Timeline and milestone events of studies on EETs. EET: eosinophil extracellular trap; EGPA: eosinophilic granulomatosis with polyangiitis; OVA: ovalbumin.

Here we provide a comprehensive overview of the diverse stimuli that induce EET formation and delineate the mechanistic variations underlying EET generation under different conditions, which collectively determine their distinct compositional profiles. Such knowledge may help identify potential therapeutic targets for diseases associated with EET-mediated pathophysiology.

EET secretion process and basic components

The investigation of ET secretion mechanisms was first established in studies on neutrophil extracellular traps (NETs). Neutrophils undergo chromatin decondensation-associated cell death when cells were exposed to various stimuli [11]. Subsequent findings demonstrated that neutrophils extrude DNA by non-lytic pathways [27]. Based on the DNA release modalities involved, NET formation is categorized into two distinct processes: suicidal NETosis and vital NET release [10,28]. Eosinophils act similarly to neutrophils in ET secretion, with parallel mechanisms designated as suicidal EETosis and vital EET release [10]. Throughout this review, the umbrella term “EET formation” will be used to encompass both suicidal EETosis and vital EET release mechanisms (Figure 2A/2B).

Figure 2.

The mechanisms of EET formation following different stimuli. A. Suicidal EETosis is characterized by the release of chromatin-derived DNA and subsequent cell death. B. Vital EET release manifests non-lytic secretion of both mitochondrial and chromatin-derived DNA. A23187: calcimycin; AMPK: AMP-activated protein kinase; cfDNA: Cell-free DNA; cysLTR: cysteinyl leukotriene receptors; EET: eosinophil extracellular traps; ERK: extracellular regulated protein kinase; FcR: Fc receptors; GM-CSF: granulocyte/macrophage colony-stimulating factor; ICAM-1: intercellular cell adhesion molecule-1; IFNγ: interferon γ; IgA: immunoglobulin A; IgG: immunoglobulin G; IL-7R: Interleukin-7 receptor; L3: third-stage larvae; LPS: lipopolysaccharide; LTC4: leukotriene C4; LysoPS: lysophosphatidylserine; MAPK: mitogen-activated protein kinase; MEK: MAPK/ ERK kinase; MLKL: mixed lineage kinase domain like pseudokinase; NADPH: nicotinamide adenine dinucleotide phosphate hydrogen; PAD4: peptidylarginine deiminase 4; PAF: platelet-activating factor; PI3K: phosphoinositide 3-kinase; PMA: phorbol 12-myristate 13-acetate; PKC: protein kinase C; PSGL-1: P-selectin glycoprotein ligand-1; RIPK3: receptor-interacting protein kinase-3; ROS: reactive oxygen species; STAT3: signal transducer and activator of transcription 3; Syk: spleen associated tyrosine kinase; TLR9: Toll like receptor 9; TSLP: thymic stromal lymphopoietin.

Suicidal EETosis refers to a cellular process occurring within 3~8 hours post-stimulation, characterized by the release of chromatin-derived DNA coupled with histones and subsequent cell death. This mechanism involves sequential molecular events: chromatin decondensation, granule protein amalgamation, plasma membrane rupture, and terminal cell demise [28]. Eosinophils demonstrate this pathway when cells are activated by stimuli such as phorbol 12-myristate 13-acetate (PMA), calcimycin (A23187), or immunoglobulin G (IgG) (Figure 2A) [18,24,25]. Vital EET release, in contrast, manifests rapidly (within <1 min) following stimulation while maintaining cell viability [20,27,29,30]. This non-lytic EET formation mechanism involves instantaneous DNA extrusion accompanied by synchronized release of granule proteins and cytokines [20,29]. The extracellular DNA originates from both mitochondria and residual chromatin following stimulation with IL-5 combined with lipopolysaccharide (LPS), complement C5a, or platelet-activating factor (PAF) [20,29,30] (Figure 2B).

Beyond the cellular viability paradigm, current research focuses on nicotinamide adenine dinucleotide phosphate hydrogen (NADPH) oxidase-mediated reactive oxygen species (ROS) generation and peptidylarginine deiminase-4 (PAD4)-dependent histone citrullination in EET formation, although less is known about the latter. The NADPH oxidase pathway serves as a principal source of intracellular ROS, critically facilitating chromatin decondensation, plasma membrane permeabilization, and granule protein exocytosis during EET biogenesis [31,32]. Concurrently, PAD4, a nuclear enzyme catalyzing arginine-to-citrulline conversion in histones, modulates chromatin decondensation [33,34]. ROS generated through the NADPH oxidase pathway directly activates PAD4, leading to loss of nuclear lobulation, marked chromatin decondensation, and nuclear envelope breakdown [34], with supporting evidence from mouse studies using PAD4-deficient eosinophils and PAD4 inhibition in human peripheral blood eosinophils [25, 35]. Under various stimulatory conditions, eosinophils exhibit differential dependency on the NADPH and PAD4 pathways for EET formation (Table 1, Figure 3).

Table 1.

Activation mechanisms and molecular origins of EETs under various stimuli.

Stimulus	Species	Activation pathway				Dominant Component		Ref
		Activating receptors	NADPH	PAD4	Other Signaling intermediates	Nucleic acid	Protein
Synthetic stimuli
PMA	human, mice	NA	dependent	dependent	PKC, PI3K, independent of ATG-5	chromatinic, mitochondrial	galectin-10, MBP	12,26,63
low-dose A23187	human, mice	NA	dependent	dependent	Ca+	chromatinic, mitochondrial	ND	35,63
high-dose A23187	human	NA	independent	dependent	Ca+	chromatinic	ND	24
Co-stimuli (require cytokine priming)
C5a	human, mice	C5aR1, C5aR2	dependent	ND	PI3K, STAT3, independent of ATG-5	mitochondrial	ECP, EPX	20,29
LPS	human, mice	TLR4	dependent	ND	PI3K, independent of ATG-5	mitochondrial	ND	20,24
CCL11	human, mice	CCR3	dependent	ND	ND	chromatinic, mitochondrial	ND	20,24
PAF	human, mice	PAFR	(in)dependent	ND	ND	chromatinic, mitochondrial	ND	26,29,62
Anti-Siglec-8 antibody	human	Siglec-8	dependent	ND	MEK/ERK	ND	EPX	46,69
Endogenous molecules
IgG	human	Fc γ R	dependent	ND	PI3K, MAPK-p38, SQSTM1-p62	chromatinic	EDN, EPX	13,16,18,24
IgA	human	Fc α R	dependent	ND	PI3K, MAPK-p38, SQSTM1-p62	chromatinic	ND	60,86
iC3b	human	CD11b/CD18	dependent	ND	RIPK, MAPK-p38,	mitochondrial	ND	22
TSLP	human	TSLP/IL-7R	dependent	ND	ND	mitochondrial	ND	30
LysoPS	human	P2Y10 independent	dependent	(in)dependent	ND	chromatinic	ND	26
Activated platelets	human	ND	(in)dependent	dependent	RIPK3/MLKL	chromatinic	ND	6,23,35
fibrinogen	human, mice	ND	ND	ND	ND	ND	ND	87
MSU	human	ND	dependent	ND	ND	ND	ND	53,84
cysteinyl leukotrienes	mice	cysteine leukotriene 1 receptor	ND	ND	ND	ND	ND	50
cfDNA	human	TLR9	ND	ND	ND	ND	ND	51
EETs	human	ND	ND	ND	ND	ND	ND	43
Microorganisms
S.aureus	human	ND	dependent	ND	ND	chromatinic	ND	35,55
S.epidermidis	human	ND	dependent	ND	ND	chromatinic	ND	35,55
Aspergillus	human	CD11b/CD18	(in)dependent	independent	Syk/PI3K/AKT/MAPK/p38, Ca+	mitochondrial, chromatinic	ND	56,57,61
C.albicans	human	ND	dependent	ND	ND	chromatinic	ND	53
S.ratti L3	human	ND	ND	ND	ND	chromatinic	ND	58
L.sigmodontis microfilariae	human, mice	dectin-1	dependent	ND	AIM2/ASC/Caspase-1/GSDMD	mitochondrial, chromatinic	ND	19,64
Influenza virus	human	ND	ND	ND	ND	ND	ND	59

A23187: calcimycin; AIM2: absent in melanoma 2; ASC: apoptosis associated speck-like protein containing a CARD; ATG5: autophagy related 5; C5aR1: C5a receptor 1; CCL11: C-C motif chemokine ligand 11; CCR3: C-C chemokine receptor 3; cfDNA: cell-free DNA; ECP: eosinophil cationic protein; EDN: eosinophil-derived neurotoxin; EET: eosinophil extracellular traps; EPX: eosinophil peroxidase; ERK: extracellular regulated protein kinase; FcR: Fc receptor; GM-CSF: granulocyte/macrophage colony-stimulating factor; GSDMD: gasdermin D; IFNγ: interferon γ; IgA: immunoglobulin A; IgG: immunoglobulin G; IL-5: interleukin-5; IL-7R: interleukin-7 receptor; L.sigmodontis microfilariae: Litomosoides sigmodontis microfilariae; L3: third-stage larvae; LPS: lipopolysaccharide; LysoPS: lysophosphatidylserine; MAPK: mitogen-activated protein kinase; MBP: major basic protein; MEK: mitogen-activated protein kinase(MAPK)/extracellular signal-regulated kinase(ERK) kinase; MLKL: mixed lineage kinase domain like pseudokinase; MSU: monosodium urate crystals; NA: Not applicable; NADPH: nicotinamide adenine dinucleotide phosphate hydrogen; ND: Not determined; PAD4: peptidylarginine deiminase 4; PAF: platelet-activating factor; PAFR: PAF receptor; PI3K: phosphoinositide 3-kinase; PKC: protein kinase C; PMA: phorbol 12-myristate 13-acetate; RIPK3: receptor-interacting protein kinase-3; SQSTM1-p62: sequestosome 1; STAT3: signal transducer and activator of transcription 3; Syk: spleen associated tyrosine kinase; TLR4: Toll-like receptor 4; TLR9: Toll-like receptor 9; TSLP: thymic stromal lymphopoietin.

Figure 3.

Factors affecting EET components. Eosinophil status, stimuli type, stimulation mechanism, stimulation dosage and time result in EETs component variability. CCL11: C-C motif chemokine ligand 11; cfDNA: cell-free DNA; cysLTR: cysteinyl leukotriene receptors; EET: eosinophil extracellular traps; GM-CSF: granulocyte/macrophage colony-stimulating factor; IgA: immunoglobulin A; IgG: immunoglobulin G; LPS: lipopolysaccharide; LysoPS: lysophosphatidylserine; MSU: monosodium urate crystals; NADPH: nicotinamide adenine dinucleotide phosphate hydrogen; PAD4: peptidylarginine deiminase 4; PAF: platelet-activating factor; PMA: phorbol 12-myristate 13-acetate; TSLP: thymic stromal lymphopoietin.

Detection of EET formation

ETs arise from multiple leukocytes. Specific cell types can be enriched using magnetic-activated cell sorting (MACS) or fluorescence-activated cell sorting (FACS), followed by stimulation to induce ET formation [36,37]. Cell-type specific proteins, such as myeloperoxidase (MPO) and neutrophil elastase (NE) from neutrophils; CD68 and F4/80 from macrophages; MBP, ECP, CLC from eosinophils; and tryptase from mast cells define the cellular origin of ETs [38]. EETs display structural and biochemical features distinct from NETs under transmission electron microscopy, including the presence of free intact extracellular granules, granule-protein-free DNA structures, and the absence of protease-mediated chromatin remodeling [12,39]. These properties give rise to thicker chromatin filaments, eosinophil-specific granule signatures, and markedly prolonged DNase I resistance, underscoring the unique biology of EETs [18,40,41]. Unlike in cultured cells, it can be challenging to identify the source of ETs in vivo. EETs can be evaluated by examining both their structural characteristics and composition [42]. Light microscopy captures dynamic morphological changes during EET formation in isolated cells or tissues. In contrast, scanning and transmission electron microscopies allow ultrastructural assessment, distinguishing early and late stages of EET formation based on nuclear envelope rupture, plasma-membrane disruption, chromatin dispersion, and granule release [43].

DNA dyes can be used to detect EET DNA components and visualize DNA-net formation with fluorescence microscopy, with fluorescence intensity providing a semi-quantitative measurement of EET levels [42]. Shaking assays can be used to distinguish EET formation from apoptosis or necrosis. [18]. EET DNA can be isolated by ultracentrifugation. PCR targeting the 16S/18S sequence ratios identifies the relative contribution of mitochondrial versus nuclear DNA in EETs [19,25]. Enzyme-linked immunosorbent assays (ELISA) and proteomic analyses can be employed to quantify EET-derived proteins [24]. DNA and protein measurements can be combined experimentally to assess EET formation. DNA, citrullinated histones, and eosinophil granule proteins can be detected together by multicolor immunofluorescence. Flow cytometry using DNA dyes together with ECP labeling helps quantify EET-forming eosinophils [25]. ELISA-based detection of ECP-DNA complexes enables quantification of circulating EETs [17].

Methodological differences can significantly influence EET detection. DNA-based assays depend on dye specificity. Nonspecific DNA dyes and mitochondria-targeted dyes yield different interpretations of DNA origins. For protein-based assays, low-throughput methods such as ELISA or immunofluorescence can quantify the release of specific eosinophil granule proteins during EET formation, while proteomic analysis enables high-throughput identification of EET-derived proteins. Citrullinated histone H3 is widely used to identify EET and serves as a marker of nuclear envelope breakdown. Histone H3 is a substrate of NET formation and has the most clearly defined function. Its arginine residues R2, R8, R17, and R26 are key citrullination sites for PAD4 [44].. Histone subtype detection in EETs, including H1, H2A/B, H3, H4, varies in multicolor immunofluorescence, with lower abundance of H3 being detected relative to other histone subtypes [24,45]. Consistently, immunofluorescence using histone H1 and H4 detects more EETs than using histone H3 in tissues from patients with eosinophilic chronic rhinosinusitis (ECRS) [45]. Therefore, the suitability of H3 as a marker of EET detection requires additional validation. Collectively, accurate characterization of EET components requires careful selection of detection methods.

Diverse classes of EETs-inducing stimuli

A diverse range of synthetic and natural components stimulate EET formation both in vitro and in vivo. Based on their biological origins, these stimulatory agents are categorized into 4 classes: synthetic stimuli, co-stimuli, endogenous molecules, and microorganisms. Mechanistically, each class of stimuli activates distinct signaling pathways to drive EET formation, thereby resulting in quantitative variations in EET-generating capacity and qualitative differences in their composition (Table 1, Figure 3).

Synthetic stimuli are compounds that are not naturally present under human physiological conditions. They act as potent inducers for EET formation and are commonly utilized as positive controls in in vitro experiments to stimulate eosinophil production of EETs. Currently, the most extensively studied synthetic stimuli include PMA and A23187 [12,24]. Co-stimulus agents, including LPS, C5a, CCL11, PAF, and anti-Siglec-8 antibody, exhibit limited capacity to induce EET formation when administered alone. These agents require cytokine priming to potentiate their EET-inducing effects [12,13,18,20,21,46]. Cytokines IL-5, IFN-γ, and granulocyte/macrophage colony-stimulating factor (GM-CSF) enhance co-stimulation by modulating eosinophil activation states and regulating cell death-related pathways [47,48].

Endogenous stimuli refer to agents that exist or are produced endogenously from cells under physiological or pathological conditions. These include proteins such as IgA, IgG, iC3b, and thymic stromal lymphopoietin (TSLP) [22,30,49]; lipid molecules such as lysophosphatidylserine (LysoPS) and cysteinyl leukotrienes (cysLTs) [26,50]; nucleic acids such as cell-free DNA (cfDNA) [51]; organic substances such as urate crystals (monosodium urate crystals, MSU) [52,53]; and activated platelets or EETs-derived extracts [6,23,54]. Various pathogenic microorganisms are capable of inducinh EET formation, including the bacteria S. aureus and S. epidermidis [35,55]; fungi Alternaria, A. fumigatus, and C. albicans [56,57]; parasitic nematodes L. sigmodontis, D. immitis, and S. ratti [19,58]; and influenza virus [59].

Types and concentrations of stimuli contribute differently to EET formation

Studies of eosinophils primed with various stimuli have revealed differential impacts on EET formation depending on the nature and intensity of the stimulus. For example, although both IL-3 and IL-5 priming exert similar effects on human eosinophils, the former demonstrates significantly stronger EET formation capacity, with the latter requiring additional treatment with bafilomycin A (an inhibitor of granule acidification and autophagolysosome formation) to generate EETs. In IL-5- or GM-SCF-primed human and murine eosinophils, stimulation with C5a induced much faster eosinophil degranulation, DNA secretion and EET formation compared to PAF or CCL11 stimulus. EET extracts can also stimulate further EET production from human eosinophils, although weaker than stimuli like PMA. Bacterial stimuli such as S. aureus and S. epidermidis exhibit differential effects on EET formation, with S. aureus generating higher EET levels in a time-dependent manner. Similarly, Aspergillus induced EET formation much more potently than did Alternaria in human eosinophils. In parasitic studies, the microfilariae from L. sigmodontis and D. immitis directly stimulated murine bone marrow-derived eosinophils and human eosinophil to produce EETs, whereas the L3 larvae from L. sigmodontis and S. ratti required heat inactivation or plasma antibody treatment in order to enhance eosinophil EET induction. Both PAF and LysoPS demonstrated dose-dependent stimulation of EET formation, with low PAF concentrations requiring additional prior cytokine priming and low LysoPS concentrations triggering degranulation rather than EET formation.

The variability in EET production under different stimuli may be attributed to the differences in eosinophil activation pathways, including intracellular intermediates and upstream receptors. Although NADPH-mediated ROS generation and PAD4-dependent histone citrullination all play critical roles in EET formation these pathways are not inherently linked. For instance, high-concentration LysoPS, A23187-activated platelets, and A. fumigatus all induce EET formation via NADPH-independent pathways but exhibit divergent PAD4 dependencies in human eosinophils. Similarly, PAD4 inhibition can completely suppress EET formation induced by A23187-activated platelets, but can only partially suppress LysoPS-induced EETs and has no effect onA. fumigatus-induced EETs [12,26,56,61,63]. Current evidence suggests that NADPH contributes to chromatin- and mitochondrial-derived DNA release, while PAD4 is exclusively associated with chromatin DNA extrusion (Figure 3).

Besides NADPH and PAD4, pathways related to cell necrosis, pyroptosis, autophagy, and apoptosis may also participate in EET formation, though further investigation is required. EET formation differs morphologically from necrosis [20] (e.g., DNA web structures, nuclear changes), but certain stimuli link EETs to necroptosis pathways. For example, isolated human eosinophils generated EETs upon iC3b stimulation through the RIPK/MLKL pathway, which could be blocked by RIPK3 or MLKL inhibitors [23]. Stimulation with Litomosoides sigmodontis microfilariae (MF) induced EET formation via an AIM-2-ASC-caspase 1 pyroptotic pathway in both human and mouse eosinophils. This process depends on AIM2 and caspase-1; inhibiting caspase1- or deleting AIM2, ASC, gasdermin D or caspase-1 blocked EET formation, while NLRP3 inhibition or deficiency had no effect, suggesting that MF-induced EET formation is NLRP3-independent [64]. In human peripheral blood eosinophils, both caspase-1 inhibitor AIM2cGAS/TLR9 inhibition completely suppressed MF-induced EET formation, with the partial suppression of PMA-induced EET formation by the latter indicating that [64].

Studies on autophagy revealed conflicting results. Upregulation of autophagy through cytokine priming, pretreatment with mTOR inhibitors or pharmacologic activation appeared to downregulate EET formation, while eosinophil-specific deletion of the core autophagy protein ATG5 increased degranulation and extracellular DNA release in mouse and human eosinophils [60]. [22]. [65]. These data suggest that autophagy plays a negative regulatory role in EET formation. However, other studies show that inhibitors of autophagy reduce EET formation, for example in a murine asthma model [21], whereas PMA, LPS, and C5a did not induce EET formation through ATG5-dependent autophagy in GM-CSF/IL-5-primed eosinophils [66]. These contradictions highlight the need for deeper explorations of autophagy-EET interactions in eosinophils. In contrast, apoptosis has been consistently excluded as a mechanism of eosinophil EET formation because stimuli such as A23187, LPS, C5a, and LysoPS failed to induce cell apoptotic markers [12,20,26]. Yet, apoptosis upstream receptors such as adhesion molecules, pathogen-associated molecular patterns (PAMP) receptors, and metabolic sensors influence EET secretion, potentially through differing mechanisms. In summary, the heterogeneity in EET induction across stimuli stems from diverse upstream and intermediate pathways [19,22,50,61].

This variability in EET production may underlie the differences in eosinophil activation states under disease conditions. For instance, nasal polyp-derived eosinophils from chronic rhinosinusitis (CRS) patients produced 2–3 times more EETs than those from controls, suggesting a correlation of EET formation with disease severity [56,67]. Eosinophils from patients with severe eosinophilic asthma exhibited enhanced LPS-induced EETs but reduced LysoPS responsiveness compared to those from patients without severe asthma [26,54]. Heterogeneity may also arise from eosinophil subtype. Single-cell RNA sequencing in murine colitis models identified two eosinophil subpopulations: active and basal eosinophils, where active rather than basal eosinophils underwent marked morphological changes in response to infection or inflammatory stimuli[68]. Genetic factors, such as LNK knockout, also enhanced baseline EET formation capacity [49]. These observations underscore how intrinsic eosinophil states, shaped by disease microenvironments or genetic alterations, critically modulate trap formation. Prior data as well as our own observations suggest that EET formation may show species specificity, with human eosinophils responding more readily than mouse cells [20,64].

EET composition heterogeneity across different stimulatory conditions

EETs form web-like structures following eosinophil activation by various stimuli, with distinct molecular compositions depending on the stimulatory context.

EET DNA source variability

Currently, chromatin is considered to be the predominant DNA source. Chromatin-derived DNA is typically identified through multiplex immunofluorescence colocalization of DNA with citrullinated histones and eosinophil granule proteins. LC-MS analyses consistently detect histones as the major nuclear component in EET, supporting chromatin origin [24,25]. Experimental stimuli such as PMA, A23187, PAF, activated platelets, urate crystals, A. fumigatus, and S. aureus induce histone-containing EETs in vitro [13,23,24,56]. In clinical samples—including nasal mucosa from eosinophilic CRS patients [12,18], vasculature from patients with eosinophilic granulomatosis with polyangiitis (EGPA) [13], skin biopsies [14], thrombi [35], and asthmatic lung tissues [25,63,69]—DNA histone colocalization further corroborates chromatin-derived DNA [6]. EETs also contain mitochondrial DNA. Early studies have demonstrated mitochondrial DNA release fromIL-5/IFN-γ-primed eosinophils upon LPS costimulationwith C5a, iC3b and TSLP also capable of promoting this process [20] [22,30]. Samples from patients with allergic asthma and from murine asthma models show evidence of mitochondrial DNA in EETs [21]. However, some studies report the presence of both chromatin-derived and mitochondrial DNA in EETs induced by various stimuli, including L. sigmodontis microfilariae and PMA. [19 [25]. However, some of these discrepances may also be attributed to methodological differences in staining or detection [29] [24]. Together, different stimuli may yield EETs with DNA from different sources, chromatin DNA or mitochondrial DNA or both with different proportions.

EET protein composition differences

In addition to the differences in DNA sources, EET protein components also differ depending on the stimulus type. For example, co-stimulation of IL-5- or GM-CSF-primed eosinophils with C5a released much higher ECP and EPX than those stimulated with PAF, CCL11, PMA, or A23187 [29,70]. PMA induced more pronounced galectin-10 and MBP release than did IgG, IgA, or PAF [13]. Cytokeratin 18 (CK18)-IgG stimulated much greater extracellular EDN release than did CK19-IgG [71]. IL-5 enhanced anti-Siglec-8 antibody-induced EPX secretion, while IL-3 synergized with IgG to boost DNA and EPX release [46].

EET protein profiles can vary even under identical stimuli, depending on the dose and duration. Prolonged low-dose PMA (10 ng/mL, 12 hours) generated EETs containing 997 proteins, whereas high-dose short-term PMA (50 nM, 4 hours) yielded EETs containing only 32 proteins [24,25]. In addition, differences in EET protein collection procedures, such as the DNase I digestion concentration used or the timing of collection may increase the risk of detecting non-EET proteins. Extended low-dose stimulation of eosinophils enriched EETs with antimicrobial proteins (myeloperoxidase, lysozyme C), chromatin regulators (histones H1.0, H1.2), cytoskeletal components (keratin type II, actin-related proteins), and metabolic enzymes (transketolase, glyceraldehyde-3-phosphate dehydrogenase) [24,25] (Figure 3). This compositional plasticity of eosinophils underscores the reshaping of the EET functional proteome by stimulatory intensity and duration, potentially tailoring their roles in immune defense or pathological damage.

Functional heterogeneity of EET components

The diversity in EET formation stimuli, heterogeneity in EET formation pathways, EET component multiplicity in both DNA sources and eosinophil protein types, and EET formation variation in response to disease severity suggests that eosinophil EETs are functionally different in the context of different stimulus environments and disease stages. In recent years, the development of biomedical engineering and gene-editing technologies have provided several novel methods to assess the pathophysiological role of EET components, allowing their functional diversities to be dissected (Table 2).

Table 2.

Functional heterogeneity of EET components.

EET components	Functional mechanisms	Pathophysiological functions	Ref
Chromatin DNA	binds to CCDC25-ILK-PKCα-CRTC1	activates goblet cells, amplifies allergic responses	25
	on pulmonary neuroendocrine cells	promotes naïve CD4+ T-cell differentiation into Tfh cells	17
Chromatin DNA	acts as scaffold	inhibits microfilarial motility	64
Mitochondria DNA	binds to CCDC25-VIRMA	increases N6-methyladenosine methylation and MAF stability, promotes Tfh cell differentiation	17
Mitochondria DNA	acts as scaffold	enhances MBP-mediated cytotoxicity	72
MBP/EPO-DNA		inhibits microfilarial motility	73
CLC	interacts with NLRP3	increases macrophage IL1β expression increases keratinocyte MMP expression	82,83
ECP		stimulate lung adenocarcinoma and epithelial cell shedding and IL6 and IL8 secretion	54
	activates IGF1R	activates airway epithelial cells	75
	regulates FGFR	regulates cardiomyocyte differentiation	76
	binds to BMPR	suppresses cardiac fibroblast TGF-β1/Smad3 and NF-κB signaling and myocardial fibrosis	2,3,4
	binds to BMPR	promotes smooth muscle cell calcification	5
EDN	triggers TLR2 and activates Myd88	activates dendritic cells and amplifies Th2 responses	77
EPO	binds to HER2	drives melanoma MUC4 signaling, activates neuron GAG/MAPK signaling, and inhibits ROS/MAPK signaling and osteogenesis	80,81
EPX		enhances EET-DNA-CCDC25 interaction	25
MBP	binds to endothelial thrombomodulin	promotes thrombosis	74
		enhances platelet activation and aggregation	6
	stimulates MRGPRX2	activates mast cells and induces bronchoconstriction	78,79

BMPR: bone morphogenic protein receptor; CCDC25: coiled-coil domain containing 25; CLC: Charcot-Leyden crystals; CRTC1: CREB-regulated transcription coactivator 1; ECP: eosinophil cationic protein; EDN: eosinophil-derived neurotoxin; EET: eosinophil extracellular trap; EGFR: epidermal growth factor receptor; EPO: eosinophil peroxidase; EPX: eosinophil peroxidase; FGFR: fibroblast growth factor receptor; GAG: glycosaminoglycan; HER2: human epidermal growth factor receptor 2; IGF1R: insulin-like growth factor I receptor; ILK: integrin-linked kinase; MAF: musculoaponeurotic fibrosarcoma; MAPK: mitogen-activated protein kinase; MBP: major basic protein; MMP: matrix metalloproteinases; MRGPRX2: Mas-related G protein-coupled receptor X2; MUC4: mucin 4; Myd88: myeloid differentiation primary response 88; NLRP3: NOD-, LRR- and pyrin domain-containing protein 3; PKCα: protein kinase Cα; ROS: reactive oxygen species; Tfh: follicular T-helper cells; TGF-β1: transforming growth factor-β1; VIRMA: Vir-like m⁶A methyltransferase-associated protein.

Chromatin-derived DNA in EETs binds to the coiled-coil domain containing 25 (CCDC25) receptor to mediate diverse pathological effects (Table 2). In murine asthma models, this DNA interacted with pulmonary neuroendocrine cells (PNECs) via the CCDC25-ILK (integrin-linked kinase)-PKCα-calcium-CRTC1 (CREB-regulated transcription coactivator-1) pathway to enhance goblet cell hyperplasia, mucus production, airway inflammatory cell infiltration, and type 2 cytokine expression, ultimately leading to the amplification of allergic immune responses [25]. In patients with bullous pemphigoid, both eosinophil numbers and EET contents correlate with disease severity. In this context, the interaction of chromatin-derived EET DNA with CCDC25 promotes naïve CD4⁺ T-cell differentiation into follicular T-helper cells, at least partly through modulating the stability of the transcription factor musculoaponeurotic fibrosarcoma (MAF) [17]. Studies using mouse knockout models and in vitro analyses have shown a role for EPX in enhancing the binding of EET-DNA to CCDC25, whereas MBP lacks this functional capacity [25]. Mitochondrial DNA in EETs as characterized by hypomethylated CpG motifs act as a scaffold to facilitate MBP-induced pore formation in lipid bilayers, disrupt intracellular potassium balance, and enhance MBP-1-mediated cytotoxicity [72] (Table 2).

Besides EET DNA, eosinophil granule proteins in EETs also exert distinct effects (Table 2). MBP and EPO inhibited microfilarial motility in a dose-dependent manner when combined with EET DNA [73]. In infective endocarditis, MBP bound to the endothelial thrombomodulin (TM) and inhibited the capacity of TM to generate the natural anticoagulant activated protein C (APC), thereby promoting thrombosis [74]. During arterial thrombosis, chemokines such as CCL5 and CCL11 promote eosinophil chemotaxis. Activated platelets release both surface-bound and soluble P-selectins that bind to P-selectin glycoprotein ligand-1 (PSGL-1) on eosinophils, triggering eosinophil activation and EET formation [8]. In murine studies of arterial thrombosis, MBP but not EPX enhanced platelet activation and aggregation via EETs, thereby stabilizing thrombi6]. In transformed and normal human lung cell lines and primary cells,, EETs stimulated cell shedding and increased IL-6/IL-8 secretion. Anti-ECP antibodies reversed this inflammatory response, whereas anti-MBP antibodies did not [54] (Figure 4, Table 2). These studies highlight the divergent pathophysiological effects of EET components, though direct mechanistic evidence remains limited.

Figure 4.

Functional heterogeneity of different EET components in disease. A. In asthma, the granule protein EPX enhances the binding of EET DNA to CCDC25 and subsequently amplifies allergic inflammation, whereas MBP lacks this ability. Anti-ECP antibodies could reverse the inflammatory response in airway epithelial damage, whereas anti-MBP antibodies could not. B. In arterial thrombosis, MBP but not EPX exposed on EETs promotes thrombus development. CCDC25: coiled-coil domain containing 25; ECP: eosinophil cationic protein; EET: eosinophil extracellular traps; EPX: eosinophil peroxidase; MBP: major basic protein; PSGL-1: P-selectin glycoprotein ligand-1; TSLP: thymic stromal lymphopoietin.

Eosinophil granule proteins act through specific receptors and pathways, providing indirect insights. For example, ECP activates airway epithelial cells via insulin-like growth factor I receptor (IGF1R) [75] and regulates cardiomyocyte differentiation through the epidermal growth factor receptor (FGFR)/ERK pathways [76]. ECP also suppresses TGF-β1/Smad3 and NF-κB signaling in myocardial fibrosis while promoting arterial calcification via the BMP-Smad-Runx2 axis in smooth muscle cells [2–5]. EDN triggers dendritic cell TLR2-myeloid differentiation primary response 88 (MyD88) signaling to amplify Th2 responses [77]. MBP stimulates mast cell Mas-related G protein-coupled receptor X2 (MRGPRX2) receptors to induce bronchoconstriction in asthma [78,79]. EPO binds to the human epidermal growth factor receptor 2 (HER2)/ERK/cell surface associated mucin 4 (MUC4) signaling in melanoma, modulates neural function via glycosaminoglycan (GAG)-ERK/MAPK activation, and suppresses osteoclastogenesis by inhibiting ROS/MAPK pathways [80,81]. CLC activates the NOD-, LRR- and pyrin domain-containing protein 3 (NLRP3) inflammasome in macrophages to release IL-1β and upregulates matrix metalloproteinases (MMPs) in keratinocytes, thereby exacerbating inflammation [82,83]. In addition to influencing effector cells, these granule proteins also regulate eosinophil development [84] and interact with each other [85]. The relative proportions and temporal release patterns of these proteins during EET formation, combined with their interactions, likely determine the heterogeneous pathophysiological roles of EETs in different disease contexts (Table 2).

Concluding remarks and future perspectives

The formation of EETs represents a critical pathobiological function of eosinophils, with their heterogeneity and complexity underpinning multifaceted roles in immune responses, allergic reactions, and antimicrobial defense. Growing evidence suggests that the pathobiological effects of EETs are intrinsically linked to their molecular composition, with functional outcomes varying significantly across stimulatory contexts. Yet, our understanding of EET formation, components, and biological and pathological functions remain limited (see Outstanding Questions). Mechanistic studies are required to elucidate stimulus-specific signaling networks and identify key targetable regulatory nodes in EET biogenesis. Omics-based high-throughput technologies can be employed to map the compositional landscape of EETs in disease-specific conditions. EET formation or activity can be targeted by precision strategies as a therapeutic approach to treatt allergic, infectious and autoimmune disorders. These advances will also require validation in vivo to fully map the physio-pathological landscape of EET function. As the studies of eosinophils and EETs enter a phase of rapid advancement, the pleiotropic functions of EETs position these extracellular traps as pivotal players across biomedical disciplines. Deciphering the mechanistic basis of EET heterogeneity promises to advance novel therapeutic avenues and catalyze precision medicine breakthroughs in eosinophil-associated diseases.

Outstanding questions.

• What are the main components in EETs? Are EETs identical complexes under various conditions?

• What are the differences between the two EET secretion processes—suicidal EETosis and vital EET release?

• What are the main methods currently used to detect the structural characteristics and compositions of EET? Do methodological differences influence EET detection? In EET detection, is it representative enough to use citrullinated histone H3 to identify EET and serve as a marker of nuclear envelope breakdown? In the future, standardized, high-throughput EET detection methods are warranted.

• What are the major classes of stimuli for EET formation? Apart from different stimulus types (synthetic stimuli, co-stimulus, endogenous stimuli, and microorganisms), duration and doses, what other factors affect EET formation?

• Two eosinophil activation pathways NADPH-induced ROS generation and PAD4-dependent histone citrullination contribute to EETs formation heterogeneity. Are they both associated with chromatin- and mitochondria-derived DNA release?

• Further investigation is required on the relationship between EET formation and cell necrosis, pyroptosis, autophagy, and apoptosis. How do these cellular processes get involved in the initiation or modulation of EET formation?

• How do differences in EET protein composition and DNA origin translate into functional heterogeneity across distinct disease contexts and inflammatory environments? Although classical granular proteins of EETs have been discovered for many years, filling the gaps in our knowledge of their specific receptors and signaling pathways in the future will help to better elucidate the functional heterogeneity of EETs.

• Can targeting specific EET components or formation pathways provide new therapeutic opportunities for different EET-associated inflammatory and immune-mediated diseases?

Highlights.

• Like neutrophil extracellular traps (NET), eosinophil extracellular trap (EET) formation also involves two distinct processes: suicidal EETosis and vital EET release.

• EET contains chromatin and/or mitochondrial DNA, granular proteins; nuclear histonevariants; cytosolic mediators; cytoskeletal proteins; organelle proteins; and cell membrane proteins.

• There are at least four major classes of stimuli that contribute to EET formation via NADPH-induced ROS generation and/or PAD4-dependent histone citrullination pathways.

• Cell necrosis, autophagy, and apoptosis, disease status and severity, and eosinophil subtypes all affect EET formation.

• The diversities of DNA sources and granule protein types in EET all determine EET functional heterogeneity depending on stimulation environments and disease status.