Tissue Factor, Blood Coagulation, and Beyond: An Overview

Emerging evidence shows a broad spectrum of biological functions of tissue factor (TF). TF classical role in initiating the extrinsic blood coagulation and its direct thrombotic action in close relation to cardiovascular risks have long been established. TF overexpression/hypercoagulability often observed in many clinical conditions certainly expands its role in proinflammation, diabetes, obesity, cardiovascular diseases, angiogenesis, tumor metastasis, wound repairs, embryonic development, cell adhesion/migration, innate immunity, infection, pregnancy loss, and many others. This paper broadly covers seminal observations to discuss TF pathogenic roles in relation to diverse disease development or manifestation. Biochemically, extracellular TF signaling interfaced through protease-activated receptors (PARs) elicits cellular activation and inflammatory responses. TF diverse biological roles are associated with either coagulation-dependent or noncoagulation-mediated actions. Apparently, TF hypercoagulability refuels a coagulation-inflammation-thrombosis circuit in “autocrine” or “paracrine” fashions, which triggers a wide spectrum of pathophysiology. Accordingly, TF suppression, anticoagulation, PAR blockade, or general anti-inflammation offers an array of therapeutical benefits for easing diverse pathological conditions.


Introduction: Tissue Factor Biology
Tissue factor (TF), also known as factor III, essentially provides additional protection to vital organs prone to mechanical injury; its strategic location is considered as a hemostatic envelope for arresting bleeding from vascular beds. High TF expression is found in highly vascularized organs (cells) such as the brain (e.g., astrocytes), placenta (e.g., trophoblasts), and the lungs (e.g., alveolar cells) followed by the heart (e.g., cardiac myocytes, pericytes, fibroblasts), kidney, intestine, testes, and uterus (e.g., epithelial cells surrounding the organs). The low expression is detected in the spleen, thymus, and liver [1]. Circulating (blood-borne) TF is mainly derived from its expression in blood cells (e.g., monocytes, macrophages, granulocytes, and platelets), platelet-free microparticles containing TF shed from cells, or even soluble TF protein; the serum level can be easily measured by TF antigen, ELISA, TF procoagulant activity (PCA), and so forth (for review, see [2]).
Full-length TF (Figure 1), a membrane integral glycoprotein (46 kDa), is a 263-amino acid single-chain polypeptide classified as CD142 (Type II cytokine receptor) with a 219amino-acid extracellular N-terminus and a 23-amino-acid transmembrane domain followed by an intracellular 21amino-acid C-terminus [3,4]. The extracellular region contains FVII/VIIa binding domains. Extracellular soluble form (sTF) could be released from EC [5] in response to proinflammatory cytokines. The intracellular domain could undergo serine phosphorylation(s), which could modify its function [6][7][8]; for instance, the cytoplasmic domain negatively regulates TF expression, which is mediated by suppressed Erk1/2 phosphorylation.
TF initiates the extrinsic coagulation that plays an integral role in blood coagulation, thrombin (FIIa) generation, and thrombi formation in close relation to thrombosis and cardiovascular dysfunctions [9,10]. Such extracellular TF signaling proceeds with the sequential generation of coagulant mediators (FVIIa, FXa, and FIIa: active serine proteases) and fibrin production, all of which are proinflammatory [4]. TF extends its roles to diverse biological phenomena related to either ro both of these two major thrombotic and inflammatory events. Emerging evidence shows TF involvement in wound repairs, embryonic development, angiogenesis, tumor metastasis, cell adhesion/migration, innate immunity, and many pathological conditions.

TF-Initiated Extrinsic Coagulation
In a classical view, TF initiates the extrinsic blood coagulation, which proceeds as Ca+2-dependent extracellular signaling to sequentially activate zymogens: FVII, factor X (FX), and prothrombin (FII) for the generation of coagulant mediators (active serine proteases): FVIIa, FXa, and thrombin (FIIa), respectively. As a result, FIIa cleaves off fibrinogen (FBG) into fibrin monomers that cross-link to produce insoluble blood clots. The extrinsic pathway plays an integral role in blood coagulation complemented by the intrinsic pathway that ensures FIIa regeneration and clot production ( Figure 2, left panel) (for review, see [3,4,10,67]). The intrinsic pathway merging with TF-initiated extrinsic coagulation at FX activation is beyond the focus of this paper.

TF-Dependent FVII Activation.
The ability of FVII to bind its cofactor (TF) has been reported with a-1 : 2 stoichiometric ratio. It has long been established that Ca+2 and membrane anionic phospholipids are required for TFdependent FVII activation. Gla, EGF-1, EGF-2, and protease domain (PD) in FVII make essential contributions to the optimal interaction/binding with its counterpart: extracellular sTF1-219. It is said that zymogen FVII affinity for sTF causes secondary conformational changes of the PD, dictating the protease activity. EPR study shows multiple contacts between two proteins; the Gla binds sTF158 and 207, the EGF-1 binds sTF22 and 140, and the PD binds sTF45 and 94. In contrast, Gla-domainless-FVII shows a rapid loss in FVII binding affinity for TF. FVII affinity is also altered upon modification/conformational changes involving the EGF-1 region. Accordingly, any FVII global conformational misfolding/unfolding disrupting the binding sites could result in impaired FVII activation. Furthermore, there are high-affinity Ca+2 binding sites in Gla, EGF-1, and PD. It has been reported that one Ca+2 molecule binds to PD, another Ca+2 binds to the EGF-1 domain at a high-affinity site, and seven more Ca+2 molecules bind with variable affinity for the Gla domain. Thus, it seems likely that Ca+2 could play a critical role in FVII binding to TF. In addition to PS being essential for TF-dependent FVII activation, cholesterol enrichment of primary human monocyte-derived macrophages also drastically increases TF PCA [68].

Downstream Sequential
Activations. The catalytic function of the binary complex TF/FVIIa relying in mutual binding conformation is believed to be directly responsible for FX and FIX activation. FXa acts as a molecular switch not only receiving the upstream (extrinsic and intrinsic) signals but also dictating the downstream coagulation. Strategically, FXa is an active enzyme component coupled with FVa in prothrombinase complex located at the center of the blood coagulation cascade, which converges the clotting signals derived from both the extrinsic (FVII activation) and intrinsic (FIX activation) pathways. FXa also undertakes a feedback activation of FVII. Finally, FIIa derived from FII cleavage by FXa assumes the main coagulant function at the termination stage; it directly catalyzes FBG cleavage releasing fibrinopeptides for fibrin clot production upon cross-linking. In addition, FIIa activates FXIII, FXI, FVIII, or FV, assuring the propagation of blood coagulation.

TF Hypercoagulability Leading to Thrombosis
As a consequence of TF hypercoagulation, thrombosis featuring fibrin overproduction is a direct outcome 4 International Journal of Inflammation (3) Figure 3: Coagulation-inflammation-thrombosis circuit. TF hypercoagulability results in direct thrombotic actions (1). TF also plays converging and diverging roles in driving the coagulation-inflammation cycle ((2) coagulation-dependent inflammation and (3) inflammation-dependent coagulation). Namely, TF hypercoagulability could result in enormous inflammation as the result of continuously refueling the cycle in which coagulation and inflammation promote each other upon the cycle gaining its initial momentum. Thrombosisinflammation connection (4) is incorporated into the coagulation-inflammation cycle to form a complete coagulation-inflammationthrombosis circuit, which manifests diverse pathological conditions in relation to inflammation and thrombosis, including cancers, APS, cardiovascular dysfunctions, diabetes, obesity, and DIC.
FIIa activates platelets mainly through protease-activated receptor (PAR) and glycoproteins (GPs) IIb/IIIa, and GPIb. PAR-1 is a primary receptor for FIIa by which platelets are activated to aggregate [71]. Platelet aggregation constitutes thrombus formation involving cross-linking of adjacent platelets mediated by the interaction of activated GP IIb/IIIa with distinct amino acid sequences, LGGAKQAGDV, and/or RGD, at each end of dimeric FBG molecules [72]. Alternatively, FIIa-induced platelet activation could result from polymerizing fibrin, which involves the recognition sites in the cross-linking of polymerizing fibrin and surface integrins via GP Ib. In fact, GP Ib acts as an FIIa-binding site and promotes platelet activation by low FIIa concentrations [73].
In addition, FIIa activates FXIII, and FXIIIa facilitates the stabilization and cross-linking of fibrin clots.

Coagulation-Dependent Inflammation
Several lines of evidence reveal in vivo coagulationdependent inflammation. PARs generally mediate inflammation derived from coagulant mediators (e.g., FVIIa, FXa, and FIIa) and fibrin ( Figure 2; right panel). Moreover, deficiencies in natural anticoagulants (e.g., tissue factor International Journal of Inflammation 5 pathway inhibitor (TFPI), antithrombin (AT III), and activated protein C (APC)) are often susceptible to sepsis [77], disseminated intravascular coagulation (DIC) consequences [78], and inflammation [79]. Consistent with such notion of coagulation-dependent inflammation, anticoagulation readily results in anti-inflammatory effects in vivo and in vitro (discussed in Section 10).  [80], which is characterized by elevated plasma IL-6 and paw swelling accompanied by fibrin overproduction and platelet aggregation. TF mediates IL-1β-induced vascular permeability, an inflammatory index [81]. Conversely, TF deficiency reduces inflammation [82]. The ability of anti-TF Ab to prevent septic shock [83] and depress macrophage expression of adhesion molecule CD18 [84] is consistent with the proinflammatory function of TF.
The receptor activation involves a proteolytic cleavage of the extracellular domain, resulting in the formation of a new N terminus that in turn acts as a tethered ligand to interact with exoloop 2 Glu 260 and then activate heterotrimeric G proteins, triggering an array of intracellular signaling cascade. For instance, the involved sequences of PAR-1 (TLDPR 41 S 42 FLLRNP) and PAR-2 (SSKGR 36 S 37 LIGKY) are cleaved between R and S by serine proteases such as FIIa that also cleaves PAR-3 (TLPIKTFRGAP) and PAR-4 (LPAPRGYPGQV) at K/T and R/G, respectively [113].
The ability of PAR per se to mediate inflammatory responses [4,[104][105][106] is readily in line with coagulationdependent inflammation. It is now clear that PARs transmit clotting signals for proinflammation ( Figures 2 and 3(2)). PAR-1, 3, or 4 is responsible for FIIa signaling. PAR-2 or 3 mediates FXa signaling, while PAR-2 enables FVIIa signaling ( Taken together, it is evident that coagulant mediator (e.g., FVIIa, FXa, and FIIa) generation and fibrin production in the extracellular compartment via PARs signal transduction and intracellular activations result in the productions of cytokines, adhesion molecules, growth factors, and other proinflammatory components.
Thus, it is clear that TF initiates cross-talks of hypercoagulable states with inflammatory outcomes ( Figure 2). Furthermore, TF hypercoagulability results in enormous inflammation as the result of continuously refueling the coagulation-inflammation cycle (( Figure 3(2) and (3)) upon gaining its initial momentum such as local or systemic inflammatory/infectious conditions.

Thrombosis-Inflammation Connection
Thrombosis and inflammation are two major consequences of blood coagulation, both of which cross-talk and promote each other. Clinical association of thrombosis with inflammation has been reported in many cases [141]. Such inflammation-thrombosis connection ( Figure 3(4)) provides an alternative pathway that blood coagulation via its inflammatory consequence indirectly contributes to thrombosis. Several lines of evidence reveal thrombosis-dependent inflammation based on the ability of fibrin and its fragments to elicit IL-1β, IL-6, and IL-8 expression [98][99][100][101][102][103]. Further, platelet activation/aggregation participates in complement activation resulting in inflammatory responses. P-selectin as a C3b-binding protein sufficiently leads to C3a generation and C5b-C9 formation, which supports a novel mechanism of local inflammation in vascular injury sites [69,141].
Conversely, in vivo inflammation-dependent thrombogenesis also exists. IL-8 enhances fibrosis in rats [142]. In support of this notion, activation and antagonism of proinflammatory PARs, respectively, trigger and reduce thrombogenesis (for review, see [69]). For instance, PAR-4 activation [80] and PAR4-activating peptides [143] trigger platelet aggregation; consistently, PAR antagonism attenuates platelet activation/aggregation (for details, see Section 10.6) in line with such inflammation-dependent thrombogenesis. An earlier study has shown that P-selectin causes leukocyte accumulation to facilitate fibrin deposition [144], complementing thrombotic episodes. P/E/L-selectins, ICAM, and VCAM are responsible for leukocyte adhesion/rolling/recruitment interacting with platelets and VEC to enhance thrombus formation [145]. In parallel, selectin-deficient mice lacking the activation of the extrinsic pathway are defective in fibrin production [140]. Antibodies to cytokines and adhesion molecules attenuate venous thrombosis [146]. LYP20, an antibody against P-selectin, blocks leukocyte adhesion to EC and platelets [147] and modifies thrombosis [148], and Pselectin inhibition decreases vein wall fibrosis [149].
In addition, there is a general perception of inflammation-dependent thrombogenesis, which is supported by the observations that anti-inflammatory agents are of antithrombotic benefits. For instance, nonsteroid antiinflammatory drugs readily block thrombosis. COX-1 inhibitor such as low doses of aspirin suppresses platelet aggregation [150]. Similarly, COX-2 inhibition downregulates VEC/leukocyte activation [151]. (Figure 3(4)) is integrated into the coagulation-inflammation vicious cycle (Figure 3(2) and (3)), thus rounting a complete circuit to link among coagulation, inflammation, and thrombosis. Concomitant with suppressed TF expression by COX inhibitors [35][36][37], the antiinflammatory and antithrombotic properties of COX-2 inhibitors [150,151] seem likely to be in agreement with the involvement of TF hypercoagulability in driving the coagulation-inflammation-thrombosis circuit. Further, activated platelets stimulate TF expression [152], while antiplatelet agent (dilazep) inhibits TF expression [153]. Both observations are in favor of the thrombosisinflammation connection (Figure 3(4)) being part of the operative blood coagulation-inflammation-thrombosis circuit. The paradigm has also been observed in lung [154] and inflammatory bowel syndrome [155] while closely relating to cardiovascular risks [9,69].

Coagulation-Dependent Events: Thrombosis/Inflammation-Associated Conditions
Mounting evidence reveals that TF hypercoagulability plays pathogenic roles closely relating to its not only inflammatory but also thrombotic actions. By driving the circuit (Figure 3), TF hypercoagulability is readily involved in an array of metabolic syndromes (e.g., atherosclerosis, hypertension, diabetes II, and obesity) and other clinical manifestations International Journal of Inflammation 7 (e.g., cancers, antiphospholipid syndrome (APS), and fetal loss).

TF in Sepsis/DIC.
Hypercoagulation is often observed in septic shock including endotoxemia or systemic inflammatory responses after trauma, which mainly results from TF overexpression [156][157][158]. The ability of TF blockade to ease septic shock [83] or organ injury [159] points to a fundamental pathogenic role of TF in sepsis. Extrinsic infection/inflammation upregulating TF expression mediates enormous local or systemic intrinsic inflammation as well as a thrombotic condition via the operational circuit (Figures 2 and 3). A common manifestation presents DIC, an acquired disorder with hemostatic imbalance; excessive FIIa formation leads to fibrin deposition in microcirculation and consequent ischemic organ damage. Thus, such autocrine or paracrine TF signaling could lead to substantial tissue damages or multiple organ failure.

TF in Cancers.
TF overexpression has been reported in ovarian cancer [160], endometriosis [161], breast cancer [162], nonsmall cell lung carcinoma [163], prostate cancer [164], pancreatic cancer [165], melanoma [166], colorectal cancer [167], gastric cancer [168], esophageal cancer [169], hepatocellular carcinoma [170], brain tumor glioblastoma [171], leukemia [172], and lymphoma [173]. Accordingly, TF overexpression could be considered a biomarker for solid tumors [174]. The roles of TF in cancer have been demonstrated with severalfold relevance in relation to thrombotic condition, tumorigenesis per se and TF signaling (i.e., coagulationdependant inflammation). Cancer linked with hypercoagulability and thrombotic risk has long been recognized by Armand Trousseau since 1865. The American Society of Hematology calling for a special session on "cancer and thrombosis" addresses its complex clinical interface of prothrombotic association with malignancies and prophylactic approaches. Cancer certainly could be recognized as a prothrombotic risk factor, leading to, for instance, venous thromboembolism and its complication of pulmonary embolism and mortality. Namely, cancers readily induce thrombosis [175]. Enhanced TF expression typically accounts for the mode of mechanism of thrombosis accompanied by suppressed TFPI [176] and defective APC anticoagulation system. Not only tumor cellular membrane-bound TF, but also microparticle-associated TF [177] links cancer to thrombosis. In addition, the similar hypercoagulable state exists in cancer stem cells [178].
The critical role of TF in tumorigenesis is supported by the observations that inhibited TF expression blocks tumor growth, metastasis [179], angiogenesis [180], cell invasion [181], and many other cancer characteristics. TF per se plays important roles in cell proliferation, tumor development, and progression apart from the accompanying coagulationdependent inflammatory environment including MMP-9 [182], growth factors (VEGF, EGF, PDGF, etc.), and adhesion molecules certainly promoting "autocrine" tumorigenesis. Either VEGF or EGF in trun stimulates sustained TF expression [27,28]. PTEN loss and tumor hypoxia readily induce TF expression [183], which could highlight TF as a major player in cancer progression.
Tumor-expressed TF promotes growth by increasing cell survival and/or angiogenesis. TF and VEGF expressions mutually enhance each other [184], where VEGF is a known main angiogenic factor of cancer characteristics. TF cytoplasmic domain has been shown to be critical for VEGF expression [185]; conversely, VEGF causes TF promoter activation and involves gene upregulation with transcription factor NFAT involvement [28]. It is of particular interest to note that the serine phosphorylated cytoplasmic domain inhibits cellular cytotoxicity [186], thereby leading to increased tumor survival and metastatic rate. In addition, increased TF cytoplasmic domain phosphorylation and PAR-2 activation significantly correlate to cancer relapse [181]. Thus, a cooperation of the phosphorylated TF cytoplasmic domain with protease signaling could account for diverse contributions of TF to metastasis and angiogenesis [81,187].

TF in
Obesity. TF gene overexpression in obese has been reported for more than a decade [198,199] accompanied by upregulated PAI-1, angiogenesis, cell adhesion, and so forth, all of which could stem from TF hypercoagulability. Inflammation has been proposed to engage in obesity development [200], while less is clear about the precise role of thrombosis per se in obesity. With the functional coagulation-inflammation-thrombosis circuit (Figure 3), triggered inflammation constitutes the pathogenesis of obesity with manifestation including diabetes and cardiovascular risks (e.g., atherosclerosis, hypertension).
TF signaling ( Figure 2) sets up inflammation, in part well accounting for elevated levels of IL-6 [200,201], IL-8 [201], and TNFα [200] detected in obese subjects. Among which, either local or systemic inflammation (TNFα) significantly contributes to obesity [202]. Based upon high leptin and low adiponectin levels in obesity, the ability of leptin [30] or adiponectin [47], respectively, to augment or suppress TF synthesis could imply a mechanistic role of TF in developing inflammatory obesity. Furthermore, the involvement of 8 International Journal of Inflammation PPARα agonists [52]/activators [53] in downregulating TF expression also likely underlines a positive TF function in the process of inflammatory obesity.

TF in Diabetes.
Diabetes including type I and II is a hypercoagulable state [203] with elevated plasma levels of clotting factors (FVII, FVIII, FX, FXI, FXII), D-dimers, and TAT accompanied by decreased AT III, heparin cofactor II, or APC, presenting a thrombotic condition.
Under hyperglycemia, excessive plasma glucose nonenzymatically conjugates with plasma proteins (e.g., hemoglobins) to form AGE. AGEs through their receptors exhibit biological damage in various tissues such as renal failure and vascular complications. For instance, hyperglycemia induces damage to vascular endothelial cells, which is mediated by the complex activation of MAPK, PKC, NF-κB, and ICAM-1, primarily causing hemostatic alterations [204].
Increased circulating AGEs enhance TF expression [205], making diabetes a hypercoagulable and thrombotic condition [203,205,206]. Platelet TF in diabetes II appreciably increases [207]; increased FIIa and FXa generations are also found in diabetic platelets, enhancing the thrombotic nature. TF overexpression essentially promotes diabetes progression as well as its manifestation. As a consequence of diabetic TF hypercoagulability, elevated inflammatory mediators elicit cardiovascular complications including atherosclerosis. It is estimated that 80% diabetic patients die from a thrombotic disease and 75% of which result from cardiovascular complications [208]. Diabetic complications are more threatening than hyperglycemia per se; accordingly, relief of hypercoagulability could become far more important than glycemic control. Population-based clinical trials (ACCORD [209], ADVANCE [210] as well as VADT [211]) have demonstrated no benefit to cardiovascular risk in diabetes II upon glycemic control with significant low AGE (e.g., glycated hemoglobin <6%) for 2-3.5 years. Furthermore, rosiglitazone substantially lowering glycemia surprisingly increases the risk of myocardial infarction and death from cardiovascular causes [212]. For diabetic cardiovascular events, one could not expect that glycemic control per se significantly and promptly reverses the downstream damages done by AGEs. Apparently, nonglycemic factors (e.g., hypercoagulability, hypertension, and hyperlipidemia) play important roles in such complications.
Apart from thrombotic natures, TF could assume a pathogenic role in diabetic progression in a close relation to inflammatory process [213,214]. It is likely that TF signaling (Figure 2) through the coagulation-inflammationthrombosis circuit (Figure 3) operating in diabetes could well be responsible for insulin resistance. Proinflammatory mediator TNFα is known to promote insulin resistance in which serine phosphorylation of insulin receptor substrate (IRS) is encouraged. As a consequence of preventing insulin downstream IRS tyrosine phosphorylation, TNFα thereby blocks insulin signal transduction [215]. In sharp contrast to TNFα negative effects on insulin action, adiponectin positively enhances insulin sensitivity, and hypoadiponectinemia accordingly leads to insulin resistance [215]. Notably, anti-inflammatory adiponectin suppresses TF expression [47], which could be in support of the role of TF in diabetes pathology. From the viewpoint of PPARα activation improving insulin sensitivity, the observations of PPARα agonists [52]/activators [53] downregulating TF expression also likely point to positive TF function(s) in insulin resistance involving inflammatory diabetes development.
In summary, TF function has twofold significance in diabetes. TF not only dictates diabetic hypercoagulable nature and thrombotic outcomes [203], but also overlays its signaling in proinflammation (Figure 2) for insulin resistance. The ability of insulin [40] or an antidiabetic agent (metformin) [59] to attenuate TF expression seemingly reinforces a key pathogenic role of TF in diabetes.

TF in Cardiovascular
Complications. Cardiovascular complications are a group of disorders closely associated with either inflammation or thrombosis or both. In these regards, it is not surprising that TF plays a major role in their pathogeneses [9]. TF overexpression, often correlated to gain-of-function of TF promoter polymorphism (A603G), promotes the development of cardiovascular diseases [216]. It has long been established that TF participates in the phase III of plaque rupture [217] during atherogenesis. TF expression is upregulated in atherosclerotic plaques of patients with unstable angina and myocardial infarction [218].
TF hypercoagulability driving the coagulationinflammation-thrombosis circuit (Figure 3) readily extends its diverse consequences to cardiovascular complications and vascular diseases [219] including arrhythmias [58], arterial hypertension [220], hypertrophy [221], ACS [222,223], andatrial fibrillation (AF) [224], TF hypercoagulability with elevated proinflammatory cytokines ( Figure 2) could in part well contribute to atherosclerosis known as chronic inflammatory disease [217]. In cultures, recombinant TF induces cellular apoptosis with increased caspase-3 activity and nuclear location of p53 while increasing cellular proliferation/hypertrophic growth [221]. As a consequence of accelerated cardiomyocyte turnover, TF could contribute to the induction and progression of cardiac hypertrophy. Angiotensin II stimulates TF synthesis [220], mediating hypertensive action. Histamine augments TF expression, accounting for its action in ACS [225]. In conjunction with its effects on endothelial damage/dysfunction and angiogenic actions [224], TF upregulation could well be involved in a thrombogenic state of AF [224].
In contrast, TF deficiency in mice shows cardiac fibrosis [226,227] largely based upon TF functions in normal extracellular cardiac homeostasis, extracellular matrix regulation, and vascular maintenance [227]. Apparently, cardiac bleeding/hemorrhages in TF deficiency certainly encourages its fibrosis [226] where PA involvement could also be ensured by insufficient FIIa generation [74][75][76]. It awaits further confirmation in human conditions. presence of aPL including lupus anticoagulants, anticardiolipin antibodies, and anti-β2-glycoprotein-1 (β2GPI) antibodies. It is proposed that TLR-4 mediates anti-β2GPIinduced TF expression [228]. Alternatively, TF overexpression results from APS-associated complement activation. aPL activates complement via the classical pathway; activated complement (e.g., C5a) drastically stimulates TF synthesis. This autoimmune thrombophilic condition is largely due to enhanced coagulation (e.g., TF overexpression) accompanied by attenuated downregulation of blood coagulation (e.g., inhibited APC, TFPI, and AnxA5) and suppressed fibrinolysis. Increased microparticles and TF expression are found in APS with prothrombotic conditions of various manifestations, most commonly venous and arterial thromboembolism and recurrent pregnancy loss.
In addition, TF could play a pathological role in APS manifestation. It is not surprising if APS of TF overexpression also presents a hyperinflammatory condition in view of the paradigm of coagulation-inflammation-thrombosis circuit ( Figure 3). Apart from that aPL-induced complement activation contributes to inflammation [229], TF signaling could well account for increased TNFα [230][231][232] production as major proinflammation reported in APS patients [233] in which TNFα also seems to be responsible for its manifestation: fetal damage [232].

TF in Miscarriage.
Miscarriage including fetal death, preeclampsia, and intrauterine growth restriction often closely links to APS involving complement and angiogenic actions. During trophoblast differentiation, aPL activates complement via the classical pathway. Complement activation (C3 and C5a) directly mediates placental injury and causes fetal loss and growth restriction, resulting from an imbalance of angiogenic factors (e.g., VEGF and placental growth factor) as well as their corresponding receptors that are required for normal placental development [234].
Alternatively, TF overexpression is triggered by aPLinduced complement activation, and TF signaling fulfills such miscarriage/placenta damage/fetal injury. For instance, recent research demonstrates that neutrophil activation by TF/FVIIa/PAR-2 signaling [235] mediates aPL-induced pregnancy complication. In fact, TF on neutrophils and monocytes is a critical mediator in trophoblast injury and embryo damage in aPL-dependent or independent pregnancy loss [236]. Rapid increases in decidual and systemic TNF-α level are also responsible for fetal death/loss [232], which could be in line with TF signaling (Figure 2) playing a pathogenic role. Further, FIIa-induced platelet activation/aggregation activates complements, possibly conferring the direct fetal damage [234].
Anti-TF mAb prevents aPL-induced pregnancy loss [234], while statins [237,238] may be a good treatment for women with recurrent miscarriages and intrauterine growth restriction. These clinical studies are consistent with a pathogenic role of TF in APS-induced fetal damage.

TF in Wound
Healing. Wound, including diabetic foot, healing process generally consists of three phases (inflammatory, proliferative, and remodeling phases) that continuously overlap one another during the process. Hemostasis initiates angiogenesis-dependent wound healing. TF overexpression often occurring after wounding, trauma, or surgeries in part accounts for hypercoagulability encouraging wounding healing [239][240][241]. Given that inflammation involved in the initial phase, such "autocrine" or "paracrine" TF signaling essentially ensures fibrin matrix formation, angiogenesis, production of growth factors (VEGF, PDGF, bFGF, TGFα/β, etc.), adhesion molecules, and so forth, (Figures 2 and 3), all of which significantly contribute to wound healing process.

TF in Development.
Limited evidence reveals that TF extracellular domain is essential for embryogenesis [242][243][244], which is believed to be mediated by TF-dependent FIIa generation and PAR-1 activation. Thus, TF serves as an important morphogenic factor during embryogenesis. Apparently, TF signaling with FVIIa, FXa, and FIIa generation for PAR activation/transduction triggers an array of biological events as a consequence of proinflammation ( Figure 2, right panel), among which growth factors (EGF, VEGF, PDGF, bFGF, etc.) could play major roles in development [245]. Consistently, inactivation of TF gene results in embryonic lethality in a murine model [243]. It is said that TF expression coordinated with TFPI, ATIII, and FVII levels could be critical in embryonic development [246].

TF in Other Diseases
. TF expression is often upregulated by an antibody to platelet factor 4 (PF4) upon a long exposure/treatment of heparin [247]. In heparin-induced thrombocytopenia, PF4 also impairs APC activity, making a pronounced hypercoagulable and prothrombotic condition. TF overexpression in adult onset asthma significantly correlates to the gain-of-function of TF promoter polymorphism (A603G) [216].
In response to surgical procedures, enhanced TF synthesis is reported in major surgeries such as hip replacement, cardiopulmonary bypass (CPB) [253] or transplantation [254][255][256]. Upon tissue injury, exposure to protein disulfide isomerase and PS readily activates TF [10-12] and its signaling. It is plausible that TF hypercoagulability in part accounts for postsurgical inflammatory responses.
With regard to lifestyles, smoking upregulating TF expression apart from its apparent free radical inhalation elicits diverse health problems including cardiovascular and cancer risks. High-fat diets [257], oral contraceptives [15,258], and estrogen replacement [21, 259] also promote TF expression, possibly driving the circuit (Figure 3) for diverse clinical manifestations in relation to inflammation or/and thrombosis.
In addition, TF overexpression is associated with other pathological conditions such as liver cirrhosis [260], synovial inflammation [261], sickle cell anemia [262], or hepatic necrosis during cholestasis [263]. These pathological conditions likely result from the coagulation-inflammationthrombosis circuit ( Figure 3); the precise mechanisms of action however remain to be defined.

Noncoagulation-Mediated TF Roles
The signaling function of TF cytoplasmic domain has been demonstrated although its biochemical mechanism remains unclear. For instance, cross-talk between intracellular TF domain with integin α3β1 promotes cell migration [264], while the cytoplasmic domain possibly upon phosphorylation of the three serine residues causes hyperchemotaxis [265]. The cytoplasmic domain contributes to renal albumin retention, and its renal expression protects against proteinuria. Consistently, the absence of the cytoplasmic domain is associated with increased albuminuria, increased spontaneous glomerular TNFα production, podocyte effacement/inflection, reduced podocyte numbers, resulting in albuminuria and proteinuria [266]. For cancer progression/relapse, increased cytoplasmic domain phosphorylation significantly correlates to metastasis and angiogenesis [267]. It is proposed that the cytoplasmic domain per se is critical for VEFG expression [185], an important angiogenic component in tumorigenesis.

FIIa Inhibition.
Heparin shows a variety of antiinflammatory potentials (for review, see [328]). Heparinbonded circuit prevents the increases in IL-6 and IL-8 in CPB patients [329], while heparin bolus reduces neutrophil activation without affecting platelet aggregation [330]. Heparin is also considered a treatment for pregnancy loss [331].

By
It has long been established that APC protects from sepsis, DIC, and endotoxemia [352,353]; APC is recognized as one of the effective anti-inflammatory agents in clinical applications. APC inactivates the production of IL-1, -6, -8 or TNF-α [354]. APC consistently reduces septic mortality and blocks DIC upon E. coli. infection in either animal or human models [355,356].
APC antithrombotic potential is implied by increased APC resistance [363] and the deficiency [364] or low plasma level [365] of APC observed in thrombosis. APC profibrinolytic effects by inactivation of PAI-1 [366] and TAFI [367] synergistically diminish the direct thrombotic inputs from blood coagulation cascade. However, APC antithrombotic potential remains in the experimental stage of animal studies. For instance, a recombinant human APC (LY203638) inhibits arterial thrombosis in a canine model [368]. A human APC product (CTC-111) reduces venous thrombosis in mice [369]. FLIN-Q3 diminishes A-V shuntinduced thrombosis in guinea pigs [370]. hAPC attenuates rat mesenteric occlusion [371], and rhAPC inhibits arterial thrombosis in baboons [372]. Infusion of bovine APC suppresses thrombus formation in rats [373] and rabbit microarterial thrombosis [374]. A rabbit APC-loaded stent reduces thrombus and platelet deposition in vitro and in vivo [375].

Remarks
Blood coagulation, a primitive biological phenomenon in the animal kingdom, has historically been recognized as a host defense to prevent one from bleeding to death. TF-initiated extrinsic pathway, known as being inducible compared to constitutive intrinsic pathway, plays an integral role in blood coagulation, FIIa generation, and thrombus formation (for review, see [3,10,67]). Accumulating evidence demonstrates TF diverse biological effects in local or systemic inflammation [4]. Not only does the extrinsic pathway but also intrinsic pathway results in inflammation [416]. Interestingly, TF hypercoagulability refuels a coagulation-inflammation-thrombosis circuit in "autocrine" or "paracrine" fashion ( Figure 3), thereby manifesting many pathological conditions. Such extracellular TF signaling activates cells, and its pronounced effects include proinflammatory cytokine production ( Figure 2). It has been elucidated that inflammasomal activation [417] in response to innate pathogens [418], viral [419], fungus [420], influenza [421], microbes [422], and chemicals (e.g., cholesterol [423], uric acid [424], or aluminium hydroxide [417] crystals, asbestos [425], silica [425]) is essential for cytokine secretion (for review, see [426]). It, however, remains elusive if inflammasomal activation is involved in such inflammatory process triggered by TF signaling. Thus far, there is no indication whether coagulant mediators (e.g., FVIIa, FXa, FIIa) could activate procaspase-1, facilitating proinflammatory cytokine secretion. Could PAR activation directly turn on inflammasomal activation, an interesting question seemingly further addressing the similar issues if inflammasomal activation is critical for coagulation-dependent inflammation?
Among diverse clinical conditions associated with TF overexpression and its signaling mentioned herein, the close link between TF hypercoagulability and neurological disorders is however seldom reported. Although high TF expression in the brain could in part account for thrombotic stroke consequences, it certainly warrants investigation to explore if TF and its signaling participate in other neuronal dysfunctions or CNS disorders. It would also be of particular interest to determine the biological events of coagulation concerning not only innate [248] but also adaptive immunity of B/T cell equipped and featured with such "autocrine" or "paracrine" TF signaling, if any.
In view of the paradigm of coagulation-inflammationthrombosis circuit eliciting diverse pathological events (Figure 3), targeting TF hypercoagulation is of therapeutical relevance. Apparently, the development of anticoagulants is of broad pharmaceutical interests; anticoagulation could turn into strategic approaches for intervention and cure not limiting to thromboprophylaxis. It is highly promising that anticoagulants available arresting different stages of blood coagulation cascade [427] exhibit benefits other than hemostasis. Approaches to direct FVIIa, FXa, or FIIa inhibition readily demonstrate broad clinical applications (Table 1). In these regards, TF posttranslational downregulation (including encryption) could deserve attention for interventional therapeutical relevance in prospective of such upstream downregulation of the extrinsic pathway ( Figure 2, left panel) with broad suppression of downstream proinflammatory coagulant mediators (e.g., FVIIa, FXa, and FIIa) as well as fibrin production.
The observations of anticoagulation exhibiting anticancer properties clearly demonstrate the new frontiers of the emerging therapeutical era. Direct PAR blockade could be part of therapeutically targeting coagulation-dependent inflammation and the circuit (Figure 3). Further research is needed to study if PAR antagonisms could widely exhibit an array of clinical benefits to relieve diseases including cancer, obesity, diabetes, APS, and others in addition to inflammation and thrombotic related cardiovascular complications.

16
International Journal of Inflammation Like any other therapies, anticoagulation bears certain limitations and cautions for its applications. For employing anticoagulants, bleeding or hemorrhage episodes become major concerns in relation to the safety and efficacy issues upon long-term uses or "over dosages." With cautions in mind, routine monitoring for hemostatic properties is highly recommended. Further investigations warrant addressing rationally designed anticoagulant approaches to achieving/maintaining/weighing in therapeutical benefits for diverse clinical applications.