Inflammasome-regulated cytokines are critical mediators of acute lung injury

Abstract

Rationale: Despite advances in clinical management, there are currently no reliable diagnostic and therapeutic targets for acute respiratory distress syndrome (ARDS). The inflammasome/caspase-1 pathway regulates the maturation and secretion of proinflammatory cytokines (e.g., IL-18). IL-18 is associated with injury in animal models of systemic inflammation.

Objectives: We sought to determine the contribution of the inflammasome pathway in experimental acute lung injury and human ARDS.

Methods: We performed comprehensive gene expression profiling on peripheral blood from patients with critical illness. Gene expression changes were assessed using real-time polymerase chain reaction, and IL-18 levels were measured in the plasma of the critically ill patients. Wild-type mice or mice genetically deficient in IL-18 or caspase-1 were mechanically ventilated using moderate tidal volume (12 ml/kg). Lung injury parameters were assessed in lung tissue, serum, and bronchoalveolar lavage fluid.

Measurements and main results: In mice, mechanical ventilation enhanced IL-18 levels in the lung, serum, and bronchoalveolar lavage fluid. IL-18-neutralizing antibody treatment, or genetic deletion of IL-18 or caspase-1, reduced lung injury in response to mechanical ventilation. In human patients with ARDS, inflammasome-related mRNA transcripts (CASP1, IL1B, and IL18) were increased in peripheral blood. In samples from four clinical centers, IL-18 was elevated in the plasma of patients with ARDS (sepsis or trauma-induced ARDS) and served as a novel biomarker of intensive care unit morbidity and mortality.

Conclusions: The inflammasome pathway and its downstream cytokines play critical roles in ARDS development.

Figure 1.

Critical illness modulates caspase-1, IL-1β, and IL-18 expression in peripheral blood cells. TaqMan polymerase chain reaction (PCR) results are shown for CASP1 (A), IL1B (B), and IL18 (C). RNA was obtained from medical intensive care unit admission day blood samples. Data are expressed as relative fold-change compared with systemic inflammatory response syndrome (SIRS) = 1. For statistical analysis, the Kruskal-Wallis test was performed for multiple group comparison, and intergroup differences were analyzed with Wilcoxon rank sum test, P < 0.05, n = 6 random samples/group. *Represents significant differences between sepsis/acute respiratory distress syndrome (ARDS) and SIRS samples.

Figure 2.

Plasma IL-18 levels are elevated in critical illness. (A) IL-18 plasma levels were measured in the Brigham and Women’s Hospital (BWH) Registry of Critical Illness (BWH RoCI). Samples were drawn after admission (Day 1) and at Day 3 and Day 7. Patients with sepsis/acute respiratory distress syndrome (ARDS) have higher levels of IL-18 in their plasma than intensive care unit (ICU) control subjects, patients with systemic inflammatory response syndrome (SIRS), or patients with sepsis only. (B) Patients with severe sepsis-induced ARDS (Vanderbilt cohort) display higher levels of IL-18 than SIRS (control). IL-18 levels in severe sepsis: 603.68 pg/ml ± 71.7, n = 30, coefficient of variance (CV) = 0.65; ARDS with severe sepsis: 509.48 pg/ml ± 72.82, n = 31, CV = 0.8; and SIRS: 266.59 pg/ml ± 46.49, n = 28, CV = 0.92. (C) Levels of IL-18 in patients with sepsis-induced ARDS are also elevated in the Massachusetts General Hospital and Beth Israel Deaconess Hospital (MGH/BIDMC) cohort when compared with patients without ARDS (control). IL18 level in sepsis/ARDS: 1,184.8 pg/ml ± 178.29, n = 23, CV = 0.72 and in control patients without ARDS who had either SIRS or sepsis: 386.1 pg/ml ± 43.36, n = 29, CV = 0.6. (D) Patients with sepsis/ARDS (ARDS) displayed higher levels of IL-18 when compared with patients with SIRS only (control) in the BWH cohort at Day 1. IL-18 level in sepsis/ARDS: 1,043.06 pg/ml ± 108.09, n = 21, CV = 0.47 and in SIRS: 571.73 pg/ml ± 83.38, n = 57, CV = 1.1. (E) IL-18 levels are increased in nonsepsis/ARDS BWH patients (ARDS), when compared with ICU control subjects (control). IL-18 levels in ARDS: 696.68 pg/ml ± 81.0, n = 7, CV = 0.32 and in ICU control subjects: 311.46 pg/ml ± 28.47, n = 35, CV = 0.7. By comparison, IL-18 levels of patients with trauma-induced ARDS (trauma ARDS, University of Pennsylvania cohort [U Penn]) are similar to patients with ARDS in BWH, and elevated relative to their ICU control subjects (control). IL-18 levels in trauma ARDS: 636.05 pg/ml ± 104.89, n = 20, CV = 0.74 and in their control subjects: 134.75 pg/ml ± 37.51, n = 18, CV = 1.18. *Significant differences any condition vs. control. ^#Significant difference sepsis vs. sepsis/ARDS. Student t test was performed on log10 transformed data, P < 0.05. Sample numbers/condition are listed in the extended Methods section in the online supplement.

Figure 3.

IL-18 correlates with mortality in the critically ill. (A) Increased IL-18 levels at medical intensive care unit Day1 were associated with increased in-hospital mortality among critically ill patients. Data were based on 217 patients, of whom 161 were discharged from the hospital and 56 died during hospitalization. Wilcoxon two-sample test, P = 6 × 10⁻⁷. (B) Mortality stratification based on plasma IL-18 level quartiles. *Represents significant differences among quartiles, Fisher exact test, P = 5 × 10⁻⁵, n/quartile = 54, 53, 55, 55. Changes in quartiles related to death, odds ratio = 2.02; 95% confidence interval, 1.48–2.76; P = 1 × 10⁻⁵.

Figure 4.

Inflammatory caspase-activated cytokines are regulated in ventilator-induced lung injury. Mice were mechanically ventilated (ventilation) for 8 hours with 12 ml/kg tidal volume and 2 cm H₂O positive end-expiratory pressure (A). IL-18, IL-1β, and IL-33 levels were measured in whole lung homogenates of ventilated wild-type mice and their nonventilated counterparts. Cytokine expression increased in ventilated mice when compared with control animals. IL-18 levels were similarly elevated in serum and in the bronchoalveolar lavage fluid (BALF) (B and C). Kruskal-Wallis test was performed for multiple group comparison and intergroup differences were analyzed with Wilcoxon rank sum test, P < 0.05. For all other animal studies the same statistical tests were performed. *Represents significant differences between ventilation and control, n = 8 animals/group.

Figure 5.

Mechanical ventilation (MV) increases the expression of the cleaved form of IL-18 in alveolar macrophages. (A) Lung tissue samples obtained from control and ventilated mice were stained with fluorophore-labeled antibodies against IL-18 (cleaved form, Cy3, red), macrophage marker Mac-3 (FITC, green), DAPI (blue nuclear stain) and analyzed by confocal microscopy. Representative images are shown. Magnification ×100, scale 10 μm = 10 mm. Enlarged area magnification ×200, scale 5 μm = 10 mm. Results are quantified by counting positively stained cells in five independent areas. (B) MV increased the number of cleaved IL-18 positive alveolar cells when compared with controls. (C) MV increased the number of cells that stained positive for both cleaved IL-18 and Mac-3. (D) MV increased the relative number of IL-18–positive alveolar macrophages. Numbers are expressed as the percentage of total cells in lung sections. *Represents significant differences between control and ventilation. P < 0.05.

Figure 6.

IL-18 and caspase-1 modulate indices of lung injury. IL-18 (Il18^−/−) or caspase-1 (casp1^−/−) knockout mice and their corresponding C57Bl/6 or NOD/shi wild-type mice, respectively, were mechanically ventilated (ventilation) for 8 hours with 12 ml/kg tidal volume and 2 cm H₂O positive end-expiratory pressure. After ventilation or control treatments, mice were analyzed for indices of lung injury. (A) Mechanical ventilation increased bronchoalveolar lavage fluid (BALF) total cell count in wild-type (Il18^+/+ and casp1^+/+) mice. The Il18^−/− and casp1^−/− mice show minimal increase in total cell count after mechanical ventilation. (B) Il18^−/− and casp1^−/− mice respond to mechanical ventilation with reduced alveolar neutrophil infiltration. (C) Il18^−/− and casp1^−/− mice are resistant to ventilator-induced lung injury (VILI)-induced alveolar protein leakage. (D) Lung wet-to-dry ratio measurement confirmed that Il18^−/− and casp1^−/− mice have increased resistance against alveolar pulmonary edema formation in VILI. (E) Il18^−/− and casp1^−/− mice were also protected from apoptotic cell death as measured by the number of caspase-3 positive-staining cells in lung tissue. Results were quantified and expressed as the percentage of cells positively stained for caspase-3. (F). IL-18 levels in lung tissue decreased in casp1^−/− mice when compared with wild-type mice after mechanical ventilation. *Represents significant differences between ventilation and control; n = 8/group for BALF and lung tissue measurements, n = 12/group for wet-to-dry lung weight ratio measurements, and n = 3/group for immunohistochemical analysis.

Figure 7.

IL-18–neutralizing antibody blocks ventilator-induced inflammation. Wild-type C57Bl/6 mice were mechanically ventilated (ventilation) for 8 hours with 12 ml/kg tidal volume and 2 cmH₂O positive end-expiratory pressure. (A) IL-18 antibody (IL-18AB) inhalation significantly reduced ventilator-induced inflammatory cell count in the bronchoalveolar lavage fluid (BALF) when compared with IgG-treated control mice. (B) BALF neutrophil cell count was also lower in IL-18–neutralizing antibody–treated mice. (C) IL-18AB inhalation only partially reduced protein leakage to the alveoli. (D) No differences were found in wet-to-dry lung weight ratio after ventilation between treatment groups. *Represents significant differences between ventilation and control; (n = 6) animals/group.