Influences of chloride and hypochlorite on neutrophil extracellular trap formation

Abstract

Background: The release by neutrophils of DNA-based extracellular traps (NETs) is a recently recognized innate immune phenomenon that contributes significantly to control of bacterial pathogens at tissue foci of infection. NETs have also been implicated in the pathogenesis of non-infectious diseases such as small vessel vasculitis, lupus and cystic fibrosis lung disease. Reactive oxygen species (ROS) are important mediators of NET generation (NETosis). Neutrophils with reduced ROS production, such as those from patients with chronic granulomatous disease or myeloperoxidase (MPO) deficiency, produce fewer NETs in response to inflammatory stimuli. To better understand the roles of various ROS in NETosis, we explore the role of MPO, its substrates chloride ion (Cl(-)) and hydrogen peroxide (H(2)O(2)), and its product hypochlorite (HOCl) in NETosis.

Findings: In human peripheral blood neutrophils, pharmacologic inhibition of MPO decreased NETosis. Absence of extracellular Cl(-), a substrate for MPO, also reduced NETosis. While exogenous addition of H(2)O(2) and HOCl stimulated NETosis, only exogenous HOCl could rescue NETosis in the setting of MPO inhibition. Neither pharmacological inhibition nor genetic deletion of MPO in murine neutrophils blocked NETosis, in contrast to findings in human neutrophils.

Conclusions: Our results pinpoint HOCl as the key ROS involved in human NETosis. This finding has implications for understanding innate immune function in diseases in which Cl(-) homeostasis is disturbed, such as cystic fibrosis. Our results also reveal an example of significant species-specific differences in NET phenotypes, and the need for caution in extrapolation to humans from studies of murine NETosis.

Figure 1. Extracellular chloride (Cl⁻) contributes to bacterial killing by human neutrophils.

(A) Human peripheral neutrophils were incubated in minimal media with or without Cl⁻ for 3 h (2×10⁶ cells/well). Cells were then assayed for viability using calcein (green) which identifies live cells, and ethidium homodimer (red) which binds DNA but cannot cross intact cell membranes, thus staining dead cells. Neutrophils incubated for 3 h in minimal media with and without Cl⁻ are equally viable. (n = 3 wells per condition) (B) Human peripheral neutrophils were incubated in minimal media with or without Cl⁻. Bacteria were incubated in same media, with and without neutrophils (MOI  = 0.1). Bacterial killing by neutrophils was reduced in the absence of extracellular Cl⁻. Each condition performed in triplicate. (C) Oxidative burst was assayed using intracellular DCF dye that fluoresces when oxidized. The production of oxidants was slower in the absence of Cl⁻, but similar peak production of oxidants was achieved by 100 min in the presence or absence of Cl⁻. Each condition was done in triplicate in each experiment (D) Oxidant production in activated neutrophils occurs in a series of reactions. Enzymes known to be important in these reactions, such as NADPH oxidase and MPO are also essential for NETosis.

Figure 2. NET formation requires extracellular chloride (Cl⁻).

Human peripheral neutrophils (2×10⁶ cells/well) were incubated in minimal media with or without extracellular Cl⁻, and stimulated to form NETs with PMA for 3 h. NETosis was completely inhibited in the absence of extracellular Cl⁻, as seen by (A) immunofluorescence using DAPI to counterstain DNA (blue) and anti-MPO antibody (green) to visualize NETs, and (B) quantification of released extracellular DNA (each condition done in triplicate).

Figure 3. Human neutrophils form NETs in response to exogenous oxidants.

(A) Hypochlorite (HOCl) release by activated neutrophils incubated in minimal media with or without Cl⁻ was measured using a colorimetric assay for chlorination of extracellular taurine. Production of HOCl by activated neutrophils requires extracellular Cl⁻. Each condition was done in triplicate. (B,C) NETosis increased in dose-dependent manner in response to exogenous hydrogen peroxide and sodium HOCl, as visualized using immunofluorescence and microscopy (blue: DAPI, green: anti-MPO antibody) (B), and using quantification of released extracellular DNA (C) as assays of NETosis. Neutrophils were plated at 2×10⁶ cells/well.

Figure 4. Differing roles of myeloperoxidase (MPO) in human and murine NETosis.

(A) The pharmacological agent ABAH inhibited MPO in both human, and murine bone marrow-derived neutrophils, as evidenced by significantly decreased MPO-catalyzed HOCl release after activation with PMA. All conditions were done in triplicate. (B,C) Human peripheral neutrophils were stimulated with potent trigger for NETosis (live Pseudomonas aeruginosa bacteria, MOI  = 1) in the presence of ABAH or vehicle control (blue: DAPI, green: anti-H2A/H2B/DNA complex antibody). Inhibition of MPO with ABAH almost completely blocked human NET formation even in response to potent stimuli (B, upper panels). However, inhibition of MPO with ABAH in murine bone marrow-derived neutrophils (B, middles panels) or skin pouch derived neutrophils (B, lower panels) did not block NET formation significantly. Data were obtained using (B) microscopy and immunofluorescence and (C) quantification of released extracellular DNA. As P. aeruginosa is known to release extracellular DNA itself, we included a bacteria only control, showing that bacteria alone do not significantly contribute to the elevation in released extracellular DNA. All conditions were performed in triplicate. (D,E) Bone marrow derived-neutrophils from MPO knockout mice or wild type siblings were stimulated for indicated times with live P. aeruginosa bacteria (MOI  = 1). No differences in NET formation were observed with either direct visualization with (D) immunofluorescence and microscopy or (E) quantification of released extracellular DNA. All conditions done in triplicate, experiment repeated two independent times.

Figure 5. Key role of hypochlorite (HOCl) in triggering NETs.

(A,B) Human neutrophils (2×10⁶ cells/well) were pre-treated with ABAH, and then stimulated with either H₂O₂ or NaOCl. MPO inhibition blocked H₂O₂-stimulated but not NaOCl-stimulated NETosis, as seen with (A) microscopy and (B) quantification of extracellular DNA release, indicating that HOCl is the central oxidant species that triggers NET formation in human neutrophils. All conditions were done in triplicate. (C) Stimulated human neutrophils incubated in minimal media with or without Cl⁻ were treated with exogenous H₂O₂ or NaOCl at indicated concentrations. In the absence of Cl⁻, human NETosis was rescued by addition of exogenous NaOCl, but not by H₂O₂, again illustrating that HOCl represents the central oxidant species for NET formation. All conditions were done in triplicate.