This study protocol for using human blood samples was approved by the ethics committee of The Hospital for Sick Children, Toronto. All the procedures including healthy human volunteer recruitment for blood donation were performed in accordance with the ethics committee guidelines. All the volunteers participating in this study gave their signed consent prior to blood draw.

SYTOX^® Green Nucleic Acid Stain dye, DHR123, and pH_i indicator Seminaphtharhodafluor (SNARF) were obtained from Molecular Probes (Thermo Fisher Scientific, Waltham, MA, USA). All buffers, agonists, inhibitors, and other reagents were purchased from Sigma-Aldrich unless otherwise stated. The standard medium of these experiments was RPMI 1640 medium (Invitrogen) supplemented with 10 mM HEPES buffer. The amount of HCl and NaOH was predetermined to change the pH of RPMI media to 6.6, 6.8, 7.0, 7.2, 7.4, 7.6, or 7.8, and these isotonic RPMI media with different pHs were added to the neutrophil suspension for further experiments.

The primary human neutrophils were isolated by using a slight modification of the PolymorphPrep (Axis-Shield) protocol as previously reported (4–7). Briefly, the peripheral blood from the healthy male donors was collected in K2 EDTA blood collection tubes (Becton, Dickinson and Co.). Equal volumes of blood were layered over the PolymorphPrep (Axis-Shield) and spun for 35 min at 600 × g, 25°C to separate the neutrophil band. After washing the neutrophil band, red blood cells were lysed with a 0.2% (w/v) NaCl hypotonic solution for 30 s followed by an addition of an equal volume of 1.6% (w/v) NaCl solution with 20 mM HEPES buffer to provide the isotonic condition. Isolated neutrophils were resuspended in RPMI medium (Invitrogen) containing 10 mM HEPES (pH 7.2). Neutrophil counting and viability check were done by trypan blue and using a hemocytometer. Furthermore, the purity of neutrophils was determined by Cytospin preparation and imaging. Only neutrophil preparations with >95–98% live and pure were used in the experiments. For each experiment, multiple donors were used to get enough replicates; see details in specific figure legends.

Isolated neutrophils were treated with 10 μM SNARF^®-4F 5-(and-6)-Carboxylic Acid (Thermo Fisher Scientific) pH indicator dye and incubated at 37°C for 15 min. After incubation, the cells were washed and suspended in fresh RPMI medium at 1 × 10⁶/mL cells. A volume of 50 μL, containing 50,000 neutrophils was seeded into a 96-well clear bottom black plate (BD Biosciences). Equal volumes (50 μL) of the isotonic RPMI with predetermined pH (6.6, 6.8, 7.0, 7.2, 7.4, 7.6, and 7.8) were added into the wells to adjust the pH of the corresponding wells. Neutrophils buffered with different pHs were stimulated either with only media (−ve control) or PMA. The dual emission spectra of SNARF were measured after adding media (−ve control) and PMA and considered as the zero time point reading. Further pH changes were recorded for every 10 min up to 60 min with an Omega fluorescence microplate reader. The emission spectrum of SNARF undergoes a pH-dependent wavelength shift, and therefore, the ratios (580/640 nm) of the fluorescence intensities from the dye at two emission wavelengths were used for pH_i determinations. Carboxy SNARF-4F is typically used by exciting the dye at one wavelength (between 488 and 530 nm) while monitoring the fluorescence emission at two wavelengths, typically at 580 and 640 nm. The fluorescence response of the SNARF indicators has been calibrated with pH-controlled RPMI (6.6, 6.8, 7.0, 7.2, 7.4, 7.6, and 7.8) buffered in the presence and absence of the cells. Each condition was tested with a technical duplicate. Biological replicates (n-values; donors) of independent experiments were reported in figure legends.

DHR123 dye (Scientific Inc., MA, USA) was used to measure the intracellular ROS production. The neutrophils were preloaded with DHR123 (20 μM) for 10 min as per manufacturer’s instructions with brief modification as we reported earlier (5, 7, 14). After washing the extracellular DHR123 dye, cells were resuspended in fresh RPMI (10 mM HEPES) media and 50 μL of 50,000 cells, seeded into 96-well plates. Furthermore, DHR123-preloaded neutrophils media were adjusted for pHs (6.6, 6.8, 7.0, 7.2, 7.4, 7.6, and 7.8) by adding the equal volume (50 μL) of isotonic RPMI calibrated for respective pHs. These cells were activated with either only media (negative control), PMA, or LPS for another 30 min. The fluorescence was measured every 10 min by an Omega fluorescence microplate reader (900 data points per well) to assess the kinetics of the ROS generation in different conditions. Each condition was tested with a technical duplicate. Biological replicates (n-values; donors) of independent experiments were reported in figure legends. Confocal images were acquired as described earlier by counter staining of the nuclei with DAPI in different pH conditions.

Flow cytometry was used for validating the ROS production in neutrophils at different pH_e conditions. Cells were treated with 20 μM DHR123 for 10 min at 37°C. These preloaded cells were washed and resuspended in fresh media, and their pHs were adjusted to 6.6, 7.4, and 7.8. These cells with adjusted pH_es were activated with either media only (−ve control) or PMA for 30 min, and the ROS production was analyzed by counting 10⁴ events using the 488 filter and the FITC laser of the Gallios (Beckman Coulter Inc.) flow cytometer. Data were analyzed by using Flow Jo software.

A cell impermeable DNA binding Sytox Green dye (Life Technologies) was used for measuring the real-time kinetics of NET release under different conditions. A volume of 50 μL media containing 50,000 neutrophils mixed with 5 μM Sytox green were seeded into 96-well plates. Equal volumes of the isotonic RPMI media with double the strength of different pHs were added into the respective wells to make the final pH 6.6, 6.8, 7.0, 7.2, 7.4, 7.6, and 7.8 of the corresponding wells. These neutrophils under different pHs were activated with agonists, only media (−ve control), PMA, LPS, P. aeruginosa, and S. aureus. The changes in green fluorescence signal was measured every 30 min for up to 240 min using a fluorescence plate reader (504 nm excitation and 523 nm emission, POLARstar OMEGA, BMG Labtech). By using this POLARstar OMEGA plate reader, 900 data points per well were scanned at each time point to quantify the fluorescence intensity. Total DNA (100% DNA) present in these neutrophils was determined by the fluorescence values of the cells lysed with 0.5% (v/v) Triton-X-100. To calculate percentage of NETosis in each condition, the green fluorescence at time 0 min was subtracted from the fluorescence at each time point and was then divided by the fluorescence values of cell lysed with 0.5% (v/v) Triton X-100. All the experimental values were standardized to total DNA at each time point. Each condition was tested with a technical duplicate. Biological replicates (n-values; donors) of independent experiments were reported in figure legends.

A volume of 100 μL RPMI medium containing 100,000 neutrophils were seeded into 12-well chamber slides (BD Falcon). Chamber slides were used for obtaining the high-resolution images. The pH of the respective chamber was adjusted as explained earlier by adding an equal volume of the isotonic RPMI with different pHs (6.6, 7.4, and 7.8). The neutrophils were activated with either media only (−ve control), PMA, LPS, P. aeruginosa, or S. aureus for 120 min. The cells and NETs were fixed with paraformaldehyde [4% (w/v) for 10 min; 2% (w/v) for overnight] and immunostained with various NET markers. Mouse anti-MPO antibody (ab25989; Abcam) at 1:500 dilution was used for staining MPO (with secondary antibody conjugated with a green fluorescence Alexa fluor 488 dye; 1:2,000 dilution; Thermo Fisher Scientific), while rabbit anti-citrullinated histone 3 antibody (ab5103; Abcam, Lot # GR273046-3) at 1:500 dilution was used for detecting the presence of citrullinated histone H3 (CitH3, with secondary antibody conjugated with a far-red fluorescence dye Alexa fluor 647; 1:1,000 dilution; Thermo Fisher Scientific). DNA was stained with DAPI (1:1,000 dilution). After treating the secondary antibody, slides were washed and mounted by glass cover slips (Fisher Scientific) with anti-fade fluorescent mounting medium (Dako). The images were then taken using an Olympus IX81 inverted fluorescence microscope with a Hamamatsu C9100-13 back-thinned EM-CCD camera and Yokogawa CSU × 1 spinning disk confocal scan head with Spectral Aurora Borealis upgrade, four separate diode-pumped solid-state laser lines (Spectral Applied Research, 405, 491, 561, and 642 nm). The images were taken at 40×/0.95 magnification and processed by Volocity software (version 6.3, Cell Imaging Perkin-Elmer). Immunostained confocal images were quantified by counting normal neutrophils (multilobed nuclei), decondensed (nuclei lost multilobed shape), cloud-like puffed extracellular DNA, and extended extracellular DNA (NET-like structure) to determine the degree of NETosis.

For the immunoblot analysis, the tubes containing 1 × 10⁶ cells with adjusted pHs (6.6, 7.0, 7.4, and 7.8) in each experimental condition were activated either by negative control (only media) or by PMA for 120 min. After incubation, the tubes were placed on ice for 10 min and then centrifuged at 20,000 rcf at 4°C for 10 min. The supernatant was then discarded, and the cell pellets were lysed using the lysis buffer containing 1% (w/v) Triton X-100, 25 mM NaF, 50 mM Tris, 10 mM KCl, 10 μg/mL aprotinin, 2 mM PMSF, 1 mM levamisole, 1 mM NaVO₃, 0.5 μM EDTA, 25 μM leupeptin, 25 μM pepstatin, one protease inhibitor cocktail tablets per 5 mL (Roche), and one phosphatase inhibitor cocktail tablet per 10 mL (Roche). The samples were then vortexed for 10 s followed by three times of sonication using an aquasonic sonicator (VWR, model 50D at the highest power setting), at 8–10°C, 3 min each. A quarter volume of 5 × loading dye [125 mM Tris HCl at pH 6.8, 6% (w/v) SDS, 8% (v/v) β-mercaptoethanol, 18% (v/v) glycerol, 5 mM EDTA, 5 mM EGTA, 10 μg/mL leupeptin, 10 μg/mL pepstatin, 10 μg/mL aprotinin, 10 mM NaF, 5 mM NaVO3, and 1 mM levamisole] was added followed by 10 min of heating at 95°C with 350 rcm shaking. The samples were separated in a 5% (w/v) stacking and 10% (w/v) resolving gel at 100 V and transferred on a nitrocellulose membrane for 90 min at 400 mA. After transfer, the membranes were blocked with 5% (w/v) milk or BSA in 0.05% phosphate-buffered saline (PBS) with 0.1% Tween (PBST) for 1 h at room temperature. The membranes were incubated with the primary antibody at 4°C overnight followed by three washes with PBST for 30 min. The antibodies used were as follows: anti-Histone H4 (ab16483; Abcam) rabbit pAb at 1:1,000 and anti-GAPDH (FL-335; Santa Cruz) rabbit pAb at 1:2,500. The membranes were then incubated in the secondary antibody solution for 1 h and then washed three times with 0.1% PBST for 30 min. The secondary antibodies used were as follows: donkey anti-rabbit IgG-HRP (31458; Thermo Fisher) at 1:7,500. The densitometry analysis of the blots was done using the Image Studio software (LI-COR Biotechnology) and normalized to the GAPDH.

Pseudomonas aeruginosa and S. aureus were selected from single colonies in LB-agar plates and grown overnight in sterile LB broth. Overnight bacterial growth culture was diluted by a factor of 10 and subcultured for 3 h to eliminate the dead colonies or bacteria. Bacterial culture was harvested and washed three times in 5 mL of PBS (pH 7.4) by centrifugation at 5,000 × g for 5 min at 4°C. The bacterial concentration was determined by taking optical density (OD) at 600 nm. Furthermore, the multiplicity of infection was calculated by using the colony-forming unit (CFU) formula established in the laboratory by empirical methods: (CFU) × 10⁸ = (OD₆₀₀) × 30.88 − 99.61.

Statistical analysis was performed using GraphPad Prism statistical analysis software (Version 5.0a). Student’s t-test was used for comparing two groups, and for more than two groups, ANOVA with Bonferroni’s posttest or Dunnett’s test was used where appropriate. The technical repeats and applied statistics are mentioned in each Figure legends. A p-value of ≤0.05 was considered to be statistically significant. All data are presented as mean ± SEM.

We first determined the effect of pH on spontaneous and Nox-dependent NETosis. We incubated purified peripheral blood neutrophils (PMNs) at pH above or below physiological blood pH (6.6–7.8; in 0.2 pH increments) with or without the prototypic agonist PMA to induce Nox-dependent NETosis. Monitoring Sytox green fluorescence (a proxy for extracellular NET DNA) every 30 min for 4 h showed that raising pH increased the rate and amount of NETosis (Figures 1A,B). To confirm true NETosis, we conducted confocal immunofluorescence microscopy. These images showed colocalization of MPO and NET DNA and confirmed that resting and PMA-stimulated neutrophils formed greater amounts of NETs with increasing pH (Figures 1C,D; Figure S1 in Supplementary Material). A regression line at the final time point showed a clear increase in Sytox Green readings with increasing pH (Figures 1A,B, inset). The slope of spontaneous NETosis was less steep than that of PMA-mediated NETosis, indicating that the rate of spontaneous NETosis was lower than that of PMA-mediated NETosis. Overall, pH affected both spontaneous and PMA-mediated NETosis; pH above the normal blood pH of ~7.4 promoted NETosis, whereas a more acidic pH suppressed NETosis.

To determine the mechanism that governs pH-mediated NETosis, we next tested how pH_e changes affect pH_i. We used a pH-sensitive dual-wavelength SNARF dye for this purpose. SNARF-loaded neutrophils were suspended in media with six different levels of pH_e ranging from 6.6 to 7.8, in the presence or absence of PMA. Changes in pH_i were monitored by SNARF wavelength ratios every 10 min up to 1 h. We limited our analysis to 1 h because neutrophils in both control and PMA conditions are viable up to this time point (Figure 1; Sytox assays also indicate cell permeability), avoiding the extracellular buffers entering any dead cells. Within 10–20 min, the changes in pH_i of the neutrophils reflected the changes in pH_e in both control and PMA conditions (Figures 2A,B). The pH_i values of PMA-treated neutrophils were lower than in control neutrophils (the SNARF ratio difference was 0.62; p = 0.001 and p = 0.002) over the entire pH_e range studied; however, the pH_i did not change substantially after 10 min (slope of the graphs at 10 and 60 for controls: 0.124 vs. 0.126; for PMA: 0.093 vs. 0.095) (Figures 2C,D). Activated Nox catalyzes the oxidation of NADPH and generates NADP⁺, electron (e⁻), and H⁺, concomitantly transports e⁻ out of the cytoplasm but leaves the NADP⁺ and H⁺ in the cytoplasm; hence, pH_i should be lower in PMA-activated neutrophils than control neutrophils. The data obtained in this set of experiments indicate that activated neutrophils had an acidic pH_i, which was further reduced by low pH_e.

The e⁻ generated by Nox reacts with molecular O₂ to generate superoxide among other ROS, which are important for NETosis (15, 16). Therefore, we measured intracellular ROS by DHR123 as previously described (5, 7, 14). Neutrophils were preloaded with DHR123 and activated with or without PMA at six different pH levels. Generation of oxidized green florescence R123 in these neutrophils was monitored every 10 min for 30 min. Raising pH stimulated ROS production in both control and PMA-treated neutrophils (Figures 3A,B). The magnitude and rates of ROS production were higher for PMA-treated than control neutrophils (see insets). To confirm increased ROS production, we imaged these cells as well as performed flow cytometry analyses at 30 min where ROS production was high in PMA-treated cells. These data sets show that increasing pH increased ROS production. Elevating pH also increased production of ROS in control neutrophils, albeit to a lesser degree (Figures 3C,D). Therefore, raising pH increased ROS production in neutrophils, and the pH effect was enhanced in the presence of the activators of neutrophils.

Neutrophils can generate ROS either by Nox or mitochondria (17, 18). To determine whether pH-dependent NETosis is attributable to Nox-mediated ROS production, we repeated the NETosis assays, in the presence or absence of the Nox inhibitor diphenyleneiodonium (DPI). Sytox Green plate reader assays (Figures 4A,B; Figure S2 in Supplementary Material) and confocal microscopy (Figures 4C,D) indicated that DPI suppresses both spontaneous and PMA-mediated NETosis. Therefore, like PMA-mediated NETosis, spontaneous NETosis was also dependent on Nox activity.

Taken together, these results (Figures 1–4) showed that alterations in pH_e led to pH_i adjustments within 10–20 min; elevating pH stimulated Nox activity and subsequent ROS production and thereby enhancing spontaneous and PMA-mediated NETosis (PMA > > no PMA). Higher pH facilitated these steps, whereas low pH exerted the opposite effect, suppressing both spontaneous and PMA-mediated NETosis.

In addition to ROS generation, cleavage of histones by granular proteases is a key step in Nox-dependent NETosis. Therefore, we determined H4 cleavage by Western blot analysis during spontaneous and PMA-mediated NETosis across a range of pH conditions. No substantial differences were detected during spontaneous NETosis. By contrast, a clear pH-dependent increase in H4 cleavage was detected during PMA-mediated NETosis. Standardizing H4 cleavage with GAPDH confirmed the pH-dependent increase in histone cleavage (Figures 5A,B). The intensity of all protein bands decreased with increasing pH, suggesting that neutrophil proteases were more active at higher pH. Therefore, neutrophil proteases that are more active at higher pH promote histone cleavage in PMA-mediated NETosis under more alkaline conditions.

We also examined the degree of CitH3 formation by immunocytochemistry to examine the relevance of PAD4 in the pH-dependent increase in Nox-dependent NETosis. Some degree of CitH3 formation (based on the qualitative examination of CitH3 immunostaining) was detected at higher pH conditions. The few NETotic neutrophils present at higher pH showed CitH3, whereas the intact neutrophils did not (Figure S3 in Supplementary Material). Nevertheless, CitH3 is hardly detectable in PMA-mediated NETosis compared to the positive controls (neutrophils activated with calcium ionophore A23187). As expected, the degree of CitH3 formation was much less in PMA-mediated Nox-dependent NETosis than calcium ionophore-mediated NETosis (4, 5). Therefore, raising pH induced a modest degree of CitH3 formation, but the contribution of CitH3 to Nox-dependent NETosis is low.

Gram-negative bacteria, their cell wall component LPS, and Gram-positive bacteria induce Nox-dependent NETosis (5, 7, 19, 20). Therefore, to determine the effect of pH on biologically relevant NETosis-inducing agonists, we first tested the effect of pH on LPS-mediated NETosis. Sytox Green plate reader assays and imaging showed that raising pH stimulated LPS-mediated NETosis (Figure 6A). Regression analyses conducted at the last time point showed a clear pH-dependent effect of NETosis (inset). Immunofluorescence confocal microscopy and quantifying neutrophil with different nuclear morphology and extracellular DNA confirmed the Sytox Green plate reader assay findings (Figures 6B,C). Therefore, as with PMA-mediated Nox-dependent NETosis, elevating pH stimulated LPS-mediated NETosis.

We also tested the effect of pH on NETosis induced by pathogens such as the Gram-negative bacterium P. aeruginosa and Gram-positive bacterium S. aureus. Like PMA and LPS conditions, more NETosis occurred in response to bacteria under more alkaline conditions (Figure 7). Collectively, these data (Figures 1–7) showed that raising pH increased ROS production, protease activity, and NETosis induced under baseline conditions, as well as by PMA, LPS, Gram-negative, and Gram-positive bacteria. The pH effect was much higher when the NETosis was induced by an agonist.

Different compounds can be used for adjusting pH. Bicarbonate and Tris are two compounds that can be used to increase pH in humans. Theoretically, each Tris molecule can bind to three H⁺ ions, whereas each bicarbonate ion can neutralize one H⁺ ion. Hence, we tested the efficiency of these compounds in modulating NETosis. To increase the pH of the media from 6.6 to various pHs, Tris solutions required smaller volumes than the equimolar bicarbonate solutions (Figure 8A). Neutrophils in low pH (6.6) media were activated by negative control (medium), PMA, or LPS to induce NETosis. After 30 min, the pH was adjusted by adding precalculated amounts of sodium bicarbonate or Tris to 7.4, and NETosis kinetics were recorded. The percentage of DNA release graphs show that the pH elevation from 6.6 to 7.4 enabled these neutrophils to undergo effective NETosis (Figures 8B–D). Therefore, both bicarbonate and Tris base (or THAM) could be used for increasing the pH and subsequently promoting NETosis. Tris requires much smaller volumes than bicarbonate solutions to adjust the pH, hence, this trivalent molecule may be a better treatment option than bicarbonate.