CONTRASTING EFFECTS OF PLASMINOGEN ACTIVATORS, UROKINASE RECEPTOR, AND LDL RECEPTOR-RELATED PROTEIN ON SMOOTH-MUSCLE CELL-MIGRATION AND INVASION

Smooth muscle cell (SMC) migration is an early response to vascular in jury and contributes to the development of intimal thickening. Upregul ation of several components of the plasminogen activator (PA) system h as been documented after vascular injury. Utilizing a Transwell filter assay system and human umbilical vein SMCs, we sought to define the r ole of four different PA system components on SMC migration and matrix invasion: (1) PAs, (2) plasmin, (3) PA receptors, and (4) PA clearanc e receptors tie, low density lipoprotein receptor-related protein (LRP ]). Addition of active two-chain urokinase-type PA (UPA) stimulated ra ndom migration (192+/-30% of control, 0.36 nmol/L, P<.001). The stimul ation was inhibited by pretreatment with diisopropylfluorophosphate, P A inhibitor type 1 (PAI-1), or aprotinin, a plasmin inhibitor. Augment ed migration was also observed with either low-molecular-weight UPA dr the amino terminal fragment of UPA (ATF), with the effects being addi tive. Stimulation by ATF alone, however, was not inhibited by aprotini n. The stimulatory effect was not specific for UPA, in that tissue-typ e PA (TPA) also increased migration (169+/-9% of control, 10 nmol/L, P <.001); the augmentation was inhibited by pretreatment with DFP, PAI-1 , or aprotinin and was additive to the UPA effect. Antibodies to the U PA receptor but not 5'-nucleotidase (another glycosylphosphatidylinosi tol-anchored cell surface protein) inhibited baseline and UPA-stimulat ed migration. Similarly, both UPA and TPA stimulated invasion of a col lagen gel; this augmentation was inhibited by aprotinin, whereas antib odies to the UPA receptor reduced baseline invasion. Finally, we teste d whether inhibition of LRP function, which mediates internalization o f PA/inhibitor complexes, affected either process. Both antibodies to LRP and recombinant receptor associated protein, a known inhibitor of ligand binding to the LRP, significantly inhibited migration but did n ot affect collagen gel invasion. These data demonstrate the ability of several components of the PA system to modulate SMC migration and inv asion in vitro via plasmin-dependent and -independent mechanisms.