Contribution of Factor VIII A3 Domain Residues 1793-1795 to a Factor IXa-Interactive Site

Factor (F) VIII functions as a cofactor in the tenase complex responsible for phospholipid surface-dependent conversion of FX to FXa by FIXa. FIXa binding sites were identified on A2, A3, and C2 domains of FVIII. Some earlier studies confirmed the FIXa-binding sites on residues 1811-1818 in FVIII A3 domain. According to the 3-D model of FVIIIa, the residues 1790-1798 in FVIII A3 domain form a loop and exist just close to residues 1811-1818 on the FVIIIa extended surface and might contribute to one of the FIXa binding regions. We, therefore, hypothesized that residues 1790-1798 might be one of the FIXa-binding components, and prepared synthetic peptides (1811-1818 and 1790-1798). The 1811-1818 peptide inhibited the FVIII light chain-EGR FIXa interaction (apparent K_i;10.5 ± 4.0 μM) on ELISA. The 1790-1798 peptide also inhibited these interaction (apparent K_i ;4.2± 1.2 μM), suggesting that the 1790-1798 region contributed to FIXa-interactive site. In general, the binding sites of serine protease in FVIII contain the acidic or basic rich residues. The residues 1790-1798, and 1811-1818 contain basic residues E1793, E1794, D1795, E1811, K1813, and K1818. To investigate the significance of these residues for FIXa-binding, the mutant forms of the A3 domain, converted to alanine, in BHK system and purified. Furthermore, because previous study demonstrated that F1816 might contribute to FIXa binding, we also prepared F1816A. Compared with wild type FVIII (K_d; 6.3± 0.3 nM), the binding affinity of F1816A and E1793A/E1794A/D1795A mutants for EGR FIXa were decreased by 1.4-fold (K_d; 9.1± 0.2 nM) and 1.3-fold (K_d; 8.4± 0.4 nM), respectively, on SPR-based assay, suggesting contribution of F1816, and E1793/E1794/D1795 to the binding interactions. On the other hand, the binding affinities of E1811A (K_d; 6.3± 2.1 nM) and K1818A (K_d; 5.1± 1.8 nM) mutants for FIXa were almost the same as that of wild type. Interestingly, the binding affinity of K1813A (K_d; 3.9± 0.7 nM) for FIXa was increased compared to wild type FVIII. For the functional evaluation of the association with FVIII mutants to FIXa, FVIII mutants were reacted with varying concentrations of FIXa in a FXa generation assay. F1816A and E1793A/E1794A/D1795A mutants decreased the FIXa affinity by 1.6-fold (K_m; 10.6± 1.0 nM) and 2.8-fold (K_m; 18.5± 4.3 nM), respectively, compared to wild type FVIII (K_m; 6.6 ± 1.0 nM). These data suggested that not only F1816A but E1793A/E1794A/D1795A mutations contributed high K_mand low catalytic efficiency in FXase complex. E1811A (K_m; 7.2 ± 0.7 nM), K1813A (K_m; 4.5 ± 0.7 nM), and K1818A (K_m; 4.5 ± 0.2 nM) mutants showed the similar FIXa affinities to wild type FVIII. Furthermore, to confirm that the F1816A and E1793A/E1794A/D1795A mutants were responsible for FIXa association, FIXa cleavages in these mutants were evaluated by SDS-PAGE and Western blot using anti-A1 monoclonal antibody. Compared with wild type FVIII, cleavages at Arg336 were slightly delayed in F1816A. On the other hand, those in E1793A/E1794A/D1795A were significantly delayed, indicative of contribution of these residues for FIXa bindings. After all, these results indicate that the 1790-1798 region in the FVIII A3 domain, and in particular a cluster of basic amino acids at residues 1793-1795, contributes to one of the FIXa-interactive sites.