RPMI-1640 medium, l-glutamine, penicillin, and streptomycin were purchased from Life Technologies (Gent, Belgium) and fetal bovine serum was purchased from Lonza (Verviers, Belgium). fMLF, phorbol 12-myristoyl 13-acetate (PMA), and diphenyleneiodonium (DPI) were obtained from Sigma-Aldrich (Bornem, Belgium). All other chemicals were of analytical grade and obtained from Merck (Darmstadt, Germany). The physiological salt solution (PSS) used had the following composition: NaCl 115 mM, KCl 5 mM, KH₂PO₄ 1 mM, glucose 10 mM, MgSO₄ 1 mM, CaCl₂ 1.25 mM, HEPES-Na 25 mM, pH 7.4. The following antibodies were used in this study: anti-calgranulin clone CF145 (Santa Cruz Biotechnology, Dallas, TX, USA); anti-S100A8/A9 antibody clone Mac387 (DAKO, Glostrup, Denmark), anti-calgranulin B (S100A9) clone B-5 (Santa Cruz Biotechnology, Heidelberg, Germany); anti-MRP8 (S100A8) clone EPR3554 (Abcam, Cambridge, UK); antimyeloperoxidase (anti-MPO) antibody clone 2C7 (Thermo Fischer Scientific, Erembodegem, Belgium); anti-neutrophil elastase antibody (reference 411001, Calbiochem, Overijse, Belgium), and anti-TLR4 neutralizing antibody (reference AF1478, R&D Systems, Oxfordshire, UK).

The promyelocytic cell line HL-60 (ATCC #CCL-240) was grown in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum, l-glutamine (2 mM), streptomycin (100 μg/mL), and penicillin (100 U/mL). The cells were kept at 37°C, in a humidified atmosphere with 5% CO₂ and subcultivated twice a week. The differentiation of HL-60 cells toward neutrophil-like cells was induced by addition of DMSO (1.3% v/v) for 4.5 days (27).

For the purification of human neutrophils, peripheral blood of healthy volunteers was collected into EDTA-containing tubes (Vacutainer^®, BD Biosciences, Erembodegem, Belgium). Samples were collected following the good clinical and ethical practices, which were approved by the Ethics Review Panel of the University of Luxembourg according to the guidelines of the “Comité National d’Ethique de Recherche” from Luxembourg.

The neutrophils were isolated by dextran sedimentation prior to centrifugation on a Ficoll-Hypaque gradient. Remaining erythrocytes were then lysed for 10 min in red blood cell lysis buffer (155 mM NH₄Cl, 10 mM KHCO₃, 0.1 mM EDTA, pH 7.4) (28).

Neutrophils were washed and resuspended in PBS (NaCl 137 mM, KCl 2.7 mM, Na₂HPO₄ 10 mM, KH₂PO₄ 1.8 mM, pH 7.4). Purity of isolated neutrophils was assessed by flow cytometry (BD FACSCanto II flow cytometer, BD Biosciences) using following selection markers: CD66b-FITC, CD16-BV510, CD45-PECy7, CD14-APC, and CD49d-PE (BD Biosciences) as selection markers on 10.000 events in the gated population of homogenous (FSC-A versus SSC-A), single (SSC-A versus SSC-H), and living cells (negative for Sytox Blue staining; Invitrogen, Ghent, Belgium). Purified neutrophils are positive for CD66b, CD16, and CD45 while being negative for CD49d and CD14, allowing for the discrimination from eosinophils or monocytes, respectively. Only preparations containing more than 98% of neutrophils were used for further experiments.

Human S100A8 and S100A9 were recombinantly expressed in Escherichia coli and purified using affinity column. Briefly, E. coli BL21(DE3) bacteria were transformed with pGEX-5X-1 expression vector containing S100A8 or S100A9 cDNA. Recombinant GST-S100A9 and GST-S100A8 were induced and then purified using glutathione sepharose column. The immobilized GST-protein was eluted by Xa protease cleavage directly in the column. Then, recombinant S100A8 and S100A9, without GST-tag, were obtained and used for the validation of the anti-PhosphoS100A9 antibody.

S100A8 and S100A9 were purified from human granulocytes as described previously (29). Granulocytes were prepared from human buffy coats as described above, lysed by sonication and the cytosolic fraction was obtained by centrifugation (200,000 g, 1 h, 4°C). The protein concentration of the supernatant were adjusted to 3 mg/mL with buffer A (Tris 50 mM, EDTA 1 mM, EGTA 1 mM, dithiothreitol, 1 mM, pH 8.5) and a large portion of irrelevant proteins was removed by ammonium sulfate (70% w/v) precipitation for 2 h at 4°C. Precipitated samples were then subjected to 10,000 g centrifugation for 30 min at 4°C. The supernatant, enriched in S100A8/A9 protein complex was dialized overnight at 4°C against buffer A. The final purification step is done by FPLC using an anion exchange column (HiTrap Q XL, GE Healthcare Life Sciences, Diegem, Belgium). Proteins were eluted with a gradient 0–0.4 M NaCl at a flow rate of 1 mL/min. The elution of pure S100A8/A9 and S100A8/PhosphoA9 fractions is obtained at NaCl concentrations of 0.12–0.135 and 0.25–0.27 mM, respectively. The purity of proteins was greater than 95%, as verified by SDS-PAGE under reducing conditions and visualized by Coomassie stain and mass spectrometry analyses.

For the measurement of S100A8/A9 secretion, 5 × 10⁵ cells/mL were stimulated in PSS under different conditions and the supernatants were collected. S100A8/A9 concentration in the supernatants was determined by ELISA (BMA Biomedicals, Augst, Switzerland), after appropriate dilution of the samples, according to the manufacturer’s recommendations.

For the validation of the Phospho-specific S100A9 antibody, dHL-60 were stimulated by 100 nM fMLF for 10 min and the cells were lysed in a Triton lysis buffer (15 mM Tris, 150 mM NaCl, 1% Triton, SigmaFast protease inhibitor cocktail and phosphatase inhibitor cocktail from Sigma). 75 μg of this lysate were then treated for 15 min or 1 h with calf intestinal phosphatase (CiP, 75U) in order to dephosphorylate S100A9. Loading buffer (Tris 63 mM pH 6.8; 2% SDS; 10% glycerol; 0.1% β-mercaptoethanol) was added to different lysates as well as recombinant S100A8 and recombinant S100A9 and all samples were heated for 5 min at 97°C. For cell culture supernatants, samples were treated as follows: supernatants of control or stimulated cells were collected and concentrated by centrifugation in an Amicon^® Ultra 3K Centrifugal Filter Device. Loading buffer (Tris 63 mM pH 6.8; 2% SDS; 10% glycerol; 0.1% β-mercaptoethanol) was added to the concentrated supernatants and the samples were heated at 97°C for 5 min.

Samples were run on a Tris-Tricine gel (10% acrylamide) and the proteins were then electrotransferred to a PVDF membrane (Merck-Millipore, Overijse, Belgium). Immunodetection of the cell lysate was performed using the polyclonal rabbit antibody directed against Phospho-S100A9 (custom-made by Thermo-Fischer) or the monoclonal antibody Mac387 (DAKO) and horseradish peroxidase-conjugated donkey antirabbit or rabbit antimouse secondary antibodies. The immunodetection was visualized by the non-commercial enhanced chemiluminescence solution as described by Haan and Behrmann (30).

For immunodetection of purified S100A8/A9 and S100A8/PhosphoA9 fraction, the anti-PhosphoS100A9 antibody was used alone or preincubated 1 h at room temperature with 1 or 4 μg/mL of immunogenic peptide (CHKPGLGEG[pT]P) as well as the antibody B-5 (Santa Cruz Biotechnology) and EPR3554 (Abcam). The secondary antibodies antirabbit Alexa Fluor 680 (Life Technologies) and antimouse IRDye 800 (LI-COR Bioscience, Lincoln, NE, USA) were used for fluorescent Western blot detection with the Odyssey Infrared Imaging system (LI-COR Biosciences).

For the immunodetection of concentrated supernatants, the anti-PhosphoS100A9 antibody was used as well as the antibody Mac387 (DAKO). The secondary antibodies antirabbit IRDye 800 CW and antimouse IRDye 680 LT (LI-COR Bioscience) were used for fluorescent Western blot detection with the Odyssey Infrared Imaging system (LI-COR Biosciences).

Neutrophil extracellular traps were quantified using a fluorescent plate assay. dHL-60 cells or freshly purified neutrophils were cultured in a black 96-well plate. After indicated treatments, the cell impermeable DNA binding Dye Sytox blue (ThermoFischer, Ghent, Belgium) was added to the wells at a concentration of 0.5 μM to detect extracellular DNA. Fluorescence was measured on a Clariostar plate reader (CLARIOstar, BMG LABTECH, Ortenberg, Germany). The baseline fluorescence of the media with the dye was subtracted from all readings and the results were expressed according to the fluorescence of the non-stimulated control.

Neutrophil extracellular trap formation was visualized by staining of extracellular DNA, MPO, and human neutrophil elastase. Briefly, (dHL-60) or freshly purified neutrophils were washed and 4 × 10⁵ cells were placed on polylysin-coated coverslips and placed for 1 h at 37°C + 5% CO₂. Then, cells were stimulated by different compounds for 4 h and fixed for 15 min at 4°C in PBS containing 3.7% w/v paraformaldehyde and 2% w/v sucrose. The cells were then washed and permeabilized with PBS containing 0.1% saponin and 0.2% BSA and blocked with human IgGs for 15 min. Anti-MPO, antielastase, or isotype control antibodies were applied to the cells for 30 min in a humidity chamber at RT. The cells were washed three times and then incubated with a fluorophore-labeled secondary antibody (DyeLight-conjugated anti-mouse antibody, cyanin3-conjugated anti-rabbit antibody from Jackson ImmunoResearch, Sufolk, UK) together with DAPI 0.5 μM for 30 min in a humidity chamber at room temperature. Cover slips were washed four times and mounted in Moviol/DABCO and fluorescent images were acquired using the laser-scanning confocal microscope LSM510 (Carl Zeiss, Zaventem, Belgium) with a 40x objective and a multitrack configuration.

First-strand cDNA was prepared from 0.5 μg of total RNA, by adding the reaction buffer, 0.2 μg random hexamers, 20U RNAsin Ribonuclease inhibitor (Promega), 0.5 mM dNTP, and 40 U GoScript reverse transcriptase (Promega). Reverse transcription was performed as follows: 5 min at 25°C, 60 min at 42°C, and 10 min at 70°C. PCR primers for interleukin 1 alpha (IL-1α), interleukin 1 beta (IL-1β), IL-6, TNFα, chemokine (C-C motif) ligand 2, 3, and 4 (CCL2, CCL3, and CCL4), and chemokine (C-X-C motif) ligand 8 (CXCL8 alias IL-8) as well as the primers for the three housekeeping genes beta 2 microglobulin (B2M), Beta-actin (ACTβ), and beta-glucuronidase (GUSβ) were designed based on published sequences in GenBank (see Table 1 for primer sequences). qPCRs were performed using the SYBR^® Select master Mix (Thermo Fischer Scientific, Erembodegem, Belgium) in a QuantStudio 12K Flex real-time PCR machine (Thermo Fischer Scientific). The cycling protocol was as follows: 3 min at 50°C and 3 min at 95°C followed by 40 cycles of 3 s at 95°C and 30 s at 60°C. Relative quantification of the different mRNAs, expressed in arbitrary units (AU), were calculated using three reference genes (ACTβ, GUSβ, B2M) and the Vandesompele method, based on the geometric averaging of multiple internal controls (31).

Density of dHL-60 cells was adjusted to 0.5 × 10⁶ cells/mL per stimulatory condition for subsequent quantitative measurement of cytokine secretion. Supernatants were collected and immediately used for CBA (BD Biosciences, Erembodegem, Belgium). The assay was carried out according to the manufacturer’s instructions. Briefly, the multiplexed standard curve composed of the mixed cytokine standards was set up by serial dilutions. Selected capture beads were prepared and added to the standards and the supernatants. Following beads were used: TNFα (bead C4), IL-1α (bead D6), IL-1β (bead B4), IL-6 (bead A7), CCL2 (MCP1, bead D8), CCL3 (MIP1α, bead B9), CCL4 (MIP-1β, bead E4), CCL5 (RANTES, bead D4), and CXCL8 (IL-8, bead A9). After 1 h incubation, detection reagent was added to each sample and incubated for 2 h. Samples were rinsed with wash buffer, centrifuged, resuspended in wash buffer and analyzed by flow cytometry (BD FACSCanto II, BD Biosciences). Results were quantified using the Flow Cytometric Analysis Program Array software (Soft Flow, Minneapolis, MN, USA). To distinguish secretion induced by stimulation from spontaneous release by cell death, the concentrations found in the supernatants were normalized to lactate dehydrogenase (LDH) activity (Promega, Madison, WI, USA).

The secretion of CCL4 and CXCL8 in the supernatants was measured by ELISA bought from R&D systems. ELISAs were performed according to the manufacturer’s protocol and normalized to LDH.

Statistical analyses were performed using the GraphPad Prism software (Graph Pad Software, La Jolla, CA, USA). For time-course results with multiple stimulatory conditions, two-way ANOVA analysis was first performed to analyze the time and stimulation dependence of the results. Then a Tukey multiple comparison test was used to compare at each time point the effect of stimulation. For two-group comparison, normality and homogeneity of variances were ascertained, as determined by Kolmogorov–Smirnov and F-tests, respectively, and then Student’s t-test analyses were performed. Otherwise, Mann–Whitney tests were used for two-group comparisons. For all statistical tests, p < 0.05 was considered statistically significant. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Since it is well established that PhosphoS100A9 contributes to microtubule reorganization (18) and regulation of the NADPH oxidase (5) at the intracellular level, we assume that the posttranslational modifications of S100A9 are also related to their inherent multiple extracellular functionalities. To confirm our hypothesis, we determined in a first step the phosphorylation state of S100A8/A9 secreted by neutrophil-like HL-60 cells (dHL-60) and purified neutrophils. In order to know under which stimulatory conditions S100A8/A9 are released in the cell culture supernatants, we performed a secretion time-course on dHL-60 cells or purified neutrophils stimulated with 100 nM fMLF or 100 nM PMA. Our results show no induction of S100A8/A9 secretion upon stimulation of dHL60 cells by fMLF (Figure 1A), while a 40-fold increased secretion is observed after 6 h of PMA addition (Figure 1B). Similar results were obtained on primary cells although the induction of S100A8/A9 secretion from PMA-stimulated neutrophils is more extensive than in dHL-60 cells (Figures 1C,D).

In a second step, a polyclonal phospho-specific S100A9 antibody necessary to determine phosphorylation of secreted S100A9 was validated for specificity by Western Blot. Unphos-phorylated recombinant S100A8 and S100A9 were not recognized by the anti-P-S100A9 (Figure 2A). Stimulation of dHL-60 cells with fMLF, which induces phosphorylation of S100A9 by p38MAPK (5), led to the detection of a high signal at 14 kDa. Moreover, exposure of cell lysates to a phosphatase (CiP) for 15 min and 1 h, resulted in a time-dependent decrease of the anti-PhosphoS100A9 antibody signal, confirming the selectivity of this antibody for a phospho-protein at the same molecular weight as S100A9 (Figure 2A). Finally, we confirmed the specificity of anti-PhosphoS100A9 antibody by an antigen competition in western blot on purified S100A8/A9 and S100A8/PhosphoA9 proteins. A single band was detected in the S100A8/PhosphoA9 proteins lane (Figure 2B, a) which merged, in two-color fluorescence detection with the Odyssey system, with total S100A9 proteins (Figure 2B, d). Moreover, when anti-PhosphoS100A9 antibody was preincubated with an increasing concentration of immunogenic peptide, we observed a strong decrease and finally a total loss of the signal (Figure 2B, b and c). Taken together, these results demonstrate that the anti-PhosphoS100A9 antibody is specific for the phosphorylated form of S100A9 in western blot.

After having determined the conditions for S100A8/A9 secretion and controlled the selectivity of the anti-PhosphoS100A9 antibody, we investigated under which form S100A9 is found in the secreted S100A8/A9 complex. Thus, the supernatants of PMA-stimulated dHL-60 cells or purified neutrophils (Figure 2C) were concentrated and analyzed by Western Blot. Phospho-S100A9 and total S100A9 were detected concomitantly by the two-color fluorescence detection with the Odyssey system and our results clearly demonstrate that S100A9 is secreted as a phosphorylated form by dHL-60 cells and neutrophils.

The secretion mechanism of S100A8/A9 from neutrophils is still not totally clarified and two hypotheses have been put forward: an active process dependent on an intact microtubule network (22) or a release by NETosis (23). Since we confirmed that S100A8/A9 is secreted by PMA stimulation and it is known that PMA is a typical NET inducer (32), we hypothesized that S100A8/A9 release from neutrophils may be mediated by NETosis. We visualized NET structures after 4 h of agonist stimulation in dHL60 cells (Figure 3A) and purified neutrophils (Figure 3B) by fluorescent staining of DNA, MPO, and human neutrophil elastase. The immunofluorescence images confirm that the cells, activated by PMA, release NETs as shown by the colocalization of extracellular DNA with MPO and neutrophil elastase. Cells stimulated by fMLF were not able to form NETs.

To compare the kinetics of NET formation to the ones of S100A8/A9 release, extracellular NET-DNA was monitored over time with the cell impermeant DNA dye Sytox blue in both dHL-60 cells (Figure 3C) and neutrophils (Figure 3D). PMA induces NET release from both dHL-60 cells (Figure 3C) and neutrophils (Figure 3D) beginning at 3 h of stimulation, while fMLF failed to induce NETs at any stimulation time point.

These results underline the fact that S100A8/A9 are detected in the supernatants only when NETs are formed. By plotting S100A8/A9 secretion against NET formation and performing a Pearson correlation test, we obtained p-values of 0.008 and 0.0009 for dHL-60 (Figure 4A, left panel) and neutrophils (Figure 4A, right panel), respectively, supporting the assumption that S100A8/A9 secretion could be related to the NETosis process.

Since it is well established that ROS production is a hallmark of NET formation induced by PMA (33, 34), we inhibited the NADPH oxidase activation to strengthen the link between NETosis and S100A8/A9 secretion. For that, dHL-60 cells and neutrophils were treated for 30 min with 10 μM DPI before stimulation with PMA at 100 nM for 6 h.

As expected, in these conditions, NET formation in dHL-60 (Figure 4B, left panel) and neutrophils (Figure 4B, right panel) was blocked. More interestingly, DPI treatment also abolished PMA-induced S100A8/A9 secretion from both dHL-60 (Figure 4C, left panel) and neutrophils (Figure 4C, right panel).

Taken together these results strengthen our hypothesis that S100A8/A9 release could be dependent on NETosis process.

To confirm our initial hypothesis on the proinflammatory function of extracellular PhosphoS100A9, we decided to monitor the effects of unphosphorylated and phosphorylated S100A9 on cytokine/chemokine expression, as these mediators are essential for the development and regulation of immune processes and inflammation. dHL-60 cells were left untreated (CTL) or incubated in presence of 3 μg of either S100A8/A9 (S100) or S100A8/PhosphoA9 (PS100) for 0.5, 1, 2, or 4 h, then the mRNA levels of the proinflammatory cytokines IL-1α, IL-1β, IL-6, and TNFα and the chemokines CCL2, CCL3, CCL4, and CXCL8 were quantified (Figure 5). No cytokine or chemokine mRNA expression was induced by S100A8/A9, while dHL-60 cells responded to S100A8/PhosphoA9 stimulation by a significant induction of all tested cytokines and chemokines (ranging from 12-fold induction for CCL2 to a 400-fold induction for IL-1α). The induction reached a maximum after 1 h of stimulation and decreased to basal levels after 2 h of stimulation. Only IL-6, TNFα and CXCL8 displayed elevated levels at 2 h. In addition, cytokine expression varied in magnitude, with IL-1α, IL-1β, IL-6, CCL3, and CCL4 displaying a high differential expression (induction higher than 75-fold after S100A8/PhosphoA9 stimulation), whereas TNFα, CCL2, and CXCL8 showed a more moderate induction. To exclude an endotoxin contamination in the Phospho-S100 preparations, a Limulus amebocyte assay was performed. However, this assay indicated that one of the S100A8/PhosphoA9 batches used had an endotoxin contamination of 0.1 ng/mL, which alone did not induce any cytokine mRNA expression after 1 h of stimulation. Results obtained by a non-contaminated S100A8/PhosphoA9 batch showed similar results to the contaminated batch and a costimulation of dHL-60 cells with LPS and S100A8/A9 did not induce cytokine expression either (Figure S1 in Supplementary Material). Moreover, heat inactivation (30 min, 80°C) of S100A8/PhosphoA9 triggered a strong inhibition of RNA expression of cytokines (Figure S2 in Supplementary Material). Altogether, these results indicate that the induction of the various cytokine mRNA we observed with the S100A8/PhosphoA9 is not influenced by the small endotoxin contamination, revealing that only S100A8/PhosphoA9 is functionally active on neutrophil-like dHL-60 cells and that it induces a quite rapid and short response.

mRNA levels do not necessarily correlate to the relative amount of proteins released. We thus measured the secretion of proinflammatory cytokines upon S100A8/A9 or S100A8/PhosphoA9 stimulation to appreciate the functional effect of extracellular S100A9 phosphorylation. dHL-60 cells were treated with 3 μg/mL of S100A8/A9 or S100A8/PhosphoA9 for 30 min and up to 6 h; the presence of the different cytokines and chemokines in the supernatants was measured by multiplexed CBA. The cytokines/chemokines tested were the ones found to be induced on the mRNA levels by S100A8/PhosphoA9 stimulation, namely IL-1α, IL-1β, TNFα, IL-6, CCL2, CCL3, CCL4, and CXCL8 (Figure 6). Although IL-1α, IL-1β mRNA expression was induced in S100A8/PhosphoA9-stimulated cells, no secretion was detected in any stimulatory condition. TNFα, CCL2, CCL3, and CCL4 were secreted already after 2 h of S100A8/PhosphoA9 stimulation and IL-6 and CXCL8 were released at 4 and 6 h of S100A8/PhosphoA9 stimulation (Figure 6). CCL4 and CXCL8 secretion was highly induced while the release of TNFα, IL-6, CCL2, and CCL3 was more moderate. Stimulation with S100A8/A9 induced no cytokine secretion at any time point, confirming that the phosphorylated form S100A8/PhosphoA9 is the active form of the heterocomplex. As indicated previously, to exclude an effect of the LPS contamination on S100A8/PhosphoA9-induced cytokine secretion, dHL-60 cells were stimulated with 0.1 ng/mL LPS for 4 h, and CXCL8 and CCL4 secretion was measured by ELISA. No CXCL8 or CCL4 secretion was detected after LPS stimulation, while the secretion was highly induced by S100A8-phosphoA9 in the same conditions (Figure S3 in Supplementary Material). Moreover, S100A8/PhosphoA9-induced effects could not be inhibited by incubation with the LPS inhibitor Polymyxin B at concentrations which inhibited by 75% the secretion of CXCL8 and CCL4 induced by 100 ng/mL LPS (Figure S3 in Supplementary Material). Thus, the S100A8/PhosphoA9 complex has the functional role in the induction of cytokine release from dHL-60 cells.

Toll-like receptor 4 was described to be important for S100A8/A9 signaling in various cell types (2, 35). In this view, we investigated the involvement of TLR4 signaling in the S100A8/PhosphoA9-mediated response, by using a TLR4 neutralizing antibody. TLR4 was blocked with 1 μg/mL of neutralizing antibody, dHL-60 cells were stimulated with 3 μg/mL S100A8/PhosphoA9 for 4 h and secretion of TNFα, IL-6, CCL2, CCL3, CCL4, and CXCL8 was measured. Results clearly show a reduced secretion for TNFα, IL-6, CCL4, and CXCL8 when TLR4 was blocked, while CCL2 secretion was not affected at all (Figure 7). The incubation of dHL60 cells with a goat control IgG and stimulated with S100A8/PhosphoA9 showed no difference in cytokine secretion when compared to the S100A8/PhosphoA9 stimulated cells (data not shown). Taken together, these results suggest that TLR4-mediated signaling pathways are involved in S100A8/PhosphoA9-induced cytokine secretion in dHL60 cells, but also that another/other receptor(s)/partner(s) seem(s) to participate in S100A8/PhosphoA9-induced CCL2 secretion.