Allergic airway inflammation decreases lung bacterial burden following acute Klebsiella pneumoniae infection in a neutrophil- and CCL8-dependent manner

Abstract

The Th17 cytokines interleukin-17A (IL-17A), IL-17F, and IL-22 are critical for the lung immune response to a variety of bacterial pathogens, including Klebsiella pneumoniae. Th2 cytokine expression in the airways is a characteristic feature of asthma and allergic airway inflammation. The Th2 cytokines IL-4 and IL-13 diminish ex vivo and in vivo IL-17A protein expression by Th17 cells. To determine the effect of IL-4 and IL-13 on IL-17-dependent lung immune responses to acute bacterial infection, we developed a combined model in which allergic airway inflammation and lung IL-4 and IL-13 expression were induced by ovalbumin sensitization and challenge prior to acute lung infection with K. pneumoniae. We hypothesized that preexisting allergic airway inflammation decreases lung IL-17A expression and airway neutrophil recruitment in response to acute K. pneumoniae infection and thereby increases the lung K. pneumoniae burden. As hypothesized, we found that allergic airway inflammation decreased the number of K. pneumoniae-induced airway neutrophils and lung IL-17A, IL-17F, and IL-22 expression. Despite the marked reduction in postinfection airway neutrophilia and lung expression of Th17 cytokines, allergic airway inflammation significantly decreased the lung K. pneumoniae burden and postinfection mortality. We showed that the decreased lung K. pneumoniae burden was independent of IL-4, IL-5, and IL-17A and partially dependent on IL-13 and STAT6. Additionally, we demonstrated that the decreased lung K. pneumoniae burden associated with allergic airway inflammation was both neutrophil and CCL8 dependent. These findings suggest a novel role for CCL8 in lung antibacterial immunity against K. pneumoniae and suggest new mechanisms of orchestrating lung antibacterial immunity.

FIG 1

Experimental design for combined model of OVA sensitization and challenge-induced allergic airway inflammation and acute K. pneumoniae (KP) lung infection. Mice underwent intraperitoneal injection of OVA adsorbed to ALUM or ALUM alone at day −18. On days −4 to −1, mice previously injected with OVA-ALUM were exposed to 1% OVA aerosol daily. Control mice did not undergo OVA aerosol challenge. Mice were then infected with either K. pneumoniae or mock infected with PBS and harvested on days 1 and 2 postinfection.

FIG 2

Preexisting allergic airway inflammation decreases airway neutrophils, lung Th17 cytokine expression, and the lung and spleen K. pneumoniae burden following acute K. pneumoniae lung infection. (A) Lung IL-17A expression measured by ELISA. (B) Airway neutrophils were quantified from airway lavage fluid. (C) Lung IL-17F expression measured by use of a cytokine bead array. (D) Lung IL-22 expression measured by ELISA. Lung (E) and spleen (F) quantitative bacterial burdens were determined by serial dilution. *, P < 0.001; **, P < 0.01; †, P < 0.05. n = 10 mice per group combined from two representative experiments (A to C, E, and F) and n = 5 mice per group from one experiment (D).

FIG 3

The decreased lung K. pneumoniae burden requires both OVA sensitization and OVA aerosol challenge. The lung K. pneumoniae burden in mice that underwent the protocol described in Fig. 1 was determined (white bars, ALUM-KP group; black bars, OVA-KP group). In addition, the lung K. pneumoniae burden in ALUM-injected mice that underwent OVA challenge (bars with vertical lines) and OVA-ALUM-injected mice that did not undergo OVA aerosol challenge (bars with diagonal lines) was determined. *, P < 0.001 versus all other day 1 groups; **, P < 0.001 versus all other day 2 groups. n = 10 mice per group combined from two representative experiments.

FIG 4

Allergic airway inflammation decreases K. pneumoniae-induced acute-phase cytokine and chemokine expression in the lung. The levels of TNF-α (A), IL-1β (B), IL-6 (C), G-CSF (D), GM-CSF (E), CXCL1 (F), CCL3 (G), and CCL4 (H) protein expression were measured. *, P < 0.001 versus day 1 groups; **, P < 0.001 versus day 2 groups; #, P < 0.05 versus ALUM-PBS mice; ##, P < 0.05 versus OVA-PBS mice. n = 10 mice per group from two representative experiments. KC, keratinocyte-derived cytokine; MIP-1α, macrophage inflammatory protein 1α.

FIG 5

Allergic airway inflammation decreases K. pneumoniae-induced LCN2 expression in the lung on day 1 and day 2 postinfection. *, P < 0.001; **, P < 0.01. n = 5 mice per group.

FIG 6

Allergic airway inflammation enhances survival following acute K. pneumoniae lung infection. Mice that underwent the protocol described in Fig. 1 were infected with K. pneumoniae on day 0. Survival was determined twice daily for 14 days postinfection. *, P = 0.002 (log-rank test) and P = 0.004 (Wilcoxon test) versus ALUM-KP mice. n = 30 for ALUM-KP mice and n = 33 for OVA-KP mice. Data from two representative experiments were combined.

FIG 7

The decreased lung K. pneumoniae burden associated with allergic airway inflammation does not require IL-4 or IL-5 but is partially dependent on IL-13. WT and IL-4-KO mice (A to C), WT mice intraperitoneally injected with isotype control or anti-IL-5 antibody on days −7 and −4 (D to F), and WT and IL-13-KO mice (G to I) underwent OVA sensitization and challenge or mock sensitization, were infected with K. pneumoniae, and were harvested on days 1 and 2 postinfection. (A, D, G) Lung K. pneumoniae burden; (B, E, H) numbers of airway neutrophils; (C, F, I) numbers of airway eosinophils. *, P < 0.001; **, P < 0.01; †, P < 0.05. n = 15 mice per group combined from three representative experiments (A to C), n = 5 mice per group (D to F), and n = 9 to 10 mice per group combined from two representative experiments (G to I).

FIG 8

Allergic airway inflammation decreases the lung K. pneumoniae burden in the absence of STAT6. WT and STAT6-KO mice underwent the protocol described in Fig. 1. (A) Lung K. pneumoniae burden; (B) numbers of airway neutrophils; (C) numbers of airway eosinophils. *, P < 0.001; **, P < 0.01; †, P < 0.05. n = 9 to 14 mice per group combined from two representative experiments.

FIG 9

The decreased lung K. pneumoniae burden associated with allergic airway inflammation occurs in the absence of IL-17A. WT and IL-17A-KO mice underwent the protocol described in Fig. 1 and were infected with 10³ CFU K. pneumoniae. (A) Lung K. pneumoniae burden; (B) numbers of airway neutrophils from WT and IL-17A-KO mice. *, P < 0.001; **, P < 0.01. n = 8 to 17 mice per group combined from three representative experiments.

FIG 10

The decreased lung K. pneumoniae burden associated with allergic airway inflammation is neutrophil dependent. (A to D) WT mice underwent sensitization and challenge and were harvested on day 0 prior to K. pneumoniae infection. Airway lavage neutrophils (A) and peripheral blood neutrophils (B) were quantified. The levels of lung expression of S100a8 (C) and myeloperoxidase (D) were measured by ELISA. (E to G) WT mice were intraperitoneally injected with isotype control or Ly6G antibody and underwent the combined allergic airway inflammation and lung K. pneumoniae infection protocol described in Fig. 1. The numbers of peripheral blood neutrophils (E) and airway neutrophils (F) and the lung K. pneumoniae burden (G) from mice harvested on day 2 postinfection are shown. *, P < 0.001; **, P < 0.01; †, P < 0.05. n = 4 to 6 mice per group (A to D) and n = 9 to 10 mice per group combined from two representative experiments (E to G).

FIG 11

OVA sensitization and challenge induce lung CCL8 expression in a partially STAT6-dependent manner. The lung CCL8 expression from WT mice harvested following the protocol described in Fig. 1 but prior to K. pneumoniae infection (day 0) (A), WT mice that underwent the protocol described in Fig. 1 and harvested on day 1 and day 2 postinfection (B), WT and STAT6KO mice harvested prior to infection (0 h) and 6 h post-K. pneumoniae infection (C), and WT and STAT6-KO mice harvested on day 1 and day 2 post-K. pneumoniae infection (D) is shown. *, P < 0.001; **, P < 0.01; †, P < 0.05. n = 5 mice per group (A and B), n = 4 to 5 mice per group (C) and n = 5 to 7 mice per group (D).

FIG 12

CCL8 neutralization abrogates the decreased lung K. pneumoniae burden associated with allergic airway inflammation. WT BALB/c mice that underwent the protocol described in Fig. 1 were intraperitoneally injected with isotype control or CCL8 antibodies on days −4, −2, and 0. (A) Lung K. pneumoniae burden; (B) total number of airway cells; (C to F) numbers of airway neutrophils (C), eosinophils (D), lymphocytes (E), and macrophages (F). *, P < 0.001; **, P < 0.01; †, P < 0.05. n = 12 to 13 mice combined from two independent, representative experiments.