Complement C5a Receptor 1 Exacerbates the Pathophysiology of N. meningitidis Sepsis and Is a Potential Target for Disease Treatment

Abstract

Sepsis caused by Neisseria meningitidis (meningococcus) is a rapidly progressing, life-threatening disease. Because its initial symptoms are rather unspecific, medical attention is often sought too late, i.e., when the systemic inflammatory response is already unleashed. This in turn limits the success of antibiotic treatment. The complement system is generally accepted as the most important innate immune determinant against invasive meningococcal disease since it protects the host through the bactericidal membrane attack complex. However, complement activation concomitantly liberates the C5a peptide, and it remains unclear whether this potent anaphylatoxin contributes to protection and/or drives the rapidly progressing immunopathogenesis associated with meningococcal disease. Here, we dissected the specific contribution of C5a receptor 1 (C5aR1), the canonical receptor for C5a, using a mouse model of meningococcal sepsis. Mice lacking C3 or C5 displayed susceptibility that was enhanced by >1,000-fold or 100-fold, respectively, consistent with the contribution of these components to protection. In clear contrast, C5ar1-/- mice resisted invasive meningococcal infection and cleared N. meningitidis more rapidly than wild-type (WT) animals. This favorable outcome stemmed from an ameliorated inflammatory cytokine response to N. meningitidis in C5ar1-/- mice in both in vivo and ex vivo whole-blood infections. In addition, inhibition of C5aR1 signaling without interference with the complement bactericidal activity reduced the inflammatory response also in human whole blood. Enticingly, pharmacologic C5aR1 blockade enhanced mouse survival and lowered meningococcal burden even when the treatment was administered after sepsis induction. Together, our findings demonstrate that C5aR1 drives the pathophysiology associated with meningococcal sepsis and provides a promising target for adjunctive therapy.IMPORTANCE The devastating consequences of N. meningitidis sepsis arise due to the rapidly arising and self-propagating inflammatory response that mobilizes antibacterial defenses but also drives the immunopathology associated with meningococcemia. The complement cascade provides innate broad-spectrum protection against infection by directly damaging the envelope of pathogenic microbes through the membrane attack complex and triggers an inflammatory response via the C5a peptide and its receptor C5aR1 aimed at mobilizing cellular effectors of immunity. Here, we consider the potential of separating the bactericidal activities of the complement cascade from its immune activating function to improve outcome of N. meningitidis sepsis. Our findings demonstrate that the specific genetic or pharmacological disruption of C5aR1 rapidly ameliorates disease by suppressing the pathogenic inflammatory response and, surprisingly, allows faster clearance of the bacterial infection. This outcome provides a clear demonstration of the therapeutic benefit of the use of C5aR1-specific inhibitors to improve the outcome of invasive meningococcal disease.

Keywords: C5aR1; Neisseria meningitidis; anaphylatoxins; complement system; inflammation; invasive disease; mouse model; neutrophils; sepsis; whole-blood model.

Copyright © 2018 Herrmann et al.

Figures

FIG 1

Role of complement in murine model of N. meningitidis sepsis. (A) Survival curves (top panels) and bacteremia (lower panels) of WT, C3−/−, and Hc°/° mice intraperitoneally infected with N. meningitidis strain MC58 inocula as indicated above the graphs. LOD, limit of detection; ns, not significant; *, P < 0.05; **, P < 0.01 (in Mantel-Cox test). (B) In vivo C3b deposition onto N. meningitidis (Nme) in infected mice. Blood smears of the infected mice described in the panel A legend were analyzed by immunofluorescence microscopy at ×60 magnification after staining with DAPI, rabbit anti-N. meningitidis (Alexa 488 channel), and anti-C3 (Cy3 channel). About 500 N. meningitidis cells per sample were identified by green fluorescence, and the fraction of C3-positive N. meningitidis cells was expressed as the percentage of all analyzed N. meningitidis cells. Each dot represents results from one animal. Representative immunofluorescence (IF) microscopy images are shown in Fig. S3. ns, not significant; *, P < 0.05 (in one-way analysis of variance [ANOVA] with Dunnett’s post hoc test). (C) Ex vivo C3b deposition from mouse lepirudin plasma on N. meningitidis MC58 as analyzed by whole-cell ELISA. ns, not significant; ***, P < 0.005 (in one-way ANOVA with Dunnett’s post hoc test). (D) Ex vivo uptake of N. meningitidis by neutrophils of WT, C3−/−, and Hc°/° mice. Lepirudinized whole-mouse blood was infected with 107 CFU/ml of acapsulate N. meningitidis expressing GFP (MC58Δcsb-GFP), and the mean fluorescence intensity (MFI) of neutrophils (PMN; gated as Ly6Ghi) was analyzed by flow cytometry. The graph shows means ± standard deviations of the means (SEM) of results from three independent experiments. *, P < 0.05 (in one-way ANOVA with Bonferroni’s post hoc test). ns, not significant.

FIG 2

Anaphylatoxin release during N. meningitidis sepsis. (A and C) Plasma levels of C3a and C5a, respectively, in WT mice at 12 h postinfection with 105 CFU of N. meningitidis strain MC58 as measured by ELISA. Plotted are means ± SEM. *, P < 0.05; **, P < 0.01 (in unpaired, two-tailed Student’s t test). (B and D) Correlation of C3a and C5a concentrations, respectively, with bacterial burden of infected mice. (E and F) Dose response of plasma C5a liberation in ex vivo infection of lepirudin anticoagulated whole-mouse blood (n = 3 independent samples) and whole human blood (n = 15 individual donors), respectively, with N. meningitidis MC58 (means ± SEM). *, P < 0.05; ***, <0.001; ****, <0.0001 (in one-way ANOVA with Dunnett’s post hoc test using PBS as a comparator).

FIG 3

In vivo N. meningitidis sepsis in WT versus C5ar1−/− mice. (A) Surface expression of C5aR1 as measured by flow cytometry on neutrophils (gated as Ly6Ghi) in WT mouse lepirudin-treated whole blood after 1 h of infection with 107 CFU of N. meningitidis MC58 versus uninfected control (ctrl.) and C5ar1−/− neutrophils as a staining control. Plotted is the mean fluorescence intensity (MFI) of results from three independent experiments. *, P < 0.05 (in paired matched observations per mouse; two-tailed Student’s t test). (B) Survival curves of WT and C5aR1−/− mice after intraperitoneal infection with 105 CFU of N. meningitidis MC58. ****, P < 0.0001 (in Mantel-Cox test). (C) N. meningitidis counts in blood of infected mice at indicated time points. The 18/24 h data comprise 18-h values from mice not surviving until 24 h plus 24-h values from the mice surviving until then. ns, not significant; **, P < 0.001; ***, P < 0.0001 (in unpaired, two-tailed Mann-Whitney test). (D) Plasma levels of inflammatory mediators at 12 h after intraperitoneal infection with 105 CFU N. meningitidis MC58 of WT versus C5ar1−/− mice (means ± SEM; n = 5 per genotype). ns, not significant; **, P < 0.01 (in unpaired, two-tailed Student’s t test).

FIG 4

Role of phagocytes in N. meningitidis sepsis in WT and C5ar1−/− mice. (A) Immunofluorescence microscopy (×60 magnification) of tissue sections of lung at indicated time points after intraperitoneal infection with 105 CFU N. meningitidis MC58. Blue, nuclei; red, neutrophil elastase; green, N. meningitidis (arrowheads). Background data in the green channel stem from erythrocytes, indicating the position of blood vessels. Insets are enlargements of points of interest from the same image. (B) Oxidative burst of Ly6Ghi neutrophils (n = 3) assayed by DHR123 assay in lepirudin-treated whole blood infected with 105 or 107 CFU per ml of N. meningitidis MC58. The positive control was PMA (100 nM). ns, not significant; **, P < 0.01 (in unpaired, two-tailed Student’s t test). (C) Neutrophil degranulation measured as the difference in levels of CD11b surface expression between infected (107 CFU/ml) and uninfected lepirudin-treated whole-blood samples from WT and C5ar1−/− mice after 1 h of incubation. Neutrophils were gated as Ly6Ghi cells and CD11b stained with clone M1/70. *, P < 0.05 (in unpaired, two-tailed Student’s t test). (D) Uptake of N. meningitidis MC58Δcsb-GFP by Ly6Ghi neutrophils in ex vivo infection of lepirudin-treated whole blood (means ± SEM of the geometric mean of GFP fluorescence; n = 3). ns, not significant, *, P < 0.05 (in one-way ANOVA with Bonferroni’s post hoc test). (E) Ex vivo N. meningitidis survival at different inocula in lepirudin-treated whole blood of WT and C5ar1−/− mice (means of CFU per milliliter ± SEM; n = 3). As a positive control for N. meningitidis killing, 1 µg/ml of anti-serogroup B mouse monoclonal antibody mAb735 was added. (F and H) Survival of n = 7 to 8 WT and C5ar1−/− mice, respectively, infected with 104 CFU of N. meningitidis MC58 after depletion of monocytes/macrophages (clodronate liposomes) or neutrophils (RB6-8C5) or the control (PBS). The experiment was conducted in a blind manner for depletion treatment. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 (in Mantel-Cox analysis relative to control). (G and I) N. meningitidis counts in blood of mice in panels G and I. ns, not significant; *, P < 0.05; **, P < 0.01; ****, P < 0.0001 (in one-way ANOVA, applying Bonferrroni’s post hoc test).

FIG 5

Cytokine induction in in vivo N. meningitidis infection of WT and C5ar1−/− mice and in whole-blood model. (A and B) Levels of IL-6 and CXCL-1 in plasma and peritoneal lavage fluid (PL), respectively, from n = 8 mice per genotype after intraperitoneal administration of 5 × 108 CFU of heat-inactivated N. meningitidis MC58. Plotted are means ± SEM. ns, not significant; *, P < 0.05; **, P < 0.01 (in unpaired, two-tailed Student’s t test). (C) Bacterial counts over time in ex vivo infection of lepirudin-treated mouse whole blood with 106 CFU/ml N. meningitidis. Plotted are means ± SEM of results from n = 3 independent experiments. (D) CXCL-1 concentrations (means ± SEM) in plasma of whole-blood infection from the experiment described for panel C. *, P < 0.05; **, P < 0.01 (in unpaired, two-tailed Student’s t test).

FIG 6

Inhibition of inflammatory cytokine release and neutrophil responses by C5aR1 blockade in lepirudin-anticoagulated human whole-blood infection with N. meningitidis. (A) IL-8 in human whole blood infected with 106 CFU/ml N. meningitidis MC58 in the presence of C5aR1 antagonist (C5aRAs) PMX53 or W-54011 (means ± SEM; n = 5 donors). (B) IL-8 in human whole blood infected for the indicated durations with 106 CFU/ml N. meningitidis MC58 in the presence of PMX53 (10 µM) or W-54011 (300 nM) (means ± SEM; n = 15 donors). (C) Cytokines at 90 min of infection of human whole blood with 106 CFU N. meningitidis MC58 (means ± SEM; n = 16 donors). (D) Oxidative burst measured by DHR123 fluorescence in neutrophils during infection of whole human blood. (E) Neutrophil degranulation during whole-blood infection by surface localization of CD11b normalized to “no C5aRA.” (F) Phagocytosis of MC58-GFP by PMNs in whole blood as determined by flow cytometry and expressed as percentages of PMNs with an increase in the level of FL1-H above the level measured for the noninfected control. (G) N. meningitidis viability in blood of donors with or without C5aRAs. (D to F) Infection with 107 CFU/ml of N. meningitidis MC58; lines indicate medians. (D to G) PMX53 was used at 10 µM and W-54011 at 300 nM. (A–G) *, P < 0.05; **, P < 0.01; ***, P < 0.001 (in repeated-measure ANOVA; matched observations per individual donor).

FIG 7

Pharmacologic targeting of C5aR1 ameliorates in vivo N. meningitidis sepsis. (A) Survival of WT mice after intraperitoneal infection with 105 CFU of N. meningitidis strain MC58. Mice were randomized into three treatment cohorts with intraperitoneal injections every 6 to 12 h (see Materials and Methods) with 3 mg/kg with C5aR1-antagonist PMX205 starting either 12 h before infection (“pre”; n = 8) or 4 h after infection (“post”; n = 6) or received vehicle only (“control”; n = 8). The experiment was conducted in a blind manner with respect to treatment cohorts. *, P < 0.05; ***, P < 0.001 (in Mantel-Cox test). (B) Bacteremia in the mice from the experiment described for panel A. ns, not significant; *, P < 0.05; **, P < 0.01 (in Kruskal-Wallis test with Dunn’s post hoc test). (C) Clinical scores for mice from the experiment described for panel A over time. Plotted are means ± SEM. (D) Relative body weights (means ± SEM). (E) Levels of CXCL-1, IL-6, and TNF-α in tail vein blood samples from mice from the experiment described for panel A at 12 h. ns, not significant; *, P < 0.05 (in one-way ANOVA, applying Dunnett’s post hoc test with the control as the comparator).