New Insights into Neutrophil Extracellular Trap (NETs) Formation from Porcine Neutrophils in Response to Bacterial Infections

Abstract

Actinobacillus pleuropneumoniae (A.pp, Gram negative) and Streptococcus (S.) suis (Gram positive) can cause severe diseases in pigs. During infection, neutrophils infiltrate to counteract these pathogens with phagocytosis and/or neutrophil extracellular traps (NETs). NETs consist of a DNA-backbone spiked with antimicrobial components. The NET formation mechanisms in porcine neutrophils as a response to both of the pathogens are not entirely clear. The aim of this study was to investigate whether A.pp (serotype 2, C3656/0271/11) and S. suis (serotype 2, strain 10) induce NETs by NADPH oxidase- or CD18-dependent mechanisms and to characterize phenotypes of NETs in porcine neutrophils. Therefore, we investigated NET induction in porcine neutrophils in the presence and absence of NET inhibitors and quantified NETs after 3 h. Furthermore, NETosis and phagocytosis were investigated by transmission electron microscopy after 30 min to characterize different phenotypes. A.pp and S. suis induce NETs that are mainly ROS-dependent. A.pp induces NETs that are partially CD18-dependent. Thirty minutes after infection, both of the pathogens induced a vesicular NET formation with only slight differences. Interestingly, some neutrophils showed only NET-marker positive phagolysosomes, but no NET-marker positive vesicles. Other neutrophils showed vesicular NETs and only NET-marker negative phagolysosomes. In conclusion, both of the pathogens induce ROS-dependent NETs. Vesicular NETosis and phagocytosis occur in parallel in porcine neutrophils in response to S. suis serotype 2 and A.pp serotype 2.

Keywords: Actinobacillus pleuropneumoniae (A.pp); Streptococcus (S.) suis; neutrophil extracellular traps (NETs); neutrophils; pigs; vesicular NETs.

Figure A1

Representative immunofluorescence images (overlay) of neutrophils treated for 3 h with different MOIs of A.pp (MOI = 0.5, 1, and 2). For each treatment two out of six images are presented. Comparable NET release was observed in all samples. Settings of the microscope were adjusted based on an isotype control (blue = DNA, DNA/histone-1-complexes = green, myeloperoxidase = MPO).

Figure A2

Representative immunofluorescence images (overlay) of NET induction assays with and without 2% FCS are presented. RPMI was used as unstimulated control (CTR). Staining: Blue = DNA; green = DNA/histone-1-complexes; red = myeloperoxidase. White arrows mark clumping neutrophils. All settings were adjusted to respective isotype controls.

Figure A3

Detection of cell-free DNA in the NET inhibition assay with A.pp after 1 h incubation (A) and after 3 h (B). (A) After 1 h incubation, neutrophils released in the absence of inhibitors the highest amount of cell-free DNA after A.pp infection. DPI inhibited the release of cell-free DNA after 1 h in A.pp infected neutrophils; (B) After a 3 h incubation period A.pp- and CD-stimulated neutrophils released significantly more cell-free DNA. The release of cell-free DNA from A.pp-infected neutrophils was significantly reduced after preincubation with anti-CD18a and DPI. All data were analyzed with one-tailed paired Student’s t-test (* p < 0.05, ** p < 0.01) and are presented with mean ± SD (1h incubation n = 3, 3 h incubation n = 5).

Figure A4

Representative immunofluorescence images (overlay) of NET induction assays after incubation with CD (positive control). Staining: blue = DNA; green = DNA/histone-1-complexes; red = myeloperoxidase. (A) Representative overview of the positions where the images of the used counting method and the alternative counting method were taken. (B) Six pictures were taken on two slides at predefined positions (1–6); (C) Six pictures were taken on one slide per row at the center of the slide (1–6). All settings were adjusted to respective isotype controls. In all six images, the total amount of cells was counted using ImageJ cell counter. Each counted cell is marked with a yellow cross and a number. The total number per image as well as the mean of six images are presented. Both counting methods reflect a comparable number of cells in the six pictures. In this study, all samples were analyzed with the method presented in (A).

Figure A5

Representative immunofluorescence image (overlay) of anti-CD18a attachment at the surface of neutrophils. Settings of the microscope were adjusted based on the presented isotype control (scale bar = 10 µm; blue = DNA, myeloperoxidase = MPO).

Figure 1

(A) Representative immunofluorescence images (overlay) of NET inhibition assays used to quantify activated cells are presented. In each experiment and for each sample, six randomly taken pictures from two individual slides were analyzed for quantification. All cells on the six pictures were counted and the mean of activated cells per stimulus was calculated and used for statistics. RPMI was used as unstimulated control. Methyl-β-cyclodextrin (CD) and phorbol 12-myristate-13 acetate (PMA) diluted in RPMI were used as positive controls. DPI and CD18a antibodies were used as inhibitors of ROS-dependent and independent NET release, respectively. Staining: Blue = DNA; green = DNA/histone-1-complexes; red = myeloperoxidase (except A.pp treated samples); scale bar = 100 µm. All settings were adjusted to respective isotype controls; (B) Statistical analysis of NET inhibition assay (3 h stimulation) sorted by stimulus. NET release by A.pp is significantly blocked by antiCD18a and DPI. NET release by S. suis is significantly blocked by DPI and significantly increased by antiCD18a co-incubated with S. suis. The data are presented with mean ± SD and were analyzed with one-tailed Mann–Whitney test. Per sample, six pictures were randomly taken on two slides at predefined positions (for detailed information see Figure A4, Appendix A) and the number of NET-activated cells were determined. Unstimulated, CD, PMA: n = 42 pictures from seven independent experiments; A.pp: n = 18 pictures from three independent experiments; S. suis n = 24 pictures from four independent experiments) (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001).

Figure 2

(A–D) Representative TEM images derived from one experimental run of porcine neutrophils after 30 min of A.pp infection; (A) Overview picture, (area 1) neutrophil containing NE and H3cit in the cytoplasm; (area 2) neutrophil containing NE and H3cit in vesicles (white arrows); (area 3) neutrophil releasing NE and H3cit positive vesicles in the extracellular space (white arrows). The hash marks the zoom area; (B) Neutrophil “type separated = type S”: gold-labeling of two serial sections identified neutrophils with phagocytosed A.pp (serial Section 1, left panel, white stars mark A.pp) and clearly outside all bacteria-containing phagosomes separated NE/H3cit positive vesicles (right panel, white arrows mark vesicles); (C) Neutrophil “type merged = type M”: gold-labeling of two serial sections identified neutrophils with phagocytosed A.pp (left panel, white stars mark A.pp) and these phagosome are in addition NE/H3cit positive (right panel); (D) The hash marks in the upper panel the zoom area. Neutrophils after A.pp infection showing nuclear membrane blebbing (NE/H3cit positive gold-labeled, white arrow heads) and vesicles (NE/H3cit, white arrows). Furthermore, NE positive granule and a NE/H3cit positive nucleus were identified; (A–C) and (serial Section 1, right panels with zoom pictures) and D lower panel: 5 nm gold labeling = H3-cit and 10 nm gold labeling = NE; (B,C) (serial Section 2, left panels with zoom pictures) 10 nm gold labeling = A.pp. Scale bars in A: overview = 10 µm; zoom pictures = 2.5 µm; zoom and gold-labeling = 200 nm; (B–D) 2 µm (upper panel) and 250 nm (lower panels).

Figure 3

(A–D) Representative TEM images taken during one experimental run of porcine neutrophils after 30 min of S. suis infection; (A) Overview picture, (area 1) neutrophil containing NE and H3cit in the cytoplasm; (area 2) neutrophil containing NE and H3cit in vesicles (white arrows); (area 3) neutrophil releasing NE and H3cit positive vesicles in the extracellular space (white arrows); (B) Neutrophil “type S”: gold-labeling of two serial sections identified neutrophils with phagocytosed S. suis (serial Section 1, left panel, white stars mark S. suis) and clearly outside all bacteria-containing phagosomes separated NE/H3cit positive vesicles (serial Section 2, right panel, white arrows mark vesicles); (C) Neutrophil “type M”: gold-labeling of two serial sections identified neutrophils with phagocytosed S. suis (serial Section 1, left panel, white stars mark A.pp) and these phagosome are in addition NE/H3cit positive (serial Section 2, right panel); (D) The hash marks in the upper panel the zoom area. Neutrophils after S. suis infection showing nuclear membrane blebbing (NE/H3cit positive gold-labeled, white arrow heads) and vesicles (NE/H3cit, white arrows). Furthermore, NE positive granule and a NE/H3cit positive nucleus were identified; (A–C) (right panels with zoom pictures) and D lower panel: 5 nm gold labeling = H3-cit and 10 nm gold labeling = NE; (B,C) (left panels with zoom pictures) 10 nm gold labeling = S. suis. Scale bars in A: overview = 10 µm; zoom pictures = 2.5 µm; zoom and gold-labeling = 200 nm; (B–D) 2 µm (upper panel) and 250 nm (lower panels).

Figure 4

(A) Representative TEM images of porcine neutrophils after a 30 min incubation period that were used for quantification (taken from one experimental run). The arrowheads mark vesicles within the neutrophils. The scale bars are: 2 µm (left) and 1 µm (right); (B–D) Statistical analysis of the TEM images: In total, 30 cellular profiles from randomly selected fields on the thin sections were analyzed per sample. Neutrophils stimulated with A.pp and S. suis showed a significantly higher NET release from the nuclei (“suicidal NETosis”) (B); a significantly higher NET surface coverage (C); and a significantly higher number of nuclear vesicles (“vital NETosis”) in neutrophils, with the highest amount in S. suis infected neutrophils (D). Data were analyzed with the Kruskal–Wallis multiple comparison test (n = 30) and presented with mean ± SD (* p ≤ 0.05, **** p < 0.0001).