The Hospital Universitario Virgen Macarena ethics committee approved the study and each sample donor gave written informed consent.

The Ags, available as commercial extracts, included D₁ (Dermatophagoides pteronyssinus), G₃ (Dactylis glomerata), T₉ (Olea europaea), M₆ (Alternaria alternata) and W₆ (Artemisia vulgaris) and were purchased from Diater, (Madrid, Spain). Wortmannin, PD098059, SB203580, 4-hydroxy-3-methoxyaceto-phenone (HMAP), 4-(2-aminoethyl) benzenesulfonyl fluoride (AEBSF), cyclosporin A (CsA), cell permeable NFAT inhibitor (VIVIT), tiotropium bromide, budesonide, formoterol, were from Sigma-Aldrich Co (Madrid, Spain). IL-4, IL-5, GM-CSF, IFN-γ, IL-10 were from Preprotech (Rocky Hill, NJ, USA). Ficoll-hypaque, phosphate-buffered saline, RPMI 1640, fetal bovine serum, penicillin/streptomycin and goat anti-human IgE (α-IgE) were purchased from Thermo-Fisher Invitrogen (San Diego, CA, USA). All cultured reagents had endotoxin levels of ≤ 0.01 ng/ml, as tested by the Limulus lysate assay (Coatest, Chromogenix, Mölndal, Sweden).

The study included three groups of adult donors: atopic patients with bronchial asthma (17) with no Ag-specific immunotherapy treatment (non-IT), atopic patients with bronchial asthma treated with Ag-specific immunotherapy (IT), and healthy donors (HD) (see Table 1 ). The group of asthmatic patients gave positive skin-prick test (SPT) results (Diater) and serum specific-IgE (HYTEC 288, Hycor Biomedicals, Germany) levels ≥50 KU/l to at least one of the inhalant Ags included in the routine testing battery (house-dust mites, pollens, molds and animal danders). Those Ags to which the patients had specific IgE levels ≥50 KU/l were used for the challenge in the experiments. The non-IT group did not receive specific hyposensitization and did not experience episodes of respiratory infections for the last 4 weeks before blood extraction. The IT group received Ag-specific IT (Diater) for the previous three years and continued to receive a maintenance dose ofthe highest dose of the extract. Allergic patients did not take any inhaled bronchodilators within 8 h before cell isolation and the in vitro cellular challenge, oral bronchodilators for 24 h or antihistamines, oral corticosteroids, disodium cromoglycate, or nedocromil sodium in the previous week. The healthy group had no history of allergy or bronchial symptoms, and gave negative skin-prick tests and had specific-IgE to the battery of inhalant Ags. None of the participants in this study suffered infection by SARS-CoV-2 in the month previous to the blood extraction.

Highly purified human peripheral blood neutrophils were isolated using the neutrophil isolation kit (Miltenyi Biotec S.L., Madrid, Spain) following the manufacturer’s instructions. The purity of neutrophils was on average >99% (16). Cells (10⁶ cells/300 μl) were cultured in RPMI 1640 medium supplemented with 10% (v/v) FBS, 2 mM L-glutamine, 100 U/ml penicillim, and 100 μg/ml streptomycin and maintained in an atmosphere of 95% O₂ and 5% CO₂. For the stimulation treatments, neutrophils were incubated with 10 μg/ml Ags or α-IgE antibodies at 37°C for the indicated times. Wortmannin, PD098059, SB203580, HMAP, AEBSF, CsA, VIVIT, IL-4, IL-5, IFN-γ, IL-10, GM-CSF, tiotropium, budesonide, and formoterol were added 1 h prior to stimulation and they were previously tested for the optimal concentration without affecting cell viability. In this case, cell health was assayed using the AlamarBlue kit (Thermo-Fisher) according to the manufacturer’s instructions. This kit quantifies the natural reducing power of living cells to convert resazurin to fluorescent resorufin.

ECP released was measured in the culture supernatants by ELISA (CAP system immunoassay; Phadia-Thermo scientific) according to the manufacturer’s instructions.

FEV1 was measured using a spirometer (Vitalograph, Buckingham, UK). The best value of three maneuvers was expressed as the percentage of the predicted value. The entire procedure was based on the guidelines of the American Thoracic Society of Standardization (18).

Sputum was induced and processed as previously described (19). Briefly, patients inhaled 4.5% hypertonic saline solution at room temperature which was nebulized by an ultrasonic nebulizer (Ultraair NE-U17, Omron Corporation, Japan) at maximal saline output for a 20 min period. Sputum plugs were isolated from the sample and resuspended in PBS supplemented with 100 U/ml penicillin, and 100 μg/ml streptomycin. For ECP levels quantification, samples were directly vortexed and centrifuged (4,000 x g, 10 min, 4°C), and the resulting supernatants were frozen at -80° C until ELISA determination. For cell culture, PBS resuspended plugs were diluted 1/10 in 0.2% dithiothreitol and mechanically mixed for 30 min to disperse the cells. Samples were filtered through a 50 μm strainer and centrifuged for 10 min at 4°C. Processed sputum pellets were resuspended in complete RMPI medium, cells plated on glass coverslips coated with poly-L-lysine and fibrinogen and cultured in the presence or absence of stimulus for 18h. After culture, coverslips were fixed for 30 min with 4% paraformaldehyde in PBS, permeabilized for 15 min at room temperature with 0.5% Triton X-100 in PBS and immediately blocked for 1 h with 1% bovine serum albumin serum in PBT (0.1% Triton X-100 in PBS). After blocking, mouse fluorescein isothiocyanate-conjugated α-human RNase3/ECP mAb (LSBio, Seatle, WA, USA) and mouse phycoerythrin-conjugated α-human Myeloperoxidase mAb (PE-MPO, Beckman-Coulter) diluted in blocking solution at 1:100 final concentration were added and incubated at room temperature for 2h. After washing with PBT, coverslips were mounted onto glass slides with DAPI-containing mounting medium. Cells were imaged using a Stellaris 5 laser scanning confocal microscope from Leica. For 2D, images were obtained using a 20X objective. For 3D, serial optical sections (z-stacks) were obtained using a 63X objective and deconvoluted using ImageJ/Fiji.

All statistical analyses were performed using GraphPad Prism 4.00 for Windows (GraphPad Software, San Diego, CA, USA) and R-4.2.1. Multi-group analysis was performed using two-way ANOVA, followed by post-hoc Tukey’s honest significance tests. Correlation was measured using the Pearson correlation coefficient. Data are expressed as mean ± SEM. A value of p<0.05 was considered significant.

In previous work, we reported that the crosslinking between Ags/α-IgE and IgE molecules bound to surface IgE receptors (Galectin-3> FcεRI) induces the synthesis and release of ECP by human neutrophils from allergic asthmatic patients (16). This prompted us to study the molecular mechanisms underlying this process by treating cells with specific inhibitors of key signaling pathways in neutrophils and measuring their effects on ECP release after treatment with and without Ags or a-(α-IgE).

During the first steps of neutrophil activation there is an increase in phosphoinositides through PI3K activation (20), as well as Mitogen-Activated Protein Kinases (MAPKs) (p38 and extracellular signal-regulated kinase 1/2 (ERK) activation) (21). PI3K and MAPKs regulate the functional assembly of NADPH oxidase (NOX-2) (22), an enzyme producing Radical Oxygen Species (ROS), second messengers involved in the priming and degranulation of neutrophils (23). We have previously shown that PI3K, MAPKs and NOX-2 pathways are activated in neutrophils from allergic patients by Ags/α-IgE (24, 25).

In addition, we have shown that the calcineurin phosphatase (CN)/NFAT transcription factor signaling pathway, a key player in neutrophil activation during the immune responses (26), is also activated in an IgE-dependent manner (27).

In this context, we aimed to explore the specific role of all these molecular intermediates in ECP production. To this end, we tested the effects of the following protein-specific inhibitors: wortmannin (PI3K inhibitor (28)), PD098059 (MEK inhibitor, the upstream kinase of ERK 1/2 (29)), and SB20358 (p38 MAPK inhibitor (30)), HMAP and AEBSF (NOX-2 inhibitors (31)), CsA (CN activity inhibitor (32)) and VIVIT (peptide blocking the interaction CN-NFAT, required for NFAT nuclear translocation (33)).

Inhibitor treatment (F=24.32, df=7, p<2.2x10^-16) and stimulation with Ags/α-IgE (F=327.10, df=2, p<2.2x10^-16) both significantly affected neutrophil ECP release and there was a significant interaction between these factors (F=6.32, df=14, 6.422x10^-09). Tukey’s post-hoc tests showed that wortmannin ( Figure 1A ), PD098059 ( Figure 1B ), HMAP and AEBSF ( Figure 1D ) and CsA/VIVIT ( Figure 1E ) significantly inhibited ECP release induced by Ags/α-IgE. Conversely, SB203580 did not have any significant effect ( Figure 1C ) (p=0.99 for Ags treatment and p=0.77 for α-IgE treatment). For all molecules tested, no changes in ECP release could be detected following treatment in the absence of stimulation.

Neutrophil ECP production is modulated by agonists such as Platelet Activating Factor (PAF) (16). To study whether other cytokines could also modulate ECP release by human neutrophils, we examined the effect of two groups: IL-4, IL-5 and GM-CSF, that promote allergic inflammation (34, 35), and IFN-γ (36) and IL-10 (37), that inhibit this process.

Neutrophils from allergic patients were preincubated with these cytokines 1h prior the addition of Ags/α-IgE. We then quantified ECP production in the culture supernatant. IL-4 did not modify the basal nor the IgE-dependent ECP production. The same dose of IL-5 did not have an effect on its own, but it significantly enhanced the Ags/α-IgE stimulating effect. Interestingly, GM-CSF displayed a robust response in Ags/α-IgE-treated cells, and was the only cytokine producing significant ECP release in unstimulated cells. On the other hand, IFN-γ and IL-10 significantly reduced ECP release in both stimulated and unstimulated cells ( Table 2 ).

Glucorticosteroids (GC) (38), bronchodilators (Long-Acting Beta₂-Agonists, LABAs) (39), and Long-Acting Muscarinic Antagonists, LAMAs) (40) are effective drugs for preventing allergic symptoms. We investigated their possible modulating effect on IgE-dependent ECP release. We tested the action of budesonide (a GC), formoterol and tiotropium (a LABA and a LAMA, respectively) by adding them to the culture of neutrophils from allergic patients (prior to Ags/α-IgE antibodies stimulation) and quantified ECP release.

Our experiments revealed that drug treatment (F=101.45, df=7, p<2.2x10^-16) and stimulation (F=327.10, df=2, p<2.2x10^-16) both significantly affected neutrophil ECP release and there was a significant interaction between these factors (F=28.58, df=14, p<2.2x10^-16). Tukey’s post-hoc tests showed that all three drugs, alone or in combination, led to a significant reduction in ECP release, compared to untreated cells, for both Ags and α-IgE stimulation. Considering the drugs alone, Budesonide led to the largest change, followed by formoterol and tiotropium ( Figure 2 ).

Current medical guidelines recommend the use of a triple therapy combining GC, LABAs and LAMAs for the non-controlled asthma (e.g. GINA) (41). Thus, we perfomed the same experiments using cocktails of these drugs.

The combination of formoterol and tiotropium enhanced the inhibitory effect that they produced alone. This inhibition was more significant when formoterol or tiotropium were combined with budesonide. The mix with all three drugs showed the largest decrease in ECP release ( Figure 2 ). No changes were detected for any of the treatments in absence of stimulation.

Ag-specific IT provides long-term reduction in both allergic symptoms and disease progression (42). This improvement correlates with a decline in inflammatory parameters such as ECP, for which nasal or sputum levels decrease in allergic patients after IT treatment (43, 44). This compelled us to analyze whether IT had any effect on Ags/α-IgE-induced ECP production, as well as on the modulating action of IL-5 and GM-CSF. We evaluated ECP release by neutrophils in three groups of subjects: non-IT treated allergic patients, IT-treated allergic patients and healthy donors. Measurements were performed in the culture supernatants of untreated or Ags/α-IgE treated cells, in the presence or absence of either IL-5 or GM-CSF. We found that IL-5/GM-CSF treatment (F=60.88, df=8, p<2.2x10^-16) and Ags/α-IgE stimulation (F=175.79, df=2, p<2.2x10^-16) significantly affected ECP release and there was an interaction between factors (F=13.80, df=16, p<2.2x10^-16). Further analysis using Tukey’s post-hoc tests showed that the group of IT-treated allergic patients experienced a significant reduction in ECP release in the presence of Ags/α-IgE, compared with the non-IT treated patients ( Figures 3A, B ). This reduction was ~70% (For α-IgE treatment: 29.80 ± 1.94 μg/l in IT-treated allergic patients vs 82.90 ± 5.39 μg/l in non-IT-treated allergic patients; For Ags treatment: 18.60 ± 3.15 μg/l in IT-treated allergic patients vs 69.48 ± 3.11 μg/l in non-IT-treated allergic patients). Although there was no significant difference in ECP release between IT-treated patients and healthy donors, values were higher in the first group with respect to the second (for α-IgE treatment: 29.80 ± 1.94 μg/l in IT-allergic patients vs 17.40 ± 4.23 μg/l in healthy donors; For Ags treatment: 18.60 ± 3.15 μg/l in IT-allergic patients vs 7.45 ± 0.74 μg/l in healthy donors). Note that in healthy donors, neither Ags nor α-IgE led to significant differences in ECP release compared with unstimulated cells, although α-IgE treatment did suggest a possible increase.

Regarding IL-5 and GM-CSF, we found that both cytokines induced a significant ~2-3 fold increase in Ags/α-IgE-induced ECP production for both non-IT and IT-treated allergic patients. However, ECP release was significantly lower (p<0.001) in patients who received IT treatment compared to patients who did not ( Figures 3A, B ).

In healthy donors GM-CSF led to a significant increase in ECP release after α-IgE stimulation (75.87 ± 5.11 μg/l in cells treated with GM-CSF + α-IgE vs 17.40 ± 4.23 μg/l in cells treated only with α-IgE, p < 0.005) ( Figure 3B ). For IL-5 there was also an increase but this was not significant (44.82 ± 7.47 μg/l in cells treated with IL-5 + α-IgE vs 17.40 ± 4.23 μg/l in cells treated only with α-IgE, p = 0.151) ( Figure 3A ). ECP release was significantly lower in healthy donors after stimulation than that in IT-treated allergic patients (p < 0.001) ( Figure 3 ).

The inflammatory process of the airways, typical of asthma pathophysiology, induces tissue remodeling resulting in abnormal lung function (45). In asthmatic patients induced-sputum, ECP levels are inversely correlated with lung function (46), and increase after Ags nasal challenge (47). ECP levels are used as a clinical marker of the disease, assuming that they reflect eosinophilic inflammation in the airways (2). However, the numbers of ECP⁺ cells are not always correlated with eosinophil counts (6), suggesting the presence of other sources of ECP. Thus, we conducted our next set of experiments to study the relationship between ECP and lung function. We selected a group of allergic asthmatic patients, and collected induced-sputum to measure ECP levels in relation to the lung function (measured as FEV1). Additionally, we analyzed in vitro Ags-induced ECP production by peripheral blood neutrophils isolated from the same donors. In agreement with previous data (46), we observed a significant inverse correlation between induced-sputum ECP levels and lung function ( Figure 4A ). Strikingly, ECP levels released by blood neutrophils versus the corresponding FEV1, also showed a significant inverse correlation ( Figure 4B ). These results indicate that in vitro peripheral neutrophil ECP is a potential marker of airway inflammation/asthma severity.

Sputum contains a mixed-cell population consisting of squamous epithelial cells and a small fraction of leukocytes (neutrophils, eosinophils, macrophages/monocytes and lymphocytes). ECP has been immunohistochemically detected not only in eosinophil granulocytes, but also in neutrophils (6). We could thus assume that neutrophils in the airways will behave similarly to neutrophils in the blood, releasing the same mediators in response to the same stimuli. However, neutrophils from different tissues (airways vs blood) can behave differently (48). Therefore, we tested whether neutrophils from induced-sputum contribute to IgE-dependent ECP production, similar to peripheral blood. To this end, we cultured leukocyte-enriched cell populations from induced sputum of asthmatic patients in the presence of the Ags to which they were sensitized. We analyzed ECP expression by confocal microscopy. A residual green fluorescence was detected in resting sputum neutrophils ( Figure 4C , upper panel) (identified as PE-MPO and DAPI-stained multilobulated nuclei double-positive cells), which increased after Ags challenge ( Figure 4C , lower panel). A dotted ECP signal was observed colocalizing with MPO, suggesting that ECP may be stored in azurophilic granules.