Endothelial Expression of Endothelin Receptor A in the Systemic Capillary Leak Syndrome

Abstract

Idiopathic systemic capillary leak syndrome (SCLS) is a rare and potentially fatal vascular disorder characterized by reversible bouts of hypotension and edema resulting from fluid and solute escape into soft tissues. Although spikes in permeability-inducing factors have been linked to acute SCLS flares, whether or not they act on an inherently dysfunctional endothelium is unknown. To assess the contribution of endothelial-intrinsic mechanisms in SCLS, we derived blood-outgrowth endothelial cells (BOEC) from patients and healthy controls and examined gene expression patterns. Ednra, encoding Endothelin receptor A (ETA)-the target of Endothelin 1 (ET-1)-was significantly increased in SCLS BOEC compared to healthy controls. Although vasoconstriction mediated by ET-1 through ETA activation on vascular smooth muscle cells has been well characterized, the expression and function of ETA receptors in endothelial cells (ECs) has not been described. To determine the role of ETA and its ligand ET-1 in SCLS, if any, we examined ET-1 levels in SCLS sera and functional effects of endothelial ETA expression. ETA overexpression in EAhy926 endothelioma cells led to ET-1-induced hyper-permeability through canonical mechanisms. Serum ET-1 levels were elevated in acute SCLS sera compared to remission and healthy control sera, suggesting a possible role for ET-1 and ETA in SCLS pathogenesis. However, although ET-1 alone did not induce hyper-permeability of patient-derived BOEC, an SCLS-related mediator (CXCL10) increased Edrna quantities in BOEC, suggesting a link between SCLS and endothelial ETA expression. These results demonstrate that ET-1 triggers classical mechanisms of vascular barrier dysfunction in ECs through ETA. Further studies of the ET-1-ETA axis in SCLS and in more common plasma leakage syndromes including sepsis and filovirus infection would advance our understanding of vascular integrity mechanisms and potentially uncover new treatment strategies.

Fig 1. Morphology and gene expression in BOEC from SCLS patients and controls.

(A) BOEC were expanded from peripheral blood. Image shows a representative EC monolayer immunostained with anti-VE-cadherin (green), F-actin (red), and DAPI (blue, nuclei). (B) Volcano plot of real-time qPCR array showing relative gene expression in SCLS BOEC relative to control on x axis and log p value on y axis.

Fig 2. Gene expression in BOEC from individual subjects.

(A-D) Relative expression of Ednra, Cx3cl1, Pf4, and Il1b in individual subjects as determined by real-time qPCR. *p = 0.03, Mann-Whitney test.

Fig 3. Expression and function of ETA receptors in endothelial cells.

(A-B) Relative Ednra and Ednrb expression in indicated cell lines was determined by qPCR relative to expression in EAhy cells (A) or vascular smooth muscle cells (B) set as ‘1’. (C) EAhy cells were transduced with ETA-YFP or GFP retroviruses followed by selection of flow sorting; relative fluorescence post-sort was determined by flow cytometry. (D) Expression and localization of ETA-YFP or GFP was examined by fluorescence microscopy. (E-G) Ca²⁺ concentrations were measured in the indicated cell lines following treatment with various concentrations of ET-1 by fluorimetry. In (F) cells were pre-incubated with the indicated receptor antagonists or vehicle alone as a control for one hour prior to stimulation with ET-1 (200 nM). Data are mean ± S.E.M. of 3–4 individual experiments; *p = 0.04, one-way ANOVA.

Fig 4. ETA mediates endothelial hyper-permeability.

(A-B) Translocation of FITC-albumin across EAhy monolayers plated in duplicate in Transwell inserts was monitored over time following treatment with vehicle alone, ET-1 (1 μM, A) thrombin (2 U/ml, B). Data are mean ± S.E.M. of 2–3 individual experiments. (C-D) Cells were stimulated with thrombin or ET-1 for the indicated times followed by fixation and immunostaining with phalloidin (F-actin, red, C) or VE-cadherin antibody (red, D). Images are from a single experiment representative of 3 similar experiments.

Fig 5. ETA-endothelin-1 axis in SCLS.

(A) Serum ET-1 levels in SCLS sera (basal or acute) or healthy control sera were determined by ELISA. **p = 0.01, one-way ANOVA. (B-C) Intracellular Ca²⁺ concentrations were determined by fluorimetry as in Fig 3 in BOEC left untreated or stimulated with thrombin (2 U/ml) or ionomycin (1 μM) (B) or), ET-1 (2 μM). Data are mean ± S.E.M. of quadruplicate measurements from a single experiment representative of 3 similar experiments. (D) BOEC were stimulated with thrombin (2 U/ml) or ET-1 (2 μM) for 20 min followed by fixation and staining with phalloidin (F-actin, red), anti-VE-cadherin (green), and DAPI (blue). Images are from a single experiment representative of 3 similar experiments using cells from different donors.

Fig 6. SCLS-associated mediator CXCL10 induce Ednra expression in endothelial cells.

(A-B) EAhy cells were left untreated (0 h) or stimulated with CXCL10 (40 ng/ml) for the indicated times followed analysis of Ednra expression by qPCR. Data are average of duplicate measurements from a single experiment (A) or mean ± S.E.M of four individual experiments (B); ***p = 0.0004; t test. (C) Edrna expression in BOEC from SCLS patients or controls (n = 4/group) left untreated or stimulated with CXCL10 (40 ng/ml) for 4 hr was determined by qPCR. Data are mean ± S.E.M.; ***p = 0.005, two-way-ANOVA. (D) BOEC from a subject with SCLS or healthy donor were pretreated with CXCL10 (40 ng/ml) for four hours prior to stimulation with the indicated concentrations of ET-1 or ionomycin and measurement of intracellular Ca²⁺ concentrations by fluorimetry. Bar graph is mean ± S.E.M. of quadruplicate values from a single experiment representative of 3 similar experiments.