Mice exposed to 100% O2 die after 3 or 4 d with diffuse alveolar damage and alveolar edema. Extensive cell death is evident by electron microscopy in the alveolar septa, affecting both endothelial and epithelial cells. The damaged cells show features of both apoptosis (condensation and margination of chromatin) and necrosis (disruption of the plasma membrane). The electrophoretic pattern of lung DNA indicates both internucleosomal fragmentation, characteristic of apoptosis, and overall degradation, characteristic of necrosis. Hyperoxia induces a marked increase in RNA or protein levels of p53, bax, bcl-x, and Fas, which are known to be expressed in certain types of apoptosis. However, we did not detect an increased activity of proteases belonging to the apoptosis "executioner" machinery, such as CPP32 (caspase 3), ICE (caspase 1), or cathepsin D. Furthermore, administration of an ICE-like protease inhibitor did not significantly enhance the resistance to oxygen. Additionally, neither p53-deficient mice nor lpr mice (Fas null) manifested an increased resistance to hyperoxia-induced lung damage. These results show that both necrosis and apoptosis contribute to cell death during hyperoxia. Multiple apoptotic pathways seem to be involved in this, and an antiapoptotic strategy does not attenuate alveolar damage.

ROS generation by NOX4 is a key player in epithelial cell death leading to pulmonary fibrosis.

Exposure to high oxygen concentration causes direct oxidative cell damage through increased production of reactive oxygen species. In vivo oxygen-induced lung injury is well characterized in rodents and has been used as a valuable model of human respiratory distress syndrome. Hyperoxia-induced lung injury can be considered as a bimodal process resulting (1) from direct oxygen toxicity and (2) from the accumulation of inflammatory mediators within the lungs. Both apoptosis and necrosis have been described in alveolar cells (mainly epithelial and endothelial) during hyperoxia. While the in vitro response to oxygen seems to be cell type-dependent in tissue cultures, it is still unclear which are the death mechanisms and pathways implicated in vivo. Even though it is not yet possible to distinguish unequivocally between apo-ptosis, necrosis, or other intermediate form(s) of cell death, a great variety of strategies has been shown to prevent alveolar damage and to increase animal survival during hyperoxia. In this review, we summarize the different cell death pathways leading to alveolar damage during hyperoxia, with particular attention to the pivotal role of mitochondria. In addition, we discuss the different protective mechanisms potentially interfering with alveolar cell death.

Hyperoxia-induced lung disease is associated with prominent intraalveolar fibrin deposition. Fibrin turnover is tightly regulated by the concerted action of proteases and antiproteases, and inhibition of plasmin-mediated proteolysis could account for fibrin accumulation in lung alveoli. We show here that lungs of mice exposed to hyperoxia overproduce plasminogen activator inhibitor-1 (PAI-1), and that PAI-1 upregulation impairs fibrinolytic activity in the alveolar compartment. To explore whether increased PAI-1 production is a causal or only a correlative event for impaired intraalveolar fibrinolysis and the development of hyaline membrane disease, we studied mice genetically deficient in PAI-1. We found that these mice fail to develop intraalveolar fibrin deposits in response to hyperoxia and that they are more resistant to the lethal effects of hyperoxic stress. These observations provide clear and novel evidence for the pathogenic contribution of PAI-1 in the development of hyaline membrane disease. They identify PAI-1 as a major deleterious mediator of hyperoxic lung injury. ( J. Clin. Invest. 1996. 98:2666-2673.)