Non-classical neutrophil extracellular traps induced by PAR2-signaling proteases

Abstract

Neutrophil extracellular traps (NETs) are associated with diseases linked to aberrant coagulation. The blood clotting cascade involves a series of proteases, some of which induce NET formation via a yet unknown mechanism. We hypothesized that this formation involves signaling via a factor Xa (FXa) activation of the protease-activated receptor 2 (PAR2). Our findings revealed that NETs can be triggered in vitro by enzymatically active proteases and PAR2 agonists. Intravital microscopy of the liver vasculature revealed that both FXa infusion and activation of endogenous FX promoted NET formation, effects that were prevented by the FXa inhibitor, apixaban. Unlike classical NETs, these protease-induced NETs lacked bactericidal activity and their proteomic signature indicates their role in inflammatory disorders, including autoimmune diseases and carcinogenesis. Our findings suggest a novel mechanism of NET formation under aseptic conditions, potentially contributing to a self-amplifying clotting and NET formation cycle. This mechanism may underlie the pathogenesis of disseminated intravascular coagulation and other aseptic conditions.

Fig. 1. NET formation is triggered by PAR2-activating proteases and small-molecule PAR2 agonists but not by proteases that cleave PAR2 at non-canonical sites.

A PAR2 sequence showing cleavage sites for human proteases. B Relative expression levels of PAR1 (F2R), PAR2 (F2RL1), PAR3 (F2RL2) and PAR4 (F2RL3) mRNA in human neutrophils. C The amount of extracellular DNA released by neutrophils 3 h post-incubation with trypsin, kallikrein 14 (KLK14), cathepsin G (catG) and neutrophil elastase (NE) at the indicated concentrations (μM) based on SytoxGreen staining (RFU = relative fluorescence units). D The amount of extracellular DNA released by neutrophils 3 h post-incubation with 0.05 μM trypsin, trypsin pre-incubated in a 1:1 ratio with aprotinin based on SytoxGreen staining. E The amount of extracellular DNA released by neutrophils 3 h post-stimulation with trypsin (0.05 μM) after pre-incubation with 100 μM FSLLRY-NH₂ for 10 min. F The amount of extracellular DNA released by neutrophils 3 h post-stimulation with 100 μM AC 264613 or SLIGRL-NH₂ (PAR2 agonists) or LRGILS-NH₂ (reversed amino acid sequence control peptide for SLIGRL-NH₂) based on SytoxGreen staining. G NET structures visualized by a confocal laser scanning microscopy. DNA is stained with Hoechst 33342 (blue), and NE is shown in red (scale bars = 50 μm). H Visualization of NETs formed by peritoneal neutrophils from wild-type and PAR2^–/–C57BL6/J mice after treatment with PAR2 agonists, visualized by a confocal laser scanning microscopy. I Peritoneal neutrophils from wild-type and PAR2^–/– C57BL6/J mice were incubated with trypsin and catG (0.25 μM). Statistical significance was determined by applying an unpaired t-test (C) or by one-way ANOVA (D–F, I) followed by Tukey’s multiple comparisons test. Data are means (± SEM) with a minimum of n = 3 (C–F) or n = 2 (I) independent experiments using neutrophils from different healthy donors (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns non-significant).

Fig. 2. NET formation is triggered by PAR2-activating FXa.

A PAR2 sequence showing cleavage sites for FXa. B The amount of extracellular DNA released by neutrophils 3 h post-incubation with FXa at the indicated concentrations (μM) based on SytoxGreen staining. C NET structures visualized by a confocal laser scanning microscopy. DNA is stained with Hoechst 33342 (blue), and NE is shown in red (scale bars = 50 μm). D The amount of extracellular DNA released by neutrophils 3 h post-incubation with FXa and FXa EGR native protein (active site irreversibly blocked by 1 μM of the chloromethylketone tripeptide inhibitor EGRck) based on SytoxGreen staining. E The amount of extracellular DNA released by neutrophils 3 h post-stimulation with FXa (1 µM) after pre-incubation with 100 μM FSLLRY-NH₂ for 10 min. Statistical significance was determined by applying an unpaired t-test (B) or by one-way ANOVA (D, E) followed by Tukey’s multiple comparisons test. Data are means (± SEM) with a minimum of n = 3 (B, D, E) independent experiments using neutrophils from different healthy donors (*P < 0.05, ***P < 0.001, ****P < 0.0001; ns non-significant).

Fig. 3. FXa induces NET formation in the liver vasculature and is dependent on PAR2 expression.

A Representative images of NETs acquired by spinning disk intravital microscopy (SD-IVM) in untreated (control) wild-type (WT) mice as well as wild-type and PAR2^–/– mice injected with murine FXa 4 h before imaging. Extracellular DNA (extDNA) was stained with SytoxGreen (bright green), histone H2A.X was stained with an AlexaFluor 555-conjugated antibody (red) and neutrophil elastase was stained with an AlexaFluor 647-conjugated antibody (blue). The images from each channel were overlaid (scale bars = 50 μm). B Representative 3D conformation of the NET structure after FXa treatment in WT mice. C–E Quantitative analysis of NETs in liver sinusoids: area (%) covered by (C) extDNA, (D) histone H2A.X and (E) neutrophil elastase (NE). C–E Statistical significance was evaluated by one-way ANOVA followed by Tukey’s multiple comparisons test. Data are means (± SEM) from n ≥ 3 independent experiments (****P < 0.0001).

Fig. 4. Induction of coagulation by RgpA in the liver vasculature is associated with NET formation.

A, E NET formation in the liver vasculature was estimated by SD-IVM in wild-type (WT) mice injected with RgpA 4 h before imaging. A Kyt-1, a gingipain-specific inhibitor was added immediately after activation of RgpA, before its administration to the mice, and (E) apixaban, which inhibits FXa, was administered i.p. 30 min before RgpA. DMSO was used as a control for apixaban. Extracellular DNA (extDNA) was stained with SytoxGreen (green bright), histone H2A.X with an AlexaFluor 555-conjugated antibody (red) and neutrophil elastase was stained with an AlexaFluor 647-conjugated antibody (blue). The images from each channel were overlaid (scale bars = 50 μm). B–D, F–H Quantitative analysis of NETs in the liver sinusoids: area (%) covered by (B, F) extDNA, (C, G) histone H2A.X, and (D, H) neutrophil elastase (NE). I The appearance of NETs triggered by LPS, FXa and RgpA as inducers of coagulation. B–D, F–H Statistical significance was evaluated by one–way ANOVA followed by Tukey’s multiple comparisons test. Data are means (± SEM) of n ≥ 3 separate experiments (****P < 0.0001).

Fig. 5. Biochemical and functional characterization of protease-derived NETs.

A Calcium mobilization in neutrophils exposed to AC 264613, trypsin, or FXa. B The amount of extracellular DNA estimated by SytoxGreen staining in neutrophils pre-treated for 5 min with the ERK1/2 inhibitor UO126 before stimulation with AC 264613, trypsin or FXa for 3 h. C Scheme of NETs samples preparation for mass spectroscopy analysis. D Formation of NETs released by neutrophils isolated from 3 independent donors. The amount of extracellular DNA released by neutrophils 3 h post-incubation with AC 264613 (100 μM) based on SytoxGreen staining (RFU = relative fluorescence units). E Principal component analysis (PCA) showing differences between the two analyzed groups of proteomes (control and AC 264613). F Venn diagram presenting all identified proteins in control NETs and after AC 264613 stimulation (FDR = 1%, n ≥ 2 unique peptides). Gene names for unique peptides identified for control and AC 264613 are tabulated. G List of GO Biological Processes (BP) Molecular Function (MF) KEGG Pathways (KEGG) and Reactome Pathways (Reactome) for unique 34 proteins quantified in AC 264613 induced NETs of at least one donor obtained using the Database for Annotation, Visualization and Integrated Discovery - DAVID database (p value = 0.05). H Volcano plot of differently abundant proteins after AC 264613 stimulation. Gene names for proteins with increased abundance are tabulated. I List of GO Biological Processes (BP) Molecular Function (MF) KEGG Pathways for 42 proteins with increased abundance found in AC 264613 induced NETs obtained using DAVID database (p value with Bonferroni posttest = 0.05). J Bactericidal activity of NETs induced by PAR2 activators. E. coli cultures were mixed with NETs generated by exposure to PMA (25 nM), AC 264613 (100 μM) and FXa (1 μM) or serum-free DMEM collected from neutrophils. After incubation for 2 h, the bacteria were plated and the number of colonies (CFUs) was determined. K The presence of LL-37 within NETs generated by PMA, AC 264613, FXa or trypsin was visualized by immunoblot analysis 3 h poststimulation. A representative immunoblot is shown. L, M Enzymatic activity of human neutrophil elastase (L) and cathepsin G (M) in NETs generated by PMA, AC 264613, FXa or trypsin. A, D, J, L Statistical significance was evaluated by applying an unpaired t-test, or (B) by one- way ANOVA followed by Tukey’s multiple comparisons test. A–M Data are means ( ± SEM) of n ≥ 3 independent experiments (*P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; ns non-significant).