Investigation Into the Mechanism of Action and Binding Site of BAX 499, An Aptamer Against Tissue Factor Pathway Inhibitor

Abstract 1214

Hemophilia is a bleeding disorder characterized by a deficiency in coagulation factor VIII or IX, rendering the body incapable of maintaining hemostasis. BAX 499 is a therapeutic aptamer that inhibits tissue factor pathway inhibitor (TFPI) and is under investigation for treatment of hemophilia patients. BAX 499 interferes with TFPI inhibition of both factor Xa (FXa) and the tissue factor/factor VIIa complex in vitro and improves thrombin generation and clotting parameters in hemophilia plasma. BAX 499 also improves coagulation in a hemophilia A-like cynomolgus monkey model, as determined by ex vivo thromboelastography and bleeding time measurements.

In order to understand the mechanism by which BAX 499 inhibits TFPI, we carried out studies to determine the BAX 499 binding site on TFPI and compared BAX 499 activity to domain-specific anti-TFPI antibodies. The binding site of the aptamer on TFPI was determined by performing binding experiments with truncated TFPI proteins, aptamer-antibody competition-binding experiments, and hydrogen-deuterium exchange mapping experiments. The aptamer binds tightly to full-length TFPI (K_D ∼ 3 nM), less well to a protein missing some of the C-terminal region of TFPI, very weakly to a protein that contains only the Kunitz 3 (K3) and C-terminal domains, and not at all to a protein that contains only the Kunitz 1 (K1) and Kunitz 2 (K2) domains. Domain-specific monoclonal antibodies were used as competitors in a nitrocellulose dot blot binding assay using radiolabeled aptamer and in an ELISA assay that monitors aptamer binding to plate-bound TFPI. In both assays, K1 and K3 domain antibodies competed with aptamer for binding, a C-terminal domain antibody partially competed, and a K2 domain antibody did not compete. When hydrogen-deuterium exchange was carried out in the presence of aptamer, residues within the K1, K3, and C-terminal domains of TFPI were protected (residues 15–69 and 191–272). Taken together, these data suggest that the binding epitope of BAX 499 on TFPI spans the entire protein molecule, explicitly requiring the K1 and K3 domains, and is partially dependent on the C-terminal domain.

Activity assays demonstrate that the inhibition profile of BAX 499 is similar to that observed with anti-TFPI antibodies that bind to the K1, K3, and C-terminal domains of the protein. In an assay measuring inhibition of FXa activity by TFPI, antibodies against the K2 domain fully inhibited TFPI and restored FXa activity. In contrast, BAX 499 and antibodies against the K1, K3, or C-terminal domains of the protein did not fully inhibit TFPI, resulting in partial restoration of FXa activity. Examination of aptamer binding to TFPI in the presence of FXa suggests that the aptamer and protein can bind to TFPI simultaneously, highlighting a potentially unique mechanism for TFPI inhibition.

Binding of BAX 499 to TFPI appears to involve regions across the entirety of the protein, with the exception of the K2 domain. As the K2 domain is primarily responsible for binding to and inhibiting FXa, these experiments suggest that the aptamer has a unique mechanism of interfering with TFPI inhibition of FXa activity. This mechanism of action is able to mediate a procoagulant response in models of hemophilia and may prove effective for the treatment of hemophilia patients.