Fig. 1

Cortistatin regulates the integrity of b.End5 cells exposed to different insults. a Schematic representation of the experimental design. Murine b.End5 cells were exposed to lipopolysaccharide (LPS), glucose deprivation (GD) for 24 h, or to 4 h of oxygen–glucose deprivation followed by 20 h of reoxygenation (OGD-R). Cortistatin (CST, 100 nM) was applied simultaneously to the insult (LPS + CST; GD + CST) or during recovery (OGD-R + CST). Cells incubated in normoxia/normoglycemia (NX/NG) were used as a reference. b–d Endothelial integrity was assessed by evaluating permeability to Evans Blue-Albumin (EBA) and Sodium Fluorescein (NaF) (b), and tight-junctions assembly (c, d). b Permeability is represented as the index (%) of the tracer permeability vs an empty-coated insert. N = 4–6 cultures/group. c, d Representative immunofluorescence images of the cellular distribution of ZO-1 (c, red) and claudin-1 (d, green) in b.End5 cells after GD or OGD-R in the absence or presence of cortistatin. Delocalization of ZO-1 was evaluated through the ratio of ZO-1 staining intensity in the membrane vs the cytosol (c, arrowheads indicate gaps in ZO-1 membrane location). After cortistatin addition, ZO-1 junctional rearrangement was observed in the membrane (asterisks). Claudin-1 intracellular overexpression was quantified by fluorescence intensity (d). Cytosolic expression of claudin-1 was augmented in GD/OGD-R (arrowheads). Cortistatin drastically decreased claudin-1 intracellular expression (asterisks). 25–50 selected ROIs in 4 independent fields were analysed (expressed as arbitrary units, a.u). N = 6 cultures/group. Scale bar: 20 µm. e, f Levels of inflammatory cytokines TNF-α, IL-6, chemokine MCP-1 (all in ng/ml) and nitrite (µM) were determined in culture supernatants after GD (e) or OGD-R (f) with or without cortistatin addition. N = 4–6 cultures/group. Data are the mean ± SEM with dots representing individual values of the independent cultures. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001