Glycolipid Metabolite β-Glucosylceramide Is a Neutrophil Extracellular Trap-Inducing Ligand of Mincle Released during Bacterial Infection and Inflammation

Abstract

Neutrophil extracellular traps (NETs) are implicated in host defense and inflammatory pathologies alike. A wide range of pathogen- and host-derived factors are known to induce NETs, yet the knowledge about specific receptor-ligand interactions in this response is limited. We previously reported that macrophage-inducible C-type lectin (Mincle) regulates NET formation. In this article, we identify glycosphingolipid β-glucosylceramide (β-GlcCer) as a specific NET-inducing ligand of Mincle. We found that purified β-GlcCer induced NETs in mouse primary neutrophils in vitro and in vivo, and this effect was abrogated in Mincle deficiency. Cell-free β-GlcCer accumulated in the lungs of pneumonic mice, which correlated with pulmonary NET formation in wild-type, but not in Mincle^-/-, mice infected intranasally with Klebsiella pneumoniae Although leukocyte infiltration by β-GlcCer administration in vivo did not require Mincle, NETs induced by this sphingolipid were important for bacterial clearance during Klebsiella infection. Mechanistically, β-GlcCer did not activate reactive oxygen species formation in neutrophils but required autophagy and glycolysis for NET formation, because ATG4 inhibitor NSC185058, as well as glycolysis inhibitor 2-deoxy-d-glucose, abrogated β-GlcCer-induced NETs. Forced autophagy activation by tamoxifen could overcome the inhibitory effect of glycolysis blockage on β-GlcCer-mediated NET formation, suggesting that autophagy activation is sufficient to induce NETs in response to this metabolite in the absence of glycolysis. Finally, β-GlcCer accumulated in the plasma of patients with systemic inflammatory response syndrome, and its levels correlated with the extent of systemic NET formation in these patients. Overall, our results posit β-GlcCer as a potent NET-inducing ligand of Mincle with diagnostic and therapeutic potential in inflammatory disease settings.

Figure 1.. β-GlcCer induces Mincle-dependent NET formation in-vitro.

A. Representative immunofluorescence images of murine peritoneal WT neutrophils unstimulated (NS) or stimulated with characterized Mincle ligands SAP130 (1μg/ml), mycobacterial TDM (5μg/ml ) or live KPn bacteria (MOI 10) for 4 hrs. Cells were fixed and stained with Sytox Green as described in Methods. Magnification 200X. The experiment was repeated three times with similar results. B. Representative images of NET formation in peritoneal neutrophils isolated from WT or Mincle^−/− mice. Peritoneal lavage neutrophils purified using magnetic column were stimulated in-vitro with β-GlcCer (30uM) or vehicle control (DMSO) for 4 hrs, cytocentrifuged on glass slides and stained with Sytox Green. Bar graph shows percentage of NET forming neutrophils ± SD in respective samples from 3 independent experiments. Student’s t test was used for statistical analysis (p**<0.01). C. Representative images of NETs in WT neutrophils transfected with control or siRNA targeted to Mincle 6 h prior to stimulation with β-GlcCer. NETs were stained with Sytox Green 4 h after β-GlcCer stimulation. Magnification 200X. Percentage of NET forming neutrophils ± SD from 3 independent experiments is shown in the bar graph. Student’s t test was used for statistical analysis (p*<0.05).

Figure 2.. Endogenous β-GlcCer accumulates in lungs of pneumoseptic mice and induces NETs in Mincle dependent fashion in-vivo.

A. Bar graph showing levels of cell-free β-GlcCer were measured in bronchoalveolar lavage (BAL) fluid harvested 3 dp.i. from WT or Mincle^−/− mice infected intranasally with K. pneumoniae. β-GlcCer levels were quantified by ELISA using rabbit anti-GlcCer antiserum and plotting the absorbance against standard curve of known concentrations of β-GlcCer using Gen 5 data analysis software. The bars represent average ± SD from 3 independent experiments. Student’s t test was used for statistical analysis (p**<0.01). No significant differences were found in BAL levels of β-GlcCer between KPn infected WT and Mincle^−/− mice. Lower panel shows representative fluorescence images depicting the extent of NET formation in neutrophils cytocentrifuged and Sytox-stained from corresponding BAL samples from uninfected or infected WT and Mincle^−/− mice. Magnification 200X. B. Bar graph shows percentage of NET forming neutrophils (average ± SD from 3 independent experiments) in peritoneal lavage isolated from WT or Mincle−/− mice injected intraperitoneally with β-GlcCer (300ug in 100ul oil-in-PBS : mineral oil 9%, Tween-80 1%, and PBS 90%). Vehicle control mice received the same volume of emulsion without β-GlcCer. Peritoneal lavage was isolated 12–14h after injection followed by NET quantitation as described in methods. Student’s t test was used for statistical analysis (p*<0.05; p**<0.01 ). Lower panel shows representative fluorescence images depicting NETs stained with Sytox Green (green) in indicated samples. Magnification 200X. C. Flow cytometry analysis of peritoneal lavage cells isolated from WT and Mincle^−/− mice 12 hrs after intraperitoneal injection of β-GlcCer as described in B. Cells were stained with fluorescent antibodies against indicated cell markers as described in Materials and Methods. Pseudocolor contour plots depict gated cells stained positive for each immune cell marker. Frequency of various cells types examined is shown in the bar graph. Data shown are the mean ± SD of four to six mice from two independent experiments. No statistically significant difference was found in the frequencies of various cell types in WT and Mincle^−/− mice in response to β-GlcCer administration. D. WT or Mincle^−/− mice were injected intraperitoneally with β-GlcCer followed 12 hrs later by DNAse I. After 2hrs mice were infected intraperitoneally with Kpn. Peritoneal lavage cells were collected 4 hrs after infection, cytocentrifuges and stained with Sytox Green to visualize NETs. Images shown are representatives of 2 independent experiments (6 mice per group). Magnification 200X. E. shows Bacterial burden enumerated in peritoneal lavage collected as described in (D). Data from one experiment (n=3 per group) representative of two independent experiments (6 mice total per group) with similar results is shown. Student’s t test was used for statistical analysis (p*<0.05; p**<0.01).

Figure 3.. β-GlcCer stimulation does not activate ROS formation or canonical Mincle signaling in neutrophils.

A. Reactive oxygen species were measured in purified peritoneal neutrophils from WT and Mincle^−/− neutrophils 10 min after in-vitro stimulation with indicated stimulants using a fluoro H₂O₂ detection kit as described in methods. Data from 3 independent experiments is shown. Student’s t test was used for statistical comparison between various stimulations within WT or Mincle^−/− groups (p***<0.001). No statistically significant differences were found between the levels of ROS in WT and Mincle^−/− neutrophils. B. Flow cytometry analysis of mitochondrial ROS (MitoSox) in vehicle (DMSO) treated or β-GlcCer stimulated peritoneal neutrophils from WT and Mincle^−/− mice. The numbers on contour plots represent percent MitoSox (MitoTracker) positive cells. Bar graph shows average ± SD of MitoSox positive cells from 3 independent experiments. No statistically significant differences were found between the levels of mitoROS in WT and Mincle^−/− neutrophils. C. Phosphorylation of Syk and CARD9 levels in WT and Mincle^−/− neutrophils 15min. after vehicle or β-GlcCer stimulation as determined by immunoblot analysis. Total Syk and GAPDH is shown as protein loading control. Experiment was repeated twice with similar results.

Figure 4.. β-GlcCer/Mincle mediated NET formation requires autophagy activation.

A. Representative immunofluorescence images showing the extent of LC3 punctation in peritoneal neutrophils isolated from WT and Mincle^−/− mice. Cells were stimulated for 90min. with vehicle control or β-GlcCer with or without pre-treatment with ATG4 inhibitor NSC185058 followed by visualization of. LC3 puncta (red) indicating activation of autophagy using a rabbit anti mouse LC3 antibody and secondary goat anti-rabbit Alexa546 antibody. DAPI (blue) was used to stain the nuclei. Magnification 400X. NET formation was analyzed in all samples parallelly by Sytox Green staining (Green). Magnification 200X. Representative images from 3 independent experiments are shown. B. Western blot analysis of autophagy activation as depicted by LC3 I processing to LC3 II in samples described in (A). The experiment was repeated twice with similar results (C). Quantitation of NETs in WT neutrophils stimulated in-vitro with β-GlcCer with or without with or without ATG4 inhibitor NSC185058. Bars depict percent NET forming neutrophils (mean ± SD) from 3 independent experiments. (**p<0.01).

Figure 5.. β-GlcCer mediated NET formation is blocked by glycolysis inhibition

A. NET formation quantified in WT neutrophils stimulated in-vitro with β-GlcCer alone (30μM for 4hr) or in combination with glycolysis inhibitor 2-DG (5mM, 60 min. pretreatment) with or without autophagy activator tamoxifen (6μM, 30 min. pretreatment). Bars depict percent NET forming neutrophils (Average ± SD) from 3 independent experiments. (*p<0.05); ***p<0.001). B. Representative fluorescence images of peritoneal lavage neutrophils isolated from WT and Mincle^−/− mice injected with purified β-GlcCer with or without peritoneal injection of 2-DG 24hrs before β-GlcCer. Neutrophils were cytocentrifuged, fixed and stained with Sytox Green as described in Methods. Magnification 200X. The bar graph shows quantitation of NET-forming in indicated samples as described in methods. Data presented is from 3 independent experiments with 3–4 mice each per group. Average ± SD is shown (***p<0.001)). C. Quantitation and representative fluorescence images of NET forming neutrophils isolated from BAL of WT mice infected intranasally with K. pneumoniae with or without intraperitoneal 2-DG treatment as described in methods. Bar graph depicts Average ± SD from 6 mice total per group in 2 independent experiments. (**p<0.01). D. Levels of cell-free β-GlcCer were measured in BAL samples described in C. No statistically significant difference was found in cell-free β-GlcCer levels in BAL isolated from KPn infected mice with or without 2-DG treatment. E. Representative immunofluorescence images showing LC3 punctation in WT neutrophils from experiments described in C. LC3 puncta (red) indicating activation of autophagy were visualized using a rabbit anti mouse LC3 antibody followed by secondary goat anti-rabbit Alexa546 antibody. DAPI (blue) was used to stain the nuclei. Representative images from 2 independent experiments are shown. Magnification 400X. Bottom panel shows western blot analysis to detect conversion of LC3-I to LC3-II in BAL neutrophils isolated from KPn infected WT mice with or without glycolysis inhibition by 2-DG treatment. β-actin was probed as loading control. Blots shown are representative of 2 independent experiments with 3–4 mice per group in each experiment.

Figure 6.. β-GlcCer accumulates in blood of patients with SIRS which correlates with NET formation.

Longitudinal blood samples from patients diagnosed with systemic inflammatory response syndrome were collected for up to 10 days following ICU admission. Levels of cell-free β-GlcCer (blue bars), percent NET forming neutrophils (green bars) and Mincle mRNA (purple bars) were measured in blood plasma as described in methods. Measurements of these parameters in plasma from healthy adults depicted baseline level. HA; healthy adult.