Bacterial exploitation of phosphorylcholine mimicry suppresses inflammation to promote airway infection

Abstract

Regulation of neutrophil activity is critical for immune evasion among extracellular pathogens, yet the mechanisms by which many bacteria disrupt phagocyte function remain unclear. Here, we have shown that the respiratory pathogen Streptococcus pneumoniae disables neutrophils by exploiting molecular mimicry to degrade platelet-activating factor (PAF), a host-derived inflammatory phospholipid. Using mass spectrometry and murine upper airway infection models, we demonstrated that phosphorylcholine (ChoP) moieties that are shared by PAF and the bacterial cell wall allow S. pneumoniae to leverage a ChoP-remodeling enzyme (Pce) to remove PAF from the airway. S. pneumoniae-mediated PAF deprivation impaired viability, activation, and bactericidal capacity among responding neutrophils. In the absence of Pce, neutrophils rapidly cleared S. pneumoniae from the airway and impeded invasive disease and transmission between mice. Abrogation of PAF signaling rendered Pce dispensable for S. pneumoniae persistence, reinforcing that this enzyme deprives neutrophils of essential PAF-mediated stimulation. Accordingly, exogenous activation of neutrophils overwhelmed Pce-mediated phagocyte disruption. Haemophilus influenzae also uses an enzyme, GlpQ, to hydrolyze ChoP and subvert PAF function, suggesting that mimicry-driven immune evasion is a common paradigm among respiratory pathogens. These results identify a mechanism by which shared molecular structures enable microbial enzymes to subvert host lipid signaling, suppress inflammation, and ensure bacterial persistence at the mucosa.

Figure 7. H. influenzae GlpQ hydrolyzes ChoP and contributes to evasion of PAF-mediated neutrophil defense of the airway.

(A) Diagram of pNPPC. Black arrow denotes the site of hydrolysis by S. pneumoniae Pce; white arrow indicates hydrolysis by H. influenzae GlpQ. (B) p-nitrophenol liberation after incubating pNPPC with WT (black) or mutant (gray) S. pneumoniae (Sp) or H. influenzae (Hi) (n = 5). (C) Survival of WT (black) or ΔglpQ (gray) H. influenzae during infection of the murine upper airway (n = 5–8 mice per group, LOD = 4). (D) Quantification of neutrophils obtained from mice inoculated with WT or ΔglpQ bacteria (n = 4). (E) CD11b and CD64 relative MFI on neutrophils elicited by WT or ΔglpQ bacteria, day 2 p.i. (n = 4). (F) Enumeration of day-2 WT or ΔglpQ H. influenzae CFU after treatment with α-Ly6G antibody or IgG2a isotype control (n = 5, LOD = 4). (G) WT and ΔglpQ bacterial loads from Ptafr^–/–mice, littermate controls, or neutropenic Ptafr^–/– mice on day 2 p.i. (n = 3–9, LOD = 10). (H) Bacterial killing assay of WT H. influenzae by 1,000:1 murine neutrophils pretreated with increasing concentrations of conditioned PAF media; PAF was preincubated with heat-killed WT (black) or ΔglpQ (white) bacteria before mixing with neutrophils for killing assays. Average values represent 3 biological replicates. Top dotted line: 100% bacterial survival; bottom dotted line: average survival in the absence of PAF. Statistical significance was assessed by Student’s t test for pairwise comparisons (B and H); 1-way ANOVA with Newman-Keuls post test for multigroup comparisons (C, D, F, and G); and 1-sample Student’s t test relative to null = 1 for relative MFI measurements (E). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 6. Exogenous stimulation of neutrophils in situ overwhelms Pce-mediated immune evasion.

(A) Enumeration of WT (black) or Δpce (gray) pneumococcal CFU obtained on day 7 p.i. and after daily (+1 to +6) i.n. treatments with 1 μg PAF or vehicle control (1% DMSO in PBS). The experiment was repeated after treatment with neutrophil-depleting α-Ly6G or IgG2a isotype control antibody (see Figure 4B) (n = 5–10 mice per condition). (B) Quantification of neutrophils elicited by WT (black) or Δpce (gray) P1121 pneumococci in nasal lavages of vehicle- or PAF-treated mice on day 7 p.i. (n = 5 per group). (C) Relative MFI of uptake receptor expression on luminal neutrophils elicited by WT or Δpce pneumococci (Δpce/WT) after daily treatment with vehicle (black) or 1 μg PAF (white), as described in A (n = 5). (D) Enumeration of bacterial CFU as in A, with and without daily i.n. treatment with 1 μg fMLP or vehicle (n = 5–7). (E) Quantification of elicited neutrophils as in B, with or without treatment with fMLP (n = 5). (F) Relative MFI of uptake receptor expression on neutrophils elicited by WT or Δpce pneumococci (Δpce/WT) after daily treatment with vehicle (black) or 1 μg fMLP (white), as in C (n = 5). Statistical significance was assessed by 1-way ANOVA with Newman-Keuls post test, except for C and F, wherein significance was assessed by 1-sample Student’s t test relative to a value of 1. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 5. Pce esterase hydrolyzes ChoP from conjugated substrates and directly inhibits PAF-mediated stimulation of neutrophil activation and function in vitro.

Killing of WT (black) or Δpce (gray) P1121 pneumococci in vitro by murine (A) or human (B) neutrophils at the indicated neutrophil/bacterium ratios, after preopsonization with BRS. Bacterial survival was measured relative to control assays in the absence of neutrophils (dotted line). HI, heat inactivated. (C) Kinetic time course of p-nitrophenol liberation (absorbance at 415 nm) after incubation of pNPPC with recombinant Pce enzyme (rPce, black line). Assays were repeated in the absence of Pce enzyme or pNPPC substrate or in the presence of 250 mM EDTA. K_obs, observed kinetic rate constant. (D) MFI quantification of bacterial uptake receptors CD11b and CD64 on murine neutrophils treated with PAF that was preincubated with rPce (black) or PBS (white). PAF stimulation assays were repeated with 10^–5 M of the PAFR antagonist PCA-4248 as a specificity control. Killing assays using WT pneumococci (PMN/bacterium ratio, 1,000:1) were performed after PAF-mediated murine (D) or human (E) neutrophil stimulation in the presence (black) or absence (white) of rPce. Top dotted line denotes 100% bacterial survival; bottom dotted line denotes average survival in the absence of PAF. For all panels, data averages reflect at least 3 independent experiments (3–4 independent biological replicates for cellular assays). *P < 0.05, **P < 0.01, and ***P < 0.001 by Student’s t test for all pairwise comparisons.

Figure 4. Pce is dispensable for pneumococcal persistence in the absence of infiltrating neutrophils or intact PAF signaling in the upper airway.

(A) Confirmation of neutrophil depletion. Mice were treated with neutrophil-depleting antibody (α-Ly6G, clone 1A8) or IgG2a isotype control (250 μg, i.p.) on days –1, +1, and +4 p.i. with PBS (white), WT (black), or Δpce (gray) P1121 pneumococci (n = 3–4 mice per condition). On day 7 p.i., depletion was confirmed by flow cytometric analysis of whole blood and nasal lavage. (B) Enumeration of WT (black) or Δpce (gray) pneumococcal CFU obtained from nasal lavages on day 7 p.i. and after treatment of mice (n = 5–11 mice per group) with neutrophil-depleting α-Ly6G or IgG2a isotype control antibodies. (C) Enumeration of bacterial CFU on day 7 p.i. and after daily i.n. treatment with 0.1 or 1 μg PAFR antagonist PCA-4248 (or 1% DMSO vehicle), from days +1 to +6 p.i. (n = 5–12). The experiment was repeated with neutrophil depletion as described in A. (D) WT and Δpce bacterial loads were enumerated in lavages obtained from Ptafr^–/– mice and their littermate controls on day 7 p.i. (n = 5–11). *P < 0.05, **P < 0.01, and ***P < 0.001 by 1-way ANOVA with Newman-Keuls post test.

Figure 3. Pce prevents accumulation of PAF in the lumen of the upper respiratory tract, and the absence of Pce stimulates transcription of genes important for PAF signaling.

(A) Diagram of PAF with the ChoP moiety labeled (and site of Pce-mediated hydrolysis marked with a black arrow). Detection of PAF levels in the upper airway lumen by LC-ESI/MS (limit of quantification = 0.066 nM, dashed line), quantified from pooled nasal lavages obtained from 5 mice at 3 days p.i. with PBS (Mock, white), WT (black), or Δpce (gray) P1121 pneumococci. Averages reflect 3 independent biological replicates of 5 pooled mice each. Statistical significance was assessed by 1-way ANOVA with Newman-Keuls post test. For representative LC traces from lavage fluid of mock-, WT-, and Δpce-infected mice, the top row displays PAF detected at 8.6 minutes’ retention; the bottom row displays ²H₄-PAF C16–spiked control samples. (B) qRT-PCR measurements of relative gene expression of the PAF synthetic enzyme Lpcat2, PAFR (Ptafr), and chemokines Cxcl1 and Cxcl2 from nasal lavages obtained 3 days p.i. from mice colonized with PBS, WT, or Δpce P1121 pneumococci (n = 6–10 mice per condition). All transcripts were normalized to GAPDH controls and are displayed relative to mice mock-infected with PBS (dotted lines). *P < 0.05 by Student’s t test.

Figure 2. Pce promotes invasive pneumococcal disease and bacterial transmission between mice.

(A) Survival of adult mice inoculated i.n. with WT (black) or Δpce (gray) pneumococci of invasive strain P1547 (type 6A) (n = 12 mice per condition from 3 independent experiments). Statistical significance was assessed by the Mantel-Cox test. (B) Enumeration of CFU in the blood of mice infected with WT or Δpce P1547 pneumococci as above (n = 5, LOD = 14). (C) Infant murine transmission. Upper airway lavage CFU enumerated from index (white circles, n = 3–4) and contact (black circles, n = 14–18) pups on day 14 of life and after index mice were inoculated with WT (black) or Δpce (gray) pneumococci on day 4 of life. All pups were infected i.n. with influenza on day 8. Numerical values above contact mice columns represent the percentage of acquisition (LOD = 10). Transmission data reflect 3 independent experiments. In B and C, statistical significance was assessed by 1-way ANOVA with Newman-Keuls post test. *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 1. Pce-deficient pneumococci exhibit impaired persistence in the upper airway and elicit the recruitment of more activated, viable, and durable neutrophils to the nasal lumen.

(A) Bacterial clearance in mice inoculated with WT pneumococci, strain P1121 (Type 23F), with (white circles) or without (black circles) systemic neutrophil depletion (n = 4–5 mice per condition, limit of detection [LOD] = 2). (B) Survival of WT P1121 (black) or P1121Δpce (gray) pneumococci in the murine upper airway (n = 4–14). (C) Day-7 survival of P1121Δpce mutant generated by in-frame, unmarked deletion (Δpce) and with genetic correction (Δpce::pce) (n = 5). (D) Day 7 survival of WT and Δpce pneumococci on a type 4 (T4, TIGR4) pneumococcal genetic background (n = 5). (E) Quantification of neutrophils (CD45⁺CD11b⁺Ly6G⁺) obtained from the upper airway lumen by nasal lavage before (n = 3) and after (n = 4–11) inoculation with WT (black) or Δpce (gray) pneumococci. (F) Flow cytometric characterization of luminal neutrophils elicited by infection with WT or Δpce pneumococci on day 4 p.i. (n = 6–8). Note that not all axes are continuous, and gaps in axes represent gaps in time. Dotted lines represent the LOD. Statistical significance was assessed by 1-way ANOVA with Newman-Keuls post test for comparisons of more than 2 conditions (A, B, and E), Student’s t test for 2-group comparisons (C and D), and 1-sample Student’s t test relative to null = 1 for relative MFI measurements (F). *P < 0.05, ***P < 0.001.