Role of pulmonary microvascular endothelial cell apoptosis in murine sepsis-induced lung injury in vivo

BACKGROUND: Sepsis remains a common and serious condition with significant morbidity and mortality due to multiple organ dysfunction, especially acute lung injury (ALI) and acute respiratory distress syndrome (ARDS). Sepsis-induced ALI is characterized by injury and dysfunction of the pulmonary microvasculature and pulmonary microvascular endothelial cells (PMVEC), resulting in enhanced pulmonary microvascular sequestration and pulmonary infiltration of polymorphonuclear leukocytes (PMN) as well as disruption of the normal alveolo-capillary permeability barrier with leak of albumin-rich edema fluid into pulmonary interstitium and alveoli. The role of PMVEC death and specifically apoptosis in septic pulmonary microvascular dysfunction in vivo has not been established. METHODS: In a murine cecal ligation/perforation (CLP) model of sepsis, we quantified and correlated time-dependent changes in pulmonary microvascular Evans blue (EB)-labeled albumin permeability with (1) PMVEC death (propidium iodide [PI]-staining) by both fluorescent intravital videomicroscopy (IVVM) and histology, and (2) PMVEC apoptosis using histologic fluorescent microscopic assessment of a panel of 3 markers: cell surface phosphatidylserine (detected by Annexin V binding), caspase activation (detected by FLIVO labeling), and DNA fragmentation (TUNEL labeling). RESULTS: Compared to sham mice, CLP-sepsis resulted in pulmonary microvascular barrier dysfunction, quantified by increased EB-albumin leak, and PMVEC death (PI+ staining) as early as 2 h and more marked by 4 h after CLP. Septic PMVEC also exhibited increased presence of all 3 markers of apoptosis (Annexin V+, FLIVO+, TUNEL+) as early as 30 mins – 1 h after CLP-sepsis, which all similarly increased markedly until 4 h. The time-dependent changes in septic pulmonary microvascular albumin-permeability barrier dysfunction were highly correlated with PMVEC death (PI+; r = 0.976, p < 0.01) and PMVEC apoptosis (FLIVO+; r = 0.991, p < 0.01). Treatment with the pan-caspase inhibitor Q-VD prior to CLP reduced PMVEC death/apoptosis and attenuated septic pulmonary microvascular dysfunction, including both albumin-permeability barrier dysfunction and pulmonary microvascular PMN sequestration (p < 0.05). Septic PMVEC apoptosis and pulmonary microvascular dysfunction were also abrogated following CLP-sepsis in mice deficient in iNOS (Nos2(−/−)) or NADPH oxidase (p47(phox−/−) or gp91(phox−/−)) and in wild-type mice treated with the NADPH oxidase inhibitor, apocynin. CONCLUSIONS: Septic murine pulmonary microvascular dysfunction in vivo is due to PMVEC death, which is mediated through caspase-dependent apoptosis and iNOS/NADPH-oxidase dependent signaling.