Not so fast, antithrombin!

In this issue of Blood, Ivanciu et al¹ report that catalytically active zymogen forms of factor IX (FIX) were resistant to plasma inhibitors like antithrombin (AT), and they enhanced thrombin generation and clot formation in vivo.

Blood coagulation is a highly regulated process.² Coagulation, of course, is necessary following injury, but unrestrained coagulation can be life threatening. However, there are times when easing the restraints on coagulation would be beneficial, such as in the case of hemophilia B.

Procoagulant FIXa in a complex with its cofactor FVIIIa catalyzes activation of coagulation FX, which appears to be the rate-limiting step for thrombin generation.³ FVIIIa is efficiently inactivated by activated protein C, thereby limiting the activity of FIXa.⁴ FIXa is also regulated by many anticoagulant proteins, including AT.⁵ However, AT has a relatively slow rate of inhibition of FIXa. Thus, Ivanciu et al sought to determine whether the FIXa-AT interaction significantly “contributes to hemostatic regulation in vivo.” The authors discovered that 1 mutation, V16L, which is adjacent to the FIX zymogen cleavage site, produced a near fully active protease locked in a zymogen configuration that is resistant to AT. The delayed AT inactivation of this FIXa variant enhanced clot formation in hemophilia B mice, suggesting that this may be a simple strategy for treatment of hemophilia B.

Assessing the hemostatic impact and activity of FIXa by blocking plasma inhibitors of FIXa would be difficult to interpret because there are many anticoagulants like AT,⁶ protein S,^7,8 and protein Z–dependent protease⁹ that inhibit FIXa. Ivanciu et al ingeniously solved this limitation by creating partially active, AT-resistant FIX variants. In the absence of cofactor and substrate, FIXa exists predominantly in a zymogenlike form.¹⁰ Zymogens of the chymotrypsinlike serine protease family including FIXa are converted to active proteases following cleavage between residues 15 and 16 and insertion of the newly formed N-terminus (eg, Val16-Val17-Gly18-Gly19) into an “activation pocket.” Substitutions at residues 16 and 17 produced the aforementioned catalytically active AT-resistant zymogen forms of FIXa, which enabled the authors to determine the natural effects of FIXa inhibitors in plasma and in vivo.

Ivanciu et al focused on 2 zymogenlike FIXa mutants, V17L and V17T, to determine their activities in plasma in the absence of the cofactor FVIIIa and their response to AT. The investigators also determined the activities of the mutants in thrombus formation in vivo. FIXa-V16L and FIXa-V16T had lower protease activity towards FX compared with wild-type FIXa. In the presence of FVIIIa, the rate of activation of FX was slightly lower for V16L and substantially lower for V16T compared with wild-type FIXa. In addition, the rate of AT inhibition of FIXa-V16L was ∼12-fold lower relative to wild-type FIXa, and FIXa-V16T was not able to bind to AT. Most importantly, FIXa-V16L improved clot strength and shortened clot time in hemophilia B mice, and FIXa-V16L exhibited a marked prolongation of activity when introduced in hemophilia B mouse plasma, a result that highlighted the resistance of the mutants to plasma protease inhibitors. Finally, in a tail clip bleeding assay, the total blood loss of hemophilia B mice administered FIXa-V16L was less than the loss for mice that received wild-type FIXa. This result suggested that, once activated in vivo, FIXa-V16L was inactivated by inhibitors (like AT) at a slower rate compared with inactivation of wild-type FIXa, a conjecture supported by the in vitro plasma assays. Finally, using a laser-induced injury model, the authors monitored platelets and fibrin deposition. At first, platelet and fibrin accumulation was restored similarly in hemophilia B mice. At later time points, there was a continuing decrease in the size of thrombi in mice injected with wild-type FIX but not with FIX-V16L. Likewise, platelet accumulation was somewhat greater with FIX-V16L administration. The authors attributed these effects on thrombus formation and platelets to slower inhibition of FIXa-V16L, which was surprising because FIXa-V16L had only 10% of wild-type specific activity.

In sum, the increased hemostatic potency of FIXa-V16L compared with wild-type FIXa directly demonstrated the contribution of AT regulation of FIXa to the control of coagulation at vascular injury. Furthermore, as the authors state, “It is possible to make a procoagulant FIX(a) protein by delaying its inactivation... even at the expense of reduced specific activity.” The work by Ivanciu et al enhances our understanding of the control of clot formation, and the findings suggest a novel strategy to augment clot formation in hemophilia B.

Conflict-of-interest disclosure: The author declares no competing financial interests.