Mincle-Mediated Neutrophil Extracellular Trap Formation by Regulation of Autophagy

Summary

This study identifies C-type lectin receptor Mincle as novel regulator of neutrophil extracellular trap (NET) formation by controlling autophagy, without affecting reactive oxygen species generation. It can have implications to modulate NET formation without compromising antimicrobial capacity of neutrophils.

Abstract

Background.

Neutrophil extracellular traps (NETs) constitute antimicrobial function of neutrophils but have also been linked to perpetuation of inflammation. Despite this evident physiological relevance, mechanistic understanding of NET formation is poor. In this study, we examined the mechanism by which Mincle, a C-type lectin receptor, regulates NET formation.

Methods.

NET formation, reactive oxygen species, autophagy activation and intracellular signaling pathways were analyzed in Mincle-sufficient and -deficient neutrophils stimulated in vitro with various stimuli and in vivo during Klebsiella infection.

Results.

We found that Mincle mediates NET formation in response to several activation stimuli in vitro and in vivo during pneumoseptic infection with Klebsiella pneumoniae, indicating its regulatory role in NET formation. Mechanistically, we show that attenuated NET formation in Mincle-/- neutrophils correlates with an impaired autophagy activation in vitro and in vivo, whereas reactive oxygen species (ROS) formation in these neutrophils remained intact. The requirement of autophagy in Mincle-mediated NET formation was further supported by exogenous treatment with autophagy inducer tamoxifen, which rescued the NET formation defect in Mincle-/- neutrophils.

Conclusions.

Our findings identify a previously unrecognized role of Mincle as a regulator of autophagy, which mediates NET formation without affecting ROS generation. Our study addresses a major challenge in the field by positing this pathway to be targeted for modulation of NETs while preserving ROS production, an important innate immune defense.

Neutrophils are the first responder cell type for combating a pathological insult [1, 2]. A recently established paradigm of neutrophil activation is the formation of neutrophil extracellular traps (NETs), which are decondensed chromatin fibrils coated with granular proteases and histones [3, 4]. Neutrophil extracellular traps can trap and kill extracellular pathogens by placing them in close proximity to antimicrobial components [3, 5]. On the other hand, unconstrained NET release has been linked to immunopathologies, including lupus, arthritis, acute lung injury, graft dysfunction, and preeclampsia [6–8]. This contrasting consequence of NET formation indicates that NET formation under various microenvironments needs to be tightly regulated. In spite of the proven physiological relevance of NETs, the identity of triggering receptors and the mechanisms of subcellular events of NETosis remain poorly defined.

To date, NET formation has largely been reported to be dependent on reactive oxygen species (ROS) generation by NADPH oxidase complex [9, 10]. Inhibitors of this complex abrogate NET formation in response to a variety of stimuli [11, 12]. This is further supported by studies on patients with chronic granulomatous disease, who lack NADPH oxidase activity and thus fail to form NETs [13]. However, reports showing ROS-independent NET formation or impaired NETs in spite of high ROS [12, 14] indicate that additional mediators and/or more complex activation mechanisms are likely involved in this process. In this regard, activation of autophagy, a homeostatic process regulating turnover of intracellular proteins, was recently shown to accelerate NET formation [15]. Accordingly, pharmacological inhibition of autophagy impaired NET formation [16]. However, the factors governing the activation of autophagy and the extent of cross-talk between ROS and autophagy in context of NET formation remain unclear.

Mincle is a C-type lectin receptor that has been reported to function as an activating receptor for host- as well as pathogen-associated molecular patterns (reviewed in [17, 18]). It is an inducible receptor, expressed mainly by myeloid cells such as macrophages, neutrophils, myeloid dendritic cells, and some B-cell subsets [17, 19, 20]. Although functional analysis of this receptor in macrophages has received the most attention [21–23], its role in regulating neutrophil-mediated responses is much less defined and its function in NET formation is completely unexplored.

We observed a reduced NET formation in Mincle^-/- mice during pneumoseptic infection with Klebsiella pneumoniae (KPn) [24]. In this study we examined the role of Mincle in NET formation in vitro and in vivo and dissected the mechanism by which it regulates NET formation. By using primary neutrophils, neutrophil activation stimuli, and a preclinical model of KPn-induced pneumonic sepsis, our findings implicate Mincle as a novel regulator of NET formation via autophagy and posit this pathway for therapeutic modulation of NET formation, segregated from ROS generation, in pneumonic sepsis, with implications in other NET-mediated inflammatory conditions.

METHODS

Bacterial Strain and Mice

The KPn (ATCC strain 43816) were grown to log phase in LB medium at 37°C. All in vivo experiments and isolation of neutrophils were performed using wild-type (WT) C57BL/6 or Mincle^-/- mice aged 6–8 weeks on same background obtained from the Consortium of Functional Genomics, Scripps, La Jolla, California, and bred in the animal facility of the University of North Dakota. The animals were used according to institutional and federal guidelines.

Reagents and Antibodies

Phorbol-myristate-acetate (PMA), N-formylmethionyl-leucyl-phenylalanine (fMLP), rapamycin, and tamoxifen were purchased from Sigma-Aldrich (St. Louis, MO). Other reagents included SYTOX Green Nucleic Acid Stain (Molecular Probe, Life Technologies), cytosolic ROS detection by fluoro H₂O₂ detection kit (Cell Technology), mitochondrial ROS detection using MitoSOX mitochondrial superoxide indicator (Molecular Probes), Apocynin (Santa Cruz Biotech), and siRNA targeted to Mincle and CARD9 (Santa Cruz). The anti-microtubule-associated protein light chain 3 (LC3) antibody, anti-Beclin-1 antibody, anti-tyrosine phospho-Syk, and anti-CARD-9 antibodies, phospho- and total p38, and Erk antibodies were all from Cell Signaling Technology.

Neutrophil Extracellular Traps In Vitro and In Vivo

For in vitro studies, peritoneal neutrophils were isolated from WT and Mincle^-/- mice 8–12 hours after the injection of sterile 4% thioglycollate in the peritoneal cavity of mice, and neutrophils were enriched (75%–80% pure as assessed by flow cytometry using Ly6G and CD11b antibodies). Isolated neutrophils were stimulated with PMA (50 nM) or fMLP (1 µM) for 4 hours at 37°C. For some experiments, neutrophils were treated with apocynin (10 uM) for 30 minutes, rapamycin (500 nM) for 90 minutes, or tamoxifen (6 uM) for 60 minute before stimulation with PMA. Neutrophils were cytocentrifuged on glass slides and fixed with paraformaldehyde, and NETs were stained using Sytox Green. The percentage of NET formation was manually quantitated by dividing the number of NET-forming neutrophils by the total number of cells in 8–10 random microscopic fields and multiplying the values by 100. NETs were quantified in vivo in KPn-infected mice lungs as described previously [24].

Transfection of Neutrophils

For some experiments, peritoneal neutrophils isolated from WT mice were transfected with control or test siRNA targeted to Mincle or CARD9 using Amaxa Nucleofector System (Lonza) 6 hours before stimulation with agonists. Briefly, 0.5 uM control or test siRNA was transfected in 2 × 10⁶ neutrophils using Amaxa human monocyte nucteofector kit and LONZA Nucleofector 2b Device per manufacturer’s instructions. The transfection efficiency was checked by using a pmax green fluorescent protein (GFP) vector as positive control, as well by Western blotting using respective antibodies.

Measurement of Reactive Oxygen Species

Intracellular ROS were measured by detecting hydrogen peroxide using the Fluoro H₂O₂ detection kit (Cell Technology) per the manufacturer’s instructions following stimulation with PMA for 10 minutes. The amount of ROS was deduced by plotting against a standard curve of known concentrations of H₂O₂. Reactive oxygen species measurement in BAL neutrophils was carried out by the same procedure immediately after isolation of cells from infected mice. For mitochondrial ROS measurement, neutrophils were suspended in Hank's balanced salt solution (HBSS) and stimulated with PMA for 15 minutes, followed by 10 minutes of incubation with MitoSOX Red Mitochondrial Superoxide Indicator (5 uM). Cells were then washed with HBSS and subjected to flow cytometry using a BD LSR II (Becton Dickinson, San Jose, CA). FlowJo (Tree Star) software was used to analyze all data.

Immunofluorescence Microscopy

For LC3 detection, neutrophils stimulated with PMA (50 nM) with or without rapamycin (500 nM) or BAL neutrophils isolated 3 days after infection from KPn-infected WT and Mincle^-/- mice were cytocentrifuged on glass slides and fixed with 4% paraformaldehyde. Immunostaining was performed by methods previously described [24, 25].

Western Blot Analysis

For the detection of signaling molecules, neutrophils were plated in 60-mm dishes at the density of 5 × 10⁶ cells and stimulated as described above. The cells were stimulated with PMA (50 nM) for 15 minutes with or without pretreatment with rapamycin (500 nM) or tamoxifen (6 uM). For in vivo analysis of signaling molecules, BAL neutrophils from KPn-infected mice were isolated 3 days after infection and were immediately processed as described previously [26]. Immunoreactivity of indicated antibodies was detected using super signal west Pico Chemiluminisecnt detection reagent (Thermo Scientific) and analyzed on BioRad reader using Chembio software.

Statistical Analysis

Statistical analyses were performed using the Student t test (SIGMA PLOT 8.0, Systat Software, San Jose, CA). P ≤ .05 was considered statistically significant.

RESULTS

Mincle-Mediated Neutrophil Extracellular Trap Formation in Response to Neutrophil Activation Stimuli

To examine a specific function of Mincle in NET formation, we first stimulated primary neutrophils isolated from WT and Mincle^-/- mice in vitro with PMA and fMLP, 2 stimuli extensively used to activate NET formation. Very low to no spontaneous NET formation was observed in WT or Mincle^-/- neutrophils in the absence of these stimuli (Figure 1A). Upon stimulation, WT neutrophils produced NETs, as visualized by staining with the DNA dye Sytox Green (Figure 1A, upper panel). Mincle^-/- neutrophils exhibited significantly reduced NET formation in response to both stimuli, in comparison with the WT neutrophils (Figure 1A, lower panel). This suggested that Mincle plays a role in PMA- and fMLP-induced NET formation in vitro. A direct requirement of Mincle for NET formation in response to these stimuli was confirmed by siRNA knockdown of Mincle in WT neutrophils, which resulted in similarly diminished NET formation in response to PMA (Figure 1B) as well as fMLP (data not shown). Efficiency of siRNA knockdown of Mincle was confirmed by Western blotting (Supplementary Figure 1). These results strongly suggested that Mincle is a critical regulator NET formation.

Figure 1.

Mincle regulates neutrophil extracellular trap (NET) formation in response to phorbol-myristate-acetate (PMA) and N-formylmethionyl-leucyl-phenylalanine (fMLP) stimulation. A, Representative fluorescence images of wild-type (WT) and Mincle^-/- neutrophils unstimulated or stimulated with PMA (50 nM) or fMLP (1 uM) for 4 hours. Neutrophil extracellular traps were fixed and stained with Sytox Green as described in the Methods. Magnification = 200X. The bar graph shows quantitation of NET-forming WT and Mincle^-/- neutrophils. NS; no stimulation. Data presented in the bar graph is from 4 independent experiments. (*P < .05). B, Representative images of NETs in WT neutrophils transfected with control or siRNA targeted to Mincle 6 hours before stimulation with PMA. Neutrophil extracellular traps were stained with Sytox Green 4 hours after PMA stimulation. Magnification = 200X. Percentage of NET forming neutrophils ± SD from 4 independent experiments is shown in the bar graph. **P < .01. Abbreviations: fMLP, N-formylmethionyl-leucyl-phenylalanine; NET, neutrophil extracellular trap; NS, no stimulation; PMA, phorbol-myristate-acetate; WT, wild-type.

Reactive Oxygen Species Production in Mincle^-/- Neutrophils

Owing to a well-established contribution of ROS in NET formation [4, 10], we analyzed whether impaired NET formation in Mincle^-/- neutrophils is due to a defect in ROS production. By using a Fluoro H₂O₂ detection kit, we found no effect of Mincle deficiency on ROS generation capacity of neutrophils upon PMA stimulation (Figure 2A). Mincle^-/- neutrophils in fact showed a slightly higher production of ROS compared with the WT neutrophils, albeit this difference was not statistically significant. Primary neutrophils transfected with Mincle siRNA remained similarly competent in ROS production in response to PMA stimulation (Figure 2B). These results showed that Mincle deficiency attenuated the NET formation without affecting ROS generation capacity. Furthermore, inhibition of ROS by treatment with apocynin, an NADPH oxidase inhibitor, did not affect the impaired NET formation in Mincle^-/- neutrophils (Figure 2C). Because mitochondria also contribute to cellular ROS generation [27], we also analyzed whether mitochondrial ROS is affected by Mincle deficiency. Phorbol-myristate-acetate stimulation did not increase the generation of mitochondrial ROS in WT neutrophils compared with the unstimulated neutrophils (Figure 2D), a finding consistent with previous reports [28, 29]. Mincle^-/- neutrophils exhibited similar basal levels of mitochondrial ROS as the WT neutrophils, which remained unchanged upon PMA stimulation (Figure 2D). These results indicated that mitochondrial ROS is not involved in PMA-induced and Mincle-mediated NET formation.

Figure 2.

Mincle-deficient neutrophils are fully competent in reactive oxygen species (ROS) generation. A, Reactive oxygen species were measured in wild-type (WT) and Mincle^-/- neutrophils 10 minutes after phorbol-myristate-acetate (PMA) stimulation using a fluoro H₂O₂ detection kit as described in Methods. Data from 4 independent experiments are shown. No statistically significant differences were found between the levels of ROS in WT and Mincle^-/- neutrophils. B, Reactive oxygen species measurement in WT neutrophils transfected with control or test siRNA targeted against Mincle 6 hours before stimulation with PMA for 10 minutes. Reactive oxygen species generation was measured in unstimulated or stimulated neutrophils. C, Quantitation of neutrophil extracellular trap (NET) formation by WT and Mincle^-/- neutrophils stimulated for 4 hours with PMA with or without treatment with NADPH oxidase inhibitor apocynin. Graph shows average ± SD from 3 independent experiments. *P < .05; **P < .01. D, Flow cytometry analysis of mitochondrial ROS (MitoSox) in unstimulated or PMA-stimulated WT and Mincle^-/- neutrophils. The numbers on contour plots represent percentages of MitoSox-positive cells. Bar graph shows average ± SD of MitoSox-positive cells from 5 independent experiments. Abbreviations: Apo, apocynin; NET, neutrophil extracellular trap; NS, not stimulated; PMA, phorbol-myristate-acetate; ROS, reactive oxygen species; WT, wild-type.

Impaired Neutrophil Extracellular Trap Formation in Mincle^-/- Neutrophils and Defective Autophagy

Because ROS formation was intact in Mincle^-/- neutrophils, we next analyzed the activation of the autophagy pathway in WT and Mincle^-/- neutrophils. Conversion of nascent autophagy protein LC3, also called ATG8 (LC3-I), into aggregates (LC3-PE complex; ie, LC3-II) serves as a marker of autophagy activation and autophagosome formation [30]. Immunofluorescence staining of unstimulated WT and Mincle^-/- neutrophils showed a homogenous distribution of LC3 in the cytoplasm (Figure 3A). Stimulation with PMA resulted in formation of autophagosomes in WT neutrophils, as indicated by extensive punctation of LC3-II (Figure 3A). Stimulated Mincle^-/- neutrophils on the other hand retained the uniform distribution of LC3, which, although increased in intensity upon PMA stimulation, failed to display the formation of LC3-II puncta (Figure 3A). Western blot analysis further confirmed attenuated activation of autophagy in the absence of Mincle, where Mincle^-/- neutrophils showed reduced amount of processed LC3-II upon PMA stimulation compared with their WT counterparts (Figure 3B). This defective autophagy activation was not due to a lack of activation of Syk/CARD-9 and MAP kinase pathways, canonical signaling pathways downstream of Mincle [17], and autophagy activation [31, 32] because CARD9 knockdown had no effect on NET formation and no impairment of p38 and ERK phosphorylation was observed in Mincle^-/- neutrophils (Supplementary Figure 2). Treatment with a macrolide rapamycin is known to initiate autophagy by inhibition of mTOR, a negative regulator of autophagy [33]. As shown in Figure 3B, rapamycin treatment only marginally increased the PMA-induced NET formation in WT neutrophils, which was not statistically significant compared with PMA stimulation alone. The attenuated LC3 processing in Mincle^-/- neutrophils stimulated with PMA remained unaltered by rapamycin treatment (Figure 3B). Impaired autophagy with or without rapamycin treatment correlated with a significantly reduced NET formation in Mincle^-/- neutrophils (Figure 3C). These results show that defective NET formation in Mincle^-/- neutrophils is associated with impaired activation of autophagy pathway.

Figure 3.

Mincle-deficient neutrophils exhibit impaired autophagy activation. A, Representative immunofluorescence images showing the extent of LC3 punctation in wild-type (WT) and Mincle^-/- neutrophils 30 minutes after phorbol-myristate-acetate (PMA) stimulation. LC3 puncta (red) indicating activation of autophagy were visualized using a rabbit antimouse LC3 antibody followed by secondary goat antirabbit Alexa546 antibody. DAPI (blue) was used to stain the nuclei. Arrows indicate punctation of LC3-II in WT neutrophils after PMA stimulation. B, Western blot analysis to detect conversion of LC3-I to LC3-II in WT and Mincle^-/- neutrophils stimulated with PMA with or without rapamycin treatment. Immunoblotting of housekeeping protein GAPDH is shown as loading control. Blot shown is representative of 4 independent experiments. C, Quantitation of neutrophil extracellular trap (NET) formation by WT and Mincle^-/- neutrophils stimulated with PMA for 4 hours with or without rapamycin treatment. Average ± SD from 5 independent experiments is shown. *P < .05; ***P < .001. Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; NS, not stimulated; P, phorbol-myristate-acetate alone; PMA, phorbol-myristate-acetate; R, rapamycin alone; Rapa, rapamycin; RP, rapamycin+PMA; WT, wild-type.

Rescue of NET Formation in Mincle^-/- Neutrophils by Activating Autophagy

Tamoxifen, a selective estrogen receptor modulator, has been shown to induce autophagy [34, 35]. We tested whether induction of autophagy by tamoxifen treatment could reverse the NET formation defect. As shown in Figure 4A and 4B, treatment with this drug rescued the NET formation in Mincle^-/- neutrophils upon PMA stimulation and induced NETs to the levels exhibited by WT neutrophils. To determine whether this tamoxifen-mediated rescue of NET formation in Mincle^-/- was due to autophagy activation, Western blot analysis was performed to assess LC3 processing. Indeed, the amount of LC3-II was increased in PMA-stimulated Mincle^-/- neutrophils upon tamoxifen pretreatment, illustrating activation of autophagy in these cells (Figure 4C). These data show that autophagy is critical for Mincle-mediated NET formation, and activation of autophagy rescues the NET formation defect in the absence of Mincle. The rescue of NET formation by tamoxifen, but not rapamycin, further suggests that the defect in autophagy pathway in the absence of Mincle is downstream of mTOR, a target of rapamycin.

Figure 4.

Tamoxifen treatment rescues autophagy and neutrophil extracellular trap (NET) formation in Mincle^-/- neutrophils. A, Representative fluorescence images of wild-type (WT) and Mincle^-/- neutrophils unstimulated or stimulated with phorbol-myristate-acetate (PMA) with or without tamoxifen (TMX). Neutrophil extracellular traps were fixed and stained with Sytox Green as described in Methods. Magnification = 200X. B, The bar graph shows quantitation of NET-forming WT and Mincle^-/- neutrophils stimulated with PMA with or without TMX. Data presented are from 3 independent experiments. Average ± SD is shown. *P < .05; **P < .01; ***P < .001. C, Western blot analysis to detect conversion of LC3-I to LC3-II in WT and Mincle^-/- neutrophils stimulated with PMA with or without TMX treatment. Immunoblotting of housekeeping protein GAPDH is shown as loading control. Blot shown is representative of 3 independent experiments. Abbreviations: NS, not stimulated; P, phorbol-myristate-acetate alone; PMA, phorbol-myristate-acetate; TP, tamoxifen+PMA; TMX, tamoxifen; WT, wild-type.

Mincle-Mediated Neutrophil Extracellular Trap Formation and Autophagy in an In Vivo Model of Bacterial Pneumonia

We next examined whether our findings of Mincle-mediated NET formation via autophagy could be recapitulated in vivo in a pneumonic infection with KPn. We have used a murine intranasal infection model of KPn to study pathogenesis of pneumonic sepsis [24, 25, 36]. As reported previously, in comparison with the KPn-infected WT mice, significantly reduced NET formation was observed in the lungs of Mincle^-/- mice infected with KPn (Figure 5A). This impairment of in vivo NET formation was not due to a defective ROS production, as the neutrophils isolated from BALF of KPn-infected Mincle^-/- mice exhibited similar levels of ROS as their WT counterparts (Figure 5B). We next examined the autophagy activation measured by LC3 processing and level of Beclin-1, another autophagy-related protein [37]. Neutrophils isolated from BALF of KPn-infected WT mice exhibited punctate structures of LC3-II, indicating autophagosome formation (Figure 5C). Mincle^-/- neutrophils on the other hand displayed a uniform cytosolic distribution of this protein by immunofluorescence staining. Concomitantly, Western blot analysis showed reduced LC3-II and Beclin-1 levels in Mincle^-/- neutrophils in comparison with the WT neutrophils isolated from BALF after KPn infection (Figure 5D). Taken together, these results showed that Mincle is required for NET formation in the lungs of mice during KPn pneumonia, which correlates with its regulation of autophagy, but not ROS formation in vivo during pneumonic KPn infection.

Figure 5.

Mincle^-/- neutrophils show reduced neutrophil extracellular trap (NET) formation and defective autophagy but normal reactive oxygen species (ROS) in the lungs of Klebsiella pneumoniae (KPn)–infected pneumonic mice. A, Representative fluorescence images of the neutrophils isolated 3 days after infection from BAL fluid of wild-type (WT) and Mincle^-/- mice infected with KPn, and stained with Sytox Green to label NETs (green). Magnification = 200X. The bar graph shows quantitation of NET-forming neutrophils from 3 independent experiments with 3–4 mice per group in each experiment. B, Measurement of ROS in BAL neutrophils isolated 3 days after infection from WT and Mincle^-/- mice infected intranasally with KPn. Data from 4 independent experiments with 3–4 mice per group in each experiment are shown. No statistically significant differences were found between the levels of ROS in neutrophils from WT and Mincle^-/- mice. C, Representative immunofluorescence images showing the extent of LC3 punctation in BAL neutrophils isolated from KPn-infected WT and Mincle^-/- mice 3 days after infection. LC3 puncta (red) indicating activation of autophagy were visualized using a rabbit antimouse LC3 antibody followed by secondary goat antirabbit Alexa546 antibody. DAPI (blue) was used to stain the nuclei. Arrows indicate punctation of LC3 in WT neutrophils. Magnification = 400X. D, Western blot analysis to detect conversion of LC3-I to LC3-II and the levels of autophagy protein Beclin-1 in BAL neutrophils isolated from KPn-infected WT and Mincle^-/- mice 3 days after infection. Immunoblotting of housekeeping protein GAPDH is shown as loading control. Blots shown are representative of 4 independent experiments with 2–3 mice per group in each experiment. Abbreviations: DAPI, 4′,6-diamidino-2-phenylindole; ROS, reactive oxygen species; WT, wild-type.

DISCUSSION

In this study we identify Mincle as a novel regulator of NET formation via modulation of autophagy. We show that Mincle controls NET formation in response to neutrophil activation stimuli in vitro and bacterial infection in vivo. Mincle-deficient neutrophils display defective NET formation despite being fully competent in ROS production. Instead, Mincle deficiency causes impairment of autophagy activation, and induction of autophagy with tamoxifen treatment in Mincle^-/- neutrophils reversed the NET formation defect in these cells. Collectively, our results show Mincle as a novel regulator of NET formation in vitro and in vivo by activating autophagy downstream of mTOR (target of rapamycin) and that this function of Mincle is independent of ROS generation pathway (working model shown in Figure 6).

Figure 6.

Working model of Mincle-mediated neutrophil extracellular trap (NET) formation. Mincle-mediated activation of autophagy, occurring downstream of mTOR (target of rapamycin), is required for NET formation in response to pneumonic infection as well as phorbol-myristate-acetate stimulation. This autophagy activation precedes reactive oxygen species (ROS) effect on NET formation, because Mincle deficiency causes NET formation defect despite intact ROS. Tamoxifen reverses this defect in autophagy activation in the absence of Mincle and rescues NET formation in Mincle^-/- neutrophils. Abbreviations: mTOR, target of rapamycin; NET, neutrophil extracellular trap; PMA, phorbol-myristate-acetate; ROS, reactive oxygen species.

The assembly of a multiprotein NADPH oxidase complex on plasma membrane of activated neutrophils generates ROS in the cytosol and triggers NET release [10]. The supportive evidence for the role of ROS in NET formation comes from studies on patients with chronic granulomatous disease, who lack NADPH oxidase activity and thus fail to form NETs [13]. However, a pleotropic role of NADPH complex in diverse cellular functions makes it difficult to tease out the role of ROS in NET formation independently of its other immune functions. Further, some reports of impaired NETs in spite of high ROS as well as ROS-independent NET formation indicate the possible involvement of additional mechanisms to induce NETs [12, 14]. Of import, Remijsen et al proposed that, in addition to ROS, autophagy activation is required for NET formation [16]. In our studies Mincle^-/- neutrophils exhibit intact ROS generation machinery, as evident by similar ROS levels in WT and Mincle^-/- neutrophils. Additionally, we did not find any modulation in mitochondrial ROS in response to PMA stimulation of WT or Mincle^-/- neutrophils. Instead, a defect in autophagy was observed in Mincle^-/- neutrophils. This suggests that activation of autophagy downstream of Mincle is likely segregated from ROS generation capacity of neutrophils. Interestingly, treatment of WT neutrophils with apocyanin, an NADPH oxidase inhibitor, mirrored the effect of Mincle deficiency by reducing the NET formation to levels observed in Mincle^-/- neutrophils (Figure 2C). This treatment did not exacerbate the impairment of NETs in Mincle^-/- neutrophils. This suggests that, although Mincle is not involved in the ROS pathway, autophagy and ROS likely act in concert, and a defect in either of these pathways attenuates NET formation. Furthermore, ROS possibly acts downstream of autophagy activation because Mincle^-/- neutrophils are defective in NET formation despite the production of ROS, due to impaired autophagy. Although our study does not rule out the possibility that Mincle might be functioning at the intersection of ROS cross-talk with autophagy, our results clearly demonstrate a pivotal role of Mincle in activation of autophagy independent of the ROS generation pathway. This provides an opportunity to dissect the molecular events underlying these 2 pathways independently of each other.

Autophagy, a physiological process of self-conservation, is initiated upon inhibition of mammalian target of rapamycin (mTOR), a negative regulator of autophagy [33]. This is followed by nucleation, which involves the Beclin-1 (Atg-6) complex, and auto-phagosome formation is completed by the ligation of LC3 (Atg-8) to phosphatidyl-ethanolamine, resulting in autophagy hallmark aggregates of LC3-II [38]. Mincle-deficient neutrophils exhibited an impaired activation of autophagy in response to PMA stimulation, as indicated by reduced levels of LC3-II. This defect in autophagy and NET formation was reversed by treatment with tamoxifen, but not rapamycin, which activates autophagy at the initiation step by inhibiting mTOR [30]. Although the precise molecular target of tamoxifen in the autophagy pathway is not defined, it has been shown to increase intracellular ceramide levels [39], which in turn, regulate autophagy at multiple levels [40]. Our observation of tamoxifen-mediated rescue of NET formation in Mincle^-/- neutrophils correlating with increased processing of LC3, which could not be accomplished by rapamycin treatment, suggests that Mincle and tamoxifen regulate autophagy at the level of autophagosome formation. This is further supported by our results in a physiologically relevant setting of pneumonic KPn infection where the impact of Mincle deficiency on LC3 processing was found to be much more pronounced than its effect on Beclin-1. Defective LC3 processing in the absence of Mincle indicates that this receptor controls a critical point in autophagosome formation during pneumonic KPn infection, which is crucial for NET formation. Interestingly, in a recent elegant study, tamoxifen was shown to induce NET formation as an off-target effect, although autophagy activation was not assessed in this study [41]. Our studies posit Mincle mediated autophagy and NET formation, as well as tamoxifen, as attractive targets to correct autophagy and NET formation defects in a variety of disease conditions.

Information on Mincle signaling in neutrophils and its role in neutrophil-mediated responses is extremely limited. In contrast with the reports in macrophages (reviewed in [42]), we observed an increased phosphorylated Syk and CARD-9 in Mincle^-/- neutrophils stimulated with PMA (Supplementary Figure 2). Further, CARD-9 knockdown did not affect the NET formation in WT neutrophils upon PMA stimulation. Similarly, Mincle deficiency did not affect phosphorylation of p38 or ERK in PMA-stimulated neutrophils. Because Syk and MAP kinase are reported to regulate ROS generation [43–45], in light of our observation that ROS generation is not defective in Mincle^-/- neutrophils, it is likely that Syk and MAP kinase activation is involved in PMA-induced ROS generation and that signals distinct from ROS production are required to overcome the defect in NET formation consequent to Mincle deficiency. An impaired NET formation despite the presence of ROS in Mincle^-/- neutrophils strongly supports a pivotal role of Mincle-mediated autophagy in driving the NET formation.

To summarize, NETs play a protective role in many infectious diseases, including Klebsiella infection [3, 24, 46]. However, exuberant NET formation can lead to unwanted inflammation [6]. Our results showing Mincle as an important regulator of NET formation by controlling autophagy are expected to identify novel targets and opportunities to modulate/fine-tune the process of NET formation according to its pathophysiologic relevance. Normal ROS generation but impaired autophagy in Mincle^-/- neutrophils provide insights into the activation and interplay between these 2 pathways in context of NETs. Because modulation of NET formation by pharmacological inhibition of NADPH oxidase has shown unintended consequences [47] due to intricate involvement of ROS in neutrophil-mediated protective functions, our studies posit Mincle as a tangible therapeutic target for selectively modulating NET formation without compromising ROS generation.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Financial support. This work was supported by grants from the National Institues of Health (NIH; R01AI121804-01 to J. S.; R21DE024300-02 to B. B. M.; 1P20GM113123-01 to B. B. M. and J. S. as project leaders). The Flow Cytometry Core Facility at the University of North Dakota (UND) is supported by INBRE and COBRE grants (P20GM103442 and P20GM113123), and the Imaging Core Facility at UND is supported by a COBRE grant (5P30GM103329) from the NIH. A. S. is partially supported by a postdoctoral salary support grant to J. S. from the office of Vice-President of Research at UND.

Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.