Decryption of tissue factor

Abstract

Tissue factor (TF) is a transmembrane protein which, in complex with factor (F)VIIa, initiates blood coagulation. Numerous studies have determined TF epitopes and individual amino acids which play an important role in the TF/FVIIa complex formation and its activity towards natural substrates. However the subject of cell-surface TF activity remains controversial. It has been almost commonly accepted that TF on the cell surface has low (if any) activity, i.e. is encrypted and requires specific conditions/reagents to become active, i.e. decrypted. One of the leading theories suggests that cell membrane lipid composition plays a crucial role in TF decryption, whereas another assigns the key role to the formation of the Cys¹⁸⁶-Cys²⁰⁹ disulfide bond. Despite a number of studies published from several laboratories, the role of this bond in the activity of the TF/FVIIa complex remains elusive and controversial. One of the causes of this controversy could be related to the lack of specificity of the reagents used for the cell treatment leading to possible alterations in other cell surface proteins and cell membrane environment. In conclusion, the influence of the Cys¹⁸⁶-Cys²⁰⁹ this bond on cell surface TF function remains unclear.

Activity of tissue factor

Tissue factor is a transmembrane protein expressed in various cells and found at the highest concentrations in brain, lungs and placenta[1–3]. Upon vascular injury or agonist stimulation (primarily with inflammation-related cytokines), tissue factor becomes exposed to the blood flow, binds circulating factor VIIa and initiates processes leading to thrombin generation and clot formation[4–7]. In healthy individuals with no pathologic conditions, there is no detectable active tissue factor present in contact with blood[8–10]. Tissue factor by itself has no proteolytic/amidolytic activity but increases activity of factor VIIa by several orders of magnitude upon the factor VIIa/tissue factor complex formation[11, 12]. Binding of tissue factor to factor VIIa is calcium-dependent, although the enzymatic complex of these two proteins can be formed in the absence of divalent metal ions as well[11, 12]. However, the affinity of this complex in the absence of calcium ions is quite low. The activity of the factor VIIa/tissue factor complex can be divided into surface-independent and surface-dependent. The former group of reactions occurs in solution and is characteristic for low molecular weight substrates, which bind directly to the active site of factor VIIa. In these reactions, tissue factor increases amidolytic activity of factor VIIa by 1-2 orders of magnitude[12–14] and localization of the factor VIIa/tissue factor complex on the membrane surface does not influence amidolytic activity of the complex[15]. However for the physiologically-relevant proteolysis of natural substrates of the factor VIIa/tissue factor complex (factors IX, X and VII), anchoring of the complex to the membrane surface by tissue factor becomes essential[16, 17]. The complex formed by factor VIIa and tissue factor proteins lacking the ability to bind to the membrane cannot efficiently (if at all) proteolyze its natural substrates and could become an inhibitor of blood coagulation by tightening circulating factor VIIa into an inefficient complex[18]. Although there is a common agreement about an essential role of the membrane surface, particularly of the cell membrane, for tissue factor activity in vivo, the mechanism regulating activity of cell tissue factor has been a subject of discussion for the last two decades.

Decryption of tissue factor

The prevailing hypothesis related to cell tissue factor activity is that many cells, including those in contact with blood, contain tissue factor on their surface. However under normal physiologic conditions that tissue factor is not active, i.e. “encrypted”, and needs “decryption” to express the procoagulant activity[19].Several mechanisms, often contradictory, have been hypothesized in attempts to explain “encryption-decryption” of tissue factor activity.

One of the suggested methods for the decryption of tissue factor on the cell surface consists of the treatment of tissue factor-bearing cells with calcium ionophore[20–26]. Upon such treatment, calcium ions are released from the internal stores followed by the influx of extracellular calcium across the membrane leading to an increase in the concentration of cytosolic calcium ions. Suggested increases in cell surface tissue factor activity, caused by the calcium ionophore treatment, range from 2 to 10-fold. However other studies assign this calcium ionophore-induced increase in tissue factor activity to an increased expression of tissue factor protein[20] or to changes in the cell membrane environment, particularly in an increased expression of acidic phospholipids [21, 23], 26]. In some studies, changes observed in the cell membrane environment are associated with the cell death[22, 25]. A few studies published from several laboratories suggested a role for the cell membrane lipid rafts, particularly for cholesterol accumulated in those rafts, as contributors to the encryption-decryption of tissue factor activity[27–29], although there are certain discrepancies between the observed results and suggested mechanisms for this process. An increase in tissue factor activity has been reported when lipopolysaccharide-stimulated monocytes were treated with platelets[30–32]. This observed increase in activity was quite limited (2 to 3-fold) and could be (at least in part) assigned to an increase in tissue factor antigen expression by monocytes[32].

In our laboratory we investigated the subject of tissue factor encryption-decryption quantitating tissue factor antigen and activity on blood monocytes and platelets, and cultured monocytic cells and purified platelets[9]. No tissue factor antigen or activity was observed on non-stimulated cells. Upon the stimulation of purified monocytic cells with lipopolysacchrides, the vast majority of them expressed both tissue factor antigen and activity. Changes in antigen concentration over time overlapped with the changes in activity (Figure 1), suggesting that no decryption was required to explain the increase in tissue factor activity upon stimulation. No tissue factor antigen or activity was observed on platelets stimulated either in blood or in a buffer environment[9]. Publications from several other laboratories also showed a correlation between the tissue factor antigen expression and its activity on cell surface[33–35] raising a question whether tissue factor decryption is common for all cells or it is a property of certain cell types only.

Figure 1.

Stimulation of THP-1 monocytes with LPS. Increase in monocyte TF antigen (■) and clotting activity (♦) over time upon stimulation with 1 µg/mL LPS.

Oxidation/reduction of cysteines

It has been suggested by several groups of investigators that cell surface tissue factor decryption is related to the Cys¹⁸⁶-Cys²⁰⁹ disulfide bond formation. The key data leading to this hypothesis were based on the mutational studies by Edgington’s group[36]. The authors mutated either Cys⁴⁹ and Cys⁵⁷ (which potentially could form a disulfide bridge in the N-terminal part of the extracellular domain of tissue factor) or Cys¹⁸⁶ and Cys²⁰⁹ (which could form a disulfide bond in the vicinity of the membrane) substituting them with serine residues. Mutations of Cys⁴⁹ and Cys⁵⁷ had no effect on activity of the factor VIIa/tissue factor complex (extrinsic factor Xase) , whereas mutations of Cys¹⁸⁶ and Cys²⁰⁹ decreased the activity of this complex by approximately 3-fold due to an impaired binding of factor VIIa to this tissue factor mutant. Based on these data it was concluded that the N-terminal pair of cysteines plays no role in tissue factor activity, whereas the disulfide bridge in the C-terminus of the extracellular domain is essential for the decryption of tissue factor. However, the Cys186Ser and Cys209Ser mutant, in contrast to the wild-type protein and Cys49Ser and Cys57Ser mutant, lacked glycosylation, which could be the contributing factor to reduced tissue factor activity[37].

The role of the Cys¹⁸⁶-Cys²⁰⁹ disulfide in the regulation of cell-surface tissue factor has been the subject of debates for the last several years. An increased tissue factor activity has been suggested upon the treatment of cells with mercuric chloride, an oxidizing agent[38]. The authors concluded that this reagent restores the Cys¹⁸⁶-Cys²⁰⁹ disulfide bond leading to the observed changes in tissue factor activity. However, no data (experimental or theoretical) are provided supporting the re-formation of the disulfide bridge hypothetically present in a reduced form on untreated cells. Existing publications related to the protein treatment with mercuric chloride show that it oxidizes only a single thiol group[39, 40] and that a similar effect could be achieved by treating tissue factor-bearing cells with other metal compounds[41].

In several publications from the laboratory of W. Ruf, it has been suggested that protein disulfide isomerase (PDI) is responsible for the regulation of cell tissue factor activity via its effect on Cys¹⁸⁶-Cys²⁰⁹ disulfide bond oxidation/reduction[42–44]. Similarly, studies from the laboratory of B. Engelmann proposed that tissue factor activation by PDI occurs due to the isomerization of a mixed disulfide and an intramolecular Cys¹⁸⁶-Cys²⁰⁹ bond formation[45, 46]. In contrast to these publications, data acquired in several other laboratories showed that the tissue factor activity-enhancing effect of PDI is related to the presence of acidic phospholipids either as a contaminant in the PDI preparations[47, 48] or due to its relocation to the tissue factor-bearing cell surface upon treatment with PDI or mercuric chloride[49–52]. Moreover, Bach and Monroe suggested that Cys¹⁸⁶ and Cys²⁰⁹ are not available for interaction with PDI when tissue factor is bound to factor VIIa, i.e. at physiologically relevant conditions[53]. Lately, it has been suggested in a report published by Kothari et al. that the Cys¹⁸⁶-Cys²⁰⁹ disulfide bond is not essential for the cell tissue factor procoagulant activity[54].

There are several possible explanations for the contradictions observed in publications describing the influence of PDI on tissue factor activity: 1) a variability of reagents and procedures used in different laboratories; 2) the use of different cell lines; and 3) the lack of specificity of PDI for tissue factor. In addition to the effect on cell membrane lipid composition[50, 51], PDI targets multiple proteins in the cell, catalyzing thiol-disulfide exchange[55–57]. Both of those processes could alter tissue factor activity without changing the status of thiols/disulfides. Additionally, PDI can alter thrombin generation in a tissue factor-independent manner via coagulation factor ligation to platelets[58] or by catalyzing complex formation with participation of thrombin and antithrombin[59].

Conclusion

The subject of tissue factor encryption-decryption in general and the role of the Cys¹⁸⁶-Cys²⁰⁹ disulfide bond in that process in particular remains controversial, and further studies are necessary to resolve this conundrum.