Factor V (FV) circulates as procofactor with little or no procoagulant activity and is activated upon proteolytic removal of a central B-domain. Recently, we have shown that discrete B-domain sequences stabilize the inactive procofactor state and that their deletion drives the expression of procoagulant function without the need for proteolytic processing (

JBC
2007
;
282
:
15033
). While the B-domain length is highly conserved in most mammals (∼800 aa), recent genomic data indicates that in some vertebrates the FV B-domain is dramatically shortened. The most striking example is found in an Australian snake family (P. textilis and O. scutellatus). These snakes, which are among the most venomous in Australia, have a powerful prothrombin activating complex in their venoms consisting of FXa-like and FVa-like components. Remarkably, both plasma FV and venom FV sequences of these snakes predict a B-domain of only 45 and 46 residues, suggesting that snake FV may have lost B-domain sequences that maintain the protein as procofactor. Alternatively, snake FV could use a different strategy to preserve the procofactor state. To gain insight into this, we expressed and purified venom-derived P. textilis FV (pt-FV) in BHK cells. Pt-FV expressed very well (4mg protein/L media), and SDS-PAGE analysis revealed that it migrated as a single-chain protein with a molecular mass of ∼180 kDa. Upon incubation with thrombin (IIa), pt-FV was completely processed to pt-FVa, yielding fragments similar to the human FVa heavy and light chains. Surprisingly, pt-FVa migrated as a single band on a non-reducing gel, indicating that the heavy and light chains are held together by a disulfide bond. Sequence analysis suggests that this disulfide bond is located between the A2 and A3 domains. Functional analysis using a PT-based clotting assay with human FV-deficient plasma demonstrated that pt-FV has low activity compared to human FVa. This is consistent with previous studies which showed that venom FV is a relatively ineffective cofactor when using bovine factor Xa (FXa), suggesting that it needs snake-derived FXa to optimally function. Despite the low reactivity towards human plasma, we were able to demonstrate that processing of pt-FV to pt-FVa did not significantly increase cofactor activity (activation quotient of 2 compared to 10–15 for human FV). This suggests that unlike all nonhuman FV variants characterized so far, pt-FV is expressed as a constitutively active cofactor. Consistent with this, assessment of cofactor activity using a purified prothrombinase assay with either human FXa or transiently expressed O. scutellatus FXa revealed that the initial rates of prothrombin activation were equivalent for pt-FV or pt-FVa. To further confirm these results we expressed and purified pt-FV in which IIa cleavage sites Arg709 and Arg1545 were eliminated (pt-FV-QQ). SDS-PAGE analysis showed that pt-FV-QQ was not cleaved by IIa, yet functional measurements revealed that its activity was similar to pt-FV or pt-FVa. These data indicate that pt-FV has constitutive cofactor activity and thus bypasses the normal proteolytic processing step to function within prothrombinase. This is in agreement with our previous findings that discrete sequences within the FV B-domain are necessary to maintain FV as procofactor. Thus, venom-derived P. textilis FV represents the first example of a naturally occurring FV variant that does not require proteolytic processing to be active.

Author notes

Disclosure:Employment: Martin F. Lavin - employed by QRxPharma Ltd.

Sign in via your Institution