Abstract

INTRODUCTION: Acute lung injury is a severe condition developing in patients with acute sepsis characterized by lung edema with extravasation of plasma proteins, infiltration by inflammatory cells and pulmonary hemorrhage. Vascular endothelial growth factor (VEGF), also known as vascular permeability factor, is known to not only promote angiogenesis but also increase vascular permeability and has been linked to pathological conditions like retinopathy and acute lung injury. Patients with sepsis and acute lung injury have increased VEGF levels in their plasma. As shown recently, mice treated with VEGF develop histological changes comparable to acute lung injury in septic patients. Endostatin (ES), a endogenous inhibitor of angiogenesis, has been shown to inhibit tumor growth in various models. The mechanisms by which ES inhibits endothelial cell proliferation and function are not clear, yet. It was proposed that one mechanism is the inhibition of VEGF-induced activation of VEGF receptor 2 (VEGFR-2). This study was performed in order to examine whether ES may antagonize VEGF-induced effects leading to increased permeability and lung injury.

METHODS: We established an in vitro permeability model using transwell chambers with human dermal microvascular endothelial cells (HDMEC) seeded in the upper chamber on a porous membrane. VEGF-induced permeability was determined by measuring methylene blue to the lower chamber. In order to test the effect of ES on VEGFR-2 activation porcine aortic endothelial cells expressing human VEGFR-2 were incubated with 50 ng/ml VEGF with or without 5 ug/ml ES for 30 min ice. Cell lysate was precipitated with conacavalin A, separated by SDS-PAGE, blotted and analyzed for KDR phosphorylation using phosphotyrosine specific and KDR antibodies. We generated cells with strong expresssion of either VEGF or mouse ES. The cells were encapsulated in alginate beads and injected s.c. to SCID mice divided in the following groups: A) control, B) VEGF, C) ES, D) VEGF + ES with 5 animals per group. After 5 days lungs were harvested and analyzed by H&E staining of tissue sections.

RESULTS: In an in vitro permeability assay VEGF enhanced permeation of dye through a monolayer of endothelial cells. ES significantly inhibited VEGF induced permeability (fig.1). ES alone had no effect compared to controls. On the molecular level VEGF causes phosphorylation of its receptor VEGFR-2 as seen in VEGFR-2 expressing cells (fig.2). This effect was abolished by coincubation with ES showing a direct antagonism of VEGF signalling by ES. In a SCID mouse model animals were treated with VEGF, ES or the combination of both (fig.3). Animals in the VEGF group developed general edema and lung injury resembling acute lung injury with extravasation, accumulation of inflammatory cells and hemorrhage. The animals treated with a combination of VEGF and ES had less generalized edema. The lung sections showed alterations less pronounced than in the VEGF group. ES alone had no effect.

CONCLUSIONS: Our results demonstrate that ES inhibits VEGF-induced permeability by blocking VEGFR-2 activation. ES treatment partially restored VEGF-induced lung injury in vivo. The incomplete inhibition may be due to excess VEGF protein levels. Taken together, VEGF blocking with ES may serve as a useful new treatment option for conditions with increased vascular permeability like acute lung injury.

Author notes

Corresponding author