Effect of Subacute Angiotensin II Elevations on Sphingolipid Levels and Lung Injury in Mice

ABSTRACT

Background: There is a paucity of therapies for acute lung injury (ALI) induced by respiratory viruses. A previously demonstrated key mechanism of ALI, particularly in the setting of severe acute respiratory syndrome coronavirus infections, has been ascribed to decreased cell surface angiotensin converting enzyme 2 (ACE2) leading to increased circulating levels of angiotensin II (Ang2). In turn, supraphysiological Ang2 levels trigger a cascade of events that culminates with endothelial injury in the systemic circulation via acid sphingomyelinase (ASMase) activation. ASMase has been implicated in several models of ALI, but its specific involvement in Ang2-induced ALI is unknown. ASMase hydrolyzes sphingomyelin to pro-apoptotic, edemagenic ceramide, which can be metabolized to endothelial-protective sphingosine-1-phosphate (S1P). Therefore, the ratio of ceramide/S1P can determine endothelial cell fate and lung vascular permeability. We hypothesized that ceramide levels are increased relative to S1P in mice with Ang2-induced ALI. Methods: Following a published protocol of Ang2-induced ALI (Wu et al, 2017), we delivered Ang2 via osmotic pumps (1 ug/kg/min, 7 days;Ang2-mice), using saline (sham) or untreated C57BL/6 mice as controls. We evaluated pulmonary function (FlexiVent);albumin, IgM (ELISA), and inflammatory cell abundance in bronchoalveolar lavage fluid (BALF);and lung parenchyma inflammation and fibrosis (Ashcroft score) on H/E-stained lungs. Sphingolipid levels in lungs and plasma were measured by tandem liquid chromatography/mass spectrometry. Results: Inspiratory capacity, lung compliance, and body weight all decreased in Ang2-mice (by 13-14%, p<0.05 each) compared to sham. Lung pressure-volume loops exhibited a right-shift in Ang2- vs. sham or untreated mice. There was no significant change in BALF albumin, IgM, or inflammatory cells, or in lung histology inflammation or fibrosis scores in Ang2-mice. Compared to sham, S1P levels were significantly increased in plasma and unlavaged lung in Ang2-mice, decreasing ceramide/S1P ratios (from 3.1 to 2.0, and 26 to 20, respectively, p<0.05 each). Conclusions: Sustained subacute systemic elevations of Ang2 increased lung stiffness, but did not cause severe ALI in mice. Lung and circulatory elevations of S1P but not ceramide may have protected against lung edema and inflammatory injury. Although the cause of increased lung stiffness in this model remains to be elucidated, it is notable that chronic (months) supraphysiological elevations of either Ang2 or S1P have been associated with lung fibrosis. In conclusion, a second-hit injury may be necessary to augment the susceptibility of murine lung to Ang2-induced endothelial damage and inflammation relevant to coronavirus.