A cofactor for factor XI activation

During formation of a blood clot, the key enzyme thrombin is formed through the coordinated activities of a group of plasma proteases (factors VIIa, IXa, Xa, and XIa). Central to this process are protein “cofactors” (tissue factor, and factors Va and VIIIa) that facilitate the activation and/or activity of the plasma proteases on membranes of platelets and tissues. In this issue of Blood, Choi and colleagues present intriguing results showing that a nonprotein, inorganic polyphosphate (PolyP), can enhance thrombin generation by serving as a cofactor for formation of factor XIa.¹ 

Factor XI, the precursor of factor XIa, differs structurally in several important respects from the vitamin K–dependent (VKD) coagulation protease zymogens (the thrombin precursor prothrombin and factors VII, IX, and X).²  For example, factor XI lacks the Gla-domain through which the VKD proteins bind to phospholipid membranes. In addition, factor XIa does not appear to require a cofactor to mediate its primary hemostatic function, activation of factor IX. These properties reflect the different natural histories of factor XIa and the VKD proteases.

The VKD proteases and their cofactors form the core of an ancient hemostatic mechanism that is common to all vertebrate organisms. Factor XI, in contrast, is a relative newcomer, appearing during mammalian evolution as the result of a duplication of the gene for plasma prekallikrein.³  The protease zymogens prekallikrein and factor XII, along with high molecular weight kininogen, comprise the plasma kallikrein-kinin system (KKS). The KKS participates in a number of homeostatic and host-defense functions, including the innate immune response to invading microorganisms.^4,5  KKS components assemble and are activated on the surface of microorganisms, generating antimicrobial peptides and contributing to complement activation.⁵  The capacity of the KKS to bind to surfaces is also important for initiating blood coagulation in vitro in the activated partial thromboplastin time assay (aPTT) used in clinical practice. In the aPTT, anionic substances such as purified earths trigger reciprocal activation of factor XII and prekallikrein in a process called contact activation. Activated factor XII (factor XIIa) then propagates clotting by activating factor XI. Given its close ties to the KKS, then, it is not surprising that factor XI activation by factor XIIa is enhanced in vitro by a variety of polyanions, including the bacterial product dextran sulfate and glycosaminoglycans such as heparin. The importance of factor XIIa-mediated factor XI activation to hemostasis has been questioned, justifiably, because of the absence of a bleeding disorder in persons lacking factor XII. Other proteases including various forms of thrombin activate factor XI and may be more physiologically relevant activators.²  However, regardless of the activating protease, factor XI activation proceeds slowly in the absence of a polyanion, strongly suggesting that a cofactor (perhaps with features of a polyanion) is required to promote the reaction.

Choi et al have now convincingly demonstrated that PolyP secreted from activated platelets is a potent enhancer of factor XI activation by the α and β forms of thrombin.¹  Previously, PolyP has been shown to influence blood coagulation by (1) induction of factor XII activation (contact activation), (2) acceleration of factor V activation by factor Xa, and (3) enhancing fibrin fibril thickness.^1,6,7  PolyP is a linear polymer of inorganic phosphate groups linked by high-energy phosphoanhydride bonds. A minimum PolyP chain length is required to support factor XI activation by thrombin, suggesting a template mechanism in which factor XI and thrombin bind to the polymer in proximity to each other. PolyP also supports factor XI autoactivation, consistent with previous descriptions of the effects of polyanions such as dextran sulfate and heparin on factor XI in solution.

In 1972, Walsh observed that collagen-stimulated platelets possessed a procoagulant activity that required factor XI but not factor XII, and proposed that this activity could explain why factor XII deficiency is not associated with a bleeding disorder.⁸  Subsequent work from several groups showed that activated platelets support factor XI activation by factor XII–dependent and –independent mechanisms. These findings are entirely consistent with the properties of PolyP presented by Choi et al, and with their observation that activated platelets and platelet releasates support factor XI activation in a manner that is blocked by a PolyP specific binding protein.¹  Taken as a whole, this work strongly supports the hypothesis that platelet PolyP can function as a physiologic cofactor for factor XI activation.

PolyP ranging in size from several to more than 1000 phosphate units can be found across all taxonomic kingdoms of organisms.^9,10  Interestingly, PolyP in bacteria are often > 200 phosphate units long, a size that efficiently supports activation of the KKS. Platelet PolyP, on the other hand, is of a somewhat smaller size (∼ 75 units) that facilitates factor XI activation, while having less effect on the KKS. It is tempting to speculate that the size of phosphate polymers in platelet granules is optimized to promote factor XI activation during hemostasis without excessive activation of the KKS. It will be interesting to determine whether dysregulation of PolyP synthesis (either amount or polymer size) is associated with pathologic consequences such as bleeding or an increased risk of thrombosis.