The contact activation system (CAS) represents a group of plasma proteins, including factor XII (FXII), plasma prekallikrein (PPK), and high molecular weight kininogen (HK) that promotes inflammation and coagulation upon contact of blood with an activating surface or protease (1, 2). CAS activation is initiated by interactions of FXII with a negatively charged surface or a protease that induces a conformational change in FXII leading to its proteolytic cleavage and the generation of the serine protease FXIIa (3, 4). The two primary substrates for FXIIa are PPK and factor XI (FXI). FXIIa-mediated cleavage of PPK results in its zymogen activation to the serine protease plasma kallikrein (PKal), which converts FXII to FXIIa and thereby provides positive feedback amplification of the CAS. The effects of the CAS on inflammation, vascular permeability, and edema are primarily attributed to the Kallikrein–Kinin System. This system involves PKal-mediated cleavage of HK to generate the nonapeptide hormone bradykinin (BK), which activates B2 receptors (B2R) that are expressed on a variety of vascular, neuronal, and immune cell types. Binding of BK to B2R activates proinflammatory signaling pathways that dilate vessels, induce chemotaxis of neutrophils, and increase vascular permeability (5, 6). C1-inhibitor (C1-INH) is the primary physiological inhibitor of both FXIIa and PKal, and C1-INH deficiency facilities CAS activation and BK-mediated angioedema (7). Both PKal and the B2R are clinically significant mediators of hereditary angioedema (HAE) (8, 9). In addition to its effects on inflammation and vascular permeability, BK has been shown to induce expression of both tissue factor (TF) and tissue plasminogen activator (tPA), which activates the extrinsic coagulation pathway and fibrinolysis, respectively (10–13) (Figure 1).

The acute effects of the CAS on coagulation have been primarily attributed to FXIIa-mediated activation of FXI and thereby the intrinsic coagulation pathway leading to fibrin generation. Although the biochemical roles of PKal and FXIIa as upstream triggers for the intrinsic coagulation system were recognized over 50 years ago (14), the clinical significance of these factors on thrombosis and hemorrhage is not fully understood. It is well documented that CAS blockade results in prolonged coagulation times in the activated partial thromboplastin time (aPTT) assay (15). The absence of thrombotic and hemostatic abnormalities in individuals with genetic PPK or FXII deficiency has suggested that the CAS plays a minimal role in physiological coagulation. However, information from rare case studies of individuals with deficiencies in individual components of the CAS has provided limited insight into the potential clinical significance of PKal and FXIIa in coagulation in specific diseases. This is further complicated by the interpretation of acquired and mild CAS component deficiencies, which may reflect the activation of the system rather than its suppression. Key observation from studies of humans with genetic FXII and PPK deficiencies is the absence of spontaneous bleeding disorders and increased bleeding risk during surgery, which are observed in individuals with genetic mutations in downstream in the intrinsic pathway, such as FIX (hemophilia B) and FXI (hemophilia C). The main evidence suggesting that CAS inhibition can reduce pathological thrombosis without increasing bleeding has emerged from studies using PPK and FXII-deficient animals with acute and artificially provoked coagulation (16–20). These clinical and experimental findings have generated increased interest in the potential therapeutic opportunities of targeting the CAS in thrombosis. Indeed the observations of the blockade of the CAS results in reduced thrombosis without a concomitant increase in bleeding in preclinical studies have suggested that therapies that inhibit the CAS may offer novel strategies for safe and effective antithrombotics. Moreover, recent studies suggest that blockade of the CAS can improve hemostasis in the retina and brain. In this review, we discuss emerging understanding of the role of CAS on bleeding and hemostasis and the clinical implications of this research.

Hemostasis is the process that prevents blood loss following vascular injury. The mechanisms of hemostasis involve both serial and parallel events, mediated by the interactions between circulating factors in blood with components of activated or injured vascular tissue. Given the critical nature of hemostasis, the system has evolved both redundant and reinforcing mechanisms that facilitate both rapid and stable cessation of blood loss. The sequence of events and the relative contributions and requirements of individual coagulation components may vary according to the affected tissue and type of injury, and possibly also the animal species. While thrombosis and hemostasis share a host of coagulation factors, the etiology of these processes has important distinctions, which may create therapeutic opportunities to selectively target thrombosis without affecting hemostasis.

Although animal models do not fully recapitulate the complexity of human physiology and disease, studies on animals with targeted deficiencies in coagulation factors have provided insight into contributions of individual components of coagulation in hemostatic mechanisms. Multiple approaches are used to quantify hemostatic responses in rodents, including tail tip bleeding, cuticle bleeding, and tail vein transection assays. Deficiencies in FXII and PPK in healthy animals exert little or no effect on tail bleeding time despite aPTT prolongation (Table 1), in contrast to prothrombin deficiency, which markedly increases blood loss in tail wound models (21).

Deficiencies in downstream factors in the intrinsic cascade (such as FIX and FXI) result in bleeding disorders, demonstrating their role in physiological hemostasis. Paradoxically, deficiencies in the CAS factors FXII, PPK, or HK, which trigger coagulation in vitro and contribute to thrombosis in animal models, are not associated with a bleeding diathesis. However, while surgical interventions provide a general assessment of hemostasis, it does not fully model the potential contributions of the CAS in all hemorrhagic conditions. Indeed, recent studies have suggested that the CAS may play an important role in hemorrhage in the retina and brain in certain diseases (17, 20, 30).

Most reports have indicated that individuals with severe deficiencies in either PPK or FXII do not display abnormalities in either thrombosis or bleeding (Table 2), and often go undiagnosed for decades without apparent clinical phenotype. The absence of a bleeding diathesis in FXII and PPK deficiencies appears to contradict the key role of these proteins in coagulation via the intrinsic pathway in vitro, and this has left the physiological significance of the CAS in coagulation and thrombosis unclear. However, CAS could be a trigger for thrombosis associated with extracorporeal circuits (46) and cardiopulmonary bypass (47) by exposing blood to artificial surfaces, non-physiological shear stress, and osmotic forces. Administration of an antibody that neutralizes FXIIa (3F7) has been shown to have a similar effect as heparin in preventing thrombosis during extracorporeal circuits (48).

Activation of FXII by kaolin (negatively charged aluminum silicate particulate) provides the basis for the aPTT, a clinical clotting assay, which is widely used to assess the integrity of the intrinsic pathway and to monitor anticoagulation with heparin. Deficiencies in FXII, PPK, and HK results in markedly prolonged aPTT and no increases in spontaneous bleeding or impairment in hemostasis (Table 2). Normalization of a severely increased aPTT (>120 s) after prolonged preincubation with aPTT reagent occurred in plasma deficient in PPK but not in plasma deficient in FXII, HK, FXI, FIX, FVIII, and Passovoy trait plasma (a deficiency characterized by abnormal coagulation affecting the intrinsic coagulation system) or plasma containing lupus anticoagulant. Autoactivation of FXII in PPK-deficient plasma in the presence of kaolin paralleled the normalization of aPTT. The addition of OT-2, a monoclonal antibody neutralizing FXII, prevents the normalization of aPTT (66). These results suggest that autoactivation of FXII is responsible for normalization of a severely prolonged aPTT upon increased preincubation time in PPK-deficient plasma.

In humans, PPK deficiency is not associated with hemostatic disorders, indicating PKal is not required for hemostasis. Although some case reports have associated FXII deficiency with prothrombotic events, a clinical study failed to demonstrate a significant correlation between increased risk of thrombosis and FXII deficiency (78). Girolami and colleagues (79) re-evaluated case reports on FXII deficiency and thrombosis. They showed that, in most cases, FXII deficiency was associated with other congenital or acquired prothrombotic risk factors. While these studies indicate that the CAS is not essential for either thrombosis or hemostasis, individuals with genetic deficiencies in the CAS are rare and it is not possible to ascertain from the anecdotal clinical information whether PKal and FXIIa could have significant roles in thrombosis or hemostasis associated with specific clinical indications.

C1-inhibitor is the primary endogenous inhibitor of both PKal and FXIIa. HAE is a rare genetic disease caused by deficiencies in C1-INH concentration (Type I) or C1-INH activity (Type II) (80). HAE has a prevalence of approximately 1:50,000 and is characterized by episodes of vasogenic edema, which vary in severity and affected tissue, and can become life threatening when affecting the larynx. The primary mechanism for this disease is the uncontrolled activity of PKal, which results in increased cleavage of HK, and increased production of BK. While the molecular and physiological triggers for the onset of attacks are not fully understood, increased levels of FXIIa is observed in this disease (81) which suggest that attacks are mediated by the CAS. Since HAE is not associated with coagulopathies, this disorder provides an opportunity to evaluate the effects of the CAS on downstream coagulation pathways, independent of confounding input associated with comorbidities of thrombosis and hemostasis. Indeed, plasma obtained from HAE subjects during edematous attacks display shortened aPTT, compared to plasma collected from these subjects during remission (82, 83). In addition, FXIa activity, prothrombin fragment F1 + 2, thrombin-anti-thrombin complex, and FVIIa are elevated in subjects with HAE during an attack (81, 83, 84). The absence of apparent clinical coagulopathies in HAE during attacks could be due to compensatory mechanisms that occur during attacks and/or requirements of additional factors not activated by the CAS to initiate or promote thrombosis formation.

Consistent with PPK and FXII deficiencies in humans, PPK, and FXII-deficient mice display prolonged aPTT (Table 1). Pharmacological inhibition of FXIIa and deficiency in either FXII or HK genes have been shown to protect mice from experimentally induced thrombosis (16, 24, 26, 28), indicating the potential role of the CAS in thrombotic disease. Cheng and colleagues (41) demonstrated that a neutralizing antibody to FXI (14E11) had a comparable effect to FXI deficiency in a FeCl₃ model of thrombosis. Antibody 14E11 binds FXI and interferes with FXI activation by FXIIa, suggesting that thrombus formation in this model requires FXI activation by FXIIa. These data have generated interest in developing strategies to therapeutically inhibit FXIIa and contact activation to treat or prevent thromboembolic disorders. Interestingly, thrombosis protection in mice was greater with FXII deficiency than with FXI deficiency, suggesting that FXII may exert effects on coagulation that do not require FXI (41).

Factor XII-deficient mice showed defective thrombus formation (16) and protective effect against experimental ischemic stroke (85) and pulmonary embolism (86). Intravital microscopy showed that the initial adhesion of platelets at the site of injury is not affected by FXII deficiency, although the formation and stabilization of three-dimensional thrombi is impaired (16). Similar defective thrombus formation was observed in FXI^−/− mice (87, 88). Furthermore, FXII deficiency or pharmacological inhibitor of FXIIa showed decreased infarct volume without the presence of intracerebral hemorrhage (ICH) (85). Reconstitution of FXII-deficient mice with human FXII protein restored susceptibility for ischemic stroke. A recent report by Nickel and colleagues (25) has shown that prostasomes released from prostate cancer activated FXII and that both FXI- and FXII-deficient mice are protected against cancer-associated thrombosis (25). However, the observation that PPK-deficient mice were not protected in the prostasome-induced thrombosis model is surprising and appears different than other CAS-associated thrombosis models, which require PPK. Taken together, these findings suggest that coagulation mediated by intrinsic pathway through FXII is critical for pathological thrombus formation in certain conditions without being relevant for hemostasis.

In addition, the CAS mediates effects on thrombosis via the kallikrein–kinin system. Genetic ablation of Kng1 in mice showed a delayed time to carotid artery occlusion in a laser injury model (28). BKB2R^−/− mice have prolonged bleeding time and delayed carotid artery occlusion times in the rose Bengal thrombosis model and these effects were attributed to increased expression of angiotensin receptor 2 and elevated nitric oxide and prostacyclin levels (29). These findings suggest that the BK system contributes to the effects of the CAS on coagulation.

Components of the CAS are normally restricted from contact with the vascular basement membrane by the endothelium. Vascular hyperpermeability enables components of the CAS to leak into the subendothelial space and gain contact with the basement membrane. The potential functions of the CAS in the subendothelial space have received relatively little attention. PKal binds to collagen (89), which may contribute to the retention of its catalytic activity in the extracellular matrix (ECM) at the site of injury and thereby enable local amplification of its inflammatory and edematous actions. Moreover, PKal in this subendothelial space may have additional substrates and functions, which are not apparent in the plasma.

Currently available anticoagulants used for prevention or treatment of thromboembolic events [heparins, vitamin K antagonists (for example, warfarin), and inhibitors of thrombin or factor Xa] all target enzymes of the coagulation cascade that are essential for the formation of fibrin, a protein necessary for controlling injury-related blood loss. As a result, currently used anticoagulants increase the risk of bleeding and are associated with an increase risk in potentially life-threatening hemorrhage (111). Bleeding is the primary complication of anticoagulation therapy and a significant risk of all currently used anticoagulants, even when maintained within their therapeutic ranges (112). The CAS exerts effects on coagulation at multiple levels. CAS activation of PPK has been implicated in promoting spontaneous microvascular bleeding and the impairment of collagen-induced platelet activation. In addition, both PPK and FXII, via activation of the intrinsic coagulation cascade promotes clot stabilization and growth. Feedback activation of FXI by thrombin contributes to clot stabilization and may explain the lack of dependence of hemostasis on FXII and PPK. PKal has been implicated in promoting spontaneous microvascular bleeding and the impairment of collagen-induced platelet activation. Pharmacological blockade of PKal has been shown to provide beneficial effects on cerebral hemostasis in animal models. Although inhibition of the CAS reduces provoked arterial and venous thrombosis in animal models, the clinical significance of the CAS in thrombosis is not yet available. The clinical indications for targeting the CAS for thrombosis will require the identification of thrombotic processes that are dependent on the CAS, which potentially include interactions with artificial surfaces and disease processes that generate factors that activate FXII. In addition, inhibition of the CAS may provide an opportunity to reduce cerebral hemorrhage, which is one of the primary concerns of increased bleeding risk associated with current antithrombotics.

FS and EF performed literature searches and wrote the manuscript.

EPF is an employee of KalVista Pharmaceuticals Inc. (Cambridge, MA, USA). FS declares no competing financial interests.