To the editor:
Neutrophil extracellular traps (NETs) have been recently identified as major contributors of several hematological and vascular diseases. These disorders include thrombosis, small vessel vasculitis, systemic lupus erythematosus, autoimmunity, pneumonia, sepsis, and blood transfusion–related acute lung injury.1-4 NETs are DNA-based extracellular traps that not only trap and kill invading microbes but also injure host tissues.1,5-7 Therefore, regulating NETosis is important to prevent many pathological conditions.1 However, key molecules that switch neutrophil death from NETosis, which is proinflammatory, to apoptosis, which is anti-inflammatory, have not been clearly established.
Nicotinamide adenine dinucleotide phosphate oxidase 2 (NOX2)-dependent reactive oxygen species (ROS) production in neutrophils can induce either NETosis or apoptosis. Phorbol 12-myristate 13-acetate (PMA) has been extensively used as an agonist to activate NOX2-mediated ROS production to study NETosis.6,8 A seminal study showed that PMA induces autophagy and that both autophagy and PMA-mediated ROS production are required for NETosis.8 This inference was made based on the inhibitory effect of a protein kinasae C inhibitor (wortmannin) on PMA-mediated autophagy and NETosis. In another study, rapamycin was used for directly suppressing mammalian target of rapamycin, a well-established regulator of autophagy. These studies show that mammalian target of rapamycin regulates NETosis via modifying hypoxia-inducible factor HIF1-α.9 However, the identities of other key kinases that regulate NETosis-apoptosis pathways remain elusive.
Akt is a well-known inhibitor of apoptosis.10 Inhibition of Akt using pharmacological inhibitors promotes apoptosis in many cell types. Hence, it is an excellent candidate to act as a direct molecular switch for regulating the NETosis-apoptosis axis. Here we show that PMA activates Akt during the induction of NETosis (Figure 1A, lanes 1 and 2), whereas the NOX2 inhibitor diphenyleneiodonium (DPI) completely suppresses Akt activation (Figure 1A, lane 3). Therefore, Akt activation is dependent on NOX2-mediated ROS production. Flow cytometry analysis confirms the production of ROS in these cells and shows that DPI, but not Akt-specific inhibitor (Akt-i) XI, inhibits PMA-induced ROS production (Figure 1B). In a Sytox Green plate reader assay, 2 different Akt inhibitors, M2206 and XI, inhibit DNA release by activated neutrophils in a dose-dependent manner (Figure 1C-D). Therefore, activation of Akt is essential for NOX2-mediated NETosis.
![Figure 1. Akt shifts NETosis to apoptosis in human neutrophils. (A) Immunoblot analysis for the activation of Akt as determined by its phosphorylation. Cells were lysed after 1 hour of PMA (25 nM) activation. DPI (20 μM) was used for inhibiting NOX2-mediated ROS production. Total Akt and glyceraldehyde-3-phosphate dehydrogenase were used as loading control (n = 6 donors; −ve, negative control; DPI → PMA, neutrophils were pretreated with DPI [20 μM] and then activated with PMA). (B) Analysis of cells for their production of ROS by flow cytometry. Prior to the activation with PMA (25 nM), cells were preincubated for 30 or 60 minutes in the presence or absence of Akt inhibitor XI (Akt-i XI; 10 μM; Akt-i XI → PMA) or DPI (20 μM; DPI → PMA) for 30 minutes. In all conditions, cells were also incubated with dihydrorhodamine 123 as a probe for ROS production. Cells were gated using forward and side scatter and confirmed with Hoechst 3342 as a counterstain. Shown is a representative of 4 independent experiments with cells from 4 individuals. (C-D) Fluorescence plate reader assay for NETosis. Cells were cultured in a 96-well culture plate (3 × 105 cells per well) in the presence of Sytox Green (5 μM; Invitrogen), a cell impermeable DNA-binding dye, to monitor release of NET DNA. Varying concentrations (0-10 μM) of 2 different Akt-i, (C) Akt inhibitor XI (Millipore), and (D) MK2206 (Sellekchem), were added to the cells 30 minutes prior to the activation of cells with PMA (25 nM). Numbers beneath the graphs represent the concentration of Akt-i used prior to activation with PMA. Extracellular DNA release was monitored at t = 0, 30, and 60 minutes and every hour for a total of 5 hours. Shown is the fluorescence intensity at 5 hours. NETotic index was calculated as percentage of total fluorescence given off by PMA-only positive control. (n = 4-7 donors). (E) Differential quantification of live, NETotic, and apoptotic (pyknotic) nuclei. Cells were cultured in an imageable special optics 96-well plate in the presence or absence of Akt-i (0-10 μM) for 30 or 60 minutes prior to the activation of PMA (25 nM). The numbers in parentheses represent the concentration of Akt-i used prior to activation with PMA. Live and apoptotic nuclei are not stained by Sytox Green dye unless the cells are fixed. Thus, the cells from the plate reader assay were fixed at the end of the assay in the presence of the dye, and cells were differentially quantified on the basis of their nuclear morphology. Representative images of live, NETotic, and apoptotic nuclei quantified are shown below the graph. At least 100 cells were quantified in each condition (n = 4-5 individual donors). (F) Immunofluorescence staining for myeloperoxidase (MPO) and cCasp3. Neutrophils were incubated with H2O2 (8 mM) to induce necrosis. In other conditions, cells were activated with PMA (25 nM) with or without preincubation for 30 minutes with Akt inhibitor XI. Cells were stained with MPO (mouse α-MPO, 1:250; Abcam) as a marker for NETs and cCasp3 (rabbit α-cleaved caspase 3, 1: 150; Cell Signaling) as a marker for apoptosis. Arrows, NETotic DNA and nuclei; arrowheads, cCasp3-positive cells. Bar: 10 µm (n = 4 donors). All data are expressed as mean ± standard error of the mean where appropriate. *P < .05 and ***P < .001 compared with PMA only controls (C-D). Analysis of variance with (C-D) Dunnett's or (E) Bonferroni post-tests was used for determining statistical significance. (G) Proposed role of Akt in regulating the switch between NETosis and apoptosis.](https://ash.silverchair-cdn.com/ash/content_public/journal/blood/123/4/10.1182_blood-2013-09-526707/4/m_597f1.jpeg?Expires=1724243496&Signature=UXCivWji7PB14pqyyby9IrmU1838pqVS0xUCFia092eOiiSsygcJOLmn2-Yca9TDBHRSULchXlpWXvjedEL78NS39vnwUnhWMUqjBOPcdfBJihVBzoeq~w2f~uUwuoNgzb7013ppM6HnUoIX7zOkg6BRyEcLR~Id2IdsoaxDeK65NZ~VES~uTt279oofhXsA9gfosexkUk571yMn0C4uqzIqiHe7FOiC3GDW~E-MPpP57A8eG6CvaK7U-onte2pQHjceC2MUs636qreq~cxz2X1jGLX3-fVWd7WhRskNqvpGNr5fkHDxF0-XuoTlFB2wRf5WG1ANraFhXHsk1AxEGw__&Key-Pair-Id=APKAIE5G5CRDK6RD3PGA)
Akt shifts NETosis to apoptosis in human neutrophils. (A) Immunoblot analysis for the activation of Akt as determined by its phosphorylation. Cells were lysed after 1 hour of PMA (25 nM) activation. DPI (20 μM) was used for inhibiting NOX2-mediated ROS production. Total Akt and glyceraldehyde-3-phosphate dehydrogenase were used as loading control (n = 6 donors; −ve, negative control; DPI → PMA, neutrophils were pretreated with DPI [20 μM] and then activated with PMA). (B) Analysis of cells for their production of ROS by flow cytometry. Prior to the activation with PMA (25 nM), cells were preincubated for 30 or 60 minutes in the presence or absence of Akt inhibitor XI (Akt-i XI; 10 μM; Akt-i XI → PMA) or DPI (20 μM; DPI → PMA) for 30 minutes. In all conditions, cells were also incubated with dihydrorhodamine 123 as a probe for ROS production. Cells were gated using forward and side scatter and confirmed with Hoechst 3342 as a counterstain. Shown is a representative of 4 independent experiments with cells from 4 individuals. (C-D) Fluorescence plate reader assay for NETosis. Cells were cultured in a 96-well culture plate (3 × 105 cells per well) in the presence of Sytox Green (5 μM; Invitrogen), a cell impermeable DNA-binding dye, to monitor release of NET DNA. Varying concentrations (0-10 μM) of 2 different Akt-i, (C) Akt inhibitor XI (Millipore), and (D) MK2206 (Sellekchem), were added to the cells 30 minutes prior to the activation of cells with PMA (25 nM). Numbers beneath the graphs represent the concentration of Akt-i used prior to activation with PMA. Extracellular DNA release was monitored at t = 0, 30, and 60 minutes and every hour for a total of 5 hours. Shown is the fluorescence intensity at 5 hours. NETotic index was calculated as percentage of total fluorescence given off by PMA-only positive control. (n = 4-7 donors). (E) Differential quantification of live, NETotic, and apoptotic (pyknotic) nuclei. Cells were cultured in an imageable special optics 96-well plate in the presence or absence of Akt-i (0-10 μM) for 30 or 60 minutes prior to the activation of PMA (25 nM). The numbers in parentheses represent the concentration of Akt-i used prior to activation with PMA. Live and apoptotic nuclei are not stained by Sytox Green dye unless the cells are fixed. Thus, the cells from the plate reader assay were fixed at the end of the assay in the presence of the dye, and cells were differentially quantified on the basis of their nuclear morphology. Representative images of live, NETotic, and apoptotic nuclei quantified are shown below the graph. At least 100 cells were quantified in each condition (n = 4-5 individual donors). (F) Immunofluorescence staining for myeloperoxidase (MPO) and cCasp3. Neutrophils were incubated with H2O2 (8 mM) to induce necrosis. In other conditions, cells were activated with PMA (25 nM) with or without preincubation for 30 minutes with Akt inhibitor XI. Cells were stained with MPO (mouse α-MPO, 1:250; Abcam) as a marker for NETs and cCasp3 (rabbit α-cleaved caspase 3, 1: 150; Cell Signaling) as a marker for apoptosis. Arrows, NETotic DNA and nuclei; arrowheads, cCasp3-positive cells. Bar: 10 µm (n = 4 donors). All data are expressed as mean ± standard error of the mean where appropriate. *P < .05 and ***P < .001 compared with PMA only controls (C-D). Analysis of variance with (C-D) Dunnett's or (E) Bonferroni post-tests was used for determining statistical significance. (G) Proposed role of Akt in regulating the switch between NETosis and apoptosis.
Akt shifts NETosis to apoptosis in human neutrophils. (A) Immunoblot analysis for the activation of Akt as determined by its phosphorylation. Cells were lysed after 1 hour of PMA (25 nM) activation. DPI (20 μM) was used for inhibiting NOX2-mediated ROS production. Total Akt and glyceraldehyde-3-phosphate dehydrogenase were used as loading control (n = 6 donors; −ve, negative control; DPI → PMA, neutrophils were pretreated with DPI [20 μM] and then activated with PMA). (B) Analysis of cells for their production of ROS by flow cytometry. Prior to the activation with PMA (25 nM), cells were preincubated for 30 or 60 minutes in the presence or absence of Akt inhibitor XI (Akt-i XI; 10 μM; Akt-i XI → PMA) or DPI (20 μM; DPI → PMA) for 30 minutes. In all conditions, cells were also incubated with dihydrorhodamine 123 as a probe for ROS production. Cells were gated using forward and side scatter and confirmed with Hoechst 3342 as a counterstain. Shown is a representative of 4 independent experiments with cells from 4 individuals. (C-D) Fluorescence plate reader assay for NETosis. Cells were cultured in a 96-well culture plate (3 × 105 cells per well) in the presence of Sytox Green (5 μM; Invitrogen), a cell impermeable DNA-binding dye, to monitor release of NET DNA. Varying concentrations (0-10 μM) of 2 different Akt-i, (C) Akt inhibitor XI (Millipore), and (D) MK2206 (Sellekchem), were added to the cells 30 minutes prior to the activation of cells with PMA (25 nM). Numbers beneath the graphs represent the concentration of Akt-i used prior to activation with PMA. Extracellular DNA release was monitored at t = 0, 30, and 60 minutes and every hour for a total of 5 hours. Shown is the fluorescence intensity at 5 hours. NETotic index was calculated as percentage of total fluorescence given off by PMA-only positive control. (n = 4-7 donors). (E) Differential quantification of live, NETotic, and apoptotic (pyknotic) nuclei. Cells were cultured in an imageable special optics 96-well plate in the presence or absence of Akt-i (0-10 μM) for 30 or 60 minutes prior to the activation of PMA (25 nM). The numbers in parentheses represent the concentration of Akt-i used prior to activation with PMA. Live and apoptotic nuclei are not stained by Sytox Green dye unless the cells are fixed. Thus, the cells from the plate reader assay were fixed at the end of the assay in the presence of the dye, and cells were differentially quantified on the basis of their nuclear morphology. Representative images of live, NETotic, and apoptotic nuclei quantified are shown below the graph. At least 100 cells were quantified in each condition (n = 4-5 individual donors). (F) Immunofluorescence staining for myeloperoxidase (MPO) and cCasp3. Neutrophils were incubated with H2O2 (8 mM) to induce necrosis. In other conditions, cells were activated with PMA (25 nM) with or without preincubation for 30 minutes with Akt inhibitor XI. Cells were stained with MPO (mouse α-MPO, 1:250; Abcam) as a marker for NETs and cCasp3 (rabbit α-cleaved caspase 3, 1: 150; Cell Signaling) as a marker for apoptosis. Arrows, NETotic DNA and nuclei; arrowheads, cCasp3-positive cells. Bar: 10 µm (n = 4 donors). All data are expressed as mean ± standard error of the mean where appropriate. *P < .05 and ***P < .001 compared with PMA only controls (C-D). Analysis of variance with (C-D) Dunnett's or (E) Bonferroni post-tests was used for determining statistical significance. (G) Proposed role of Akt in regulating the switch between NETosis and apoptosis.
To assess whether Akt is involved in redirecting NETosis to apoptosis, immunofluorescence microscopy and quantitative analyses were performed. The results show that preincubation of cells with Akt-i XI dose dependently increases the number of apoptotic cells containing pyknotic nuclei and a concomitant decrease in NETotic cells (Figure 1E). Immunofluorescence microscopy analysis of MPO and cleaved caspase 3 (cCasp3) further confirms that the inhibition of Akt switches neutrophil death from NETosis to apoptosis (Figure 1F). H2O2 (8 mM) is known to induce necrosis in neutrophils.8 Necrosis in neutrophils neither activates apoptotic caspase 3 nor precoats MPO on DNA before release (Figure 1F). Collectively, these data show that NOX2-mediated NETosis is dependent on Akt activation, and suppression of Akt switches NETosis to apoptosis.
Based on the data presented in this study, we propose that Akt is a bona fide molecular switch that regulates the NETosis-apoptosis axis (Figure 1G). Taken together, PMA-mediated NOX2-dependent activation of Akt induces NETosis while suppressing apoptosis. Suppression of Akt, on the other hand, allows for the induction of caspase-dependent apoptosis. The finding that NETosis and apoptosis are 2 opposing pathways in neutrophils and that NETosis can be redirected by targeting Akt could provide avenues for novel therapeutic strategies to treat NET-related hematological and other inflammatory disorders.
Authorship
Acknowledgments: Approval to obtain blood samples from healthy volunteers for the study was approved by the Research Ethics Board of the Hospital for Sick Children.
D.N.D. was supported by an Ontario graduate scholarship, the Ontario Student Opportunity Trust Fund/SickKids Restracomp, the Dr Goran Enhorning Award in Pulmonary Research, and the Peterborough K.M. Hunter graduate studentship. L.Y. was supported by an Ontario graduate scholarship. M.A.K. received a postdoctoral fellowship from the operating grants awarded to N.P. from Cystic Fibrosis Canada (grant 2619) and Canadian Institutes of Health Research (CIHR; MOP-111012). This work was funded by CIHR operating grant MOP-111012 to N.P.
Contribution: D.N.D. designed and conducted experiments, analyzed the data, and wrote the manuscript; L.Y. and M.A.K. did experiments, analyzed the data, and participated in manuscript revisions and editing; H.G. participated in experimental design; and N.P. conceived the project, supervised the experiments, analyzed the data, and participated in manuscript revisions and editing.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
Correspondence: Nades Palaniyar, Lung Innate Immunity Research Laboratory, The Hospital For Sick Children, 555 University Ave, Toronto, ON, Canada M5G 1X8; e-mail: nades.palaniyar@sickkids.ca.
