Members of the family Filoviridae, including Ebola virus (EBOV) and Marburg virus (MARV), cause severe hemorrhagic fever in humans and nonhuman primates. Given their high lethality, a comprehensive understanding of filoviral pathogenesis is urgently needed. In the present studies, we revealed that the exchange protein directly activated by cAMP 1 (EPAC1) gene deletion protects vasculature in ex vivo explants from EBOV infection. Importantly, pharmacological inhibition of EPAC1 using EPAC-specific inhibitors (ESIs) mimicked the EPAC1 knockout phenotype in the ex vivo model. ESI treatment dramatically decreased EBOV infectivity in both ex vivo vasculature and in vitro vascular endothelial cells (ECs). Furthermore, postexposure protection of ECs against EBOV infection was conferred using ESIs. Protective efficacy of ESIs in ECs was observed also in MARV infection. Additional studies using a vesicular stomatitis virus pseudotype that expresses EBOV glycoprotein (EGP-VSV) confirmed that ESIs reduced infection in ECs. Ultrastructural studies suggested that ESIs blocked EGP-VSV internalization via inhibition of macropinocytosis. The inactivation of EPAC1 affects the early stage of viral entry after viral binding to the cell surface, but before early endosome formation, in a phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K)-dependent manner. Our study delineated a new critical role of EPAC1 during EBOV uptake into ECs.

Rationale: To maintain vascular patency, endothelial cells (ECs) actively regulate hemostasis. Among the myriad of pathways by which they control both fibrin formation and fibrinolysis is EC expression of annexin A2 (ANXA2) in a heterotetrameric complex with S100A10 [(ANXA2-S100A10) ₂ ]. This complex is a well-recognized endothelial surface platform for the activation of plasminogen by tissue plasminogen activator. A noteworthy advance in this field came about when it was shown that the cAMP pathway is linked to the regulation of (ANXA2-S100A10) ₂ in ECs. Objective: These findings prompted us to determine whether a druggable target, namely the exchange p rotein directly a ctivated by c AMP (EPAC) pathway, plays a role in vascular luminal fibrinolysis. Methods and Results: Taking advantage of our Epac1-null mouse model, we found that depletion of Epac1 results in fibrin deposition, fibrinolytic dysfunction, and decreased endothelial surface ANXA2 in mice, which are similar to phenomena discovered in ANXA2 -null and S100A10 -null mice. We observed upregulation of EPAC1 and downregulation of fibrin in endocardial tissues beneath atrial mural thrombi in humans. Of note, our thrombosis model revealed that dysfunction of fibrinolysis in EPAC1 -null mice can be ameliorated by recombinant ANXA2. Furthermore, we demonstrated that suppression of EPAC1 using a small-molecule inhibitor (ESI09) reduces the expression of ANXA2 in lipid rafts and impedes ANXA2 association with S100A10. Endothelial apical surface expression of both ANXA2 and S100A10 were markedly decreased in ESI09-treated ECs, which was corroborated by results from a nanoforce spectroscopy study. Moreover, inactivation of EPAC1 decreases tyrosine 23 phosphorylation of ANXA2 in the cell membrane compartment. Conclusions: Our data reveal a novel role for EPAC1 in vascular fibrinolysis, by showing that EPAC1 is responsible for the translocation of ANXA2 to the EC surface. This process promotes conversion of plasminogen to plasmin, thereby enhancing local fibrinolytic activity.