Pancoll Human was purchased from PAN-Biotech GmbH (Aidenbach, Germany). Fetal calf serum (FCS) was obtained from GE Healthcare Life Science (Freiburg, Germany). GM-CSF was supplied by Novartis Pharma (Nuremberg, Germany). Human recombinant C5a was purchased from Hycult Biotech (Uden, The Netherlands). Phorbol 12-myristate 13-acetate (PMA), and diphenyleneiodonium chloride (DPI) were purchased from Calbiochem (distributed by Sigma-Aldrich (Buchs, Switzerland)). Dimethylsulfoxide (DMSO), p-nitrophenyl-2-acetamido-2-deoxy-α-D-glucopyranoside, N-formylmethionyl-leucyl-phenylalanine (fMLF), phosphate-buffered saline (PBS), bovine serum albumin (BSA), saponin, potassium bicarbonate (KHCO₃), Triton X-100, and dihydrorhodamine 123 (DHR123) were from Sigma-Aldrich (Buchs, Switzerland). Prolong Gold mounting media, Hoechst 33342, the Quant-iT PicoGreen dsDNA assay kit, propidium iodide (PI), ethylenediaminetetraacetic acid (EDTA, pH 8.0), RPMI-1640/GlutaMAX medium were obtained from Thermo Fisher Scientific (distributed by LuBioScience GmbH, Lucerne, Switzerland). X-VIVO 15 medium without phenol red and antibiotics was purchased from Lonza (Walkersville, MD, USA). Proteinase K was from Roche Diagnostics (Rotkreuz, Switzerland), and deoxyribonuclease I (DNase I) from Worthington Biochemical Corporation (Lakewood, NJ, USA). Mouse FITC-conjugated anti-human myeloperoxidase (MPO, clone MPO-7) was obtained from Agilent Dako (Santa Clara, USA). APC-conjugated anti-human CD63 (clone H5C6), FITC-conjugated anti-human CD66b (clone G10F5), and PE-conjugated anti-human CD35 (clone E11) were from BioLegend (London, UK). Polyvalent human IgG was a gift from CSL Behring (Bern, Switzerland). Normal goat serum was from Santa Cruz Biotechnology, Inc (Heidelberg, Germany), and ChromPure human IgG from Jackson ImmunoResearch Laboratories Inc (Philadelphia, PA, USA). German glass coverslips (#1 thickness, 12-mm diameter) were obtained from Hecht-Assistant (Altnau, Switzerland). Black glass-bottom 96-well plates were from Greiner Bio-One GmbH (Frickenhausen, Germany). Ammonium chloride (NH₄Cl) and Hemacolor Rapid staining kit were purchased from Merck Millipore (Darmstadt, Germany).

The sequences of TAT-syntaxin-4 and TAT-SNAP23 peptides were adapted from Uriarte et al. by fusing the t-SNARE coiled-coil homology domain of the respective SNARE proteins syntaxin-4 and SNAP23 to the TAT sequence (7) (TAT-syntaxin-4: YGRKKRRQRR RRHSEIQQLE RSIRELHDIF TFLATEVEMQ GEMINRIEKN IL; TAT-SNAP23: YGRKKRRQRR RHQITDESLE STRRILGLAI ESQDAGIKTI TMLDEQKEQL NRIEEGLDQI NKDMRETEKT LTEL). Both peptides were synthesized at a purity level of more than 95% (GL Biochem, Shanghai, China) and dissolved in 100% DMSO.

TAT-control peptide (FITC-ahx-GRKKRRQRRR PPQ) was kindly offered by Prof. Thomas Kaufmann (Institute of Pharmacology, University of Bern). The peptide was synthesized at a purity level greater than 95% (GL Biochem Shanghai, China) and dissolved in ddH₂O.

Human blood neutrophils and eosinophils were purified from peripheral human blood as previously described (11, 19). In brief, white blood cells were layered on Pancoll Human (density of 1.077 g/mL, PAN-Biotech, Aidenbach, Germany) and separated using density-gradient centrifugation (20 min, 800 x g, room temperature (RT)). The remaining erythrocytes in the granulocyte fraction were lysed with lysis buffer containing 155 mM NH₄Cl and 10 mM KHCO₃. For neutrophil experiments, only resulting cell populations with ≥ 95% purity assessed by an automated hematology analyzer (Sysmex Digitana, Horgen, Switzerland) were considered. For eosinophil experiments, an EasySep Human Eosinophil Isolation Kit (StemCell Technologies, Cologne, Germany) was used to isolate eosinophils from the granulocyte fraction by negative selection. Eosinophil purity (≥ 97%) was assessed using Hemacolor Rapid staining kit (Merck Millipore, Darmstadt, Germany) and light microscopic analysis.

Written, informed consent was obtained from all blood donors. The Ethics Committee of the Canton of Bern approved this study.

Isolated human neutrophils and eosinophils were pretreated with 5 µg/mL TAT-syntaxin-4, TAT-SNAP23, and/or TAT-control or 50 µM DPI in selected experiments at 37°C for 30 min. Subsequently, cells were primed with 25 ng/mL GM-CSF for 20 min and stimulated with 10 nM C5a or 1 µM fMLF at 37°C for the indicated time points. Unprimed cells were also stimulated with 25 nM PMA at 37°C for the indicated times.

The measurement of ROS production was adapted from a previous study (18). Briefly, freshly purified human neutrophils (0.25 x 10⁶ cells/250 µL) and eosinophils (0.5 x 10⁶ cells/250 µL) were resuspended in RPMI-1640 supplemented with 5% FCS and stimulated as described above. DHR123 was added to the cells with a final concentration of 1 µM at the end of priming with GM-CSF. Subsequently, the cells were activated and 100 µL of cell suspension was added to a black, glass-bottom 96-well plate in duplicates. The ROS activity of the samples was immediately measured every 5 min over the time period of 1h using a SpectraMax M2 plate reader (Molecular Devices) equipped with a 37°C incubator.

Staining of extracellular DNA traps was performed as previously described (19). Briefly, freshly purified human neutrophils were resuspended (0.25 x 10⁶ cells/100 µL) in X-VIVO 15 medium, seeded on 12 mm glass coverslips, and stimulated as described above. Cells were fixed with 4% paraformaldehyde for 10 min, washed with PBS, and permeabilized with 0.05% saponin in PBS for 3 min at RT. Subsequently, the cells were washed with 0.005% saponin in PBS and stained in presence of 0.01% saponin. Cells were blocked in blocking buffer containing 1% ChromPure human IgG for 30 min at RT. Direct immunofluorescence staining was performed by using monoclonal mouse FITC-conjugated anti-human MPO (1:100, clone MPO-7, Cat #F0714, Agilent Dako, Santa Clara, USA) for 1h at RT. Cells were washed in PBS, stained with 10 µg/mL PI for 10 min at RT, and mounted with Prolong Gold Antifade mounting medium.

Images were acquired by confocal laser scanning microscopy (LSM 800 with Airyscan, Carl Zeiss Micro Imaging, Jena, Germany) using a Plan-Apochromat 63x/1.40 Oil DIC objective and analyzed with Imaris software (Bitplane AG, Zurich, Switzerland). Ten representative pictures from each condition were subjected to analysis to measure mean fluorescence intensity (MFI) of MPO (green channel) intracellularly and within extracellular traps using the surface module of Imaris software.

Freshly purified human neutrophils and eosinophils were resuspended (1 x 10⁶ cells/500 µL) in X-VIVO 15 medium and stimulated as described above.

In eosinophils, in the last 2 min of activation time, 2.5 U/mL DNase I and 0.2 mg/mL proteinase K were added, as previously described (11). In neutrophils, 2.5 U/mL DNase I was added in the last 10 min of activation time, similar to a previous study (19).

Subsequently, 2.5 mM EDTA (pH 8.0) was added to the cells to stop reactions. Cells were pelleted at 1’400 rpm for neutrophils and 13’000 rpm for eosinophils for 5 min. 100 µL supernatant was transferred to black, glass-bottom 96-well plates in duplicates. The fluorescent activity of PicoGreen dye bound to dsDNA was excited at 480 nm and fluorescence emission intensity was measured at 520 nm using a spectrofluorometer (SpectraMax M2, Molecular Devices) according to the instructions described in the Quant-iT™ PicoGreen™ assay kit.

Freshly purified human neutrophils and eosinophils were resuspended (0.25 x 10⁶ cells/200 µL) in X-VIVO 15 medium and incubated with 5 µg/mL FITC-conjugated TAT-control peptide for the indicated time points at 37°C. Subsequently, cells were washed, seeded on 12 mm glass coverslips, and fixed with 4% paraformaldehyde for 10 min. Cells were washed with PBS, stained with Hoechst 33342 (1 µg/mL) for 10 min at RT, and mounted with Prolong Gold Antifade mounting medium. Cells were acquired by confocal laser scanning microscopy (LSM 800, Carl Zeiss Micro Imaging) using a Plan-Apochromat 63x/1.40 Oil DIC objective and analyzed with Imaris software (Bitplane AG).

Freshly purified human neutrophils and eosinophils were resuspended (0.5 x 10⁶ cells/200 µL) in X-VIVO 15 medium and incubated with 5 µg/mL FITC-conjugated TAT-control peptide at 37°C for the indicated time points. Subsequently, cells were washed and acquired by flow cytometry (FACSVerse, BD Biosciences). The MFI of FITC was analyzed using FlowJo software (Tree Star).

Cell viability was measured as previously described (17, 22). Briefly, freshly purified human neutrophils and eosinophils were resuspended (0.1 x 10⁶ cells/100 µL) in RPMI-1640 supplemented with 2% FCS and stimulated as described above. Cell death was assessed by the uptake of 10 µg/mL PI. Cells were acquired by flow cytometry (FACSVerse, BD Biosciences) and analyzed using FlowJo software (Tree Star).

GraphPad Prism 8 software (GraphPad Software Inc., La Jolla, CA, USA) was used for the analysis of the data. Data are presented as mean values ± SEM. To compare groups, one-way ANOVA with Tukey’s multiple comparisons test was used. Two-way ANOVA with Dunnett’s multiple comparisons test was applied to compare groups at different time points. p values ≤ 0.05 were considered statistically significant.

Syntaxin-4 and SNAP23 were previously reported to be involved in the exocytosis of tumor necrosis factor (TNF)-primed and fMLF-stimulated human neutrophils (7). In order to investigate the contribution of syntaxin-4 and SNAP23 on the kinetics of GM-CSF/C5a-induced neutrophil degranulation, we assessed the surface levels of surrogate markers for azurophilic granules (CD63), specific granules (CD66b), and SVs (CD35) by flow cytometry in a time-dependent manner ( Figures 1A, C, E ). Freshly isolated blood neutrophils were pretreated with or without SNARE peptides (5 µg/mL) for 30 min, and degranulation and ROS activity were measured in the presence and absence of GM-CSF/C5a, or PMA alone. Our findings revealed that CD63 surface expression reached its peak within 2 min of C5a activation followed by a gradual decline ( Figure 1A ), possibly due to its tightly regulated trafficking between cellular locations (25). In contrast, CD66b and CD35 surface levels remained stable after a strong upregulation in the first 2 min of stimulation ( Figures 1C, E ). The TAT-fusion peptides TAT-syntaxin-4 and TAT-SNAP23, both separately and in combination, resulted in reduced CD63 surface levels, particularly within the first 15 min of stimulation ( Figure 1A ). We confirmed the strong inhibition observed in the surface expression of CD63 following competitive inhibition of syntaxin-4 and SNAP23 by measuring the release of NAG, a protein found in the azurophilic granules (18, 19) ( Figure 1B ). After 60 min of stimulation, TAT-syntaxin-4 and TAT-SNAP23 considerably decreased the release of NAG, both individually and in combination ( Figure 1B ). Furthermore, we detected decreased surface expression of CD66b upon treatment with TAT-syntaxin-4 and TAT-SNAP23, with statistical significance achieved over a time period of 60 min only when both peptides were employed together ( Figure 1C ). A similar trend of inhibition was observed when assessing the release of MMP-9, a key component of the gelatinase granules (26), but to a lower extent ( Figure 1D ). In contrast, CD35 surface levels remained unaltered ( Figure 1E ). Taken together, these results indicate that syntaxin-4 and SNAP23 contribute to the release of azurophilic and specific granules in GM-CSF/C5a-stimulated human neutrophils, while they may also have some involvement in gelatinase granules. However, their impact on the release of SVs appears to be negligible.

A previous study demonstrated that a significant proportion of the membrane-bound nicotinamide adenine dinucleotide phosphate (NADPH) oxidase subunit gp91^phox is associated with granule membranes of resting human neutrophils (27). Furthermore, the process of granule exocytosis leads to the translocation of gp91^phox to the plasma membrane, facilitating the assembly of the NADPH oxidase, and subsequent production of ROS (28, 29). Given the involvement of syntaxin-4 and SNAP23 in neutrophil degranulation, we investigated their impact on ROS production in GM-CSF/C5a-activated human neutrophils in a time-dependent manner ( Figure 1F ). Human neutrophils demonstrated increased production of ROS over time, eventually reaching a plateau after 20 min of treatment ( Figure 1F ). We observed a significant reduction of ROS levels in human neutrophils pretreated with TAT-syntaxin-4 and TAT-SNAP23, both individually and in combination ( Figure 1F ).

To confirm that human neutrophils efficiently internalize the TAT-fusion peptides, we investigated the uptake of the TAT-control peptide by confocal microscopy ( Figure S1A ). Neutrophils isolated from human peripheral blood were treated with FITC-conjugated TAT-control peptide for 10 to 30 min, followed by staining with Hoechst 33342 to visualize the nucleus. Using confocal microscopy, we observed intracellular localization of the TAT-control peptide (green) as early as 10 min after treatment of neutrophils ( Figure S1A ). Additionally, we quantified the uptake of the TAT-control peptide in human neutrophils using flow cytometry ( Figure S1B ). We observed strong increase in FITC levels, confirming the efficient internalization of TAT-fusion peptides already after 10 min. Furthermore, the FITC levels of the TAT-control peptide were slightly increased after 30 min of treatment compared to 10 min, implying that the TAT-fusion peptides were not degraded in this time period. Collectively, our findings demonstrate the rapid and stable uptake of TAT-fusion peptides by human neutrophils.

While neutrophils release toxic granule content by degranulation, they also employ extracellular traps as an active defense mechanism (8). However, unlike the release of granule proteins, the mechanism by which dsDNA is released remains largely unidentified. To evaluate the role of syntaxin-4 and SNAP23 in the release of dsDNA, we assessed NET formation using confocal microscopy and stained DNA with PI and the azurophilic granule protein MPO with anti-MPO antibody ( Figure 2A ). We observed that the capability of human neutrophils to form extracellular traps was not impaired upon competitive inhibition of syntaxin-4 and/or SNAP23 ( Figure 2A ). In accordance with degranulation assessed by upregulation of the surrogate marker CD63 ( Figure 1A ), TAT-syntaxin-4 and TAT-SNAP23 significantly inhibited the release of MPO in GM-CSF/C5a stimulated human neutrophils ( Figure 2B ). Interestingly, the amount of MPO found embedded within the dsDNA scaffold remained unaffected by TAT-syntaxin-4 and/or TAT-SNAP23 pretreatment ( Figure 2C ). Additionally, we assessed the release of dsDNA in the extracellular space from human neutrophils with the same time-dependent stimulation as for the degranulation kinetics ( Figure 2D ). We observed a gradual release of dsDNA over time, with levels rising continuously until reaching their peak after 45 to 60 min ( Figure 2D ). In agreement with the confocal microscopy data, neither of the TAT-fusion peptides demonstrated a significant overall impact on the release of dsDNA. It is worth to note that TAT-syntaxin-4, but not TAT-SNAP23, appears to partially delay dsDNA release from GM-CSF/C5a-stimulated neutrophils in the early time points (5 to 15 min) of stimulation, although the overall dsDNA levels after 60 min are the same for the different conditions ( Figure 2D ). These findings suggest a delayed response rather than complete prevention of dsDNA release in response to TAT-syntaxin-4. Moreover, we confirmed that neither the formation of extracellular traps nor the treatment with TAT-syntaxin-4 and/or TAT-SNAP23 led to differences in the viability of human neutrophils ( Figure S2A ). In summary, we demonstrate that while syntaxin-4 and SNAP23 affect the degranulation of azurophilic and specific granules, they do not significantly contribute to the formation of NETs from human neutrophils.

To further investigate the involvement of syntaxin-4 and SNAP23 in degranulation, ROS production, and dsDNA release, we adopted the same approach with additional agonists such as GM-CSF/fMLF ( Figure 3 ) and PMA ( Figure S3 ), both of which were reported to trigger NET formation (18, 19, 30). Based on our previous degranulation findings ( Figures 1A-E ), we investigated the surface levels of surrogate markers for azurophilic granules (CD63), specific granules (CD66b), and SVs (CD35) by flow cytometry in neutrophils stimulated with GM-CSF/fMLF or PMA for 5 min. Similar to GM-CSF/C5a, GM-CSF/fMLF-stimulated neutrophils demonstrated a strong increase in CD63 ( Figure 3A ) and CD66b ( Figure 3C ) surface expression after 5 min, which was partially inhibited following TAT-syntaxin-4 and/or TAT-SNAP23 pretreatment. In contrast, CD35 surface upregulation was not impaired upon competitive inhibition of syntaxin-4 and SNAP23 ( Figure 3E ). Furthermore, the release of NAG contained in azurophilic granules was also significantly decreased after 60 min of stimulation ( Figure 3B ), which is consistent with our flow cytometry data ( Figure 3A ), while the release of MMP-9 from gelatinase granules remained unaltered despite an apparent trend ( Figure 3D ). In addition, we noticed a similar pattern of ROS production in neutrophils stimulated with GM-CSF/fMLF ( Figure 3F ) compared to GM-CSF/C5a ( Figure 1F ). Similar observations were made regarding the release of dsDNA in the supernatant, which showed a slight delay after 15 min of stimulation in TAT-syntaxin-4-pretreated neutrophils that ultimately resulted in the same amount of released dsDNA after 60 min ( Figure 3G ). Interestingly, neutrophils stimulated with PMA demonstrated similar trends for degranulation ( Figures S3A-E ) and dsDNA release ( Figure S3G ), although to a lesser extent. Moreover, ROS production was not inhibited following TAT-syntaxin-4 and/or TAT-SNAP23 pretreatment, which contrasts with data obtained from neutrophils stimulated with GM-CSF/C5a ( Figure 1F ) or GM-CSF/fMLF ( Figure 3F ). Taken together, we demonstrate that syntaxin-4 and SNAP23 play a significant role in degranulation of azurophilic granules and secondary granules, as well as in ROS production, while they are not involved in the release of gelatinase granule, SVs, and dsDNA in neutrophils stimulated with GM-CSF/fMLF and GM-CSF/C5a.

SNARE proteins, which are widely recognized for their conserved function across all eukaryotic cells (31), are also present and expressed in eosinophils (32). Human eosinophils are also capable of degranulation and extracellular trap formation similar to human neutrophils (33). Therefore, we sought to explore the role of syntaxin-4 and SNAP23 in eosinophil functions ( Figure 4 ). To assess eosinophil degranulation, we evaluated the time-dependent release of crystalloid granules using CD63 as a surrogate marker ( Figure 4A ), employing a similar approach as in human neutrophils ( Figure 1A ). We observed maximum release of CD63⁺ crystalloid granules within 5 min of C5a stimulation. In contrast to neutrophils, pretreatment with either TAT-syntaxin-4 and/or TAT-SNAP23 did not result in any inhibition of GM-CSF/C5a-induced degranulation in human eosinophils.

To demonstrate the capability of human eosinophils to efficiently take up the TAT-fusion peptides, we investigated the uptake of the TAT-control peptide using confocal microscopy ( Figure S1C ). Eosinophils were isolated from human peripheral blood, treated with the FITC-conjugated TAT-control peptide for 10 to 30 min, and stained with Hoechst 33342 for visualization of the nucleus. As in neutrophils ( Figure S1A ), we observed an intracellular localization of the TAT-control peptide (green) already 10 min after exposure to eosinophils ( Figure S1C ). Additionally, we assessed the intracellular content of TAT-control peptide in human eosinophils using flow cytometry ( Figure S1D ). Like in neutrophils, we observed strong uptake of the TAT-control peptide within 10 min and a further increase after 30 min, indicating that the TAT-fusion peptides are efficiently taken up by eosinophils and not degraded over time ( Figure S1D ). Collectively, these data provide evidence that TAT-fusion peptides are taken up both rapidly and stably by eosinophils confirming previous findings (34).

The generation of ROS plays a role in multiple processes of eosinophils, including the formation of eosinophil extracellular traps (EETs) (9), and other host defense mechanisms (35). Accordingly, we explored the role of syntaxin-4 and SNAP23 in the production of ROS in human eosinophils by assessing the kinetics of ROS production following GM-CSF/C5a and PMA stimulation in a time-dependent manner ( Figure 4B ). Over time, we observed a progressive rise in ROS production in human eosinophils. Consistent with our findings in eosinophil degranulation ( Figure 4A ), neither of the SNARE peptides exhibited relevant effects on the levels of ROS ( Figure 4B ). While TAT-SNAP23 alone and in combination with TAT-syntaxin-4 resulted in enhanced ROS production, TAT-syntaxin-4 individually lead to reduced levels of ROS, although the differences were not of statistical significance ( Figure 4B ).

To investigate whether syntaxin-4 and/or SNAP23 participate in the release of dsDNA within EET formation, we assessed the amount of extracellular DNA in the supernatant of human eosinophils following GM-CSF/C5a stimulation in a time-dependent manner ( Figure 4C ). Similar to human neutrophils ( Figure 2D ), we detected a gradual increase in the levels of released dsDNA from human eosinophils over time ( Figure 4C ). Moreover, the release of dsDNA was not significantly impaired by TAT-syntaxin-4 and/or TAT-SNAP23 treatment. Additionally, we demonstrated that the viability of human eosinophils remained unaffected by EET formation or treatment with TAT-syntaxin-4 and/or TAT-SNAP23 ( Figure S2B ). Collectively, our findings suggest that syntaxin-4 and SNAP23 do not contribute to degranulation, respiratory burst, and EET formation of human eosinophils stimulated with GM-CSF/C5a.