Neutrophil Extracellular Traps Mediate in vitro and in vivo Degradation of α-Synuclein Amyloid Fibrils

Abstract

Purpose

Neutrophil extracellular traps (NETs) are chromatin-based structures decorated with neutrophil enzymes such as elastase and myeloperoxidase (MPO). We previously demonstrated that amyloid fibrils (AFs), regardless of their protein composition, induce NET release in human neutrophils in vitro through a mechanism dependent on reactive oxygen species generated by NADPH oxidase 2 (NOX2). In this study, we investigated whether α-synuclein amyloid fibrils (αSF) induce NET formation in vivo and whether NETs modulate amyloid-associated pathology.

Methods

αSF were instilled intratracheally into wild-type (WT) and gp91phox knockout (KO) mice, which lack NOX-2 activity and are unable to form NETs. Both male and female mice were analyzed. Eight hours post-instillation, lungs were assessed for neutrophil infiltration, NET formation (citrullinated histones and myeloperoxidase), amyloid fibril persistence (Congo red staining), and lung mechanics (elastance and resistance).

Results

αSF induced robust neutrophil recruitment and lung inflammation in both WT and KO mice. However, NET formation was detected exclusively in WT animals. Congo red–positive amyloid-like structures persisted in KO lungs but were largely absent in WT mice, indicating NET-associated proteolysis promotes fibril clearance. Functionally, αSF impaired lung elastance in both genotypes at early time points. As fibrils were cleared in WT mice, lung function recovered. In contrast, KO animals exhibited persistent fibril deposition and sustained elastance impairment. Additionally, NET structures were observed in post-mortem brain tissue from patients with Parkinson’s disease.

Conclusion

These findings demonstrate that NETs act as an innate immune mechanism capable of sensing and degrading amyloid fibrils in vivo, thereby promoting their clearance and preserving tissue function. This mechanism may play a relevant role in modulating amyloid-associated pathology, including in human neurodegenerative diseases such as Parkinson’s disease.

Graphical Abstract

Introduction

Neurodegenerative diseases such as Parkinson’s disease (PD) and Alzheimer’s disease (AD) can be classified as forms of amyloidosis, as they share a common hallmark: the accumulation of protein aggregate species, including amyloid fibrils (AFs) and oligomers.^1,2 A wide range of peptides and proteins can adopt this fibrillar conformation, including the Aβ peptide and α-synuclein (αSyn), which are primarily associated with AD and PD, respectively.^3–5 Regardless of the protein of origin, aggregation follows a conserved pathway in which proteins undergo conformational changes that first give rise to small toxic oligomers during the nucleation phase, which then progress into protofibrils and ultimately to mature AFs through exponential growth.^1,6,7 Importantly, oligomers are not merely transient intermediates but are actively involved in a variety of toxic cellular processes, including the formation of membrane pores that promote cytotoxicity, synaptic dysfunction, endoplasmic reticulum stress, and impaired autophagy, among others.^7,8

The presence of protein aggregates triggers inflammatory responses and promotes the release of cytokines, chemokines, and other immune mediators that recruit and activate effector cells.^9–12 Neutrophils, the most abundant leukocytes in the blood, are among the first responders to tissue injury and play a crucial role in host defense, sustained by their highly diverse antimicrobial arsenal stored within distinct granule subsets.^13–15 Through this arsenal, neutrophils deploy key defense mechanisms include degranulation, phagocytosis, and the formation of neutrophil extracellular traps (NETs), a specialized response involving the release of DNA-histone complexes together with granular proteins into the extracellular space.^14,16

NETs are chromatin-decondensed structures decorated with a wide array of different proteins, including elastase, myeloperoxidase (MPO), as well as antimicrobial peptides, and other bioactive molecules.^17,18 This highly conserved immune response has been observed across multiple species and was initially described as a mechanism for trapping and killing invading microorganisms.^16,19 Nowadays, the involvement of NETs has been recognized far beyond their antimicrobial function. Excessive or dysregulated NET formation, as well as impaired clearance, has been linked to tissue damage and to the pathogenesis of a wide range of inflammatory and autoimmune diseases.^20–22

Depending on the stimulus, NET formation can proceed through either ROS-dependent or ROS-independent mechanisms.²⁰ In the ROS-dependent pathway, NET release is primarily driven by superoxide (O₂⁻) production catalyzed by the NADPH oxidase complex (NOX-2). NOX-2 is a transmembrane enzyme complex whose catalytic subunit, gp91phox, reduces molecular oxygen to superoxide by oxidizing NADPH.^23,24 A broad range of stimuli can trigger NET release by elevating ROS levels, including pathogens of different origins (viruses, bacteria, fungi), activated platelets, pro-inflammatory cytokines, and even AFs.^17,24–28

Our group was the first to demonstrate that human neutrophils stimulated with AFs of three distinct proteins, αSyn, transthyretin (TTR), and Sup35, release extracellular traps in vitro in a NOX-2 dependent manner.^27,28 Additionally, we found that elastase, a major component of NETs, can degrade TTR-AFs into toxic oligomers. We also observed that NET markers, including elastase and extracellular DNA, colocalized with amyloid deposits in lungs and skin biopsies from patients with systemic amyloidosis, supporting the relevance of this interaction.

In vivo, AF clearance is likely mediated by multiple, partially overlapping mechanisms, including phagocytic uptake by macrophages, microglia, and other myeloid cells, lysosomal degradation after internalization, extracellular proteolysis, autophagy-dependent processing of intracellular aggregates, and tissue drainage pathways such as BBB transport, perivascular/glymphatic clearance, and lymphatic removal. In addition, NETs may contribute to extracellular amyloid processing through NET-associated proteases, although this mechanism may generate smaller amyloid fragments with potential toxicity. Thus, amyloid clearance should be viewed as a balance between aggregate removal, inflammatory activation, proteolytic remodeling, and possible generation of toxic intermediates.²⁹

Building on previous evidence, our study investigated whether NETs induced by αSyn fibrils (αSF) contribute to amyloid-associated pathology in vivo. To this end, αSF were instilled into the lungs of wild-type (WT) and gp91phox-knockout (KO) mice, which lack the catalytic subunit of NOX-2. These animals are a well-established model of chronic granulomatous disease and are widely employed to study NET deficiency in settings where the stimulus is ROS-dependent.^30,31 Our findings revealed that although both WT and KO mice showed neutrophilic infiltration and inflammation following AF instillation, only WT animals were able to trigger NET release and clear αSF from the lungs. In addition, fibril clearance in WT animals was accompanied by the restoration of pulmonary function, whereas lung dysfunction persisted in KO mice where fibrils remained. These findings suggest that NETs contribute not only to the inflammatory response but also to the clearance of amyloid deposits. Moreover, analysis of substantia nigra sections from patients with Parkinson’s disease revealed the presence of NETs, supporting the occurrence of NET formation in the human brain. To our knowledge, this is the first report demonstrating both the deleterious impact of amyloid accumulation in the lungs and the critical role of neutrophil extracellular traps in mediating amyloid degradation (amyloidolysis) and promoting recovery of organ function.

Materials and Methods

Ethic Committees’ Approvals

All animal experiments in this study were approved by the Ethics Committee on the Use of Animals (CEUA) of the Health Sciences Center at the Federal University of Rio de Janeiro. Procedures were conducted under process number 01200.001568/2013-8, protocol n. 004/23, in accordance with established ethical guidelines for animal research of CONCEA (National Council for the Control of Animal Experimentation linked to the Ministry of Science, Technology and Innovation (MCTI), Brazil.

All in vitro experiments involving human neutrophils were approved by the Research Ethics Committee (CEP) of the Clementino Fraga Filho University Hospital, Federal University of Rio de Janeiro (HUCFF-UFRJ) (protocol no. 4261.015400/0052-57). Human neutrophils were isolated from peripheral blood obtained from healthy adult volunteers who provided written informed consent, in accordance with protocols approved by the institutional ethics committee.

Animal Model

In this study, we used male and female C57BL/6 mice (2–4 months old), including both wild-type and gp91phox knockout animals. gp91phox encodes the catalytic subunit of the NOX-2 complex, and its deletion abolishes the generation of ROS via NOX-2. Because the gp91phox gene is located on the X chromosome, male knockout mice are hemizygous (gp91phox^y/–), whereas females are homozygous (gp91phox^–/–). For consistency, all gp91phox deficient animals are referred to as gp91phox KO mice throughout the paper, irrespective of sex. The sex of the animals used in each experiment is indicated in the figure legends and methods sections.

Animals were housed in the animal facility of the Health Sciences Center of our university under standard conditions, with controlled temperature and humidity, a 12 h light/dark cycle, and free access to food and water. gp91phox knockout mice were kindly provided by the Federal University of Minas Gerais, Brazil. All procedures involving animals were conducted in accordance with institutional guidelines for animal care and handling and followed the ARRIVE guidelines.

In vitro Experiments

Expression and Purification of Human Wildtype α-Synuclein

Purification of α-Synuclein was performed as previously described.³² To do this, human wildtype αSyn was expressed in Escherichia coli BL21 (DE3) pLysS cells transformed via heat shock with the corresponding plasmid. Protein expression was induced with 0.5 mM IPTG (isopropyl β-D thiogalactopyranoside, Sigma-Aldrich) and cultures incubated for 4 h under the same conditions. Cells were harvested (6000 rpm, 15 min, 4 °C; Himac CR 21G2, Hitachi), resuspended in 20 mL lysis buffer (20 mM Tris, 5 mM EDTA, pH 8.0) with protease inhibitor cocktail (Thermo Fisher Scientific), and lysed by sonication in ice for 1 h (59 sec on/59 sec off, 60% amplitude; Vibra Cell™, Sonics). The lysate was centrifuged (16,000 rpm, 20 min, 4 °C), and the pH of the supernatant was lowered to 3.5 with HCl, followed by a second centrifugation and readjustment to pH 7.4 with NaOH. Protein was precipitated with 2 mM ammonium sulfate, and the pellet dialyzed overnight against 20 mM Tris, 1 mM EDTA (pH 8.0), then in Milli-Q water for 4 h before lyophilization. The lyophilized protein was resuspended in PBS (pH 7.4), filtered (0.22 μm), and quantified at 280 nm (NanoDrop. 2000, Thermo Fisher Scientific; ε = 5,900 M⁻¹.cm⁻¹). Residual lipopolysaccharide was removed using Pierce™ High-Capacity Endotoxin Removal Spin Columns (Thermo Fisher Scientific) according to the manufacturer’s instructions.

αSynuclein Aggregation and Formation of Pre-Formed Fibrils

To induce fibril formation, monomeric αSyn was incubated under constant agitation at 800 rpm for 7 days at 37°C. The presence of AFs was confirmed by light scattering (excitation and emission at 330 nm), ThT binding assay (10 μM of αSyn; 50 μM of ThT; excitation at 450 nm and emission collected from 470 to 520 nm; maximum emission at 482 nm) both collected in a Jasco FP8200 spectrofluorometer (Jasco Corp., Tokyo, Japan), and transmission electron microscopy (negative staining, 1% uranyl as contrast agent). The characterization of αSF is shown in Supplementary Figure 1.

Purification of Mouse Bone Marrow Neutrophils

Murine neutrophils were isolated from the bone marrow of euthanized mice following exposure to 5% isoflurane. Femurs and tibias were carefully removed and placed in Hank’s balanced salt solution (HBSS). Bone marrow cells were flushed from the cavities using a 1 mm needle with 3 mL of HBSS, dissociated, and centrifuged at 400 x g for 5 min. Erythrocytes were lysed by incubation in 10 mL of 0.2% NaCl for 40s, after which osmolarity was restored by adding 10 mL of 1.6% NaCl. Cells were centrifuged and resuspended in 5 mL of HBSS before being layered over a two-step Percoll gradient (65% and 72%). Gradients were centrifuged at 1000 x g for 30 min with the brake off. Neutrophils were collected from the 65–72% interface, washed twice with HBSS to remove residual Percoll, and finally resuspended in RPMI medium. Neutrophil purity was assessed by cytocentrifugation followed by panoptic staining.

Assessment of NET Production via Extracellular DNA Detection

To evaluate NET production, purified neutrophils (5x10⁵ cells/well) from WT and gp91phox KO animals were seeded in a 96-well plate (Costar), and stimulated with 10 μM αSyn fibril (αSF), 100 nM PMA (Sigma-Aldrich - P8139) or 1μg/mL of LPS O111:B4 (Sigma- Aldrich, #L3023) for 3 h. NETs were measured using Sytox Green (Invitrogen, Cat. No. S7020) according to the manufacturer’s instructions and imaged in a EVOS M5000 Imaging System (Thermo Fisher Scientific). For experiments with hydrogen peroxide (H₂O₂), it was pre-diluted at 1 μM in RPMI medium and added to the cells accordingly.

Determination of Reactive Oxygen Species (ROS) Production

To assess ROS production, 5×10⁵ neutrophils from WT or gp91phox KO mice were incubated with 100 nM phorbol myristate acetate (PMA, positive control), 10 μM αSF, and 1.2 μM dihydrorhodamine 123 (DHR, Sigma-Aldrich). Fluorescence intensity was measured using a SpectraMax microplate reader (Molecular Devices), with excitation at 500 nm and emission at 540 nm.

Production of Supernatant of Human PMA-Activated Neutrophil (SPAN)

Purified neutrophils (8 × 10⁶ cells/well) from two donors were cultured in 24-well plates (Corning⁠) and incubated with 100 nM PMA (Sigma-Aldrich⁠) in RPMI 1640 medium (Thermo Fisher Scientific⁠) at 37 °C in 5% CO₂ for 4 h. The pooled supernatants were then collected and centrifuged at 400 × g for 10 min. The pellet was discarded, and the supernatants were aliquoted and stored at −80 °C until further use. NET quantification in the SPAN was performed using the PicoGreen dsDNA kit (Invitrogen Life Technologies), and dsDNA concentration was determined by interpolation from a DNA standard curve. To ensure consistency across independent SPAN preparations and minimize batch-to-batch variability, all SPAN samples used in α-syn fibril digestion experiments were normalized to a final dsDNA concentration of 1 μg/mL.

In vitro Digestion of Amyloid Fibrils by Neutrophil Elastase and SPAN

To evaluate the digestion of αSF by neutrophil elastase and SPAN, 5 μM of αSF were incubated with 10 μg/mL NE (Sigma-Aldrich, #324681) and 1 μg/mL SPAN at 37 °C for 4 h. Aliquots were withdrawn at 0, 15, 30, 60, 90, 120, 180, and 240 min, and αSF degradation was monitored by 15% SDS-PAGE followed by Western Blotting using a PVDF membrane previously activated with methanol. The membrane was blocked with Intercept (PBS) Blocking Buffer (LI-CORⓇ) for 1 h, incubated with anti-human α-Syn antibody (1:5,000; Thermo Fisher Scientific) overnight, and with the corresponding secondary antibody (IRDyeⓇ 800CW, 1:5,000, LI-CORⓇ) for 2 h. Visualization was performed on the LI-CORⓇ Odyssey Scanner 9120, and quantification was done on ImageJ software (National Institutes of Health). Fibril digestion was further evaluated by Transmission Electron Microscopy (TEM) using negative contrast with 2% uranyl and by Thioflavin T binding using the ChronosDFD ISS spectrofluorimeter (ISS Inc., Champaign, IL, USA) with excitation at 450 nm and emission from 470 nm to 600 nm.

To determine whether H₂O₂ produced by NET-associated MPO contributes to amyloid fibril fragmentation, αSF was incubated with 10 μM H₂O₂. ThT fluorescence was monitored for 180 min and compared with the ThT signal of αSF incubated in the absence of H₂O₂.

In vivo Experiments

Instillation of Amyloid Fibrils in the Lung

The animals were anesthetized by aerosolization of isoflurane (4%) (Cristália, São Paulo) and instilled intranasally with 50 μg of α-synuclein fibrils diluted in 40 μL of sterile saline (NaCl 0.9%). Control animals received the same volume of saline. After instillation, the animals were subjected to tests to analyze lung function and leukocyte infiltrate.

Invasive Assessment of Respiratory Mechanics

Animals were anesthetized via intraperitoneal (i.p). injection with ketamine (20 mg/kg) and xylazine (10 mg/kg), followed by a single i.p. dose of sodium pentobarbital (60 mg/kg) diluted in distilled water. Tracheostomized animals were mechanically ventilated, and lung function was assessed. The trachea was cannulated and the cannula connected to a pneumotachograph. Airflow and transpulmonary pressure were recorded with a FinePointe R/C Buxco Platform (DSITM, Minneapolis), which calculated resistance (cmH₂O.s/mL) and dynamic lung compliance (mL/cmH₂O) in each breath cycle. Elastance was calculated as the inverse of compliance values. Animals were allowed to stabilize for 5 min and increasing concentrations of methacholine (9–162 mg/mL) were aerosolized for 5 min each. Baseline pulmonary parameters were assessed with aerosolized PBS.

Bronchoalveolar Lavage and Leukocyte Analysis

Mice were killed by an overdose of sodium pentobarbital (500 mg/kg, i.p.; Sigma Aldrich, São Paulo, Brazil). Leukocytes in the airways were assessed by bronchoalveolar lavage fluid (BALF) collection. Briefly, a tracheal incision was made, and a polyethylene catheter (1.7 mm OD) was inserted. The airways were washed three times with 0.5 mL of phosphate-buffered saline (PBS) containing 10 mM EDTA. BALF was centrifuged (300g, 10 min, 4 °C), and cell pellets were resuspended in 250 μL PBS with 10 mM EDTA. For total leukocyte quantification, BALF samples were diluted in Türk solution (2% acetic acid) and cells were counted using a Neubauer chamber under a light microscope (BX40; Olympus, Center Valley, PA). Differential cell counts were performed on cytospin smears stained with May-Grünwald Giemsa and analyzed by light microscopy (BX40; Olympus).

Determination of Myeloperoxidase (MPO) in the Lung

Myeloperoxidase activity was determined using one-third of the tight lung lobe. Cryopreserved lung tissue samples were homogenized (Homogenizer; Omni International, Kennesaw, GA, USA) and centrifuged at 10,000 x g for 10 min. The resulting pellets were resuspended in PBS supplemented with 0.5% HTAB and 5 mM EDTA, followed by centrifugation to isolate the supernatants, which were incubated in a reaction mixture containing PBS-HTAB-EDTA, HBSS, O-dianisidine dihydrochloride (1.25 mg/mL), and 0.4 mM H₂O₂. After 15 min at 37 °C with agitation, the reaction was stopped by adding 1% NaN₃. MPO activity was quantified spectrophotometrically at 460 nm, normalized to total protein content, and expressed as optical density per mg of protein.

Histological Analysis

The left lung lobes were removed and fixed in Millonig buffer solution (pH 7.4) with 4% paraformaldehyde, following standard procedures: Dehydration in a graded ethanol series (70% to 100%); clearing with xylene; embedding in paraffin; sectioning using a microtome (5 μm thickness), and staining with Hematoxylin and Eosin (H&E) for general lung tissue architecture; Picrosirius Red (PS) for collagen deposition, and Congo Red for detection of amyloid fibrils. The slides were coded, and the analyses were carried out blindly. Random airways were photographed using a light microscope Olympus BX40 (Olympus, PA, USA) at x 400 and a digital camera.

Congo red-positive aggregates were quantified from fluorescence images acquired in the red fluorescence channel (Texas Red filter set), in which Congo red-bound amyloid exhibits maximal fluorescence emission, using ImageJ (NIH, Bethesda, MD, USA). Brightfield images were used solely for histological reference. All fluorescence images were acquired using identical microscope settings (laser power/exposure, gain, and offset) across experimental groups. Images were converted to 8-bit grayscale, and a single intensity threshold was applied uniformly to all images within the experiment to identify Congo red-positive signal. Total fluorescence intensity and aggregate size were quantified using the Analyze Particles function after thresholding. The same threshold and particle analysis parameters were applied to all samples.

Assessment of Lung Function Following Amyloid Fibril Instillation

To evaluate lung injury induced by amyloid fibril instillation, animals were pre-anesthetized via intraperitoneal (i.p). injection with ketamine (20 mg/kg) and xylazine (10 mg/kg), followed by a single i.p. dose of sodium pentobarbital (60 mg/kg) diluted in distilled water. Animals were then positioned on a heated platform within a plethysmograph chamber to maintain normothermia. A tracheostomy was performed, and a cannula was inserted to enable mechanical ventilation, ensuring constant flow and tidal volume parameters. Pulmonary function was assessed using an invasive whole-body barometric plethysmography system (Buxco System – DSI). Lung elastance (mL/cmH₂O) and airway resistance (cmH₂O.s/mL) were measured at baseline and following intratracheal aerosolization of increasing concentrations of methacholine (9–162 mg/mL in PBS). Each dose was delivered over a 5-min period to evaluate airway hyper responsiveness.

Immunohistochemistry for NET Detection in Lung Tissue

To assess the presence of NETs in lung tissue, immunohistochemistry was performed following standard deparaffinization and antigen retrieval procedures, as previously described.³³ Tissue sections were blocked with 5% bovine serum albumin (BSA, Sigma-Aldrich #A7906) and incubated with primary antibodies against myeloperoxidase (1:500; R&D Systems, AF3667) and citrullinated histone H3 (CitH3, 1:250; Abcam AB5103) overnight. Afterward, tissue sections were washed and incubated with fluorophore-conjugated secondary antibodies Alexa Fluor 647 and Alexa Fluor 488 (1:500; Thermo Fisher Scientific A-21447 and A-11008) for 2 h.

Parkinson’Disease Human Brains

Post-mortem brain tissue from two male patients with Parkinson’s disease (PD) was used for immunofluorescence analyses. One patient was 76 years old with a clinical diagnosis at 66 years of age, and the other was 69 years old with a diagnosis at 65 years of age. Tissue samples were obtained from the Multiple Sclerosis and Parkinson’s Tissue Bank (UK). All procedures involving human tissue were approved by the Research Ethics Committee of the CHU de Québec (CA), protocol A13-02-1138.

Immunofluorescence

Parafilmed brain sections of the human PD substantia nigra were warmed at 60–70°C for 20 min. Brain sections were sequentially washed in citrisolv (Decon Laboratories, #1601), ethanol (100%, 100%, 95%, 95%, 70%, 50%) and milliQ water. Then, brain slices were incubated with citrate buffer (pH 6.2) for 30 min at 80 °C. Following three washes (8 minutes/each wash) in phosphate-buffered saline (PBS), the brain slices were blocked for 30 min at room temperature in a solution containing 4% normal donkey serum, 0.1% Triton X-100 (Sigma, #T8787), 1% BSA (BioShop, #9048-46-8) and PBS. The slices were then incubated overnight at 4°C with primary antibodies: anti-MPO (R&D Systems, #AF3667, 1:500), anti-citrullinated Histone 3 (citH3, Abbovax, #630-180, 1:250). The following day, the substantia nigra sections were washed three times (8 minutes/each wash) with PBS and subsequently incubated with secondary antibodies: donkey anti-goat 647 for MPO (Invitrogen, #A21447, 1:500) and donkey anti-rabbit 488 for citH3 (Invitrogen, #A21206, 1:500) for 2.5 h at room temperature. Brain sections were then washed three times for 8 min in PBS, following by an incubation with amytracker 520 (1:1000) diluted in PBS for 15 min. After 3 washes in PBS, substantia nigra slices were incubated with 4′6-diamidino-2-phenylindole (DAPI) nuclear stain (Invitrogen, #D35571, 1:2000) diluted in PBS for 7 min. Following three washes with PBS (8 minutes/each wash), brain slices were exposed to Sudan black 0.1% diluted in ethanol 70% for 5 min. Finally, sections were rinsed in ethanol 70%, washed with PBS (three times, 5 min/each wash) and coversliped on slides (Fisher Scientific, #12-550-15) using Fluoromount-G (Invitrogen, #00-4958-02).

Statistical Analysis

Statistical analyses were performed using GraphPad Prism version 9.0.1 (GraphPad Software, San Diego, CA, USA). Data are presented as mean ± standard error of the mean (SEM), unless otherwise indicated. Normality was assessed using the Shapiro–Wilk test. For comparisons between two groups, an unpaired two-tailed Student’s t-test was used for normally distributed data. Comparisons involving multiple groups were performed using one-way or two-way analysis of variance (ANOVA), as appropriate, followed by the recommended post hoc multiple-comparison test indicated in each figure legend. Statistical significance was defined as p < 0.05. Female and male mice were analyzed independently. Statistical analyses were performed separately for each sex, and sex was not included as a covariate or factor in the statistical models.

Results

As previously reported, AFs act as potent ROS-dependent inducers of NETs in vitro, using human neutrophils as a model. We also detected the presence of NETs in the lungs and skin of a post-mortem patient with primary amyloidosis.²⁷ Based on these observations; we sought to determine whether NET formation exerts beneficial or detrimental effects in the context of amyloid diseases in an in vivo setting. To address this, we employed a mouse strain deficient in NOX-2-mediated ROS production as an experimental model for NET deficiency. This strain carries a deletion in the gene encoding the catalytic subunit of NOX-2 (gp91phox), which is located on the X chromosome. Consequently, both male (gp91phox^y/-) and female (gp91phox^−/−) mice lack functional NOX-2 activity. Both sexes were included in this study, as specified in the experiments.

AFs Induce NET Formation in WT Murine Neutrophils but Not in gp91phox KO Neutrophils in vitro

As a first step in our characterization, we sought to determine whether AFs could trigger the release of NET, a phenomenon previously demonstrated only in human neutrophils. To validate the functional deficiency of NOX-2, we evaluated ROS production in bone marrow neutrophils isolated from gp91phox KO and WT mice following stimulation with PMA, a well-established ROS inducer, or pre-formed αSF, which properties are shown in Supplementary Figure 1. As expected, neutrophils from gp91phox KO mice (red symbols) failed to generate ROS in response to both stimuli, whereas WT neutrophils (black symbols) showed robust ROS production, confirming the absence of functional NOX-2 activity in gp91phox KO animals (Figure 1A and inset).

Figure 1.

Amyloid fibrils induce neutrophil extracellular trap (NET) formation in vitro in a NADPH oxidase-2 (NOX-2) dependent manner. (A) ROS production in wild-type (WT) and gp91phox knockout (KO) neutrophils stimulated with PMA or αSF for 1 h. Inset: area under the curve (AUC) analysis. (B) Time course of NET release after stimulation with 10 μM αSF or 100 nM of PMA for 3 h as measured by Sytox green. (C) Representative images of WT and KO neutrophils stimulated with αSF or αSF + H₂O₂ for 2 h. Extracellular DNA was visualized using Sytox Green staining. (D) Quantification of NET formation corresponding to the conditions in panel C, based on Sytox Green fluorescence Scale bar = 200 μm. Data are expressed as mean + SEM. *p < 0.05; **p < 0.01 vs control from 5 independent replicates. Images shown are representative of at least 5 independent replicates. Legends are presented in the right (circles = non stimulated; squares = PMA and triangles = αSF).

Next, we performed a time-course analysis of NET production in WT and gp91phox KO neutrophils stimulated with αSF or PMA. As shown in Figure 1B, WT neutrophils (black curves) exhibited increase of approximately 10-fold in Sytox Green fluorescence, an indicative of NET formation, after 3 h of stimulation with both αSF or PMA. In contrast, gp91phox KO neutrophils (red symbols) failed to form NETs in response to either stimulus. However, when KO neutrophils were costimulated with αSF and a low concentration of H₂O₂ (1 μM), they underwent NETosis, as demonstrated by the increase in Sytox Green fluorescence (Figure 1C and quantified in D) and the presence of web-like structures characteristic of NETs visualized by microscopy (Figure 1C). These results demonstrate that murine neutrophils, like human neutrophils, are susceptible to AF-induced NET formation in a NOX-2-dependent manner. gp91phox KO neutrophils fail to generate NETs in response to αSF or PMA unless exogenous H₂O₂ is provided (Figure 1D), highlighting the critical role of NOX-2-derived ROS in this process.

αSF Trigger the Recruitment of Immune Cells to the Lungs Independently of NOX-2 Activity

Next, to gain better insight into whether AFs can trigger the recruitment of immune cells, we instilled αSF into the lungs of both WT and gp91phox KO mice. The choice of the lung was based not only on its association with systemic amyloidosis, but also on its widespread use in acute inflammation models. As shown in Figure 2, both WT and gp91phox KO mice, female (Figure 2A–F) and male (Figure 2G–L), recruited leukocytes (A and G), mononuclear cells (B and H), and neutrophils (C and I) to their lungs 8 h after αSF instillation, although at different levels. Notably, in both sexes, gp91phox KO mice (red bars) exhibited markedly greater leukocyte recruitment than WT mice (black bars) following αSF instillation. This increase was particularly evident for neutrophils, which were approximately 2.5-fold higher in females and 4-fold higher in males than in WT mice. Interestingly, male mice recruited substantially more neutrophils than female mice in both WT and gp91phox KO backgrounds. A striking observation was the absence of mononuclear cell recruitment in female WT mice following fibril instillation due to reasons that we do not know at the present.

Figure 2.

αSyn amyloid fibrils (αSF) instilled into the lung of WT and gp91phox KO animals recruit leukocytes. 50 μg of αSF were intratracheally instilled into the lungs of the animals and after 8 h cellular recruitment was assessed. (A and G) Total leucocytes; (B and H) Mononuclears; (C and I) Neutrophils and (D and J) MPO activity were quantified in the BALF. The data are shown separately for females (A–F) and males (G–L). The lower images show representative BALF images with the recruited immune cells for female (E and F) and male (K and L) mice. White arrows point to neutrophils. Scale bar: 100 μm. Results shown as mean ± SEM. * p < 0.05 compared to the control group instilled with saline (Ctl); # p < 0.05 compared to the WT group instilled with αSF. For female mice, sample sizes were n = 8 for controls and n = 9 for αSF-instilled animals. For male mice, wild-type groups included n = 10 for both controls and αSF-instilled, while gp91phox^y/− mice included n = 3 for controls and n = 6 for αSF-instilled animals. Scale bar = 100 μm.

Representative images of the bronchoalveolar lavage fluid (BALF) of female and male mice are shown in Figure 2E, F, K and L, respectively. To further support the neutrophil recruitment observed in the BALF, we measured MPO levels in the lungs (Figure 2D and J). MPO was significantly increased in both gp91phox KO and WT mice (female and male) instilled with AFs compared to saline-treated controls, reinforcing the conclusion that AFs promote neutrophil recruitment.

Clearance of αSF Is Impaired in gp91phox KO Mice

To further investigate the inflammatory processes induced by αSF instillation in the lungs, we employed a series of histological staining techniques. Hematoxylin and eosin (H&E) staining of lung sections revealed a densely populated pulmonary parenchyma and alveolar infiltration in both WT and gp91phox KO mice (Figure 3A). Picrosirius staining further demonstrated collagen deposition (red) following αSF instillation in both genotypes, consistent with an acute inflammatory response. Notably, we observed several “toothpick-like” structures, predominantly in gp91phox KO lungs, morphologically resembling AFs (red/yellow structures pointed with white arrows). Higher magnification images are provided for improved visualization in the bottom of the Figure.

Figure 3.

Characterization of lung injury induced by αSF instillation in WT and gp91phox KO mice. (A) Histological analysis of lung sections collected 4 h after αSF instillation, stained with hematoxylin and eosin (H&E) to evaluate tissue architecture, Picrosirius red for collagen deposition, and Congo red for amyloid detection. Lower panels show magnification of the areas highlighted in the rectangles, where it is possible to see (arrows) the presence of toothpick-like structure with fibrillar appearance in the lungs of gp91phox KO mice. (B) Quantification of Congo red fluorescence intensity, indicating the extent of amyloid deposition in lung tissue. (C) Measurement of aggregate size in lung sections from each experimental group. Data present in (B and C) are shown as mean ± SEM of 5 animals per condition. Statistical analysis was performed using one-way ANOVA in (B): * = p < 0.05; ** = p < 0.01 (compared to the control group) and Student’s t-test in (C): **** = p < 0.01 compared to the control group. In these experiments, only male mice were used. Scale bar = 200 μm. Data were analyzed using one-way ANOVA followed by Tukey’s multiple comparisons test. Statistical significance was set at p < 0.05.

To confirm the amyloid nature of these structures, we performed Congo Red staining, which specifically labels β-sheet–rich aggregates. Quantification of Congo Red–positive structures showed a significantly greater number and larger average size of aggregates in gp91phox KO lungs compared to WT mice at 4 h post-instillation (Figure 3A–C). In WT lungs, only small aggregates were detected at this time point.

Together, these findings suggest that αSFs are efficiently cleared from WT lungs within 4 h of instillation, whereas in gp91phox KO mice, fibrillar structures persist and accumulate, potentially exacerbating lung inflammation.

αSF Trigger NETs Formation in vivo in a NOX-2-Dependent Manner

Having established that AFs induce NET formation in WT murine neutrophils in vitro, and that AFs instilled into the lungs of WT and gp91phox KO mice recruit immune cells, including neutrophils, we next investigated whether AFs could trigger NET formation in vivo. For this, we stained αSF-instilled lungs with anti-H3Cit and anti-MPO antibodies, two NET associated proteins that, when colocalized, suggest NETosis. As shown in Figure 4, colocalization of H3Cit and MPO was observed exclusively in the lungs of WT mice instilled with αSF, confirming NET formation in response to αSF instillation. We did not notice the presence of NET markers in the lungs of αSF-instilled gp91phox KO mice, which is expected since these animals are ineffective in ROS-dependent NET production.

Figure 4.

αSF induces NET formation in vivo in a NADPH oxidase 2-dependent manner. Immunofluorescence analysis of male lung sections of WT and gp91phox KO mice 4 h after intratracheal instillation with either saline or 50 μg of αSF. NETs were identified by co-localization of citrullinated histone H3 (green; anti-CitH3 antibody) and myeloperoxidase (red; anti-MPO antibody). Nuclei were counterstained with DAPI (blue). Scale bar = 200 μm.

NETs Mediate Lung Recovery After Amyloid Fibril Instillation

To determine whether αSF deposition affects lung function, we evaluated lung pulmonary parameters both in male and female of WT and gp91phox KO mice 8 h after αSF instillation. For this, resistance and elastance were evaluated using invasive whole-body plethysmography, following exposure to increasing doses of the bronchoconstrictor agent methacholine. As shown in Figure 5, female and male of gp91phox KO mice instilled with αSF exhibited a significant increase in lung elastance, beginning at 9 mg/mL and 27 mg/mL of methacholine, respectively (red bars in Figure 5A and C). Notably, except for male gp91phox KO mice instilled with αSF and treated with 127 mg/mL of methacholine, no significant changes in lung resistance were observed in WT or KO animals, female or male, following αSF instillation, even when high doses of methacholine were used (Figure 5B and D). This selective increase in elastance, without a corresponding change in resistance, suggests that the damage is primarily localized to the lung parenchyma, without airway obstruction. In contrast, male and female WT mice 8 h after AFs instillation did not display significant alterations in either lung elastance or resistance, even at higher methacholine concentrations (Figure 5A–D).

Figure 5.

αSF trigger alterations in pulmonary function of mice. Lung injury was assessed by measuring elastance (A and C) and resistance (B and D) 8 h after instillation of 50 μg of AF, following increasing doses of the bronchoconstrictor agent methacholine. Results are presented separately by sex (A and B, female; C and D, male). (E) Elastance measurements with male WT mice instilled with AFs as a function of time. Note that at 3h elastance if compromised in these lungs. Results expressed as mean ± SEM. * p < 0.05 compared to the control group instilled with saline (Ctl); # p < 0.05 compared to the WT group instilled with αSF; Female: n = 8 for Ctl and 9 for instilled with αSF in both groups. Male: n = 10 for WT in both groups, 3 for gp91phox^y/- Ctl and 6 for gp91phox^y/- αSF. Kinetic data with males: n = 5–6.

Since αSFs were no longer detectable in the lungs of WT mice 8 h after instillation, the time point at which impaired lung function was observed in KO animals, we evaluated lung elastance in male WT mice at earlier time points (Figure 5E). Remarkably, a significant impairment in lung function was already evident 3 h after αSF instillation in WT mice, with progressive recovery over time. This finding suggests that αSF cause a transient disruption of lung mechanics in WT animals, which is resolved as the fibrils are effectively cleared.

Are NET’s Components Able to Digest AFs into Small Species?

Since NETs are composed of various proteases, including elastase, next we investigated whether the proteases associated with NETs could degrade αSF into smaller species in vitro. To this end, αSF were incubated with SPAN (Supernatant of PMA-Activated human Neutrophils), a suspension enriched in NETs or with purified neutrophilic elastase (Figure 6).

Figure 6.

SPAN and neutrophilic elastase digest αSF into small fragments. (A) Thioflavin T (ThT) fluorescence emission spectra of aSF at different times in the presence of SPAN (left), neutrophilic elastase (middle), and 10 M H₂O₂ (right). (B) Quantification of the ThT emission spectra extracted from panel (A) SPAN (continuous line), Elastase (dotted line), and H₂O₂ (dashed line). The intensities of αSyn monomers bands, quantified from the Western blots presented in Supplementary Figure 2 are displayed in blue (SPAN, continuous line and Elastase, dotted line). (C) TEM images for the mature SF (left panel), fragments generated after 2 h in the presence of SPAN (SF +SPAN), Elastase (SF + NE), or H₂O₂ (SF + H₂O₂). Scale bar = 500 nm. ThT emission: Ex = 450 nm, Em = 482. [SPAN] = 1 μg/μL; [Neutrophilic Elastase] = 10 μg/μL; [H₂O₂] = 10 M, and [aSF] = 5 μM.

As shown in Figures 6A, ThT fluorescence, an indicator of amyloid fibrils, decreased in a time-dependent manner upon incubation of αSF with SPAN (left panel) or elastase (right panel), indicating progressive fibril degradation while untreated αSF maintained stable ThT fluorescence over time (not shown). Transmission electron microscopy (TEM) images (Figure 6C) further confirmed the structural disruption of αSF after 2 h of incubation with SPAN or elastase (middle and right panels, respectively), revealing the transformation from intact fibrils to fragmented structures. Western blot analysis using an anti-α-synuclein antibody (Supplementary Figure 2) showed a reduction in the monomeric α-synuclein band, consistent with proteolytic processing by SPAN and elastase, and their quantification together with ThT changes are displayed in Figure 6B.

To exclude the possibility that ROS produced, for instance, by NET-associated MPO contribute to fibril fragmentation, mature αSF were incubated with H₂O₂, and ThT fluorescence emission was monitored over time. As shown in Figure 6A and B, no change in ThT emission was observed up to 180 min, suggesting that this ROS does not promote amyloidolysis under the conditions tested. These observations were confirmed by TEM images (Figure 6C).

NETs are Present in the Substantia Nigra of Brain Sections of PD Patients

To confirm that NETs are formed in human PD brains and to strengthen the translational relevance of our findings, we evaluated post-mortem brain sections from two PD patients for the presence of NET markers Strikingly, we were able to detect NET formation in both cases, as evidenced by the colocalization of MPO and H3Cit (Figure 7). Moreover, NET markers colocalized with amyloid aggregates, as revealed by Amytracker staining, a dye specific for amyloid structures.

Figure 7.

Colocalization of NET markers and aggregates in human PD substantia nigra. (A and B) Representative confocal photomicrographs of brain slices labeled by immunofluorescence to detect MPO (red), citH3 (green), Amytracker (Orange) and DAPI-nucleus (blue) in 2 PD patients. Scale bar = 20 μm. All data are shown as mean ± SEM with individual points representing values for each mouse.

Abbreviations: citH3, citrullinated histone 3; MPO, Myeloperoxidase; PD, Parkinson’s disease.

Discussion

Since the seminal description of NETs by Brinkmann and coauthors,¹⁶ these structures have been implicated in a wide range of pathological conditions.²⁰ Our group was among the first to propose a direct link between NET formation and neurodegenerative diseases caused by amyloid deposition.²⁷ Specifically, we demonstrated that amyloid fibrils of various compositions, including α-synuclein, can induce ROS-dependent NET release from human neutrophils in vitro. Not only AFs were able to trigger NETosis, but they were also digested by proteases, mainly elastase embedded within these traps, into toxic oligomeric species. We reported the simultaneous presence of AFs and NET markers in human tissues, specifically in lung and skin biopsies from patients who died from primary systemic amyloidosis.²⁷

However, despite this striking observation, it remained unclear whether AFs could actively recruit neutrophils or directly induce NET release in vivo. To address this, we employed a mouse model deficient in NOX-2 activity, a key enzyme required for the generation of ROS in phagocytes, mainly neutrophils. These mice, originally developed as a model for chronic granulomatous disease, are unable to produce ROS and, consequently, fail to generate ROS-dependent NETs, as demonstrated in this study (Figure 1).³⁰ This makes them a valuable tool to investigate the contribution of NETs in pathologies where ROS-mediated mechanisms drive NET formation. ROS-independent NET formation has been described,^24,34 but we did not address this possibility in the present study, mainly because in our previous publication, it was very clear that AF induced NET formation in ROS dependent manner with DPI completely abolishing it.²⁷

Consistent with our in vitro observations in human neutrophils, we found that αSF stimulated NET formation in murine neutrophils in a ROS-dependent manner (Figure 1A). This response was evident in WT mice but absent in gp91phox-deficient animals, as confirmed by NET formation kinetics measured through Sytox Green and fluorescence images (Figure 1B and C). Supplementing gp91phox-deficient neutrophils with low concentrations of hydrogen peroxide (1 μM H₂O₂) restored their ability to release NETs upon αSF stimulation, reaching levels comparable to WT controls (Figure 1C and D). Previous studies have shown that modulation of the neutrophil microenvironment, such as extracellular pH, can spontaneously induce NET release.³⁵ In our experiments, the low dose of H₂O₂ used did not induce NETs by itself, in contrast to higher concentrations previously reported to do so.³⁵ Together, these findings strongly suggest that the impaired NET formation in gp91phox KO neutrophils results from ROS deficiency rather than from intrinsic defects in the molecular machinery regulating NETosis.

Once we confirmed that αSF–induced NETosis occurs in vitro in murine neutrophils, we next investigated whether αSF could recruit neutrophils in vivo. To this end, 50 μg of αSF were instilled intratracheally into the lungs of both WT and gp91phox-KO mice of both sexes. Eight hours later, leukocyte recruitment was observed in both sexes (Figure 2A and G), with neutrophils representing a major proportion of the infiltrating cells (Figure 2C and I). Notably, gp91phox-KO mice of both sexes recruited significantly more leukocytes than WT controls, likely as a compensatory response to their ROS deficiency (Figure 2A, C, G, and I). This enhanced neutrophil influx was also accompanied by increased MPO activity (Figure 2D and J). Elevated blood levels of MPO and other neutrophil-derived enzymes have previously been linked to cognitive decline in AD, underscoring a possible pathological relevance of dysregulated neutrophil activity in neurodegeneration.³⁶

Sex-specific differences were also observed. Male WT mice recruited mononuclear cells at levels comparable to KO animals, whereas female WT mice displayed no detectable mononuclear cell recruitment following αSF instillation. Although the basis of this difference remains unclear, similar sex-dependent immune responses have been reported.^37,38 Furthermore, male KO mice recruited more immune cells than their female counterparts, possibly reflecting their hemizygous status for the X-linked gp91phox gene. A comparable phenomenon was described previously in these mice and reported enhanced immune cell recruitment, including neutrophils, into the peritoneal cavity after thioglycolate challenge compared to WT controls.³⁰

Lung sections from both WT and gp91phox KO mice, examined 4 h after αSF instillation, showed intense cellular infiltration (H&E) and alveolar obstruction by instilled AFs (confirmed by CR staining) (Figure 3B and C). These alterations were more pronounced in gp91phox KO animals than in WT (Figure 3A). In addition to cell infiltration, both genotypes displayed increased collagen deposition within the parenchyma, a process known to represent an early hallmark of acute pulmonary inflammation triggered by antigenic stimulation or injury.³⁹

The colocalization of H3Cit, a hallmark post-translational modification of NETs, with MPO confirmed NET formation in vivo in WT mice.⁴⁰ In contrast, H3Cit was undetectable in gp91phox-KO mice, either alone or in colocalization with MPO, indicating that ROS generation is an upstream requirement for amyloid fibril–induced NET release (Figure 4), consistent with findings for other NET stimuli.^18,24 Collectively, these findings demonstrate that αSF recruit neutrophils and induce NETosis in vivo in a NOX-2-dependent manner.

As previously demonstrated in our in vitro studies with transthyretin fibrils, NET associated proteases, particularly neutrophil elastase, can degrade AFs into smaller, toxic species.²⁷ In line with this, 4 h after αSF instillation, large toothpick-like structures were still evident in the lungs of KO mice (Figure 3A), whereas WT animals displayed fewer and smaller Congo red–positive amyloid-like deposits, suggesting a possible role of NETs in amyloid clearance. However, it is not possible to exclude the contribution of other AF clearance mechanisms operating in this animal model, such as phagocytosis by professional immune cells, proteolytic degradation by extracellular or cell-associated proteases among others.

Given that our earlier work had focused on TTR fibrils, we next examined whether NET-associated proteases could similarly digest αSF. For this, αSF was incubated in vitro with SPAN or neutrophilic elastase for 3 h, and fibril degradation was monitored by Western blotting (Supplementary Figure 2) and ThT fluorescence (Figure 6). The progressive disappearance of the αSF-specific band, together with a time-dependent decrease in ThT fluorescence, indicated amyloidolysis and loss of amyloid structure (Figure 6B). TEM further confirmed this breakdown, revealing that SPAN- and neutrophilic elastase-mediated digestion produced shorter and less clustered fibrils (Figure 6C). To exclude a direct effect of ROS produced by NET-associated MPO on fibril integrity, we examined whether addition of H₂O₂ would promote fibril fragmentation. Under the conditions tested, no fibril fragmentation was observed (Figure 6). This finding is consistent with reports showing that ROS can promote αSyn aggregation rather than amyloid fibril degradation.^41,42 Taken together, these results strongly suggest that, like TTR, αSF can be digested by NET-associated proteases into smaller, potentially toxic species.

In addition, to evaluate the impact of aggregates and the NETs induced by them in the lung, we performed functional assays using plethysmography to measure elastance and resistance. Our analysis revealed no evidence of pulmonary dysfunction 8 h after αSF instillation in the WT mice (Figure 5A–D). However, a transient reduction in lung elastance was detected at earlier time points (Figure 5E), reflecting inflammation at the level of the lung parenchyma. Regarding the KO mice, even 8 h after αSF instillation, lung elastance was still altered. These results indicate that the presence of αSF causes acute lung injury and that their efficient elimination in WT mice protects the lungs from sustained damage, the persistence of αSF in the KO mice, which results in lung injury. Whether these smaller species can disseminate to other organs and tissues and seed pathology remains to be determined.

The involvement of neutrophils in the prognosis of amyloidosis has been previously reported. For instance, neutrophil infiltration in the brain parenchyma was identified in two transgenic mouse models of Alzheimer’s disease (5xFAD and 3xTg-AD).⁴³ In addition to the presence of neutrophils, the authors detected NET-associated markers within the brain tissue and postmortem human brain tissue form AD patients.⁴³ Depletion of neutrophils in these animal models significantly restored cognitive function, and ameliorated disease-related symptoms. In another study using the APP/PS1 mouse model of AD, researchers reported elevated levels of neutrophils in the brain compared to wild-type controls.⁴⁴ However, it remains unclear whether the observed improvement was due to the absence of neutrophils themselves or the lack of NET production.

More recently, the role of neutrophils in the clearance of human AL amyloid extracts was analyzed using murine amyloidome models composed of ALλ, which is refractory to clearance, and ALκ, which is readily cleared in mice.⁴⁵ Their data showed that more neutrophils were recruited to ALκ amyloid masses as compared to the ALλ material, which was devoid of neutrophils. Interestingly their data showed by ex vivo analyses that neutrophils do not efficiently phagocytose amyloids directly but, instead, in the ALκ amyloidome there was an abundant presence of NETs, which were absent in the ALλ amyloidome. These findings are in accordance with ours, which suggest the participation of NETs in amyloid clearance.

As we recently showed, NETs may act as a double-edged sword.⁴⁶ On the one hand, as demonstrated here, NETs can degrade AFs, thereby protecting the lungs or other tissues from the detrimental accumulation of these aggregates. On the other hand, fragments generated by NET-induced fragmentation of TTR amyloid fibrils were shown to be toxic to cultured hamster kidney cells and immortalized human liver cells.²⁷ Thus, we cannot exclude the possibility that fragments generated by NET protease-mediated digestion of AFs may be toxic to pneumocytes.

Besides, another aspect that must be considered is the damage that NET components cause to the surrounding tissues. In the peripheral nervous system, our group previously showed that NET-associated components, such as MPO and histones, are highly toxic to neurons and Schwann cells dissociated from dorsal root ganglia.⁴⁶ Thus, NET components can also contribute to lung tissue damage. All these aspects must be balanced in the course of an amyloid disease: deposition of AF versus fibril clearance induced by NETs with generation of toxic fragments; and toxicity of NETs components to the surroundings. These aspects are currently under investigation in our group.

Beyond the data presented here supporting a role for neutrophils and their extracellular traps in amyloidosis, it is important to acknowledge certain limitations of the model employed. To the best of our knowledge, α-synuclein aggregates have not been detected in the lungs of patients with PD. Our choice of the lung as a model for amyloid clearance was guided by several considerations: (i) pulmonary involvement is relatively common in systemic amyloidosis, although it rarely manifests with overt clinical symptoms;⁴⁷ (ii) the lung is a well-established model for investigating acute inflammation⁴⁸ and its functionality is easy to assess (elastance and resistance); (iii) its compartment provides direct access for immune recruitment and interactions; and (iv) the lung–brain axis has recently emerged as an area of growing interest, with increasing evidence that these organs may communicate through neuroanatomical, endocrine, and immunological pathways.^49–53 We envision that a similar scenario may occur in the brain of patients with neurodegenerative amyloid diseases, since neutrophil infiltration into the brain parenchyma is well documented and, as previously shown, microglia can also release extracellular traps.^43,54

In fact, we identified NET markers, such as MPO and citrullinated histones, in the brains of two PD patients (Figure 7), where they colocalized with an amyloid probe (Amytracker). Although we acknowledge that this represents a very small sample size, these findings open the possibility for future investigation.

Conclusion

Our study demonstrates that αSF are potent inducers of NET formation in vivo in a NOX-2–dependent manner and that this response may plays a protective role in limiting amyloid-associated tissue damage. While both WT and KO mice exhibited robust neutrophil recruitment following αSF instillation, NET formation occurred exclusively in WT animals, confirming the requirement of functional NOX-2 activity for this process. Importantly, WT mice efficiently cleared amyloid structures and recovered lung function, whereas KO animals showed persistent fibril deposition and sustained impairment in elastance. These findings indicate that NET-associated proteases contribute to amyloid fibril degradation in vivo and mitigate their deleterious functional consequences. Furthermore, the detection of NET markers in post-mortem brain tissue from PD patients supports the clinical relevance of this mechanism. Together, our data position NET formation not merely as a pro-inflammatory event, but as an innate protective response against amyloid burden, with important implications for understanding immune–amyloid interactions in neurodegenerative and systemic amyloid-associated diseases.

Funding Statement

This work was supported by Brazilian Funding Agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and Capes and by the State Funding Agency of Rio de Janeiro, Faperj. T.R.C.P. was supported by a doctoral scholarship from the CNPq and a postdoctoral fellowship from the Canadian Institutes of Health Research (CIHR). E.L. was supported by a doctoral scholarship from Fonds de recherche du Québec en santé (FRQS). F.C. receives funding from the Canadian Institutes of Health Research (CIHR), the Huntington Society Canada (HSC) and the Michael J. Fox Foundation for her research.

Data Sharing Statement

Data and materials supporting results are available upon request to the corresponding author.

Author Contributions

TRCP: Conceptualization, Visualization, Methodology, Investigation, Supervision, Data curation, Formal analysis, Software and Writing original draft.

ASM: Visualization, Methodology, Investigation, Data curation, Formal analysis, and writing original draft.

FVG, PMRS and TRFM: Visualization, Methodology, Investigation, Data curation, Formal analysis.

EL and MSP: Investigation, Data curation, and Formal analysis.

FC and ES: Resources, Funding acquisition, Conceptualization, Supervision, Project Administration.

DF: Resources, Funding acquisition, Conceptualization, Study design, Formal analysis, Supervision, Project Administration, Writing original draft.

All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors report no conflicts of interest in this work.