Integrins are involved in pro-uPA activation. (A) Gelatin zymography of endothelial cell lysates. The 62-kDa MMP-2 band significantly increases in intensity upon VEGF₁₆₅ (50 ng/mL) or S-RGD (10 μM) stimulation, while 2 mM manganese (Mn⁺⁺) blocks the VEGF₁₆₅ effect. (B) Pro-uPA levels measured by cell ELISA. Mimicking integrin ligand binding by soluble tetrapeptide S-RGD (10 μM) resulted in pro-uPA activation, while 2 mM manganese (Mn⁺⁺) inhibited the VEGF₁₆₅-induced decrease of surface pro-uPA. Wortmannin (100 nM) did not affect the S-RGD-induced pro-uPA activation; also GDP-β-S (20 μM) had no effect on VEGF₁₆₅-induced pro-uPA activation. Mean ± SDs; n = 4; ^**P < .01. (C) VEGF₁₆₅-induced increase in fibrinolytic activity was measured by a d-dimer ELISA. The soluble tetrapeptide S-RGD (10 μM) increased the fibrinolytic activity, while manganese (2 mM) or coincubation of manganese and wortmannin (100 nM) inhibited the VEGF₁₆₅ effect. Mean ± SDs. ^*P < .05; ^**P < .005. (D) Fluorometric detection of β₁ integrin in “on”-conformation on endothelial cells by an activity-specific anti-β₁ integrin antibody (HUTS-4). Significant change in conformation of β₁ integrins by VEGF₁₆₅ (50 ng/mL, 60 minutes) shows PI3-kinase dependency (100 nM wortmannin). In the presence of 2 mM manganese (Mn⁺⁺) the effect of VEGF₁₆₅ was blocked. The soluble tetrapeptide S-RGD (10 μM) decreases HUTS-4 immunoreactivity significantly (n = 3). GDP-β-S (20 μM) could not influence VEGF₁₆₅-induced decrease of HUTS-4 immunoreactivity. For statistical analysis, one-way analysis of variance and Dunnett tests as posttests were used; significance was assigned to P values less than .05 (^*P < .05, ^**P < .01). Mean ± SDs; n = 7.