Differential Role of Plasma and Platelet-Derived Factor V In Vivo

Coagulation factor V (FV) has a profound effect on coagulation and deficiency states lead to bleeding. In blood, there are two pools of FV with 80% found in plasma and 20% contained in the α-granules of platelets. The biological basis for this separation is not clear; however, structural distinctions between the pools have been identified suggesting they may have unique effects or play different roles following injury. Broadly however, in vivo evidence for this is lacking. Transgenic mice with synthesis limited to liver or megakaryocytes have been generated to study the role of plasma and platelet FV (Sun et al, Blood 102:2856, 2003). Either pool is sufficient for basal hemostasis and tail-bleeding times were only slightly prolonged compared to wild-type (wt) mice. The impact of the two pools following different hemostatic challenges in different vessels is not known. To expand our in vivo understanding of the FV pools on clot formation, we further studied mice with only platelet FV (platelet-FV⁺) or plasma FV (plasma-FV⁺). As a control, we also generated mice with essentially little to no total FV (FV-low). To generate platelet-FV⁺ mice, wt mice were treated with a single dose of a liver targeted antisense oligonucleotide (ASO) to knockdown plasma FV (ASO-FV; 40 mg/kg). After 3 weeks of ASO-FV treatment, the plasma FV levels assessed with a murine FV-specific ELISA were at the lower limits of detection while platelet FV levels were normal. This confirms prior work showing platelet FV originates from megakaryocyte synthesis in mice, which is different from humans. To generate plasma-FV⁺ mice, we used previously established transgenic mice only expressing FV (~30% of wt) from the liver with no platelet FV. To create FV-low mice, plasma-FV⁺ mice were treated with ASO-FV; this resulted in mice with markedly reduced (<0.5%) plasma FV and no platelet FV. Using whole blood from these mice, ROTEM assessment showed that compared to wt mice (n = 5; 5.9 ± 0.3 min), the clot time (CT) for plasma-FV⁺ mice (n = 3; 5.9 ± 0.4 min) was normal while the CT was prolonged 2-fold for platelet-FV⁺ mice (n = 10; 13 ± 1.2 min). As expected, the almost complete lack of both pools (FV-low) substantially increased the CT (n = 3; 24 ± 2.2 min). Evaluation of these mice in several injury models probing different vessels was next pursued to examine the hemostatic potential of either pool. Using either a partial (mild) or total (severe) tail injury model, total blood loss for both wt (n = 5) and platelet-FV⁺ mice (n=10) were similar. In contrast, plasma-FV⁺ mice (n=5, p<0.001) exhibited a 2- (mild injury) or 5-fold (severe injury) increase in blood loss compared to wt mice. As expected, FV-low mice (n=5; p<0.001) had the highest levels of blood loss. Further, we used a microcirculatory model of hemostasis to image the distribution of plasma vs. platelet FV following laser injury and assess fibrin and platelet deposition. In plasma-FV⁺ mice, a rapid and robust fluorescence signal for murine FV/Va was detected on the vessel wall surrounding the microscopic injury and on a subset of platelets adherent at the injury site. For platelet-FV⁺ mice, murine FV/Va was detected only on core subset of platelets, which we speculate are fully activated. Further, analysis of the median integrated florescence as a function of time in multiple mice (n=3-5/group; 15-25 injuries) revealed that the absence of either platelet or plasma FV did not considerably alter the kinetics of platelet accumulation (~2-fold less than wt). However, in contrast to plasma-FV⁺ mice, platelet-FV⁺ mice had significantly reduced and delayed fibrin deposition following laser injury suggesting the FV released from platelets is not sufficient to produce normal amounts of fibrin in this model. Our data indicate that the overall impact and contribution of the platelet and plasma FV pools is different depending on the type of injury and/or vessel type. In the microcirculation, plasma FV appears to play an important role in binding to damaged endothelium and contributing to robust thrombin formation, while platelet FV may play a reduced role. However, in more severe injury models, platelet FV plays a prominent role. Collectively these data highlight potentially different roles for the FV pools and/or indicate that the injury models employed impact FV activation or release from platelet α-granules to different extents.