The Role of Abl Family Kinases in Acute Respiratory Distress Syndrome

The Abl family kinases, c-Abl (Abl1) and Abl related gene (Arg, Abl2) are highly likely mediators of vascular permeability due to their ability to regulate cellular shape and adhesion dynamics in multiple cell types; however, their roles in this context have not been thoroughly investigated. Inhibitors of these kinases are used clinically for the treatment of chronic myelogenous leukemia (CML) and other malignancies. Recent work has demonstrated that these inhibitors attenuate vascular permeability induced by vascular endothelial growth factor (VEGF), thrombin, histamine, and oxidative stress. In the present study, we aimed to determine the effects of the Abl family kinase inhibitor imatinib mesylate (STI-571, Gleevec) in murine and cell culture models of lipopolysaccharide (LPS)-induced lung injury and ventilator-induced lung injury (VILI). Additionally, we aimed to determine the mechanisms by which c-Abl and Arg mediate inflammatory vascular leak in the context of these injury models. We observed that imatinib pre-treatment attenuates LPS-induced Acute Lung Injury (ALI) as measured by bronchoalveolar lavage (BAL) fluid protein content, total cell counts, neutrophil counts, and cytokine levels (IL-6 and KC). These barrier protective and anti-inflammatory effects were confirmed in vitro using human pulmonary artery endothelial cells (HPAEC) as measured by transendothelial electrical resistance (TER), an avidin-biotin based gap formation assay, vascular endothelial cadherin (VE-Cadherin) expression, nuclear factor kappa B (NFκB) activation, and upregulation of vascular cellular adhesion molecule (VCAM-1) and inflammatory cytokines including interleukin (IL)-6 and IL-8. However, contrary to these protective effects, imatinib pre-treatment exacerbates VILI as measured by the same measures of lung injury severity used in our LPS-induced ALI model. These deleterious effects of imatinib were confirmed in HPAEC challenged with pathologic CS (18%), as measured by avidin-biotin affinity based gap formation assay, the expression and localization of VE-Cadherin, and the upregulation of VCAM-1, and inflammatory cytokines. Thus, imatinib attenuates LPS-induced ALI, but exacerbates VILI in both murine and cell culture models. Due to the necessity of mechanical ventilation (MV) in the treatment of ARDS patients, these data suggest that Abl kinase inhibition should be approached with caution in ARDS. In order to address this concern we next tested the effects of imatinib in a “two-hit” ALI model that incorporates both LPS and VILI. We observed that imatinib attenuates vascular leakage, as measured by BAL protein, lung tissue albumin, wet:dry lung weight, and Evan’s blue dye (EBD) extravasation, and inflammation, as measured by BAL total cell counts, neutrophil counts, and inflammatory cytokine production. These results demonstrate that the beneficial effects of imatinib in LPS-induced ALI outweigh its deleterious effects in VILI. Subsequently, due to the difficulty in identifying ARDS in its early stages, we next tested the effects of imatinib in a more therapeutically relevant model in which the drug was given four hours after LPS challenge. Imatinib retained its protective properties in the context of pre-existing lung injury, suggesting that it may be a potential therapeutic strategy in ARDS. Given the profound effects of imatinib in LPS-induced ALI, we next aimed to determine the targets of imatinib that mediate the barrier protective and anti-inflammatory effects that we observed. LPS treatment increases the mRNA levels of c-Abl, but does not alter the mRNA levels of Arg or the protein levels of either kinase. However, we observed an increase in the kinase activity of both c-Abl and Arg after LPS challenge, as measured by phosphorylation of the substrate CrkL at an Abl family kinase specific site. We found that silencing c-Abl, but not Arg, attenuates LPS-induced inflammation, as measured by VCAM-1 upregulation and pro-inflammatory cytokine release, indicating that the anti-inflammatory effects of imatinib in LPS-induced ALI are mediated by c-Abl. However, silencing Arg, but not c-Abl, attenuates LPS-induced endothelial cell (EC) barrier dysfunction, as measured by gap formation assay and VE-Cadherin localization, indicating that the barrier protective effects of imatinib in LPS-induced ALI are mediated by Arg. Together these results indicate that 1) c-Abl and Arg have distinct roles in the regulation of inflammatory vascular leak and 2) Abl kinase inhibition may be a novel therapeutic strategy in the treatment for ARDS.