Thrombin uses PAR1-dependent and -independent mechanisms to regulate the angiogenic functions of microvascular and large vessel endothelial cells

CE Farrar1, P Kitsche1, R Herranz2, J Mitchell3, CPD Wheeler-Jones1. 1The Royal Veterinary College, London, UK, 2NHLI, Imperial College London, London, UK, 3Instituto de Qu í mica Mé dica, Madrid, Spain

There is emerging evidence that protease-activated receptors (PARs), novel G-protein coupled receptors implicated in inflammatory signalling, are important in the reparative angiogenesis associated with inflammation. The serine protease, thrombin, is known to cleave and activate both PAR1 and PAR4 with previous evidence supporting a prominent role for PAR1 as a mediator of thrombin’s functional effects on endothelial cells (ECs). We and others have shown that thrombin increases capillary-like tube formation by human umbilical vein ECs (HUVEC) on Matrigel. However, the effects of thrombin on the angiogenic functions of microvascular ECs (MEC) have yet to be defined and the contribution of individual PARs established.

To begin to define roles for PARs in MEC function the kinetics of MEC and HUVEC differentiation on Matrigel were analysed using real-time imaging. HUVEC and human lung MECs (HLMEC) were serum-deprived for 1 hour and plated (15 000 cells/well) onto growth factor-reduced Matrigel. Cells were exposed to thrombin (0.01 - 1 U/ml), or to PAR-1- (TFFLR; 10 µM) or PAR-4 (AYPGKF; 50 µM)-selective agonist peptides (PAR1-AP and PAR4-AP, respectively). Cells were imaged continuously over 16 hours and the number of inter-branches formed over time monitored. A two-way ANOVA was used to assess statistical significance of treatments which were performed in triplicate. PAR1-AP significantly inhibited (0.8-fold; p<0.0001) HUVEC differentiation on Matrigel whereas thrombin and PAR4-AP significantly increased capillary-like tube formation by 1.5- and 1.6-fold, respectively (p<0.001 and p<0.0001). PAR4-AP showed a trend towards promoting (1.4-fold), whereas PAR1-AP showed a trend towards inhibiting (0.4-fold), HLMEC differentiation. In contrast, thrombin markedly suppressed tube formation by HLMEC in a concentration-dependent manner; treatment with 1 U/ml thrombin resulted in a 0.8-fold inhibition in differentiation (p<0.05). The difference between the effects of thrombin versus PAR1-AP on tube formation by HUVEC, as well as the opposing actions of thrombin on HUVEC and HLMEC, raises questions about the receptor(s) mediating thrombin’s cellular effects. Hirudin (5 U/ml), an inhibitor of thrombin’s protease activity, partially reduced the effects of thrombin on tubulogenesis by both HUVEC and HLMEC, confirming the involvement of proteolysis in thrombin’s actions. The highly selective PAR-1 antagonist RWJ-58259 (10 µM) completely reversed the inhibitory effects of PAR1-AP (p<0.0001) on tube formation by HUVEC but did not modify the stimulatory effect of thrombin on EC differentiation. The PAR4 pepducin antagonist, p4-pal10 (10 µM), however, completely abrogated thrombin’s pro-angiogenic effects on HUVEC (p<0.01). A preliminary experiment demonstrated that the anti-angiogenic actions of thrombin on HLMEC were partially rescued by treatment with RWJ-58259 (inter-branches: control – 48 ± 2; RWJ – 38 ± 8; thrombin (1 U/ml) – 12 ± 2; thrombin plus RWJ – 30 ± 3). These data suggest that the repertoire of PARs utilized by thrombin to influence the angiogenic functions of ECs differs depending upon EC origin and provide evidence for PAR-1-independent regulation of EC angiogenic potential.