Intraventricular haemorrhage (IVH) occurs in up to 50% of patients with primary intracerebral haemorrhage and aneurysmal subarachnoid haemorrhage. It is a significant and independent contributor to mortality and morbidity in these intracranial haemorrhages. Using a model of isolated IVH, we assessed the morphological changes induced by intraventricular bleeding and investigated the effects of intraventricular fibrinolytic treatment following IVH. IVH was induced in 32 pigs by intraventricular infusion of 10 ml autologous blood along with thrombin. The treatment group received an intraventricular injection of 1.5 mg (1 mg/ml) tissue plasminogen activator (tPA) following the injection of blood. The placebo group received the same volume of normal saline. Morphological examinations of the brains were carried out 7 days and 6 weeks following IVH. The ventricles were incompletely filled with blood and significantly enlarged in the placebo group 7 days after the IVH. In contrast, no residual intraventricular clots were visible in the animals treated with tPA, and the diameters of the lateral ventricles had returned to normal within 7 days. Marked losses of the ependymal covering of the ventricular walls were found in the placebo-treated animals, while the ependymal layer was largely intact in the animals treated with tPA. No haemorrhages induced by tPA were observed. The results indicate that intraventricularly administered tPA significantly enhances the lysis of intraventricular blood clots, accelerates the resolution of acute posthaemorrhagic hydrocephalus, and preserves the integrity of the ependymal layer.

Upregulation of urokinase receptors (uPAR) is common during tumor progression and thought to promote metastasis. UPAR directly binds urokinase (uPA) and a set of beta1 integrins, but it remains unclear to what degree uPAR/integrin binding is important to beta1 integrin signaling. Using sitedirected mutagenesis we have identified single amino acid residues in uPAR domains 2 and 3 that fail to associate with either alpha3beta1 (D262A) or alpha5beta1(H249A) but associate normally with uPA.To study the effects of uPAR mutations on beta1 integrin function, we first stably silenced uPAR by expression of hairpin RNAi in multiple tumor cell lines: HT1080, H1299, and H838. Knockdown of uPAR in tumor cells resulted in markedly reduced fibronectin (Fn) and alpha5beta1-dependent ERK phosphorylation, MMP9 expression, and lamellipodia formation. UPAR cDNAs were then constructed that were insensitive to RNAi knockdown and used to express wild type or mutant uPARs in knockdown cells. Re-expression of wt or D262A mutant uPAR but not the non-alpha5beta1 binding H249A uPAR reconstituted Fn responses. Because uPAR/alpha5beta1 binds Fn in exons 12-14 whereas alpha5beta1 binds in exons 9-11, we tested whether signaling and MMP9 responses were binding site dependent. Only cells engaging both Fn exons 9-11 and exons 12-14 activated ERK and upregulated MMP9. Thus uPAR is required for maximal alpha5beta1-dependent responses to Fn and this operates through enhanced signaling following integrin attachment to separate sites on Fn by alpha5beta1 and uPAR-bound alpha5beta1. These findings provide a direct link between uPAR expression and the metastatic phenotype.