Hyperoxia causes angiopoietin 2-mediated acute lung injury and necrotic cell death

Abstract

The angiogenic growth factor angiopoietin 2 (Ang2) destabilizes blood vessels, enhances vascular leak and induces vascular regression and endothelial cell apoptosis. We considered that Ang2 might be important in hyperoxic acute lung injury (ALI). Here we have characterized the responses in lungs induced by hyperoxia in wild-type and Ang2-/- mice or those given either recombinant Ang2 or short interfering RNA (siRNA) targeted to Ang2. During hyperoxia Ang2 expression is induced in lung epithelial cells, while hyperoxia-induced oxidant injury, cell death, inflammation, permeability alterations and mortality are ameliorated in Ang2-/- and siRNA-treated mice. Hyperoxia induces and activates the extrinsic and mitochondrial cell death pathways and activates initiator and effector caspases through Ang2-dependent pathways in vivo. Ang2 increases inflammation and cell death during hyperoxia in vivo and stimulates epithelial necrosis in hyperoxia in vitro. Ang2 in plasma and alveolar edema fluid is increased in adults with ALI and pulmonary edema. Tracheal Ang2 is also increased in neonates that develop bronchopulmonary dysplasia. Ang2 is thus a mediator of epithelial necrosis with an important role in hyperoxic ALI and pulmonary edema.

Figure 1

Effect of hyperoxia on Ang-1, Ang-2 and Tie 2. Mice were exposed to room air (RA) or 100% O₂ for up to 72 h. (a–c) Pulmonary Ang and Tie2 mRNA (a), pulmonary and extrapulmonary organ Ang mRNA (b) and BAL fluid Ang2 protein (c) were assessed by RT-PCR or western blotting. (d) In situ hybridization was used to localize Ang2 mRNA in the airways and parenchyma of mice exposed to room air or 100% O₂ for 72 h. Arrows highlight airway epithelial cells (top) and type II cells (bottom). (e,f) Immunohistochemical analysis was used to localize Ang2 protein in mice exposed to room air or 100% O₂ for 72 h. Single labeling experiments highlight prominent staining in airway and alveolar epithelial cells (e). Double labeling experiments highlight staining with anti-Ang2 (red) and anti–SP-C (green) antibodies (f). Arrow indicates a cell labeled with both antibodies. Original magnification, ×40 (d–f).

Figure 2

Effect of hyperoxia on survival, inflammation, permeability and tissue injury. Wild-type (Ang2^+/+), Ang2^+/– and Ang2^–/– mice were exposed to 100% O₂, and survival (a), BAL total cell recovery (b), differential cell recovery (c), BAL protein (d) and lung tissue injury (assessed by light microscopy, hematoxylin and eosin stain; e) were assessed. In b–e, mice were exposed to room air or 100% O₂ for 72 h. Data represent assessments in a minimum of n = 8 mice. Original magnification, ×40 (e). *P < 0.01, #P ≤ 0.02, **P = 0.03, ##P < 0.04.

Figure 3

Role of Ang2 in hyperoxia-induced oxidant and DNA injury. (a,b) Wild-type and Ang2^–/– mice were exposed to room air (–) or 100% O₂ (+) for 72 h, and subjected to both staining for 8-OHdG (a) and TUNEL evaluation (b). (c) Percentage of TUNEL-positive cells. Original magnification, ×10 (a, b, top); ×40 (a, b, bottom). *P < 0.0001, **P < 0.03.

Figure 4

Effects of rAng2 and Ang2 siRNA on hyperoxia-induced responses. (a–c) Wild-type mice were exposed to 100% O₂ and were randomized to receive rAng2 or vehicle control. After 72 h in hyperoxia, vascular congestion (a), BAL cellularity (b) and TUNEL-positive cell death (c) were evaluated. (d–g) Mice were randomized to receive an Ang2 or control (scrambled) siRNA, and Ang2 mRNA (d,e), BAL cellularity (f) and TUNEL-positive cell death (g) were evaluated. Two doses of Ang2 siRNA, low (LD) and high (HD), were used (Methods). Data represent assessments in a minimum of n = 5 mice. Original magnification, ×40 (a). *P < 0.0001, **P ≤ 0.01, #P ≤ 0.02, ##P < 0.03.

Figure 5

Role of Ang2 in hyperoxia-induced alterations in apoptosis and angiogenic regulators and the effect of rAng2 on MLE-12 cell survival. (a–d) Wild-type, Ang2^+/– and Ang2^–/– mice were exposed to room air or 100% O₂ for 72 h, and the indicated mRNAs (a) and caspase-3, caspase-8 and caspase-9 bioactivity (b–d) were assessed. (e,f) MLE-12 cells were cultured for up to 48 h in 5% CO₂ and air, or 95% O₂ in the presence and absence of the indicated concentration of rAng2. (e) Apoptosis and necrosis were evaluated by Annexin V and propidium iodide (PI) staining and expressed as a percentage of the total cell number. (f) Percentage of cells undergoing pure necrosis (PI-positive, Annexin V–negative). *P < 0.01, #P ≤ 0.02, **P ≤ 0.03, ##P < 0.05.

Figure 6

Ang2 in biological fluids from human adults and neonates. (a) Concentration of Ang2 in the plasma and undiluted AEF of individuals with ALI and healthy controls. Con, healthy adults, Hyd, adults with hydrostatic edema; ALI, adults with acute lung injury. n = 3–4 per group; *P < 0.0001, #P = 0.005, **P = 0.02, ##P = 0.05. (b) Concentration of Ang2 in the tracheal aspirate of premature babies affected with RDS with and without an adverse outcome (bronchopulmonary dysplasia and/or death). n = 5–9 per group; *P < 0.01.