Sepsis-induced inflammasome impairment facilitates development of secondary A. baumannii pneumonia

Abstract

Background: Acinetobacter baumannii has become one of the most critical pathogens causing nosocomial pneumonia. Existing animal models of A. baumannii pneumonia are not relevant to the majority of critical care patients. We aimed to develop a novel model of secondary A. baumannii pneumonia in post-sepsis mice.

Methods: A two-hit model of sepsis induced by cecal ligation and puncture followed by A. baumannii pneumonia on day 5 was established. In addition, the two-hit model was established in humanized mice. A period of 2 h of mechanical ventilation followed by observation was used in additional experiments. Lung histopathology, bacterial cultures, and cellular infiltration were analysed as well as markers of the inflammasome activity in vivo and ex vivo.

Results: A. baumannii infection caused mortality and loss of body weight and temperature in post-sepsis mice. Increased lung bacterial burden and dissemination together with signs of enhanced inflammatory injury were observed in post-sepsis mice but not control mice that were challenged with A. baumannii. Post-sepsis mice were unable to mount inflammasome activation in response to secondary pneumonia to the level of control mice. Transfer of wild-type but not capsase-1 KO alveolar macrophages was able to restore the pulmonary protection against A. baumannii. Mechanical ventilation exacerbated the pathological response to pneumonia in post-sepsis mice but enhanced inflammasome signalling in non-sepsis mice with pneumonia.

Conclusions: We established a novel model of A. baumannii pneumonia that revealed sepsis-induced impairment of inflammasome activation in alveolar macrophages is critical for the control of secondary A. baumannii pneumonia.

Keywords: Sepsis; VAP; inflammasome; nosocomial infection; translational research.

Figure 1.

Development of the two-hit model. A. Scheme of the model. B. Survival curves of mice subjected to CLP. B. Changes in body temperature. C. Changes in body weight. E. The results of bacterial cultures from the lung homogenates and F. Spleen homogenates of mice challenged with A. baumannii. B-D: n = 20; E, F: n = 6. Data are reported as the means ± SDs. Groups were compared with the Mann–Whitney test. **p < 0.01, ***p < 0.001.

Figure 2.

Pneumonia pathology. A. Histopathological assessment of A. baumannii pneumonia in control mice (left) and post-sepsis mice (right). B. Total cell count in bronchoalveolar lavage (BAL) fluid. C. Frequencies of the populations of major myeloid cell populations in the lung homogenates of mice. n = 6. Data are reported as the means ± SDs. Groups were compared with ANOVA with Tukey’s post hoc test except for the CD45 + cells and alveolar macrophages figures, where comparisons were made to the healthy control mice (without CLP and without A. baumannii) using one-way ANOVA. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 3.

The inflammasome response in secondary pneumonia. A. mRNA levels of the selected inflammasome-related genes were analysed by qPCR and normalized to Gapdh expression. B. Activity of caspase-1 in Siglec-F + CD11b^dim alveolar macrophages was evaluated with the FAM-FLICA probe. On the left: representative histograms of FAM-FLICA staining in control and post-CLP mice infected with A. baumannii, x axis represent intensive of fluorescence and y axis represent number of cells, on right: graph presents the mean values of the geometric mean fluorescence (GMF) for FAM-FLICA. C. Concentration of IL-1$\mathrm{\text{β}}$ in lung homogenates. D. Immunoblots for the cleaved (active) forms of caspase-1 and gasdermin d (GSDMD) are shown. E. Ability to release IL-1$\mathrm{\text{β}}$ after ex vivo stimulation of alveolar macrophages with LPS and nigericin. F. Transfer of alveolar macrophages from WT, but not Casp1^{-/ –} mice reduced the bacterial load after 24 h in post-septic mice infected with A. baumannii. Data are reported as the means ± SDs. Groups were compared with ANOVA with Tukey’s post hoc test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 4.

Effects of mechanical ventilation on pneumonia. A. Histopathological assessment of lungs from mice subjected to A. baumannii and mechanical ventilation: control (left) and postsepsis (right). B. The results of bacterial cultures from the lung homogenates. C. Frequencies of alveolar macrophages and neutrophils. D. Expression of inflammasome-related genes. E. Immunoblots for the cleaved (active) forms of caspase-1 and gasdermin D (GSDMD). F. Concentration of IL-1$\mathrm{\text{β}}$ in lung homogenates. G. Immunofluorescence staining for ASC and cleaved caspase-1 in the lungs of infected mice. Arrows indicate puncta of inflammasome assembly (ASC-caspase-1 complex, yellow dots). n = 4-6. Data are reported as the means ± SDs. Groups were compared with Student’s t test or ANOVA with Tukey’s post hoc test. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.00001.

Figure 5.

Secondary pneumonia in humanized mice. A. Changes in body temperature, asterisk for comparisons at 120 h. B. The results of bacterial cultures from the lung homogenates. C. Histopathological assessment of lungs from mice subjected to sepsis and/or secondary pneumonia. D. Bacterial cultures from lung homogenates of mice subjected to mechanical ventilation. E. Frequencies of human alveolar macrophages (hCD45 ⁺ hCD206 ⁺ CD169 ⁺ CD14^low) and monocytes (hCD45 ⁺ CD33⁺ CD14⁺). n = 4-6. Data are reported as the means ± SDs. Groups were compared with the Mann–Whitney test or ANOVA with Tukey’s post hoc test. *p < 0.05; **p < 0.01; ***p < 0.001, ****p < 0.0001.

Figure 6.

The inflammasome response to pneumonia in humanized mice. A. mRNA levels of the selected inflammasome-related genes were analysed by qPCR and normalized to Gapdh expression. B. Activity of caspase-1 in human alveolar macrophages was evaluated with the FAM-FLICA probe. The upper panel shows representative histograms of FAM-FLICA staining in control and post-CLP mice infected with A. baumannii, and the lower graph presents the mean values of the geometric mean fluorescence (GMF) for FAM-FLICA. C. Two groups of humanized mice were distinguishable: hypothermic (Low) and normo-/hyperthermic (High) which clustered with the hIL-1$\mathrm{\text{β}}$ concentration in lung homogenates 6 h after A. baumannii instillation. D. Comparison of the expression levels of selected inflammasome genes in regard to body temperature changes. E. Immunofluorescence staining for human CD45 and ASC in the lungs of infected humanized mice. n = 4-6. Data are reported as the means ± SDs. Groups were compared with the Mann–Whitney test or ANOVA with Tukey’s post hoc test. *p < 0.05.