Abstract

Protein C (PC) activation by thrombin on endothelial cells normally involves thrombomodulin (TM) and the endothelial cell protein C receptor (EPCR), and dysfunction of this mechanism is implicated in a variety of pathologies. Platelet factor 4 (PF4), a releasable platelet alpha-granule protein, stimulates PC activation in vitro and in vivo; PF4 stimulates PC activation in the presence but not in the absence of TM and this requires the PC Gla-domain (Slungaard et al). In contrast to PC activation, we found that thrombin/TM-mediated activation of thrombin-activatable fibrinolysis inhibitor (TAFI) is inhibited by PF4. In assays where TAFI-dependent inhibition of clot lysis was promoted by thrombin/TM-mediated TAFI activation, PF4 dose-dependently inhibited TAFI’s anti-fibrinolytic effects. In the absence of TM, PF4 did not affect TAFI activation by thrombin alone and did not alter the clot-lysis times in normal plasma. Thus, the effects of PF4 on thrombin/TM actions are enhancing for PC activation and inhibitory for TAFI activation such that as PF4 concentration increases, the ratio for relative rates of PC and TAFI activation changes more than 40-fold to favor selective PC activation. The current model to explain PF4 stimulation of PC activation involves electrostatic bridging of the negatively charged Gla-domain of PC and the negatively charged glycosaminoglycan moiety of TM by the highly positively charged PF4. In assays using purified soluble reactants components, EPCR abrogated PF4 stimulation of PC activation by thrombin/TM. On 293 cells expressing TM but not EPCR, PF4 enhanced PC activation by thrombin, and this effect was inhibited by soluble EPCR (IC50 ~ 1.0 μM). Based on these and previous observations, we hypothesized that the binding sites for EPCR and PF4 on the Gla-domain of PC may overlap and, if so, that PF4 could interfere with EPCR-dependent APC cytoprotective activities involving protease activated recpotor-1 activation. Indeed, studies showed that PC and APC binding to 293 cells expressing EPCR was inhibited by PF4 (IC50 ~ 1.0 μM) and that PF4 concentrations which inhibited APC binding to EPCR significantly impaired EPCR-dependent anti-apoptotic activity of APC in assays of staurosporine-induced endothelial cell apoptosis. PF4 enhanced EPCR-dependent PC activation on endothelial cells at PF4 concentrations which did not significant inhibit PC binding to EPCR. On K293 cells transfected with both TM and EPCR, PF4 at low concentrations enhanced EPCR-dependent PC activation. Anti-EPCR antibodies abrogated both EPCR-dependent and PF4-dependent stimulation of PC activation on cells containing TM and EPCR, further suggesting a direct interaction between PF4 and EPCR. Proof for this interaction came from experiments showing that PF4 bound to immobilized soluble EPCR. Thus, these results suggest that on endothelial cells, PF4 enhances APC generation by enhancement of the ability of both TM and EPCR to promote PC activation while inhibiting TAFI activation. Therefore, the current model for stimulation of PC activation by PF4 involving thrombin/TM-PF4-PC quaternary ternary complex interactions does not adequately predict the effects of PF4 on PC activation on cells in the presence of EPCR. We propose a new model to explain the effects of PF4 on the generation of APC on endothelial cells in which the tetrameric character of PF4 promotes multiple simultaneous interactions of PF4 in a macromolecular complex with thrombin, TM, PC and EPCR.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author