Patients, included in the study (n=9, Table 1 ), were diagnosed with RA based on the 2010 ACR/EULAR criteria and were over 18 years old. All patients were diagnosed with RA-ILD, as confirmed by clinical examination, CT scans, and pulmonary function tests. UIP was the predominant ILD pattern observed. Despite their lung conditions, none of the patients exhibited active arthritis or other extra-articular manifestations. Moreover, five of the nine patients received the antifibrotic agent nintedanib and were assessed before and after 16 weeks of treatment. Additionally, 9 age- and sex-matched healthy individuals served as controls. Neutrophils, sera, and plasma were isolated from patients with RA-ILD and healthy individuals for analysis. In vitro stimulation experiments were conducted on human pulmonary fibroblasts (HPFs). The study protocol was approved by the Ethics Review Board of the University Hospital of Ioannina (Ethics Review Board protocol number:12916,31/5/22), and all subjects provided written informed consent before participating in the study.

To isolate serum, venous blood was collected in appropriate blood collection tubes (BD Vacutainer^® SST II Advance Tubes, Becton, Dickinson and Company, Franklin Lakes, NJ, USA). To isolate plasma EDTA (PE), venous blood was collected in blood collection tubes with K3EDTA (BD Vacutainer^® EDTA tubes, Becton, Dickinson and Company, Franklin Lakes, NJ, USA). Serum was collected after a 10 min centrifugation at 2000x g, according to manufacturer’s instructions, whereas for PE collection a centrifugation at 500x g for 15 min was performed. All samples were stored at -80°C until further analysis (20, 26).

Peripheral blood neutrophils were isolated from heparinized blood by Histopaque double-gradient density centrifugation (11191 and 10771, Sigma-Aldrich, St Louis, MO, USA). Following the manufacturer’s instructions, whole blood was centrifuged at 700x g at 20-25°C for 30min. Isolated neutrophils were then washed once with 1x Dulbecco’s phosphate-buffered saline solution (1x PBS) (200x g, 10 min, at 20-25oC). Isolated neutrophil population purity exceeded ≥98%, as assessed by flow cytometry.

Primary HPFs (Cat#: C-12360, PromoCell, Heidelberg, Germany) were cultured at 37oC with 5% CO2 in low glucose Dulbecco’s Modified Eagle Medium (DMEM, PAN Biotech, Aidenbach, Germany), supplemented with 10% v/v Fetal Bovine Serum (FBS, Capricorn Scientific, Ebsdorfergrund, Germany), 100 U/mL Antibiotics-Antimycotic solution (Biosera, Cholet, France) and 5% v/v MEM Non-Essential Amino Acids Solution (Thermo Fisher Scientific, Waltham, MA, USA). Once HPFs reached 80-85% confluency, they were sub-cultivated by using Trypsin/EDTA solution (Capricorn Scientific, Ebsdorfergrund, Germany) for cell detachment. Cells from passages 4-8 were used for this study (23).

For ex vivo NET structures generation, a total of 2 × 106 neutrophils (RA-ILD or control) were cultured in a 6-well cell culture plate (SPL Life Sciences, Kyonggi-do, Republic of Korea) in RPMI, supplemented with 2% v/v heterologous healthy donor serum, for 3h at 37oC with 5% CO2. Following, the cell culture medium was removed, and neutrophils were washed once with pre-warmed RPMI. After vigorous agitation of the culture plate and centrifugation at 20× g for 5 min, NETs were collected in the supernatant phase. Control neutrophils were used as negative control (control NETs) (20).

Peripheral blood neutrophils were cultured in a 24-well cell culture plate (SPL Life Sciences, Kyonggi-do, Republic of Korea) on Poly-L-Lysine coverslips (Biocoat, NY, USA), to evaluate their capacity to release NETs and examine the protein profile of NETs. Following a 3-hour incubation (37oC, 5% CO2), neutrophils were fixed with 10% formaldehyde solution (Biognost, Zagreb, Croatia) for 30 min at 4oC. Nonspecific binding sites were blocked with 6% normal goat serum (Thermo Fisher Scientific, Waltham, MA, USA) in 1x PBS (blocking solution). The samples were stained with a primary antibody solution, consisting of an anti-human tissue factor (TF) monoclonal antibody (mAb) (1:200 dilution, BioMedica Diagnostics, Windsor, Canada), an anti-human IL-17A mAb (1:50 dilution,R&D System, Minneapolis, MN, USA) or an anti-human neutrophil elastase (NE) mAb (1:100 dilution, Abcam, Cambridge, UK) in blocking solution, for 1h at room temperature (RT). Following, incubation with a polyclonal anti-rabbit IgG AlexaFluor647 antibody (Invitrogen, Waltham, MA, USA) or a polyclonal anti-mouse IgG AlexaFluor488 antibody (Invitrogen, Walthan, MA, USA), diluted in blocking solution according to manufacturer’s instructions, was performed. Finally, cells were stained with DAPI solution (Sigma-Aldrich, St Louis, Missouri, USA) and mounted on microscope slides (Knittel Glass, Braunschweig, Germany) using a hardening mounting medium (Thermo Fisher Scientific, Waltham, MA, USA) (20, 21).

Sample visualization was performed on a Nikon ECLIPSE Ti2 Inverted Microscope (Nikon, Melville, NY, USA) with a 40× oil lens (1.30NA) and image acquisition was achieved, using NIS-Elements software (Nikon, Melville, NY, USA). Images were analyzed in Fiji software version 2.9.0 (27).

Concentration of CitH3 was measured in serum samples, in accordance with manufacturer’s instructions (Cayman Chemical, Ann Arbor, MI, USA).

To assess the levels of thrombin, the concentration of TAT was measured in: (a) PE from RA-ILD patients and healthy individuals and (b) ex vivo-isolated NET structures. The assay was performed based on the manufacturer’s instructions (Abcam, Cambridge, UK).

IL-17A ELISA was applied to measure IL-17A concentration in: (a) serum samples and (b) ex vivo-isolated NET structures. The ELISA kit was utilized in accordance with manufacturer’s protocol (R&D Systems, Minneapolis, MN, USA).

Human soluble terminal complement complexes (i.e., sTCC or sC5b-9) were quantified in PE from RA-ILD patients and healthy individuals, by applying a commercially available ELISA kit in line with manufacturer’s guidelines (Hycult Biotech, Uden, The Netherlands).

As formerly described, RNA isolation was performed using TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, USA) according to manufacturer’s instructions. cDNA was synthesized with an appropriate kit (Takara Bio, Shiga, Japan) and RT-qPCR was conducted using a commercially available SYBR green RT-qPCR master mix (Kapa Biosystems, Wilmington, MA, USA). The expression of TF, IL-17A and RAR-related orphan receptor C (RORc) was examined in neutrophils. To evaluate the activation of HPFs, the expression of smooth muscle actin alpha 2 (ACTA2) was studied. To normalize the expression of the abovementioned genes, glyceraldehydes-3-phosphate dehydrogenase (GAPDH) was utilized, following the housekeeping gene normalization method. RT-qPCR primers were designed using Beacon Designer version 4.0. Further details regarding the primers and the RT-qPCR conditions are shown in Supplementary Table S1 . Data analysis was performed by applying the 2-ΔΔCt method (19, 21, 28).

To evaluate the migratory/wound healing capacity of HPFs, cells were seeded in a 24-well cell culture plate (SPL Life Sciences, Kyonggi-do, Republic of Korea). When cells reached 90% confluency, stimulation/inhibition studies were performed. Wound healing was evaluated after 20hr incubation and assessed by May-Grünwald Giemsa stain. Assay was performed following manufacturer’s instructions and recommendations (21).

MGG stain was performed to visualize the migrating/wound healing capacity of HPFs. Firstly, cells were incubated with May-Grünwald stain for 5min at RT. After washing away, the excess of May-Grünwald stain with water, a 20-minute incubation (RT) with Giemsa stain (1:10 dilution) followed. Finally, Giemsa stain was removed and HPFs were washed with water (18). Stained cells were observed under an OLYMPUS IX73 Inverted Microscope (OLYMPUS Corporation, Tokyo, Japan) with a 4x air lens (0.10NA). Images were acquired using OLYMPUS cellSens Entry software version 1.14 (OLYMPUS Corporation, Tokyo, Japan) and final images were produced with Fiji software version 2.9.0 (22).

Comparisons between two independent groups were performed by using the non-parametric Mann-Whitney U test (two-tailed), whereas the statistical analysis between two paired groups was achieved by applying the non-parametric Wilcoxon matched-pairs signed rank test (two-tailed). Statistical comparisons between three paired groups were performed via the Friedman test. Spearman’s rank correlation coefficients test at 95% confidence intervals (CI) was utilized for bivariate correlation analysis. Data in all graphs are presented as Mean ± standard deviation (SD) and the cut-off value for statistical significance was set to P < 0.05. Statistical analysis of the experimental data was conducted with the use of GraphPad Prism version 8 (GraphPad Software, Inc., San Diego, CA, USA).

Inflammatory mediators were assessed in serum samples collected from patients diagnosed with RA-ILD and healthy controls ( Table 1 ). Using protein immunoassays, we observed elevated levels of circulating NETs, measured by CitH3 ELISA, in active RA-ILD patients compared to controls ( Figure 1A ). Additionally, RA-ILD sera exhibited increased TAT activity ( Figure 1B ) and IL-17A expression ( Figure 1C ). Importantly, these inflammatory proteins correlated positively with NET levels ( Figures 1D, E ). Moreover, RA-ILD patients showed significantly elevated plasma levels of sC5b-9, detected by TCC ELISA ( Figure 1F ), underscoring a pro-inflammatory microenvironment conducive to disease progression.

Further characterization of neutrophils from active RA-ILD patients revealed spontaneous release of NETs enriched with functional TF, as assessed by confocal microscopy ( Figure 2A ) and TAT assay in ex vivo-isolated NET structures ( Figure 2B ). Similarly, in vitro experiments demonstrated that RA-ILD serum induced TF mRNA expression in control neutrophils ( Figure 2C ).

Additionally, NETs from RA-ILD patients were found to be coated with IL-17A, confirmed by immunofluorescence ( Figure 3A ) and IL-17A NET ELISA ( Figure 3B ). Control neutrophils stimulated with RA-ILD serum showed intracellular overexpression of IL-17A and RORc, a transcription factor associated with IL-17 regulation ( Figures 3C, D ). These findings collectively highlight the increased formation of TF- and IL-17A- bearing NETs in active RA-ILD, implicating neutrophils in the induction of disease-associated inflammation.

To understand the impact of inflammatory NETs on tissue-resident cells, HPFs were incubated with NET structures released by active RA-ILD patients (RA-ILD NETs). RA-ILD NETs triggered the activation of HPFs, as evidenced by the up-regulation of ACTA2 ( Figure 4A ). In addition, after this stimulation, HPFs showed increased proliferation/migration rates ( Figure 4B ).

On the other hand, disassembly of RA-ILD NETs with DNase I abolished the fibrotic potential of HPFs ( Figures 4A, B ), suggesting a specific effect of NETs on the fibrotic process. In addition, protein components of RA-ILD NETs, namely TF and IL-17A, also exert a direct effect on the fibrotic activity of HPFs. Particularly, the pretreatment of cells with the FLLRN peptide, which blocks thrombin signaling, resulted in a significant reduction of HPFs fibrotic potential ( Figures 4A, B ). The neutralization of IL-17A on RA-ILD NETs with a monoclonal antibody also led to a similar result ( Figures 4A, B ). Consequently, these observations indicate RA-ILD NETs as potent mediators of tissue damage and suggest multiple targets for therapeutic interventions in RA-ILD.

Prompted by recent studies discussing the effects of novel antifibrotic agents on the progression of RA-ILD (10), we next performed a paired analysis in samples derived from five RA-ILD patients, with samples taken before and 16 weeks after starting treatment with nintedanib ( Table 1 ). Blood serum obtained from patients with RA-ILD under treatment with nintedanib (treated patients) showed a decrease in circulating CitH3 compared to the same patients before the initiation of the antifibrotic therapy, as verified by ELISA immunoassay ( Figure 5A ). In addition, treated RA-ILD patients were characterized by reduced levels of sC5b-9 in plasma, as evidenced by TCC ELISA assay ( Figure 5B ).

The effect of nintedanib was also assessed on ex vivo-isolated NET structures. RA-ILD-treated patients showed lowered NET formation compared to the same patients before the initiation of therapy, as evidenced by CitH3 ELISA ( Figure 5C ). In contrast, the presence of TF ( Figure 5D ) and IL-17A ( Figure 5E ) in ex vivo-isolated NETs was not found to be reduced, as indicated by ELISA assays. This observation was further verified by mRNA studies ( Supplementary Figure S1 ). Together, these observations support that nintedanib could have a regulatory effect on NET formation and sC5b-9 release in patients with RA-ILD.