t-PA–mediated increase in BBB permeability is plasmin-dependent and unique to t-PA and its highly related variant tenecteplase. (A) Increasing concentrations of human t-PA were added to the luminal chamber of the in vitro BBB and the transfer of FITC-BSA to the abluminal chamber assessed after 24 hours (n = 5; ***P < .001, **P < .01 compared with “no cells”). (B) t-PA–mediated increase in permeability requires catalytically active t-PA. t-PA (1μM) and ct-PA (1μM) were added to the BBB as described in panel A, and permeability was assessed after 24 hours. As shown, inactive t-PA does not increase permeability of the BBB (n = 3). (C) t-PA–mediated increase in permeability is plasmin-dependent. t-PA (1μM) was added to the BBB as described in panel A in the absence or presence of α2–anti-plasmin (α₂AP; 1μM), and permeability was assessed after 24 hours. α₂AP blocked the ability of t-PA to increase permeability (n = 4). (D) Plasmin cannot substitute for t-PA to increase permeability. t-PA (1μM) or human plasmin (500nM) was added to the luminal chamber of the BBB, and permeability was assessed after 24 hours. As shown, plasmin alone fails to promote an increase in permeability (n = 3). (E-H) t-PA (1μM) was added to the luminal chamber of the BBB, and the extent of permeability compared at 24 hours with tenecteplase (TNK, 1μM, n = 4; E), reteplase (1μM), and its vehicle (n = 3; F), urokinase (u-PA, 1μM, n = 6; G), or desmoteplase (DSPA, 1μM, n = 6; H). Only t-PA and its structurally related variant TNK-tPA can augment BBB permeability. In all panels, bars represent mean ± SEM (in B-H, ***P < .001, *P < .05 compared with all other groups).