Characterization of new angiogenic molecules for vascular targeting

CD93 is a single-pass type I transmembrane glycoprotein, which together with Endosialin/TEM1, thrombomodulin and CLEC14A, belongs to the type 14 family of C-type lectins. CD93 is involved in apoptosis and inflammation and has been recently recognized as an important regulator of the angiogenic process able to act as both a soluble factor or a transmembrane protein. Much evidence suggests that CD93 promotes adhesion of endothelial cells. However, it is not known whether the extracellular domain of CD93 binds to extracellular matrix proteins. Here, we identified MMRN2, an extracellular matrix glycoprotein specifically expressed in the endothelium and involved in angiogenesis, as a ligand for CD93 and demonstrated the key role that the binding of CD93 to MMRN2 plays in the regulation of several EC functions including adhesion, migration and tube formation. In order to characterize the MMRN2-binding region, we generated deletion mutants of the extracellular domain of CD93 and identified the smallest CD93 fragment displaying interaction with MMRN2. Then, we used the recombinant protein fragment to inhibit the CD93-MMRN2 interaction in competition binding experiments. The recombinant fragment of CD93 reduced endothelial cell migration and tube formation confirming the role of this interaction in the regulation of angiogenesis. We found that the two proteins co-localized in the endothelium of several human cancer types, and disruption of the CD93-MMRN2 interaction, affected angiogenesis in vitro, making the binding of CD93 to MMRN2 an ideal target to block pathological angiogenesis. With the aim to generate new antiangiogenic molecules able to inhibit the interaction between CD93 and MMRN2 we performed model structure and docking studies to envisage residues hotspots directly or indirectly involved in the binding. By site directed-mutagenesis we observed that the F238 of CD93 represents an indispensable residue implicated in the complex formation. Consistent with this role, we showed that the expression of the F238T CD93 mutant in endothelial cells reduced their ability to migrate on the MMRN2 substrate. Since angiogenesis represents an underlying process in several human diseases including cancer, it is evident that the suppression of this process could be a valid approach for the treatment of many pathological conditions. We propose a molecular model in which the coiled-coil domain of MMRN2 is engaged by F238 of CD93, providing the basis for the development of new antiangiogenic molecules.