Abstract
The complex and dynamic interplay between the intrinsic and extrinsic pathways of blood coagulation is incompletely understood. The mediator of prothrombin cleavage, Factor X (FX), plays a pivotal role as part of both the extrinsic and intrinsic tenase complexes. Moreover, the existence of naturally occurring Factor X mutations that can be asymmetrically activated through one but not both of these pathways affords one strategy for analyzing the relationship of the two pathways. The Factor X Roma (FXRoma) variant, originally described in a patient with mild bleeding tendency (severe following trauma, De Stefano et al., 1988), is due to a missense mutation (Thr318←Met) in exon 8. Coagulation testing revealed markedly decreased activity (1–3% wild-type) in the intrinsic pathway as measured by aPTT, but substantially higher activity (30–50% wild-type) in the extrinsic pathway as measured by PT. We chose to generate a mouse model of FX asymmetric activation to further probe the extrinsic-intrinsic pathway physiological relationship in hemostasis and thrombosis. For this, we used both an in vitro and an in vivo approach. We first constructed and purified the mouse homolog of FXRoma (mFXRoma) as well as wild-type mFX. Using a clotting-based assay, mFXRoma exhibited intrinsic and extrinsic activity comparable to that reported for the human mutation (5% and 18%, respectively). The reduced intrinsic and extrinsic activity of mFXRoma was not due to a secretion defect, based on Western blot analysis of supernatant and cell extracts from mFXRoma and mFX stably-transfected human embryonic kidney (HEK-293) cell lines. Mice homozygous for the analogous mutation (Thr315←Met) in exon 8 of the murine FX gene were generated by using a plug-and-socket approach. This resulted in the endogenous mFX exon 8 sequence being replaced with the mutated one, thus affording gene expression under the endogenous promoter. Analysis of mFXRoma homozygous mice showed a 6.4% and 19.2% intrinsic and extrinsic activity relative to wild-type littermates, respectively, confirming our in vitro data. The reduced activity in these mice resulted in a slight reduction in levels of the thrombin-antithrombin (TAT) complex. To determine any physiological defect of this mutation on the two pathways of coagulation, we performed two hemostatic challenges of the macrocirculation (tail clip and FeCl3-induced thrombus formation). In the tail-clip assay, blood loss showed no statistical difference between wild-type (n=5) and mFXRoma (n=6) mice. In contrast, following FeCl3-induced injury on the carotid artery (larger vessel diameter that in the tail), mFXRoma mice (3/3) failed to result in vessel occlusion (up to 30 min of observation), whereas wild-type littermates showed stable vessel occlusion (3/4) within ∼6 min of FeCl3 application. Although the type of injury was different, these data suggest that an impeded intrinsic activity of FX does not appear to affect hemostasis of the macrocirculation at relatively small diameter vessels but is essential for thrombus formation in large diameter vessels, and a relatively normal extrinsic activity does not compensate for this defect. This mouse model will aid in determining the safety and efficacy of therapeutic approaches based on impeding the intrinsic pathway of coagulation.
A mouse model of severe von Willebrand disease: Defects in hemostasis and thrombosis