Abstract
Integrin αIIbβ3 is the essential platelet receptor for hemostasis and thrombosis. The plexin-semaphorin-integrin (PSI) domain is an approximately 54 amino acid sequence located near the N-terminus of the β3 subunit. We recently reported that the PSI domain possesses two CXXC motifs and contains endogenous thiol-isomerase activity. Targeting the PSI domain via our novel monoclonal antibodies (mAbs) reduced platelet aggregation in anticoagulated platelet-rich plasma in-vitro and inhibited thrombus formation under non-anticoagulated conditions in-vivo (Blood, 2017). This suggests that these anti-PSI mAbs may attenuate the PSI domain-mediated blood coagulation. Notably, the L33P polymorphism (Human Platelet Antigen 1) within the PSI domain has been associated with a greater risk of cardiovascular disease. However, the roles of the PSI domain and its L33P polymorphism in blood coagulation were not conceived and have never been explored.
Recombinant PSI (rPSI) and recombinant L33P (rL33P) proteins were generated using a BL21 E.Coli. Using a combination of reduced RNase, insulin β-chain reduction and MPB binding assays, we found that rL33P possessed greater thiol isomerase activity compared to rPSI. Using thromboelastography and clot retraction assays, we found that clot formation time was decreased in rPSI-treated blood, with rL33P further enhancing this phenomenon. Bacitracin attenuated these changes, elucidating that the role of PSI within coagulation is due to its thiol isomerase activity. Using scanning electron microscopy imaging to examine fibrin formation and structural phenotypes, we found smaller fibrin strand formation and increased branching for rPSI treatment groups. Moreover, rL33P treatment further increased branching, representing increased end-point coagulation. Interestingly, using APTT tests, an intrinsic coagulation pathway test, we found no significant difference between rPSI or rL33P-treated platelet-poor-plasma clot formation. However, PT assays, an extrinsic coagulation pathway-specific test, displayed significantly faster clotting times for rPSI and rL33P-treated platelet-poor-plasma. Recombinant tissue factor addition further enhanced extrinsic coagulation pathway activation for rPSI and further for rL33p. Excitingly, using biochemical binding assays isothermal titration calorimetry and biolayer interferometry, we discovered rPSI and rL33P directly bind tissue factor. Additionally, using circular dichroism, we found calcium to be integral in these interactions. In all models, rL33P presented tighter binding to tissue factor than rPSI, possibly through its enhanced thiol isomerase activity. Alphafold and pie-mole theoretical docking software's modeled alternative binding orientations between PSI and L33P proteins with tissue factor and cations.
Our results first demonstrate the PSI domain as a novel contributor to blood coagulation (i.e. cell-based model of coagulation) with the L33P polymorphism representing a gain of function mutation likely through enhanced thiol isomerase activity. This pro-coagulant state observed by L33P may contribute to the increased risk of cardiovascular events in patients carrying this integrin polymorphism. These findings elucidate integrins as a novel regulator of the coagulation cascade and support the development of therapeutics targeting the PSI domain.
