© Schattauer 2014 Thrombosis and Haemostasis 112.5/2014

909Theme Issue Article

Differential cellular effects of old and new oral anticoagulants: consequences to the genesis and progression of atherosclerosisLeon J. Schurgers; Henri M. H. SpronkDepartment of Biochemistry, Cardiovascular Research Institute Maastricht, Maastricht University, Maastricht, The Netherlands

SummaryThe main purpose of anticoagulants is to diminish fibrin formation, thereby decreasing the risk of venous or arterial thrombosis. Vitamin K antagonist have been used for many decades in order to achieve re-duced thrombotic risk, despite major drawbacks of this class of drugs such as cumbersome dossing and monitoring of anticoagulant status. To overcome these drawbacks of VKA, new classes of anticoagulants have been developed including oral anticoagulants for direct in-hibition of either thrombin or factor Xa, which can be administrated in a fixed dose without monitoring. Coagulation factors can activate cel-lular protease-activated receptors, thereby inducing cellular processes as inflammation, apoptosis, migration, and fibrosis. Therefore, in-hibition of coagulation proteases not only attenuates fibrin formation, but may also influence pathophysiological processes like vascular cal-

cification and atherosclerosis. Animal models revealed that VKA ther-apy induced both intima and media calcification and accelerated plaque vulnerability, whereas specific and direct inhibition of throm-bin or factor Xa attenuated atherosclerosis. In this review we provide an overview of old and new oral anticoagulants, as well discuss po-tential pleiotropic effects with regard to calcification and athero-sclerosis. Although translation from animal model to clinical patients seems difficult at first sight, effort should be made to fully understand the clinical implications of long-term oral anticoagulant therapy on vascular side effects.

KeywordsArterial thrombosis, arteriogenesis, atherothrombosis, vitamin K- dependent factors, atherosclerosis

Correspondence to:Leon J. Schurgers, PhDDepartment of BiochemistryMaastricht UniversityP.O. Box 6166200 MD Maastricht, The NetherlandsTel.: +31 433881681, Fax: +31 433884159E-mail: l.schurgers@maastrichtuniversity.nl

Received: March 25, 2014Accepted after major revision: September 16, 2014Epub ahead of print: October 9, 2014

http://dx.doi.org/10.1160/TH14-03-0268Thromb Haemost 2014; 112: 909–917

Oral anticoagulants

Vitamin K-antagonists (VKA) and heparins have been anticoagu-lants for prevention and treatment of thromboembolitic events for some 70 years. VKA are still the most widely used anticoagulant drugs to treat patients at risk of arterial and venous thrombosis. Due to unfavorable pharmacokinetics of VKA, direct thrombin and factor Xa (FXa) inhibitors (non-vitamin K-antagonist oral anticoagulants, NOACs) have been introduced into the clinic as al-ternatives to VKA. Today, it is still a matter of debate which treat-ment is superior with respect to bleeding risk. Anticoagulant drugs also influence non-haemostatic activities of coagulation factors. Experimental data indicate that thrombin can induce an array of pro-atherogenic and plaque-destabilizing effects. Together with the clinical safety of long-term treatment with VKA this suggests that they have also protective effects on the vasculature, possibly via protease-activated receptors (PARs) and via vascular vitamin K-dependent proteins. Knowledge of these as yet unknown effects of VKA and of NOAC will help to improve the long-term prophy-laxis and treatment of arterial and venous thrombosis, and will support personalised medicine.

History

The story of oral anticoagulants starts in the early 1920s. Oral anti-coagulants were discovered in the malady of cattle involving fatal bleeding (1). Cattle being dehorned that had eaten spoiled hay had little to no clotting power resulting in fatal haemorrhages after 30–50 days. It was Roderick who showed in 1931 that the abolish-ed coagulability of the blood was due to a prothrombin deficit.

In the late 1930s Campbell (2) isolated and purified crystalline dicoumarol from the hay. After synthetic dicoumarol became available in quantity, the essentials of its physiologic action were quickly established. Due to the lag period of dicoumarol, probably due to variability in absorption kinetics, it was anticipated that the action would largely vary within human individuals (3). Dicouma-rol was used in trials for its rodenticidal purpose. However, the ac-tivity of dicoumarol was not high enough, which was due to poor and variable absorption kinetics, weak antagonist of vitamin K, and the vitamin K content of grains and green foods. In the follow-ing years, over 100 related dicoumarol compounds were prepared, with number 42 being the one with the highest anticoagulant ac-tivity, and this was promoted for rodent control under the auspices of Wisconsin Alumni Research Foundation. Hence, the name

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

Thrombosis and Haemostasis 112.5/2014 © Schattauer 2014

910 Schurgers, Spronk: Coagulation meets atherosclerosis

WARFarin with “arin” derived from coumarin. Warfarin became the leader in the rodenticide field. The hospitalisation of an army inductee who had taken warfarin over a period of five days was the starting point for warfarin being a clinical drug (3). It was clear that warfarin was not an effective suicide drug, as it allowed too much time for thinking. Today we appreciate the ineffectiveness of warfarin because of its delayed action, its low acute toxicity to a single dose (as evidenced by its high LD50) together with the ready reversibility of the coagulopathy by vitamin K as antidote. In the hospital he was treated with high dose vitamin K, and recovered uneventfully. This event propelled the clinical use of coumarins for the treatment of hypercoagulability.

Mechanism of vitamin K-antagonists

VKA interfere with vitamin K-metabolism, as the name suggests. Vitamin K was discovered in the 1930s by Henrik Dam. It was ser-endipity that Dam found this “antihaemorraghic factor” as he was investigating cholesterol metabolism in chickens which developed unexpectedly large subcutaneous and intramuscular haemor-rhages (4). Dam coined this micronutrient Koagulation factor (ab-breviated as vitamin K), because of its requirement for normal haemostasis. It took until 1974 before the unequivocal role of vit-amin K action was established as cofactor for the posttranslational carboxylation of glutamate residues (5). The substrates of gamma-glutamyl carboxylase are the so-called vitamin K–dependent pro-teins, which are involved in various physiological processes such as blood coagulation (▶ Figure 1), bone and soft tissue minerali-sation, and cellular proliferation. The coagulation factors II (pro-thrombin), VII, IX, and X all have procoagulant activity, whereas proteins C and S inhibit blood coagulation (reviewed in [6]). Four key components are involved during the post-translational modifi-cation of the vitamin K–dependent proteins: the enzymes gamma-glutamyl carboxylase and vitamin K epoxide reductase complex

(VKOR), vitamin K, and a precursor protein (▶ Figure 2). The dis-covery of the unusual amino acid γ-carboxyglutamic acid (Gla) (7) in prothrombin (8) as the product of vitamin K-action was the first hint on the action of VKA. Matschiner et al. showed that adminis-tration of warfarin resulted in a significant increase of the epoxide form of vitamin K in rats (9). In brief, vitamin K derived from food products (vitamin K quinone) has little to no cofactor activity and needs to be converted into the active cofactor vitamin K hydroqui-none (KH2, ▶ Figure 2). During carboxylation, vitamin KH2 is converted into vitamin K epoxide (KO), which subsequently will be reduced to vitamin K and vitamin KH2 in two reactions by the action of the enzyme vitamin K-epoxide reductase (VKOR). The VKOR gene was identified in 2004, when both the lab of Stafford (10) and Oldenburg (11) published the purification of the gene coding for the enzyme VKOR. This recycling mechanism is called the vitamin K cycle and the VKA specifically inhibit the enzymatic activity of VKOR, thereby causing a decrease in the active cofactor vitamin KH2, resulting in attenuated gamma-glutamyl carboxy-lation. Through the action of gamma-glutamyl carboxylase, the vitamin K-dependent coagulation factors (procoagulants: pro-thrombin, factors VII, IX, and X, and anticoagulants: protein C and S) gain a calcium-specific membrane binding domain (also known as Gla-domain), enabling the high affinity binding to an anionic-rich phospholipid bilayer. Assembling the key players in coagulation on an anionic phospholipid bilayer increases the local concentrations of both enzyme and substrate, thereby decreasing the apparent Km far below their plasma concentration. However, this drop in Km is not translated into an increase in zymogen con-version and the cofactors Va and VIIIa are needed in order to en-hance the enzymatic activity (Vmax). In summary, organisation of the coagulation factors on a anionic phospholipid bilayer and the activities of cofactors V and VIII, causes a dramatic acceleration of the velocity at which factor X and prothrombin are activated by the tenase (factors IXa, VIIIa, and X) and prothrombinase

Figure 1: Effects of vitamin K antagonists (VKA) and direct oral anticoagulants (NOAC) on coagulation, e.g. fibrin formation, and vascular proteins. Whereas VKA interfere with synthesis of all vitamin K- dependent proteins – coagulation factors pro-thrombin (II), X, VII, IX, protein C, and protein S, matrix Gla protein (MGP, and Gas6), NOACs were developed for the specific inhibition of either thrombin (IIa) or factor Xa.

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

© Schattauer 2014 Thrombosis and Haemostasis 112.5/2014

911Schurgers, Spronk: Coagulation meets atherosclerosis

complex (factors Xa, Va, and prothrombin), respectively. There-fore, inhibition of gamma-glutamyl carboxylation leading to dis-rupted membrane-binding capacity for coagulation factors results in diminished haemostasis.

Coumarins

Since the mid 1950s, VKA has been given to patients worldwide and still form the treatment basis of contemporary thrombopro-phylaxis (12). Today, several VKAs are used in the clinical setting all belonging to the class of compounds called 4-hydroxycouma-rins or 1,3 indandiones (13). The major 4-hydro-xycoumarin anti-coagulants in current use are warfarin, acenocoumarol, and phen-procoumon. All VKAs act indirectly on inhibition of carboxy-lation of vitamin K-dependent proteins. The most important com-plication of treatment with VKA is haemorrhage, which may be life-threatening (14) as it increases the risk of major bleeding by 0.5%/year (15).

Despite the effectiveness of VKA, treatment has several short-comings. First, the activity of VKA on anticoagulation potential has to be monitored by blood testing. Secondly, many commonly used medications interact with VKA. Thirdly, large amounts of vit-amin K in particular in green leafy vegetables are well absorbed in the intestinal tract, thereby counteracting VKAs in the liver (16). Finally, VKA are not restricted to vitamin K-dependent coagu-lation factors and inhibit vitamin K-dependent proteins involved in other physiological processes, such as bone metabolism and vascular function (17), as well. The best studied side-effect of VKA is on matrix Gla-protein (MGP), a strong inhibitor of vascular cal-cification (reviewed in [18]).

Due to these unfavourable pharmacokinetics and actions of VKA, direct thrombin and factor Xa (FXa) inhibitors (NOACs) have been introduced as alternatives to VKA.

Non vitamin K antagonist oral anticoagulants

Non vitamin K antagonist oral anticoagulants (NOACs) are small synthetically made molecules against one clotting factor (i.e. thrombin (or factor IIa) or factor Xa, ▶ Figure 1). They bind selec-tively and with high affinity to the active site of the enzyme, thereby inhibiting the turnover of the natural substrates. Direct thrombin inhibition is straightforward by inhibition of the key regulator of coagulation. Factor Xa inhibition is based on reducing the thrombin burst in the propagation of coagulation.

Thrombin inhibitors

The story of thrombin inhibitors goes back to the late 1950s when hirudin, produced by the medicinal leech Hirudo medicinalis, was isolated and biochemically characterised as a potent thrombin in-hibitor. Hirudin forms a 1:1 stoichiometric complex with throm-bin through binding to exosite 1 (anion-binding or fibrinogen binding site) and the active catalytical domain of the protease (19). With a dissociation equilibrium constant (Ki) of 20 fM, hirudin is still the most potent thrombin inhibitor. Using recombinant DNA

technology structural variants of the native hirudin were produced (reviewed in [20]). Although hirudin analogs are approved for clinical use, their use is limited because they are not orally active. Ximelagatran, a pro-drug of melagatran, was the first orally active thrombin inhibitor. Melagatran directly targets the active site of thrombin. Ximelagatran has been evaluated extensively for the re-placement of warfarin in atrial fibrillation (AF) patients. Despite favorable results, ximelagatran did not make it for clinical use due to hepatotoxicity. Dabigatran etexilate, a prodrug of dabigatran, was the second orally thrombin inhibitor. It is a potent, competi-tive, reversible direct thrombin inhibitor directed against the active site of thrombin, and capable of inactivating both fibrin-bound and unbound thrombin.

Direct factor Xa inhibitors

Factor Xa inhibitors were discovered in the early 1980s. The first factor Xa inhibitor, called Antistasin, was isolated from the salivary glands of the Mexican leech Haementeria officinalis. Antistasin is a polypeptide with a slow and tight-binding, and thus potent factor Xa inhibitor (21). A second naturally occurring factor Xa inhibitor, isolated from extracts of the soft tick Ornithodoros moubata, is the tick anticoagulant peptide (TAP), a single-chain, 60 amino-acid peptide. Inhibition of factor Xa by recombinant TAP showed sus-tained in vitro and in vivo inhibition of clot-associated procoagu-lant activity. These data provided the support for the concept of factor Xa inhibition. Subsequently, chemical companies inspired by this coagulation inhibition of factor Xa, initiated chemistry pro-grams to develop small synthetic factor Xa inhibitors, such as riva-roxaban, apixaban, betrixaban, and edoxaban. Rivaroxaban is the first most advanced synthetic factor Xa inhibitor with a 10,000-fold higher selectivity for factor Xa than other serine pro-teases. Inhibition of factor Xa is concentration dependent and af-fect prothrombinase complex bound as well as clot-associated fac-tor Xa.

Stroke preventionin AF patients: VKA vs NOACs

Atrial fibrillation (AF) is the predominant indication for long-term anticoagulant treatment. According to AF-guidelines, indi-cation for long anticoagulant treatment is based on the stroke risk, which can be estimated using the CHADS2 score (22). CHADS2 stands for congestive heart failure, hypertension, age ≥ 75 years and stroke/transient ischaemic attack. Each risk factor of the CHADS2 score accounts for one point except for stroke/transient ischaemic attack which accounts for two points. The higher the CHADS2 score the higher the risk of thrombo-embolic compli-cations and as a result long-term anticoagulant treatment is indi-cated in patients with a CHADS2 score of ≥1 according to Euro-pean and Canadian guidelines and in patients with a CHADS2 score of ≥2 according to American guidelines (23). Several major trials designed to test whether direct thrombin or factor Xa in-hibitors were non-inferior to VKAs with respect to efficacy are

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

Thrombosis and Haemostasis 112.5/2014 © Schattauer 2014

912

Figure 2: Anticoagulants have different effects on activation of pro-tease activated receptors (PARs). A) The vitamin K cycle. Vitamin K-qui-none (K) is first reduced to vitamin K-hydroquinone (KH2), and then oxidised to vitamin K-epoxide (KO). The vitamin K-epoxide reductase (VKOR) enzyme converts the epoxide form back to the quinone form, thereby completing the vitamin K cycle. During the oxidation step glutamic acid (Glu) is reduced to gamma-carboxy glutamic acid (Gla) by the enzyme gamma-glutamyl car-boxylase. The VKOR enzyme is inhibited by vitamin K antagonsists (VKA), causing diminished carboxylation. As a result, the liver secretes the vitamin

A

K-dependent proteins in their precursor form, without the correct membrane binding domain. Activated coagulation factor X without the membrane bind-ing domain retained the capability to activate PARs. B) Direct factor Xa in-hibitors attenuate the activity of the protease, which as a consequence can not activate prothrombin and the cellular PARs. C) Direct thrombin inhibitors attenuate the activity of the protease, which as a consequence can not acti-vate cellular PARs, whereas factor Xa remains unaffected and can activate the PARs.

B

C

Schurgers, Spronk: Coagulation meets atherosclerosis

published (24–26). In the RELY trial 18,113 subjects with AF were enrolled and randomly assigned to receive fixed dose of dabigatran (direct factor IIa inhibitor), 110 mg or 150 mg twice daily, or dose adjusted VKA (26). Administration of 110 mg dabigatran showed similar results as compared to VKA with respect to stroke and sys-temic embolism; however, with lower rates of major haemorrhage. The 150 mg dose of dabigatran was associated with lower rates of stroke and systemic embolism, but comparable rates of major haemorrhage. In the ROCKET-AF study 14,264 randomly as-signed patients with AF who were at increased risk for stroke re-ceived rivaroxaban (direct Xa inhibitor) or dose adjusted VKA (27). Rivaroxaban showed to be non-inferior to VKA for the pre-vention of stroke or systemic embolism. And although no signifi-

cant difference in risk for major bleeding was found, less intra-cranial and fatal bleedings occurred in the rivaroxaban group. These results were reinforced by the ARISTOTLE trial in which 18,201 subjects with AF and at least one risk factor for stroke were enrolled and randomly assigned to receive apixaban 5 mg twice daily (factor Xa inhibitor) or VKA (24). Instead of finding non-in-feriority of apixaban to VKA, the investigators found that apixa-ban was superior to VKA with respect to reduction of stroke risk or systemic thromboembolism by 21 % and risk of bleeding by 31%. Furthermore, all cause death, as compared to VKA, was re-duced by 11%. Thus far, dabigatran etexilate, rivaroxaban and api-xaban all have shown to be non-inferior to VKAs with apixaban being superior to VKA. Co

agul

atio

n pr

otea

ses

and

CVD

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

© Schattauer 2014 Thrombosis and Haemostasis 112.5/2014

913Schurgers, Spronk: Coagulation meets atherosclerosis

Major advantages of all NOACs over VKA are the stable phar-macokinetics and –dynamics, which allows fixed dosing in all pa-tients. Consequently frequent laboratory monitoring is redundant. However, there are a number of clinical situations in which labora-tory monitoring might be advisable: such as in non-compliance, therapy failure and major bleeding. Moreover, in patients with ex-tremely low or high body mass index, in poly-pharmacia or in the presence of comorbidity the possibility to check the activity of a prescribed drug might be advantageous. However, at the moment there are no laboratory tests available to monitor the direct antico-agulants in practice. In addition, unlike for VKA no antidotes exist to reverse the effects of direct thrombin or Xa inhibitors.

Coagulation and atherosclerosis

Although the pharmacological effects of anticoagulants are clear and have been described in detail elsewhere, long-term treatment with VKA or the NOACs, such as dabigatran and rivaroxaban, may have unwanted side effects. In a recent review published on the haemostatic system as a modulator of atherosclerosis, throm-bin and factor Xa are believed to be potential modulators of plaque phenotype, as they are involved in inflammation, endothelial func-tion, vasomotor tone and many more (28). Both factors signal through a family of cellular PARs (PAR1 to –4). PARs are activated through proteolytical cleavage and all PARs, except PAR2, are acti-vated directly or signal directly in response to thrombin. However, thrombin-cleaved PAR1 can donate its tethered ligand to transac-tivate PAR2, which in turn has been shown to modulate the PAR1-induced hyperplastic response to arterial damage preceding stenosis (29) and PAR1-mediated tumour cell migration and metastasis (30). However, direct thrombin-induced signalling re-sponses through PAR1 activation seem to be different from the transactivation of PAR1 and PAR2 heterodimers (31). Also, formation of PAR1 and PAR4 heterodimers dependent on throm-bin-induced cleavage of both receptors has also been reported (32). Factor Xa-dependent cellular responses can occur through PAR1 and/or PAR2 cleavage, and these processes may depend on the receptor-specific cell expression pattern, ligand concentration, solubility, or association with other coagulation factors (33). Over-all, the proteolytic activation of PARs by either thrombin or factor Xa results in the activation of a canonical G-protein pathway and, consequently, of downstream signalling pathways that influence multiple transcription-regulated, cell-specific events including proliferation, inflammation, migration, adhesion, and apoptosis (28). Through this modulation of cell behavior, activation of PARs by thrombin or factor Xa may contribute to the development and progression of pathologic conditions including arthritis, fibrotic lung disease, cancer, and atherosclerosis (33–35). In relation to atherosclerosis, diminished coagulation has been shown to attenu-ate atherosclerosis (reviewed in [36]), and these observations are supported by experiments demonstrating that the direct cellular effects of factor Xa and thrombin can be responsible for promoting inflammation, leukocyte transendothelial migration, angiogenesis, narrowing of blood vessels, as well as migration and proliferation

of VSMCs (28). PAR4-mediated activation of platelets by throm-bin has been suggested to have no contribution to development of early atherosclerotic lesions (37). Not only does PAR4 require high concentrations of thrombin or PAR3 as a cofactor in activation (38), a study with PAR4 deficient mice on an atherosclerotic back-ground showed no significant differences in the percentage of aor-tic lumenal surface covered by plaques and in the cross-sectional area of plaques compared to ApoE-/- mice after five and 10 weeks on a Western diet (37).

Supporting the theory of a relation between coagulation and development and progression of atherosclerosis is the presence of coagulation factors and corresponding proteolyical activity in atherosclerotic lesions (39, 40), with increased activities in early atherosclerotic lesions compared with lesions at a later stage (41). Indeed, increased (hypercoagulability) and diminished (hypoco-agulability) coagulation have been suggested to contribute to de-velopment of atherosclerosis, as suggested from several animal studies. A study from our group showed enhanced atherosclerosis with a higher plaque vulnerability, spontaneous atherothrombosis, enhanced oxidative stress, neutrophil intraplaque infiltration and apoptosis, and diminished VSMC proliferation in animals on a combined atherogenic and hypercoagulable background (42). A hypocoagulable phenotype on an atherogenic background showed reduced atherosclerosis, decreased leukocyte infiltration and alter-ed collagen and VSMC content (42). These results confirmed ear-lier observations showing that hypocoagulable mice double defi-cient in factor VIII on an atherogenic background showed signifi-cantly less early-stage atherosclerotic lesions (43) and that hyper-coagulable mice (either deficient in the natural anticoagulant he-parin cofactor II or tissue factor pathway inhibitor) showed a larger total atherosclerotic plaque area or atherosclerotic burden while on an atherogenic background (44, 45). Remarkably, a 50% reduction of tissue factor expression in ApoE-/- mice did not result in a change of the atherosclerosis extent compared to wild-type mice (46). Besides animal studies, several clinical studies suggest both a hypercoagulable state and active thrombin generation in as-sociation with cardiovascular disease and atherosclerosis (re-viewed in [47]). However, a direct relationship between a hyperco-agulable state and development and progression of atherosclerosis is not immediately evident from clinical studies. The association between hypercoagulability-related common thrombophilic gene defects, such as prothrombin G20210A mutation, FV Leiden mu-tation, and deficiencies in protein C, protein S, or antithrombin (AT), and cardiovascular disease are modest. Besides, reported as-sociations are mainly observed in young patients (age < 45 – years), at onset of disease before the age of 55 years (and in the ab-sence of usual risk factors), or with early occlusion after revascu-larisation procedures (48, 49). Comparable conclusions can be drawn for hypocoagulability and association with cardiovascular. Although, haemophilic patients have a two- to three-fold in-creased overall mortality from bleeding disorders, liver diseases, or viral infections, the effects of factor deficiency on cardiovascular mortality resulting from ischaemic heart remain controversial. One possible confounder in the different observations might be the therapeutic administration of coagulation proteins in patients

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

Thrombosis and Haemostasis 112.5/2014 © Schattauer 2014

914 Schurgers, Spronk: Coagulation meets atherosclerosis

with severe haemophilia, potentially obscuring the effects of the original factor deficiency. Hypercoagulability can also be defined as having elevated circulating markers of activated coagulation such as prothrombin fragment 1.2, thrombin:antithrombin com-plexes, or D-dimer. Overall, plasma levels of markers of active co-agulation are elevated in relation to the extent of atherosclerosis, and have been shown to be predictive of cardiovascular events in some studies (50). More precisely, several clinical studies have demonstrated that the specific fibrin degradation product D-dimer is associated with an increased risk of severe athero-sclerosis and vascular complications (51–59). Prothrombin frag-ment 1.2 has been shown to be independently associated with ca-rotid intinma media thickness (IMT) in adults without clinically overt atherosclerotic disease (60). Furthermore, the presence of coronary atherosclerosis was associated with increased levels of the prothrombin activation markers prothrombin fragment 1.2 and thrombin:antithrombin complex (61) and the size of athero-sclerotic plaques positively correlated with prothrombin fragment 1.2 in patients with a previous myocardial infarction (62) and first ischaemic stroke (63). Data from our own group showed that elev-ated thrombin:antithrombin levels were independently associated with the presence and severity of coronary atherosclerotic plaques, but also with the degree of vascular calcification, as defined by cor-onary computed tomographic angiography (64).

In conclusion, a clear contribution of coagulation proteases to the development and progression of atherosclerosis has been dem-onstrated, whereas in humans with atherosclerosis increased co-agulation activity is present in circulation and within athero-sclerotic lesions. It remains, however, to be elucidated if hyperco-agulability in humans is causally related to structural changes in the arterial vessel wall contributing to atherosclerosis. Neverthe-less, given the strong evidence obtained from both experimental studies and animal models, it seems reasonable to hypothesise that long-term treatment with anticoagulants might have beneficial or harmful outcomes.

Pleiotropic effects of anticoagulantsCalcification

As mentioned above long term treatment with VKA induces and accelerates vascular calcification. The first pre-clinical data are from the mid-1990s, in which the combined treatment of vitamin K with VKA resulted in a severe vascular vitamin K deficiency, with concomitant medial vascular calcification (65). Recently it was shown that in apoE-/- mice bearing atherosclerotic plaques VKA also induced accelerated intimal calcification and increased features of plaque vulnerability (66). Additionally, mice on VKA treatment showed augmented aortic peak velocity, aortic valve–peak gradient, and carotid pulse-wave velocity (67).

It took until 2004 before the effect of VKA on vascular calcifi-cation in the clinical setting was revealed (68). Today, the associ-ation between VKA use and vascular calcification has been con-firmed in several trials (17). As mentioned above, AF is the pre-dominant indication for long-term anticoagulant treatment. In pa-

tients with atrial fibrillation coronary artery calcification was in-vestigated in relation to treatment and duration of VKA treatment (69). Low-risk AF patients (those with a CHADS2 score below 1) using a VKA had more calcification and atherosclerosis than those patients not treated with VKA. Moreover, duration of treatment was also significantly correlated with increased coronary artery calcium score. However, due to the high efficacy of VKA in stroke prevention, the potential harmful effects of VKA-associated vascu-lar calcification might be offset by the beneficial effect, especially in patients with high stroke risk. For that reason the clinical im-portance of VKA-induced vascular calcifications remains unclear, and needs to be more personalised.

Unlike VKA the direct anticoagulant drugs possibly do not in-terfere with functionality of MGP (18) and prospective observa-tional studies including AF patients treated with either VKAs or new anticoagulant drugs might reveal the importance of VKA as-sociated vascular calcifications in the long term (17). Some degree of vascular inflammation is concomitant with most forms of cal-cification. Monocytes and macrophages have been observed to en-hance vascular calcification via both cell–cell interaction and the production of soluble factors. Inflammatory cytokines such as in-terleukin (IL)-1β, IL-6, IL-8, insulin-like growth factor (IGF)-1 and tumour necrosis factor (TNF) induce the osteogenic switch of vascular smooth muscle cells (VSMCs), thereby enhancing calcifi-cation (70–72). Nadra et al. (71) showed that basic calcium phos-phate crystals are taken up by macrophages, and that this was as-sociated with release of TNF, IL-1β and IL-8. Additionally, calcium phosphate induced macrophage IL-1β secretion through acti-vation of the NLRP3 inflammasome (73). Also VSMCs are profes-sional phagocytic cells, and thus can take up calcium crystals. This results in apoptosis for VSMCs thereby increasing the inflamma-tory response in the vasculature (74). Acceleration of calcification by VKAs administration may exert anti-inflammatory effects as well.

Although the direct relationship between vascular calcification and coagulation is still open for investigation, we can only specu-late on how coagulation and calcification influence one another. One possible relationship can go via microvesicles (MVs) secreted by VSMCs. MVs, which are increasingly secreted upon calcifi-cation, express phosphatidylserine (75) that can form the negatively charged surface to increase coagulation and even may contribute to activation of factor XII as demonstrated previously for platelet- and erythrocyte microparticles (76).

Atherosclerosis

Pleiotropic effects of NOACs on atherosclerosis or atherothrom-bosis have been suggested from both animal studies and clinical trials. An increased frequency of myocardial infarctions has been reported in some (but not all) trials using directs thrombin in-hibitors (26, 77). Moreover, in patients treated with ximelagatran increased markers of inflammation, such as IL-10 and C-reactive protein (CRP), were observed (78). Nonetheless in the latest AF guidelines, dabigatran is recommended instead of VKA or as an alternative for VKA for the prevention of stroke. Animal models

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

© Schattauer 2014 Thrombosis and Haemostasis 112.5/2014

915Schurgers, Spronk: Coagulation meets atherosclerosis

have shown beneficial effects for NOACs on the development and progression of atherosclerosis. One of the first studies exploring the effects of direct thrombin inhibition on atherosclerosis was performed in ApoE-/- mice applying ximelagatran. The direct in-hibition of thrombin caused enhanced stability of advanced atherosclerotic lesions characterised by thicker fibrous caps, smaller necrotic cores, and a decrease in matrix-metalloprotei-nase-9 (MMP-9) (79). Treatment of ApoE-/- mice with dabigatran resulted in comparable results with impaired formation and re-duced size of atherosclerotic lesions (80). On a combined hyper-ocoagulable and atherogenic background dabigatran reduced the pro-inflammatory and pro-atherogenic phenotype of athero-sclerotic lesions, resulting in enhanced plaque stability (42). In this later model, leukocyte recruitment was attenuated and severe pro-gression of atherosclerosis and atherothrombosis was prevented through direct thrombin inhibition. Comparable animal studies with direct factor Xa inhibitors showed comparable effects on atherosclerosis. For instance, selective inhibition of factor Xa through recombinant antistasin or tick anticoagulant peptide in rabbits was shown to reduced restenosis after balloon angioplasty of atherosclerotic femoral arteries and limit the narrowing of these vessels by atherosclerotic lesions (81). Direct factor Xa inhibition through administration of rivaroxaban has been shown to increase the stability of advanced atherosclerotic plaques in apolipoprotein E-deficient mice, as suggested by the presence of thicker protective fibrous caps and decreased erosion of the plaques (35). In addition, factor Xa inhibition reduce the expression of inflammatory medi-ators, such as IL-6, TNF-α and monocyte chemoattractant protein (MCP)-1, pointing towards atherosclerotic plaque stabilisation. However, rivaroxaban plasma levels in this model were rather low and the effects of clinical plasma concentrations needs to be estab-lished.

Conclusions

Although both VKA and NOACs act as anticoagulants, their mode of action suggests differences in pleiotropic effects. Whereas VKA treatment is mainly associated with enhanced calcification, NOACs treatment might have an effect on atherosclerosis progres-sion in humans via PAR signalling. Coagulation factors produced in the presence of VKA are synthesised without the Gla domain, which is needed for binding to negatively charged cell membranes. These (so called) proteins induced by vitamin K antagonism or ab-sence (PIVKAs) (82) cannot bind to negatively charged cell mem-branes. As a consequence, PIVKA-factor Xa will only convert minimal amounts of prothrombin into thrombin. However, the ac-tivation of PARs by coagulation factors has been suggested to occur in a membrane-binding independent way (83), suggesting that PIVKA factor Xa can still activate PARs (▶ Figure 2). Treat-ment of cultured fibroblasts with full-length factor Xa or a trun-cated form of factor Xa lacking the membrane-binding domain (e.g. PIVKA factor Xa) revealed that both forms of the protease ac-tivated PARs (84), suggesting that factor Xa synthesised in the presence of VKA can still induce cellular responses (▶ Figure 2).

In contrast inhibition of factor Xa using a direct inhibitor will in-hibit not only the conversion of prothrombin to thrombin but also the factor Xa-mediated activation of PARs. In the case of a direct thrombin inhibitor, fibrin formation and other functions of thrombin will be diminished, whereas factor Xa can still be gener-ated by the tissue factor:factor VIIa complex (▶ Figure 2). These differences in anticoagulant mechanisms provide in part an expla-nation for the observed differences in pleiotropic effects.

With an increasing demand for risk assessment of the individu-al patient and upcoming of new anticoagulants, the interplay be-tween coagulation, inflammation and calcification deserve more studies to address its implications in risk. Although inhibiting fac-tor Xa and thrombin may provide additional therapeutic potential over the VKA, data is still limited and mainly derived from experi-mental animal studies. The translation from animal to humans needs to be made, especially for the long-term effects of anticoagu-lant therapy. Therefore, further investigation is needed to fully understand the complexity and possible long-term benefits or drawbacks of both direct and indirect inhibition of coagulation proteases.

Conflicts of interestNone declared.

References1. Schofield FW. A brief account of a disease in cattle simulating hemorrhagic sep-

ticemia due to feeding sweet clover. Can Vet Rec 1922; 3: 74–79. 2. Campbell M. Haemorrhagic disease in cattle. J Biol Chem 1941; 18: 21–33. 3. Link K. The discovery of dicumarol and its sequels. Circulation 1959; 19:

97–107. 4. Dam H. The antihemorrhagic vitamin of the chick: occurrence and chemical

nature. Nature 1935; 135: 652–653. 5. Stenflo J, Ferlund P, Egan W, et al. Vitamin K dependent modifications of glu-

tamic acid residues in prothrombin. Proc Natl Acad Sci USA 1974; 71: 2730–2733.

6. Versteeg HH, Heemskerk JWM, Levi M, et al. New fundamentals in hemostasis. Physiol Rev 2013; 93: 327–358.

7. Burnier J, Borowski M, Furie B, et al. Gamma-carboxyglutamic acid. Mol Cell Biochem 1981; 39: 191–207.

8. Esmon C, Sadowski J, Suttie J. A new carboxylation reaction. The vitamin K-de-pendent incorporation of H-14-CO3– into prothrombin. J Biol Chem 1975; 250: 4744–4748.

9. Matschiner JT, Bell RG, Amelotti JM, et al. Isolation and characterization of a new metabolite of phylloquinone in the rat. Biochim Biophys Acta 1970; 201: 309–315.

10. Li T, Chang C, Jin D, et al. Identification of the gene for vitamin K epoxide re-ductase. Nature 2004; 427: 541–544.

11. Rost S, Fregin A, Ivaskevicius V, et al. Mutations in VKORC1 cause warfarin re-sistance and multiple coagulation factor deficiency type 2. Nature 2004; 427: 537–541.

12. Shapiro SS. Treating thrombosis in the 21st century. N Engl J Med 2003; 349: 1762–1764.

13. Levi M, Eerenberg ES, Kampuisen PW. Anticoagulants. Old and new. Hamosta-seologie 2011; 31: 229-235.

14. Mannucci PM, Levi M. Prevention and treatment of major blood loss. N Engl J Med 2007; 356: 2301–2311.

15. Schulman S, Beyth RJ, Kearon C, et al. Hemorrhagic complications of antico-agulant and thrombolytic treatment: American College of Chest Physicians Evi-dence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; pp. 257S–298S.

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

Thrombosis and Haemostasis 112.5/2014 © Schattauer 2014

916 Schurgers, Spronk: Coagulation meets atherosclerosis

16. Holmes MV, Hunt BJ, Shearer MJ. The role of dietary vitamin K in the manage-ment of oral vitamin K antagonists. Blood Rev 2012; 26: 1–14.

17. Chatrou MLL, Winckers K, Hackeng TM, et al. Vascular calcification: the price to pay for anticoagulation therapy with vitamin K-antagonists. Blood Rev 2012; 26: 155–166.

18. Schurgers LJ, Uitto J, Reutelingsperger CP. Vitamin K-dependent carboxylation of matrix Gla-protein: a crucial switch to control ectopic mineralization. Trends Mol Med 2013; 19: 217–226.

19. Corral-Rodríguez MA, Macedo-Ribeiro S, Pereira PJB, et al. Leech-derived thrombin inhibitors: from structures to mechanisms to clinical applications. J Med Chem 2010; 53: 3847–3861.

20. Markwardt F. Historical perspective of the development of thrombin inhibitors. Pathophysiol Haemost Thromb 2002; 32 (Suppl 3): 15–22.

21. Nutt EM, Jain D, Lenny AB, et al. Purification and characterization of recombi-nant antistasin: a leech-derived inhibitor of coagulation factor Xa. Arch Bio-chem Biophys 1991; 285: 37–44.

22. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibril-lation. J Am Med Assoc 2001; 285: 2864–2870.

23. Wasmer K, Eckardt L. Management of atrial fibrillation around the world: a comparison of current ACCF/AHA/HRS, CCS, and ESC guidelines. Europace 2011; 13: 1368–1374.

24. Granger CB, Alexander JH, McMurray JJV, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2011; 365: 981–992.

25. Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvu-lar atrial fibrillation. N Engl J Med 2011; 365: 883–891.

26. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in pa-tients with atrial fibrillation. N Engl J Med 2009; 361: 1139–1151.

27. Fox KAA, Piccini JP, Wojdyla D, et al. Prevention of stroke and systemic embol-ism with rivaroxaban compared with warfarin in patients with non-valvular at-rial fibrillation and moderate renal impairment. Eur Heart J 2011; 32: 2387–2394.

28. Borissoff JI, Spronk HMH, Cate Ten H. The hemostatic system as a modulator of atherosclerosis. N Engl J Med 2011; 364: 1746–1760.

29. Sevigny LM, Austin KM, Zhang P, et al. Protease-activated receptor-2 modulates protease-activated receptor-1-driven neointimal hyperplasia. Arterioscler Thromb Vasc Biol 2011; 31: e100–106.

30. Shi X, Gangadharan B, Brass LF, et al. Protease-activated receptors (PAR1 and PAR2) contribute to tumor cell motility and metastasis. Mol Cancer Res 2004; 2: 395–402.

31. Lin H, Trejo J. Transactivation of the PAR1-PAR2 heterodimer by thrombin eli-cits β-arrestin-mediated endosomal signalling. J Biol Chem 2013; 288: 11203–11215.

32. Arachiche A, Mumaw MM, la Fuente de M, et al. Protease-activated receptor 1 (PAR1) and PAR4 heterodimers are required for PAR1-enhanced cleavage of PAR4 by α-thrombin. J Biol Chem 2013; 288: 32553–32562.

33. Borensztajn K, Spek CA. Blood coagulation factor Xa as an emerging drug tar-get. Expert Opin Ther Targets 2011; 15: 341–349.

34. Borensztajn K, Peppelenbosch MP, Spek CA. Factor Xa: at the crossroads be-tween coagulation and signalling in physiology and disease. Trends Mol Med 2008; 14: 429–440.

35. Zhou Q, Bea F, Preusch M, et al. Evaluation of plaque stability of advanced atherosclerotic lesions in apo E-deficient mice after treatment with the oral fac-tor Xa inhibitor rivaroxaban. Mediators Inflamm 2011; 2011: 432080.

36. Kalz J, Cate Ten H, Spronk HMH. Thrombin generation and atherosclerosis. J Thromb Thrombolysis 2014; 37: 45–55.

37. Hamilton JR, Cornelissen I, Mountford JK, et al. Atherosclerosis proceeds inde-pendently of thrombin-induced platelet activation in ApoE-/- mice. Athero-sclerosis 2009; 205: 427–432.

38. Patterson C, Stouffer GA, Madamanchi N, et al. New tricks for old dogs: non-thrombotic effects of thrombin in vessel wall biology. Circ Res 2001; 88: 987–997.

39. Wilcox JN, Noguchi S, Casanova J. Extrahepatic synthesis of factor VII in human atherosclerotic vessels. Arterioscler Thromb Vasc Biol 2003; 23: 136–141.

40. Wilcox JN, Smith KM, Schwartz SM, et al. Localization of tissue factor in the normal vessel wall and in the atherosclerotic plaque. Proc Natl Acad Sci USA 1989; 86: 2839–2843.

41. Borissoff JI, Heeneman S, Kilinç E, et al. Early atherosclerosis exhibits an en-hanced procoagulant state. Circulation 2010; 122: 821–830.

42. Borissoff JI, Otten JJT, Heeneman S, et al. Genetic and Pharmacological Modifi-cations of Thrombin Formation in Apolipoprotein E-deficient Mice Determine Atherosclerosis Severity and Atherothrombosis Onset in a Neutrophil-Depend-ent Manner. PLoS one 2013; 8: e55784.

43. Khallou-Laschet J, Caligiuri G, Tupin E, et al. Role of the intrinsic coagulation pathway in atherogenesis assessed in hemophilic apolipoprotein E knockout mice. Arterioscler Thromb Vasc Biol 2005; 25: e123–126.

44. Vicente CP, He L, Tollefsen DM. Accelerated atherogenesis and neointima formation in heparin cofactor II deficient mice. Blood 2007; 110: 4261–4267.

45. Westrick RJ, Bodary PF, Xu Z, et al. Deficiency of tissue factor pathway inhibitor promotes atherosclerosis and thrombosis in mice. Circulation 2001; 103: 3044–3046.

46. Tilley RE, Pedersen B, Pawlinski R, et al. Atherosclerosis in mice is not affected by a reduction in tissue factor expression. Arterioscler Thromb Vasc Biol 2006; 26: 555–562.

47. Loeffen R, Spronk HMH, Cate ten H. The impact of blood coagulability on atherosclerosis and cardiovascular disease. J Thromb Haemost 2012; 10: 1207–1216.

48. de Moerloose P, Boehlen F. Inherited thrombophilia in arterial disease: a selec-tive review. Semin Hematol 2007; 44: 106–113.

49. Houbballah R, LaMuraglia GM. Clotting problems: diagnosis and management of underlying coagulopathies. Semin Vasc Surg 2010; 23: 221–227.

50. Foley CJ, Nichols L, Jeong K, et al. Coronary atherosclerosis and cardiovascular mortality in hemophilia. J Thromb Haemost 2010; 8: 208–211.

51. Tuinenburg A, Rutten A, Kavousi M, et al. Coronary artery calcification in he-mophilia A: no evidence for a protective effect of factor VIII deficiency on atherosclerosis. Arterioscler Thromb Vasc Biol 2012; 32: 799–804.

52. Zwiers M, Lefrandt JD, Mulder DJ, et al. Coronary artery calcification score and carotid intima–media thickness in patients with hemophilia. J Thromb Hae-most 2012; 10: 23–29.

53. Bilora F, Zanon E, Petrobelli F, et al. Does hemophilia protect against athero-sclerosis? A case-control study. Clin Appl Thromb Hemost 2006; 12: 193–198.

54. Srámek A, Reiber JH, Gerrits WB, et al. Decreased coagulability has no clini-cally relevant effect on atherogenesis: observations in individuals with a hered-itary bleeding tendency. Circulation 2001; 104: 762–767.

55. Narins CR, Zaręba W, Moss AJ, et al. Relationship between intermittent claudi-cation, inflammation, thrombosis, and recurrent cardiac events among surviv-ors of myocardial infarction. Arch Intern Med 2004; 164: 440–446.

56. Lowe GDO, Sweetnam PM, Yarnell JWG, et al. C-reactive protein, fibrin D-dimer, and risk of ischemic heart disease: the Caerphilly and Speedwell studies. Arterioscler Thromb Vasc Biol 2004; 24: 1957–1962.

57. Salomaa V, Stinson V, Kark JD, et al. Association of fibrinolytic parameters with early atherosclerosis. The ARIC Study. Atherosclerosis Risk in Communities Study. Circulation 1995; 91: 284–290.

58. Smith FB, Rumley A, Lee AJ, et al. Haemostatic factors and prediction of is-chaemic heart disease and stroke in claudicants. Br J Haematol 1998; 100: 758–763.

59. Tataru MC, Heinrich J, Junker R, et al. D-dimers in relation to the severity of ar-teriosclerosis in patients with stable angina pectoris after myocardial infarction. Eur Heart J 1999; 20: 1493–1502.

60. Páramo JA, Orbe J, Beloqui O, et al. Prothrombin fragment 1+2 is associated with carotid intima-media thickness in subjects free of clinical cardiovascular disease. Stroke 2004; 35: 1085–1089.

61. Kienast J, Thompson SG, Raskino C, et al. Prothrombin activation fragment 1 + 2 and thrombin antithrombin III complexes in patients with angina pectoris: re-lation to the presence and severity of coronary atherosclerosis. Thromb Hae-most 1993; 70: 550–553.

62. Henareh L, Jogestrand T, Agewall S. Prothrombin fragment 1 + 2 is associated with intima media thickness of the carotid artery in patients with myocardial in-farction. Thromb Res 2009; 124: 526–530.

63. Di Tullio MR, Homma S, Jin Z, et al. Aortic atherosclerosis, hypercoagulability, and stroke the APRIS (Aortic Plaque and Risk of Ischemic Stroke) study. J Am Coll Cardiol 2008; 52: 855–861.

64. Borissoff JI, Joosen IA, Versteylen MO, et al. Accelerated In Vivo Thrombin Formation Independently Predicts the Presence and Severity of CT Angio-graphic Coronary Atherosclerosis. JACC Cardiovasc Imaging 2012; 5: 1201–1210.

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207

© Schattauer 2014 Thrombosis and Haemostasis 112.5/2014

917Schurgers, Spronk: Coagulation meets atherosclerosis

65. Price P, Faus S, Williamson M. Warfarin causes rapid calcification of the elastic lamellae in rat arteries and heart valves. Arterioscler Thromb Vasc Biol 1998; 18: 1400–1407.

66. Schurgers LJ, Joosen IA, Laufer EM, et al. Vitamin k-antagonists accelerate atherosclerotic calcification and induce a vulnerable plaque phenotype. PLoS one 2012; 7: e43229.

67. Krüger T, Oelenberg S, Kaesler N, et al. Warfarin induces cardiovascular dam-age in mice. Arterioscler Thromb Vasc Biol 2013; 33: 2618–2624.

68. Schurgers LJ, Aebert H, Vermeer C, et al. Oral anticoagulant treatment: friend or foe in cardiovascular disease? Blood 2004; 104: 3231–3232.

69. Weijs B, Blaauw Y, Rennenberg RJMW, et al. Patients using vitamin K antago-nists show increased levels of coronary calcification: an observational study in low-risk atrial fibrillation patients. Eur Heart J 2011; 32: 2555–2562.

70. Shanahan CM, Crouthamel MH, Kapustin A, et al. Arterial calcification in chronic kidney disease: key roles for calcium and phosphate. Circ Res 2011; 109: 697–711.

71. Nadra I, Mason J, Philippidis P, et al. Proinflammatory Activation of Macro-phages by Basic Calcium Phosphate Crystals via Protein Kinase C and MAP Ki-nase Pathways: A Vicious Cycle of Inflammation and Arterial Calcification? Circ Res 2005; 96: 1248–1256.

72. Tintut Y, Patel J, Parhami F, et al. Tumor necrosis factor-alpha promotes In vitro calcification of vascular cells via the cAMP pathway. Circulation 2000; 102: 2636–2642.

73. Pazár B, Ea H-K, Narayan S, et al. Basic Calcium Phosphate Crystals Induce Monocyte/Macrophage IL-1{beta} Secretion through the NLRP3 Inflam-masome In Vitro. J Immunol 2011; 186: 2495–2502.

74. Ewence AE, Bootman M, Roderick HL, et al. Calcium phosphate crystals induce cell death in human vascular smooth muscle cells: a potential mechanism in atherosclerotic plaque destabilization. Circ Res 2008; 103: e28–34.

75. Kapustin AN, Davies JD, Reynolds JL, et al. Calcium regulates key components of vascular smooth muscle cell-derived matrix vesicles to enhance mineraliz-ation. Circ Res 2011; 109: e1–12.

76. Van Der Meijden PEJ, Van Schilfgaarde M, van Oerle R, et al. Platelet- and ery-throcyte-derived microparticles trigger thrombin generation via factor XIIa. J Thromb Haemost 2012; 10: 1355–1362.

77. Olsson SB, Executive Steering Committee of the SPORTIF III Investigators. Stroke prevention with the oral direct thrombin inhibitor ximelagatran com-pared with warfarin in patients with non-valvular atrial fibrillation (SPORTIF III): randomised controlled trial. Lancet 2003; 362: 1691–1698.

78. Christersson C, Oldgren J, Wallentin L, et al. Treatment with an oral direct thrombin inhibitor decreases platelet activity but increases markers of inflam-mation in patients with myocardial infarction. J Intern Med 2011; 270: 215–223.

79. Bea F, Kreuzer J, Preusch M, et al. Melagatran reduces advanced atherosclerotic lesion size and may promote plaque stability in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2006; 26: 2787–2792.

80. Lee I-O, Kratz MT, Schirmer SH, et al. The effects of direct thrombin inhibition with dabigatran on plaque formation and endothelial function in apolipopro-tein E-deficient mice. J Pharmacol Exp Ther 2012; 343: 253–257.

81. Ragosta M, Gimple LW, Gertz SD, et al. Specific factor Xa inhibition reduces res-tenosis after balloon angioplasty of atherosclerotic femoral arteries in rabbits. Circulation 1994; 89: 1262–1271.

82. Liska DJ, Suttie JW. Location of gamma-carboxyglutamyl residues in partially carboxylated prothrombin preparations. Biochemistry 1988; 27: 8636–8641.

83. Bae J-S, Yang L, Rezaie AR. Factor X/Xa elicits protective signalling responses in endothelial cells directly via PAR-2 and indirectly via endothelial protein C re-ceptor-dependent recruitment of PAR-1. J Biol Chem 2010; 285: 34803–34812.

84. Spronk HMH, de Jong AM, Crijns HJ, et al. Pleiotropic effects of factor Xa and throm-bin: what to expect from novel anticoagulants. Cardiovasc Res 2014; 101: 344–351.

Coag

ulat

ion

prot

ease

s an

d CV

D

For personal or educational use only. No other uses without permission. All rights reserved.Downloaded from www.thrombosis-online.com on 2015-02-11 | ID: 1000364783 | IP: 137.120.195.207