Pre-treatment with allopurinol or uricase attenuates barrier dysfunction but not inflammation during murine ventilator-induced lung injury

Abstract

Introduction: Uric acid released from injured tissue is considered a major endogenous danger signal and local instillation of uric acid crystals induces acute lung inflammation via activation of the NLRP3 inflammasome. Ventilator-induced lung injury (VILI) is mediated by the NLRP3 inflammasome and increased uric acid levels in lung lavage fluid are reported. We studied levels in human lung injury and the contribution of uric acid in experimental VILI.

Methods: Uric acid levels in lung lavage fluid of patients with acute lung injury (ALI) were determined. In a different cohort of cardiac surgery patients, uric acid levels were correlated with pulmonary leakage index. In a mouse model of VILI the effect of allopurinol (inhibits uric acid synthesis) and uricase (degrades uric acid) pre-treatment on neutrophil influx, up-regulation of adhesion molecules, pulmonary and systemic cytokine levels, lung pathology, and regulation of receptors involved in the recognition of uric acid was studied. In addition, total protein and immunoglobulin M in lung lavage fluid and pulmonary wet/dry ratios were measured as markers of alveolar barrier dysfunction.

Results: Uric acid levels increased in ALI patients. In cardiac surgery patients, elevated levels correlated significantly with the pulmonary leakage index. Allopurinol or uricase treatment did not reduce ventilator-induced inflammation, IκB-α degradation, or up-regulation of NLRP3, Toll-like receptor 2, and Toll-like receptor 4 gene expression in mice. Alveolar barrier dysfunction was attenuated which was most pronounced in mice pre-treated with allopurinol: both treatment strategies reduced wet/dry ratio, allopurinol also lowered total protein and immunoglobulin M levels.

Conclusions: Local uric acid levels increase in patients with ALI. In mice, allopurinol and uricase attenuate ventilator-induced alveolar barrier dysfunction.

Figure 1. Uric acid levels in ventilated patients.

Levels of uric acid in bronchoalveolar lavage fluid obtained from patients with acute respiratory distress syndrome (ARDS) (n = 16) and patients without ARDS (n = 62) (A). In a different cohort of cardiac surgery patients elevated uric acid levels in lung lavage fluid were correlated with the mean pulmonary leakage index (PLI) measured, displayed in ×10⁻³ min⁻¹ (B) (n = 16). Data are presented as mean ± SEM. **p<0.01 vs no ALI.

Figure 2. Hemodynamic parameters during 5 hours of ventilation in mice.

Hemodynamic parameters of mice pre-treated with vehicle (10% dimethylsulfoxide), uricase (0.2 mg/kg), or allopurinol (25 mg/kg) 1 hour before start of mechanical ventilation. Systolic bloodpressure (A) and heart rate (B) was measured at start of ventilation, after 2.5, and after 5 hours. Data represent mean ± SEM of n = 9 mice per group.

Figure 3. Uric acid levels in murine ventilator-induced lung injury.

Uric acid levels in plasma (A) and bronchoalveolar lavage fluid (BALF) (B) of mice pre-treated with vehicle (10% dimethylsulfoxide, dark grey bars), uricase (0.2 mg/kg, light grey, striped bars), or allopurinol (25 mg/kg (white bars) 1 hour before start of 5 hours of mechanical ventilation (VILI). Spontaneously breathing, vehicle pre-treated mice served as controls (C). Data represent mean ± SEM of 4 control mice and n = 6−9 ventilated mice. **p<0.01 vs. control, #p<0.05 vs. vehicle ventilated.

Figure 4. Pulmonary inhibitory kappa-B alpha levels in murine ventilator-induced lung injury.

Fold down regulation of inhibitory kappa-B alpha (IκB-α) in lung tissue homogenate relative to total protein levels measured by Ponceau staining. Mice were pre-treated with vehicle (10% dimethylsulfoxide), uricase (0.2 mg/kg, light grey, striped bars), or allopurinol (25 mg/kg (white bars) 1 hour before start of mechanical ventilation (VILI). Spontaneously breathing, vehicle pre-treated mice served as controls (C). After 5 hours mice were killed. Data represent mean (SEM) of 4 control mice and n = 5−6 ventilated mice. ***p<0.001 vs. vehicle control.

Figure 5. Allopurinol or uricase pre-treatment do not reduce neutrophil influx.

Mice were pre-treated with vehicle (10% dimethylsulfoxide, dark grey bars), uricase (0.2 mg/kg, light grey, striped bars), or allopurinol (25 mg/kg (white bars) 1 hour before start of mechanical ventilation (VILI). Spontaneously breathing, vehicle pre-treated mice served as controls (C). After 5 hours mice were killed. Neutrophil influx into the alveolar compartment (A) and the pulmonary expression of intracellular adhesion molecule-1 (ICAM-1) (B), vascular adhesion molecule (VCAM) (C) and e-selectin (D) relative to the housekeeping gene HPRT were analyzed. Data represent mean ± SEM of 4 control mice and n = 7−9 ventilated mice. *p<0.05, **p<0.01 vs. vehicle control.

Figure 6. Allopurinol and uricase pre-treatment do not affect pulmonary and systemic cytokine levels.

Mice were pre-treated with vehicle (10% dimethylsulfoxide, dark grey bars), uricase (0.2 mg/kg, light grey, striped bars), or allopurinol (25 mg/kg (white bars) 1 hour before start of mechanical ventilation (VILI). Spontaneously breathing, vehicle pre-treated mice served as controls (C). After 5 hours mice were killed. Interleukin (IL)-6 (A,B), IL-1β (C,D) and keratinocyt-derived chemokine (KC) (E,F) levels were determined in lung tissue homogenate and plasma. Data represent mean ± SEM of 4 control mice and n = 9 mice per ventilated group. *p<0.05, **p<0.01 vs. vehicle control.

Figure 7. Allopurinol and uricase pre-treatment attenuate alveolar barrier dysfunction.

Mice were pre-treated with vehicle (10% dimethylsulfoxide, dark grey bars), uricase (0.2 mg/kg, light grey, striped bars), or allopurinol (25 mg/kg (white bars) 1 hour before start of mechanical ventilation (VILI). Spontaneously breathing, vehicle pre-treated mice served as controls (C). After 5 hours mice were killed. Lung wet-to-dry ratio (A), total protein levels in lung lavage fluid (B) and immunoglobulin M (IgM) concentrations in (C) were analyzed. Data represent mean ± SEM of 4 control mice and n = 9 for the VILI groups (for IgM n = 5−6). ##P<0.01, ###P<0.001 vs. vehicle ventilated. ***p<0.001 vs. vehicle control.

Figure 8. Histopathology of VILI in mice is affected by allopurinol pre-treatment.

Lung tissue slides of mice pre-treated with vehicle (10% dimethylsulfoxide, dark grey bars), uricase (0.2 mg/kg, light grey, striped bars), or allopurinol (25 mg/kg (white bars) 1 hour before start of mechanical ventilation (VILI). Spontaneously breathing, vehicle pre-treated mice served as controls (C). Lungs were scored for presence of edema, hemorrhage, neutrophil influx, and hyaline membranes. Total scores and edema scores are demonstrated below. Data are presented as mean ± SEM of 4 control mice and n = 9 for the VILI groups. *p<0.05 vs vehicle control H&E staining, magnification ×20.