Activation of coagulation factors (F) XII and XI support thrombogenesis through multiple pathways. FXII-deficient mice are more resistant to FeCl3-induced arterial occlusion than either FIX or FXI deficient mice, suggesting that the resistance of FXII-deficient mice to experimental thrombosis is not completely explained by the FXII-FXI-FIX pathway, suggesting the existence of a pathological FXII by-pass, in vivo.
The APTT of FXII deficient plasma is longer than the APTT of FXI, FIX, or FX deficient plasmas. We found that addition of 150 nM activated FXII (FXIIa) decreased the recalcification time of FXI or FIX-deficient plasma by up to 85%. In a purified system FXIIa could activate prothrombin but not FX. Addition of rivaroxaban, a FXa inhibitor, to FXI or FIX-deficient plasma blocked the observed procoagulant effect of FXIIa, suggesting that FXIIa promotes the activation of FX independent of FXI or FIX, but the ability of FXIIa alone to induce coagulation is insufficient in plasma, in vitro.
Addition of long polyphosphate (polyP), typically found in bacteria, but not short polyP, which is secreted by activated platelets, decreased the recalcification time of FXI or FIX-deficient plasma. The presence of either corn trypsin inhibitor (CTI), that inhibits FXIIa, or rivaroxaban blocked the procoagulant effect of long polyP, suggesting that the activation of FXII by long polyP promotes coagulation in an FXI- and FIX-independent manner.
Addition of CTI or an antibody that inhibits FIX activation by FXIa, but not addition of an antibody that inhibits activation of FXI by FXIIa, increased the time of occlusive thrombus formation in recalcified human blood that was driven through collagen and tissue factor (TF)-coated capillary tubes, consistent with the thrombogenic roles of FXIIa and feedback activation of FXI. Only CTI inhibited the prothrombotic effect of long polyP, also suggesting that FXIIa could be thrombogenic independent of FXI and FIX.
In summary, we propose that pathological FXII activation, e.g., by foreign surfaces or long polyP, is thrombogenic both in FXI/FIX-dependent and -independent manners. Provided that FXII has no significant physiological function in humans, our data support the hypothesis that inhibition of FXII activity or activation may have safe antithrombotic effects.
Morrissey:No organization, but the speaker is co-inventor on pending patent applications on the medical uses of polyphosphate: Patents & Royalties.
Asterisk with author names denotes non-ASH members.