Glucocorticoid-Augmented Efferocytosis Inhibits Pulmonary Pneumococcal Clearance in Mice by Reducing Alveolar Macrophage Bactericidal Function

Abstract

Inhaled corticosteroids (ICS) increase community-acquired pneumonia (CAP) incidence in patients with chronic obstructive pulmonary disease (COPD) by unknown mechanisms. Apoptosis is increased in the lungs of COPD patients. Uptake of apoptotic cells (ACs) ("efferocytosis") by alveolar macrophages (AMøs) reduces their ability to combat microbes, including Streptococcus pneumoniae, the most common cause of CAP in COPD patients. Having shown that ICS significantly increase AMø efferocytosis, we hypothesized that this process, termed glucocorticoid-augmented efferocytosis, might explain the association of CAP with ICS therapy in COPD. To test this hypothesis, we studied the effects of fluticasone, AC, or both on AMøs of C57BL/6 mice in vitro and in an established model of pneumococcal pneumonia. Fluticasone plus AC significantly reduced TLR4-stimulated AMø IL-12 production, relative to either treatment alone, and decreased TNF-α, CCL3, CCL5, and keratinocyte-derived chemoattractant/CXCL1, relative to AC. Mice treated with fluticasone plus AC before infection with viable pneumococci developed significantly more lung CFUs at 48 h. However, none of the pretreatments altered inflammatory cell recruitment to the lungs at 48 h postinfection, and fluticasone plus AC less markedly reduced in vitro mediator production to heat-killed pneumococci. Fluticasone plus AC significantly reduced in vitro AMø killing of pneumococci, relative to other conditions, in part by delaying phagolysosome acidification without affecting production of reactive oxygen or nitrogen species. These results support glucocorticoid-augmented efferocytosis as a potential explanation for the epidemiological association of ICS therapy of COPD patients with increased risk for CAP, and establish murine experimental models to dissect underlying molecular mechanisms.

Copyright © 2015 by The American Association of Immunologists, Inc.

Figures

Figure 1. GCAE reduced inflammatory cytokine production by murine AMø in response to stimulation via TLR4

Adherence-purified AMø from normal C57BL/6 mice were pre-treated with media alone (none); 2 μM fluticasone for 3 h (Flu), AC (at a ratio of 10 AC/AMø) for 2 h (AC); or 2 μM fluticasone for 3 h followed by AC for 2h (Flu + AC). Next, LPS at 1 ng/mL was added to all wells for an additional 24 h. Supernatants were collected and assayed by Luminex for protein concentrations of (A) TNF-α; (B) IL-6; (C) IL-12; (D) CCL3; (E) CCL5; and (F) KC. Results are mean ± SEM of four independent experiments, each using pooled AMø from two mice. *, p < 0.05; **, p < 0.01; ***, p<0.001; ****, p<0.0001; NS, not significant by ANOVA with Fisher LSD post-hoc testing.

Figure 2. Fluticasone increased in vivo uptake of AC by resident AMø

A. C57BL/6 mice received two IN administrations, of either various doses of fluticasone (100–10,000 ng/mL) or saline control, followed 6 h later by an IN administration of 1 × 107 AC. One h later, AMø were collected by BAL; cytospins were stained with H&E and ingested AC were counted under oil at 1000 × final magnification. A. Representative cytospins showing in vivo AC uptake following in vivo fluticasone treatment with either saline (top panel) or 1000 ng fluticasone. Arrows point to ingested AC. Top panel, percentage of AMø ingesting at least one AC; lower panel, efferocytic index. B. Data are mean ± SEM from three mice in a single experiment, and are representative of results of three independent experiments. Statistical testing using one-way ANOVA with Dunnett’s post-hoc testing for multiple comparisons relative to saline-only control group.

Figure 3. GCAE specifically reduced clearance of viable S. pneumoniae from the lungs in a murine model

C57BL/6 mice received two IN administrations, given 4 h apart, of either saline (indicated by minus symbol), fluticasone followed by saline, saline followed by AC, or fluticasone followed by AC. All mice were infected via the IT route 24 h after the final IN treatment using 50,000 CFU S. pneumoniae serotype 3. Lungs were collected 48 h later to assay total CFU by serial dilution on blood agar plates. Data are derived from 2–3 mice per condition assayed individually in each of three independent experiments (total n = 34). A. Log lung CFU of individual mice; symbols denote mice from different experiments. B. Fold-change in lung CFU, relative to the group pre-treated twice with saline before infection, the geometric mean of which was set to 1. Data are shown as median, 25% & 75% (box) and 5%, 95% CI (whiskers), with outliers shown individually. *, p<0.5; **, p<0.01 by Kruskal-Wallis non-parametric ANOVA with Dunn’s post-hoc testing for multiple comparisons to saline-only control group.

Figure 4. GCAE did not alter inflammatory cell recruitment to the lungs during pneumococcal pneumonia

Mice were pre-treated by the IN route and infected by the IT route with 50,000 CFU/mouse S. pneumoniae serotype 3, exactly as described in the Legend to Figure 3, except that 48 h post-infection, lungs were harvested and processed individually for flow cytometry. Hematopoietic cells were gated using light-scatter parameters and CD45 staining as described in the Results. Data are expressed as the absolute number of cells per lung for each cell type on the vertical axis (note differences in scales), as mean ± SEM of 2–3 mice per condition assayed individually in each of two independent experiments (total n = 21). There were no statistically significant differences between treatment groups by ANOVA with Fisher’s LSD post-hoc testing.

Figure 5. GCAE reduced inflammatory cytokine production by murine AMø in response to stimulation by heat-killed pneumococci

Adherence-purified murine AMø from C57BL/6 mice were pre-treated with media alone (none); 2 μM fluticasone for 22 h (Flu), AC (at a ratio of 10 AC/AMø) for 2 h (AC); or 2 μM fluticasone for 22 h followed by AC for 2h (Flu + AC). All conditions were incubated for and additional 24 h with heat-killed S. pneumoniae serotype 3 at a multiplicity of infection (MOI) of 100 (except for CCL3, for which MOI = 10). Supernatants were collected and assayed by Luminex for protein concentrations of TNF-α, IL-6, IL-12 (top row) and CCL3, CCL5 and KC (bottom row). Results are mean ± SEM of three independent experiments, each using pooled AMø from two mice. *, p < 0.05; **, p < 0.01; ***, p < 0.001; NS, not significant by ANOVA with Fisher LSD post-hoc testing.

Figure 6. GCAE significantly inhibited in vitro killing of S. pneumoniae by murine AMø

Adherence-purified AMø from normal C57BL/6 mice were treated with one of four regimens: saline (indicated by minus sign) alone twice; 2 μM fluticasone for 22 h followed by saline for 2 h; saline for 22 h followed by AC for 2 h; or 2 μM fluticasone for 22 h followed by AC for 2 h. AC were added at a ratio of 10:1 relative to AMø. Next, viable pneumococci (2 × 106 CFU) were added, and then bacterial killing was assayed as described in Material & Methods. Data are expressed as (A) the percentage of bacterial killing at T120 mins; and (B) lung CFU (in millions) at T0 mins; both are derived from four independent experiments each containing at least three mice per group. Results are depicted as a box & whiskers plot, indicating median, 25th & 75th percentiles and minimum, maximum values; *, p < 0.05; **, p < 0.01, by one-way ANOVA with Holm-Sídák post-hoc testing for multiple comparisons.

Figure 7. Fluticasone alone and GCAE significantly inhibited phagolysosome acidification

Adherence-purified AMø from normal C57BL/6 mice were treated with one of four regimens: saline alone twice; 2 μM fluticasone for 22 h followed by saline for 2 h; saline for 22 h followed by AC for 2 h; or 2 μM fluticasone for 22 h followed by AC for 2 h. Next, either pHrodo Bioparticles (A, B) or FITC Zymosan Bioparticles (C) were added and plates were cultured for an additional 90 minutes. Finally, AM were washed, harvested and analyzed by flow cytometry. Corrected MFI was calculated individually for each condition by subtracting the MFI of the well without added particles from the well with added particles in the same experiment. (A) Representative staining for pHrodo Bioparticles (FITC+) for each condition. (B) Aggregated corrected MFI for pHrodo. Results are mean ± SEM of three independent experiments (4–5 mice per experiment); each symbol represents an individual well. (C) Aggregated corrected MFI for FITC-Zymosan; note compressed range. Results are mean ± SEM from one experiment using pooled AMø from five mice; each symbol represents an individual well. *, p