Signal transduction pathway in endothelial dysfunction.

BACKGROUND: Endothelial dysfunction is an important feature of sepsis, acute respiratory distress syndrome (ARDS), and other infectious conditions. Previously, we reported an in vitro model to study endothelial dysfunction, in which endothelial cells are induced to form capillary tube networks by culturing on a basement membrane matrix (Matrigel). In this study, we defined the signal transduction pathways that lead to endothelial cell function and capillary disruption characteristic of sepsis and other infectious conditions. METHODS: Human aortic endothelial cells (HAEC) were cultured on a laminin-rich matrix to form capillary-like networks. The HAECs were treated with a protein tyrosine phosphatase inhibitor (sodium orthovanadate), a phosphoinositon-3-phosphate inhibitor (wortmannin), or a protein kinase C inhibitor (bisindolylmaleimide) before capillary tubes had formed or after the capillary tubes had matured. The degree of capillary tube formation was quantified by counting the intersection of capillary networks in triplicate wells. Statistical significance was determined by analysis of variance. RESULTS: Endothelial dysfunction occurred after inhibition of protein tyrosine phosphatase or protein kinase C. Whereas inhibition of phosphoinositon-3-phosphate did not cause endothelial dysfunction, sodium orthovanadate (2-20 microM) and bisindolylmaleimide (2-10 microM) significantly reduced capillary networks. The mean +/- SD of the number of capillary tubes in the control, sodium orthovanadate-treated, and bisindolylmaleimide-treated groups were 251.0 +/- 7.0, 65.6 +/- 9.9 (p < 0.001), and 181.7 +/- 0.1 (p < 0.001), respectively. Sodium orthovanadate (20-200 microM) and bisindolylmaleimide (10-100 microM) inhibited capillary tube formation. At higher concentrations, sodium orthovanadate (> 200 microM) and bisindolylmaleimide (>100 microM) disrupted mature capillary tubes. CONCLUSIONS: Our results suggest that PKC and protein tyrosine phosphatase play a role in endothelial dysfunction by interfering with the phosphorylation signals within endothelial cells. These mechanisms may be important in the endothelial dysfunction in sepsis and other infectious conditions.

beta1-integrin-ligand disengagement induces in vitro capillary tube disruption mediated by p38 MAPK activity.

BACKGROUND: Disrupting cell-matrix interactions may lead to capillary injury as seen in sepsis and transplant rejection. Previously, we demonstrated capillary disruption mediated by beta1-integrin-ligand disengagement. We now determine whether p38 mitogen-activated protein kinase (MAPK) and extracellular signal-regulated kinase (ERK) pathways are involved in this capillary injury. METHODS: Endothelial capillaries on Matrigel were preincubated with a p38 MAPK inhibitor (SB203580), ERK pathway inhibitor (PD98059), or dimethyl sulfoxide. Subsequently, a beta1-integrin blocking (P5D2) or an irrelevant antibody was added. After 24 hours, capillary integrity was quantified as capillary intersections/well. Antibody-treated cell lysates then were immunoprecipitated with either a phospho-p38 MAPK or phospho-ERK1/2 antibody. Kinase activity was measured with ATF-2 and Elk-1 fusion proteins as substrates for p38 MAPK and ERK, respectively, followed by Western blotting. RESULTS: P5D2 disrupted capillary tubes. Increased p38 MAPK activity at 8 hours and ERK activity at 2 and 8 hours were seen in P5D2-treated lysates. Preincubation with SB203580, but not with PD98059 or DSMO, significantly reduced capillary tube disruption. CONCLUSIONS: The beta1-integrin-ligand disengagement resulted in capillary disruption and stimulated p38 MAPK and ERK activity. In spite of activation of both pathways, the p38 MAPK but not the ERK pathway inhibitor prevented beta1-integrin antibody effects. Inhibiting p38 MAPK may mitigate capillary injury associated with sepsis and transplant rejection.