The Contribution of Hirudin-Like Acidic Sequences within the Factor Va Heavy Chain to Prothrombinase Function.

The molecular mechanism governing the ability of activated factor V (FVa) within prothrombinase to accelerate prothrombin activation is not well understood. It is possible that FVa provides a docking surface for prothrombin, and/or causes a conformational change in FXa which exposes an exosite that enhances the presentation of the substrate to the active site of the enzyme. While there is indirect evidence that FVa binds prothrombin via (pro)exosite I, it is not entirely clear whether this interaction directly leads to the enhanced function of prothrombinase. It has been speculated that the corresponding binding site for (pro)exosite I on FVa is contributed by hirudin-like acidic sequences at the C-terminus of the heavy chain (659–709). In support of this, peptide inhibition studies indicate that residues Asp⁶⁹⁵ to Tyr⁶⁹⁸ are required for optimal prothrombinase function and likely provide a binding site for prothrombin. To directly assess the role of this region in prothrombin activation we exploited a novel strategy in which recombinant FVa (rFVa) variants truncated at the C-terminus of the heavy chain were produced. The entire B-domain and heavy chain sequences were removed and a PACE-furin cleavage site was introduced between these deleted residues and the light chain generating: FVa⁶⁵⁹ (des^660–1545), FVa⁶⁷⁹ (des^680–1545), FVa⁶⁹³ (des^694–1545), FVa⁷⁰⁰ (des^701–1545), and FVa⁷⁰⁹ (des^710–1545). These single-chain proteins were intracellularly processed and secreted into the media as two-chain proteins with appropriate truncations to the heavy chain. Factor Va⁷⁰⁹ served as a control as it is chemically identical to plasma-derived (PD)-FVa and rFVa. Following stable expression and purification in high yield, SDS-PAGE and N-terminal sequence analysis suggested the derivatives were correctly processed. Factor Va⁷⁰⁹, rFVa and PD-FVa had high specific activities (~1500 U/mg) in a one stage PT-based clotting assay. Surprisingly, we found that FVa⁶⁷⁹, FVa⁶⁹³, and FVa⁷⁰⁰ exhibited specific clotting activities comparable to the wild-type cofactors. Detailed kinetic studies using prothrombin and saturating amounts of cofactor revealed that both the K_m and k_cat values for each of these variants were essentially equivalent to wild-type FVa. We observed minor differences (<4-fold) in the kinetic parameters when using prethrombin-1. Direct fluorescent and kinetic measurements also indicated that these variants bound FXa-membranes with high affinity. An exception to these findings was FVa⁶⁵⁹, which only had a minor reduction in the k_cat for prothrombin (~1.5-fold), but had ~1–2% clotting activity and an increased K_m (4-fold) and decreased k_cat value for prethrombin-1 (>13-fold). Direct binding fluorescent measurements indicated that this variant had a marked reduction in the binding affinity for FXa-membranes (>15-fold). Additionally, the maximal fluorescent signal was reduced (~2-fold), further suggesting that this derivative interacts with FXa-membranes differently compared to wild-type FVa. Surprisingly, these data demonstrate that deletion of residues 679–709 does not impair cofactor function towards the physiological substrate prothrombin. Additionally, deletion of residues 659–709 only appears to significantly compromise FXa binding, a finding that likely explains the reduced functional activity of FVa⁶⁵⁹. Our results suggest that any possible binding interactions between prothrombin and the C-terminal acidic region on the FVa heavy chain do not contribute in a detectable way to the enhanced function of prothrombinase.