The Pathophysiology of The Antiphospholipid Syndrome: A Perspective From The Blood Coagulation System

1.3.1. Anticoagulant treatment

Up to these days, standard treatment is anticoagulation with vitamin K antagonists (VKA) in patients who develop thrombosis. ² Standard treatment of CAPS is based on high doses of steroids and heparins. For patients with obstetric manifestations, standard treatment also may include the administration of aspirin in combination with unfractionated heparin or low molecular weight heparin. ² Recently, successful experiences for anticoagulation based on the use of direct oral anticoagulants namely rivaroxaban, apixaban, edoxaban, and dabigatran, has been published. ²⁵

1.3.2. Alternative therapies

Several alternative therapies in addition to anticoagulant therapy such as statins and hydroxychloroquine (HQ) have been described. HQ has some effects in vitro that may help the management of APS, for example, reducing blood viscosity and platelet aggregation. ²⁵ Also, it may have immunological effects such as inhibition of activation of intracellular Toll-like receptors (TLRs) as well as reduction of the production of IL-1, IL-2, IL-6, and TNF-α. ²⁵ Moreover, HQ reduces the activation and expression of endosomal NADPH oxidase 2 (NOX2) in human umbilical vein endothelial cells (HUVECs) stimulated with TNFα or sera from preeclamptic women. Lastly, it prevents the loss of zonula occludens 1 (ZO-1) protein, thus, reducing the TNFα or preeclamptic sera-induced increased permeability of the HUVECs monolayer.^26,27

Statins are inhibitors of 3-hydroxy-3-methylglutaryl coenzyme A reductase that inhibit the synthesis of cholesterol and enhances its clearance by raising the level of low-density lipoprotein receptors. ²⁸ It has been described that they may block some aPLs-induced effects. Indeed, some of these effects are: enhancing tissue plasminogen activator (tPA) expression while inhibiting the endothelin-1 (ET-1) expression (consequently, they may reduce the production of reactive oxygen species); inhibition of platelet adhesion and aggregation; inhibition of recruitment of inflammatory cells by repressing the expression of adhesion molecules (such as intercellular adhesion molecule 1 -ICAM-1- and P-selectin); reduction of the levels of C-reactive protein; shifting from T_H1 towards T_H2 differentiation; and inhibition of the upregulation of endothelial major histocompatibility complex class II and costimulatory molecules. All these effects could be dependent or independent of the reduction of cholesterol levels.^28,29 Furthermore, statins like mevastatin and pravastatin induce the expression of heme oxygenase (HO)-1 through the NrF2 signaling pathway ³⁰ or the PKCa/Pyk2/p3 MAPK/ JNK1/2/AP-1 signaling pathway. ³¹ The expression of HO-1 inhibits the activation of NF-κB p65 (31) and additionally, activation of NrF2 which allows the transcription of antioxidant proteins such as superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GPx). ³⁰ As pravastatin does, SM934 an artemisinin derivative, induces the translocation of NrF2 to the nucleus where it stimulates the transcription of HO-1, SOD, GPx, and CAT. ³²

4.1.1. Placental histopathology

In obstetrical APS, a failure in the spiral artery remodeling by the extravillous trophoblast induces a reduction or interruption of the maternal blood supply to the placenta generating either hypoxic/ ischemic damage or inadequate nutrient influx to the fetus and/or an increased flux that may cause placental damage. ⁶³

In aPLs-treated mice, the histopathological findings show a higher TF staining on the decidua and embryonic debris as compared with mice tissues treated with IgG from healthy subjects (HS). This increased TF was neither associated with a higher staining for fibrin nor evidence of thrombosis, a fact that strongly suggests that the TF-dependent thrombosis induced by aPLs is not the cause of fetal loss. Fetal loss, C3d deposition, and infiltration of neutrophils induced by aPLs were inhibited by inhibiting TF with specific antibodies. ⁶⁴ On the other hand, as compared with the placental of healthy women, the histopathological findings in the placenta of APS patients with obstetric manifestations were a greater decidual, vascular and intervillous necrosis; increased syncytial nodes; abnormalities in spiral artery remodeling; reduction of vasculo-syncytial membranes; placental infarction; fibrin deposition in syncytial nodes; and stromal fibrosis.^63,65

Because β2GP1 is constitutively expressed on the surface of all subpopulations of the placental trophoblast and maternal endothelial cells, it is likely to hypothesize that this protein is the target for α-β2GP1 and other aPLs. Binding of these autoantibodies to β2GP1 may induce a proinflammatory, antimigratory, and antiangiogenic phenotype in the trophoblast. ⁶³

4.1.2. Complement activation

Plasma levels of C5a and the complement terminal complex C5b-9 were analyzed in 60 APS patients (43 non-pregnant women and seventeen pregnant women) and sixteen healthy women (eight non-pregnant women and eight pregnant women). Patients were classified according to their thrombotic and/or pregnancy morbidity. Moreover, pregnant patients were further classified based on the pregnancy prognosis as favorable or unfavorable. Levels of C5a in patients with thrombotic manifestations without a history of pregnancy morbidity were similar to those with healthy women. In contrast, C5a levels were higher in patients with thrombotic manifestations and pregnancy morbidity history as compared with healthy women. Furthermore, C5a levels in patients with obstetrical manifestations but favorable prognosis were not different from the levels of controls. In contrast, C5a levels in pregnant patients with unfavorable prognosis were higher than those of control women. Moreover, levels of complement terminal complex C5b-9 and complement regulatory proteins (CD46, CD55, and C59) were analyzed in APS and their controls. In APS patients, there was an increase in the C5b-9 levels and a reduced expression of CD59. Additionally, in pregnant patients with unfavorable prognosis, increased C5b-9 levels and reduced CD55 levels were observed. As a consequence, it is likely that a reduced expression of the complement regulatory proteins may promote a hyperactivation of the complement system and the subsequent induction of inflammation, thrombosis, and tissue damage. ⁶⁶

4.1.3. Neutrophil activation mediated by aPLs

The effects of the neutrophil’s extracellular traps (NETs) on activation of HUVECs and trophoblast have been analyzed. Increased levels of myeloperoxidase (MPO), cell-free DNA, MPO-DNA complex, and neutrophil elastase (NE) were found in the serum of 22 patients with obstetric APS as compared with controls. When the ability to release NETs in APS patients versus controls was evaluated, it was found that neutrophils from APS patients released spontaneously more NETs than their counterparts. This effect was attributed to aPLs of the APS patients because incubation of neutrophils from healthy women with aPLs induced an enhanced release of NETs secondary to increased production of NADPH oxidase-dependent reactive oxygen species (ROS). ⁶⁷ On the other hand, the anti-migratory and anti-angiogenic ability of NETs on trophoblast cells and endothelial cells have been studied. ⁶⁷ In murine models, neutrophils may be directly activated by aPLs through binding of β2GP1 or, indirectly, through PLs-induced activation of the complement system and generation of C5a which, in turn, interacts with C5aR on the neutrophil’s surface or forms the TF/activated factor VII/ protease-activated receptor 2 (PAR2) complex. Main effects of neutrophil activation by aPLs were chemotaxis, NETs release, enhanced phagocytic capacity, increased intracellular calcium concentration, degranulation, decidual neutrophil infiltration, enhanced TF expression, and increased generation of ROS.^2,6

4.1.4. Role of oxidative stress in obstetrical APS

Oxidative stress is defined as a disequilibrium between the production of reactive species and antioxidant defenses leading to abnormalities in signaling pathways and potential tissue damage.^68,69 Oxidative stress has been implicated in numerous pathologies like neurodegenerative, cardiovascular^70–72 or autoimmune disease (systemic lupus erythematosus or rheumatoid arthritis)^73–75 Several authors have implicated oxidative stress in the physiopathology of the APS. In placental cultures, it was found that interaction between maPLs and placental explant induces ROS production; furthermore, these maPLs could bound in a dose-dependent manner to mitochondria isolated from placenta. The meaning of this observation is that target antigens from aPLs are present in the mitochondria and this interaction induces and enhances the respiration rate raising ROS production by mitochondrial complex I, II, and mGDPH. ⁷⁶

Activation of trophoblast by aPLs, proinflammatory cytokines, TLRs ligands such as histones and NE, causes oxidative stress, lipoperoxidation, production of proinflammatory cytokines and chemokines (mainly IL-8) by the trophoblast, induces an increased expression of prostaglandin F2 alpha-receptor and oxytocin receptor as well as an increased production of prostaglandins by cyclooxygenase-2. All these events may cause myometrial contraction, cervical ripening, membrane rupture, and uterine contractions that result in the induction of labor or fetal death.^64,77–79

In conclusion, in obstetric APS patients, aPLs binds to β2GP1 in the trophoblast surface causing its activation and the consequent production of proinflammatory cytokines and chemokines that recruits neutrophils. These cells may be either directly activated through the binding of aPLs to β2GP1 or indirectly, by complement activation products induced by aPLs binding. Their activation would raise because of autocrine mechanisms that may induce TF expression and its binding to factor VII, a complex that generates down-stream signaling through PAR2. All these activation events produce chemotaxis in the decidua, increase phagocytic capacity, enhance the production of pro-inflammatory cytokines, augment of NETs release (a positive feed-back mechanism), and increase the activation and generation of proinflammatory cytokines resulting in premature birth, fetal death or miscarriage, the characteristic findings of obstetric APS. Figure 2 shows a summary of the pathophysiological mechanism of obstetric APS.

4.2.1. Role of oxidative stress in thrombotic APS

In order to induce aPLs, in a murine model with severe combined immunodeficiency injection of hybridomas generated a reduction of paraoxonase activity, total antioxidant capacity, production of NO as well as the expression of induced oxide nitric synthase (iNOS). On the contrary, an enhanced production of nitrotyrosines was found when hybridomas secreting antibodies with no aPLs specificity were administered. ⁸⁰

In APS patients, a higher concentration of inflammatory markers such as C reactive protein, serum amyloid A, and PGE₂ has been described. Also, inflammatory cytokines and chemokines such as VEGF, IL-8, MIP1α, and tPA as well as markers of oxidative stress like 8-isoprostane and NO are elevated as compared with heathy individuals.^81,82 The augment of 8-isoprostane and PGE₂ was more pronounced in triple-positive patients was found in single and double-positive APS patients. ⁸¹ In monocytes or neutrophils from APS patients, a higher activity and expression of TF,^81,82 PAR2, and FLT1 has been described. ⁸² Moreover, as compared with healthy individuals, a higher production of superoxide and reduced concentrations of glutathione, reduced catalase, and glutathione peroxidase activity have been described. Also, as compared with controls, these reductions were associated with a reduced nuclear abundance of Nrf2 in APS patients. ⁸² However, in the study by Vaz et al no differences were found between thrombotic PAPS patients (n = 70) and controls (n = 74) when comparing total antioxidant capacity, oxidative damage and oxidative balance measured by the concentration of protein carbonyl and 8-isoprostane. ⁸³ We highlight that the aforementioned studies^81,82 did not include a homogeneous group of PAPS patients as was done in the study by Vaz et al ⁸³ The study by Sciascia et al ⁸¹ classified the APS patients (n = 45) as PAPS or APS associated with other immunological diseases, and the study by Perez-Sanchez et al ⁸² enrolled 25 APS patients with thrombotic manifestation and 19 with pregnancy morbidity. In addition, the different thrombotic treatments in patients may explain the discrepancies between studies. All APS patients in the study by Vaz et al were on warfarin treatment, while patients in the study by Perez-Sanchez et al ⁸² were under treatment with warfarin or acenocoumarol, ASA or clopidogrel, or hydroxychloroquine.

On the other hand, to evaluate whether aPLs have a causal role on the APS patients, monocytes were incubated with a pool of IgG from APS patients (IgG-APS) or healthy subjects (IgG-HS). Monocyte stimulated with IgG-APS showed a higher activity and expression of TF, an enhanced superoxide production, a reduction of GSH levels, and a decrement of the nuclear abundance of Nrf2 as compared with monocytes stimulated with IgG-HS.^82,84 These findings strongly suggest that aPLs have a causative role on the characteristic effects observed in APS patients.

In order to assess if supplementation with antioxidants would influence the concentration of inflammatory molecules and the antioxidant capacity, monocytes were treated with antioxidant before stimulation with aPLs. Incubation with antioxidants reduced the superoxide production and TF expression induced by IgG-APS. Moreover, increased levels of GSH, production of NO, and expression of iNOs were also observed. Moreover, antioxidant treatment reduced the membrane expression of TF, VEGF, and Flt2; probably, this was a consequence of a reduced activity of the p38 MAPK and NF-kb signaling pathways.^82,84 These in vitro effects were also demonstrated in APS patients supplemented with vitamin C or E in whom reduced urinary isoprostane levels and concentration and activity of in monocytes were found. ⁸⁴

The above mentioned information may suggest that elevated levels of oxidant stress markers found by some authors, and the proinflammatory /procoagulant phenotype were, at least partially induced by the oxidative stress induced by aPLs. The discrepancies found in the levels of oxidative stress markers measured in APS patients should be addressed in further studies.

4.2.2. Platelet activation mediated by aPLs

Several platelet activation mechanisms induced by α-β2GP1 have been described. For example, α-β2GP1binding to PF4 stored in alpha granules induces a conformation change that potentiates its immunogenicity. ⁴⁴ Platelets exposed to the α-β2GP1-β2GP1-PF4 complex increases the generation of thromboxane A₂ (TxA₂) as compared with platelets stimulated with the α-β2GP1-bovine serum albumin-PF4 complex. This increased platelet activation was mediated by P38MAPK phosphorylation. ⁴⁴ Another described mechanism is a direct interaction of α-β2GP1 with β2GP1 which induces dimerization of β2GP1 as well as a 100-fold increase of its phospholipid-affinity. ⁸⁵ The interaction between dimeric β2GP1 with platelets is mediated by the ApoER2'-glycoprotein (GP) Ibα complex, a fusion event secondary to the DI-DIII and DV domains from β2GP1 and the binding site of VWF in GPIbα. Generation of this protein complex results in a higher platelet sensibility for a second stimulus such as the exposition of collagen upon vascular damage. These phenomena are mediated by MAP kinase pathway which triggers the production of TxA₂.^85,86

In searching for the role of platelets on endothelial activation and fibrin generation caused by α-β2GP1, a similar expression of ICAM-1 in the presence or absence of α-β2GP1 was found, nonetheless, in the presence of platelets and α-β2GP1, expression of ICAM-1 and fibrin generation was boosted by 20-500-fold. It was shown that α-β2GP1 did not bind to endothelial cells but to platelets suggesting that first, platelet thrombus promotes the endothelial activation upon vascular damage in presence of α-β2GP1 and second, in the presence of α-β2GP1 recruitment and activation of platelets is increased. ⁸⁷ Likely explanations for these findings are an increased expression of adhesion molecules like ICAM-1 ⁸⁷ or VWF dysfunction. Indeed, VWF has both, a higher half-life in APS patients as compared with controls as well as an abnormal VWF multimeric pattern which is characterized by larger size multimers and increased concentration of ultra-large multimers (ULVWF). An explanation for these facts may be a decreased activity of the metalloproteinase with a thrombospondin type 1 motif member 13 (ADAMTS13) found in APS patients which is likely secondary to anti-ADAMTS13 antibodies. ⁸⁸ Another explanation for the presence of larger size and high concentrations of ULFVW is the high apoptosis indexes observed in APS patients which increases the availability of vimentin to bind to A2 domain of VWF. ⁸⁹ As a consequence, vimentin may compete with ADAMTS13 for the binding site on VWF. Another mechanism that increases the vimentin availability is the aPLs-induced enhanced platelet, endothelial, monocyte, and neutrophil activation, as described in the text.

An increased concentration of thrombospondin 1 (TSP1) is another mechanism that explains by which the size and distribution of ULFVW is modified in APS patients. TSP1, a 450 KDa homotrimeric glycoprotein, is synthetized by smooth muscle cells, fibroblasts, megakaryocytes/platelets, and endothelial cells. TSP-1 is stored in alpha granules and Weibel Pallade bodies (WPB), in platelets and endothelial cells, respectively. TSP1 plasma concentration is 0.1-0.3 µg/mL and it may interact with multiple receptors such as CD36, integrins, CD47, the GPIb/IX/V complex, heparan sulfate, VWF, fibrinogen, laminin, fibronectin, and collagen. ⁹⁰ TSP1 anchors the blood clot in damaged vessels preventing embolization of the clot. Also, it binds to the A3 domain of VWF generating a competitive interaction with ADAMTS13 and decelerating its proteolytic action on VWF. ⁹⁰

In APS patients, TSP1 concentration is higher than in controls. ⁹¹ Furthermore, higher concentrations of TSP-1 in APS patients with arterial or venous thrombosis as compared with patients with obstetric APS have been found. TSP1 concentrations are higher in patients with acute thrombotic events versus patients with a history of thrombosis. Elevated plasma and platelet TSP1 are likely induced by the α-β2GP1-β2GPI-PF4 complex which induces release of the stored protein from platelet alpha granules and endothelial WBPs. ⁹¹ Therefore, raised ULFVW concentrations are due to a decreased ADAMTS13 activity and increased availability of vimentin and TSP1 in response to enhanced platelet, endothelial cell, and monocyte activity as well as to the presence of α-β2GP1 which reduces de binding of β2GP1 to active VWF forms.

Another mechanism that may trigger platelet and leukocyte activation in APS patients is the presence of anti-vimentin/cardiolipin antibodies. CL, a lipid of the inner mitochondrial membrane, is redistributed to the plasmatic membrane after induction of apoptosis mediated by FAS receptors and/or TNF-α receptor. ⁹² The redistribution of CL depends on vimentin, a cytoskeletal protein expressed on mesenchymal derived cells such as neutrophils, T cells, activated macrophages, endothelial cells, smooth muscle cells, activated platelets, and apoptotic cells^43,92,93 As β2GP1, vimentin interacts with CL and, as a consequence, it may be a cofactor for the presentation of CL to the immune system increasing its immunogenicity.^92,94 In patients with seronegative APS (n = 29), 55.2% had anti-vimentin/CL antibodies as was also observed in APS patients with positive aCL titers in whom 92.5% had anti-vimentin/CL antibodies. ⁹⁴ In both studies, the presence of anti-vimentin/CL antibodies was not related to thrombotic events. Despite the lack of association between thrombosis and the presence of anti-vimentin/CL antibodies, in vitro studies strongly suggest an effect of these antibodies on hemostasis. For example, stimulation of HUVECs with anti-vimentin/CL antibodies induced phosphorylation of the interleukin-1 receptor-associated kinase (IRAK) and activation of the nuclear factor-kappa B (NF-κb), effects that were not observed after stimulation with IgG in controls. ⁹⁴ Moreover, incubation of whole blood with IgM anti-vimentin antibodies (AVA) caused platelet depletion, formation of platelet-leukocyte aggregates, expression of CD62P on the platelet surface, expression of TF on the leukocyte surface, and fibrin and C3d deposit on the platelet and leukocyte membrane. All these effects were inhibited by adding recombinant vimentin. ⁹³ AVA bind preferentially to neutrophils, monocytes, and activated platelets however, platelet activation is due to the release of inflammatory mediators such as platelets-activating factor (PAF) by leukocytes but not due to a direct effect of AVA, ⁹³ as shown after adding PAF inhibitor to the whole blood.

The origin of AVA has not elucidated however, it is proposed that induction of these antibodies in autoimmune diseases results from an excessive exposition of vimentin to apoptotic cells. Although cleavage of vimentin by caspases is a requisite for apoptosis, ⁹³ evidence regarding this dysregulation process of apoptosis in APS patients has not been proven. ⁹⁵

Finally, platelets are also activated by NETs, specifically by histones. These proteins activate platelets expressing phosphatidylserine, P-selectin, and FVa on their surface. This activation depends on the presence of ADP, P2Y12 receptor, and GPIIb/IIIa^{. 96}

4.2.3. Endothelial activation mediated by aPLs

α-β2GP1 binds to endothelial cells in the presence of platelets however, ⁸⁷ other authors claim that binding of α-β2GP1 to endothelial cells is through annexin A2 (A2).^97–99 A2 is a profibrinolytic receptor that binds to plasminogen and tPA, acting as a cofactor for plasmin generation and allowing the localization of fibrinolysis to the endothelial cell surface.^97–99 A direct activation of endothelial cells by anti-A2 and/or through the complex α-β2GP1-β2GPI-A2 has been demonstrated. ⁹⁸ Endothelial cell activation results in an increased expression of E-selectin, vascular cell adhesion molecule-1 (VCAM-1), and TF. ⁹⁹ APS patients who had anti-A2 have a 35-83% reduction of tPA-dependent plasmin production. The high variability of this reduction was probably secondary to a specific epitope on A2, the presence of multiple epitopes, and/or the presence of an alternative endothelial plasminogen receptor. ⁹⁷

β2GP1 binds tPA with high affinity, increasing 19.7-fold the plasmin generation by tPA as compared with A2 binding which increases 63-fold. The role of β2GP1 on thrombus dissolution has been evaluated. In the absence of β2GP1, dissolution kinetics was reduced and reversed when β2GP1 was added. Evaluating the effect of α-β2GP1 on tPA activity, the IgG fraction of APS patients inhibited the tPA-dependent plasminogen activation in a concentration-dependent way. ⁴³ Thus, β2GP1 and A2 are tPA synergic cofactors for plasminogen activation and, as a consequence, antibodies against them reduce plasminogen activation predisposing to thrombosis in APS patients.

Another mechanism for β2GP1-induced endothelial cell activation has been suggested. α-β2GP1 cross-recognize β2GP1 as plasminogen on the surface of endothelial cells inducing overexpression of ICAM-1 and subexpression of thrombomodulin (TM), a protein with a critical role in the protein C system, the most important natural anticoagulant mechanism. ⁴⁶ Besides, other authors have shown that α-β2GP1 also induces the generation of ROS through the activation of MAP kinases. ¹⁰⁰ The increased production of ROS triggers the production of nitric oxide (NO) by eNOS and, as a consequence, an increased production of ROS which consumes NO and decreases its release from endothelial cells,. ¹⁰¹ As a result, there is a decreased bioavailability of NO. On the other hand, ET-1 concentrations were analyzed in 27 APS patients, 11 and 16 with a history of venous or arterial thrombosis, respectively. Higher ET-1 concentrations were found in only eight out of sixteen patients with arterial thrombosis but no differences were found in patients with venous thrombosis. Moreover, aCL induced the production of preproendothelin-1 mRNA in presence of β2GP1. ¹⁰² In 2015, factors inducing ET-1 production were described including thrombin, ROS, and proinflammatory cytokines such as IL-1, IL-6, and TNFα. ¹⁰³ All of these proteins are elevated in APS patients. Therefore, all these above mentioned facts strongly suggest that APS patients have abnormal endothelial vasomotor activity due to both, a reduced bioavailability of NO caused by excessive production of ROS and an increased production of ET-1.

Considering that endothelial activation by aPLs increases the expression of adhesion molecules such as P-selectin, ICAM-1, and VCAM-1 and that it induces TF overexpression and subexpression of TM, the induction of proinflammatory cytokines, high concentration of ET-1, low NO bioavailability, and higher ROS production, results in a pro-adhesive, pro-coagulant, pro-inflammatory, and pro-aggregatory endothelium. Table 4 shows the receptors involved in the recognition of aPLs on the cell surface of different cell types including endothelial cells.

On the other hand, histones can also activate endothelial cells. Stimulation of blood outgrowth endothelial cells with histones induce exocytosis of WPB and a subsequent rise in the concentration of VWF and angiopoietin 2 as well as an increased platelet-endothelium interaction through VWF. The exact activation mechanism induced by histones is not clear yet ¹⁰⁴ however, it has been proposed that endothelial activation is mediated by TLR4. ¹⁰⁵ Finally, studies in mice demonstrated that infusion of histones induces thrombocytopenia and increased VWF; this decrease of the platelet count was not related to VWF because the degree of thrombocytopenia was not different between histone-treated VWF deficient mice and histone-treated wild type mice. ¹⁰⁴

4.2.4. Eosinophilic granulocyte activation mediated by aPLs

In 2018, a working group compared the protein composition of fibrin clots in controls, patients with deep venous thrombosis, and APS patients using liquid chromatography and mass spectrometry. Clots from controls and patients with deep vein thrombosis had higher levels of complement proteins such as C5-C9, platelet glycoproteins (GPIIb, GPIIIa), TSP1, neutrophil myeloperoxidase, histones (H2A/H2B), and major basic protein from eosinophils as compared with clots from APS patients. ¹ These findings may reflect the involvement of extracellular traps in the development of thrombosis. One year later, another group showed the role of eosinophils on the development and stability of blood clots. They showed that clots of eosinophil-deficient mice were less stable in comparison to clots from wild-type mice. Furthermore, they demonstrated that platelet-eosinophil interactions increased platelet aggregation and triggered the production of eosinophil extracellular traps (EETs) which are differ from NETs because of their major basic protein (MBP) content. This protein has effects on clot stability because MBP-deficient mice had fewer stable clots and it has been shown that MBP binds to TM inducing an inhibitory effect on PC activation. Finally, MBP promotes platelet aggregation and thrombus development. ¹⁰⁶

4.2.5. Activation of plasmatic hemostasis

The effect of β2GP1 on the activation of blood coagulation FXII induced by ellagic acid and kaolin is very well known however, this phenomenon is reverted adding α-β2GP1 in a dose-dependent way. ⁴⁵ Exploring the presence of anti-tPA antibodies in 91 APS patients (68 primary APS and 23 LES-associated SAPS patients) and 914 controls, it was observed that the anti-tPA antibody levels were higher in APS patients. In addition, the activity and concentration of tPA were assayed in 53 of 91 patients; APS patients had higher tPA concentration but lower activity when compared with controls. ¹⁰⁷ Finally, in APS patients with or without anti-FXII antibodies, it was found that 50% of these antibodies were able to reduce both, the antigenic levels and the activity of FXII ¹⁰⁸ .The meaning of this finding that another mechanism predisposing APS patients to thrombosis is the lack of fibrinolysis activation mediated by FXII.

On the other hand, the presence of anti-activated factor XI antibodies was assayed in 38 APS patients and 30 controls. A higher antibody concentration in APS patients was found and these antibodies were also capable to bind to factor IX (FIX). Although these antibodies did not affect the FIX activity they did alter the inactivation of activated FIX by antithrombin, resulting in an increased FIX activity and a procoagulant state. ⁶²

From a distinct perspective, the effects of aPLs on the structure and properties of clots have been analyzed. Clots from APS patients are less permeable, denser, with thinner fibers, and more branched than clots from controls and patients with thrombosis but without APS. Moreover, the clotting process is delayed and the clots are denser and more difficult to lysate in APS patients ¹⁰⁹ .An explanation for the higher density of the clots from APS patients could be a higher activity of factor XIII (FXIII), a transglutaminase of endothelial cell origin that cross-links fibrin-fibrin and fibrin-fibronectin fibers giving stability to the clot. In APS patients, increased activity of FXIII in comparison with aPL carriers, controls, and cardiac mechanical valve controls has been described. It is likely that, in APS patients, this high FXIII activity promotes a higher cross-linking of fibrin monomers making the cloth less sensitive to plasmin. ¹¹⁰

The effects of aPL on plasma factors can be assessed through the blood coagulation tests. APS patients have elevated levels of cell-free DNA and NETs which correlate with the α-β2GP1 titers and that these titers are more increased in LA positive patients and even more in triple-positive patients (aCL⁺, α-β2GPI⁺, LA⁺) ¹¹¹ .The increased NETs levels are promoted by α-β2GP1 from APS patients by a mechanism involving TLR4 ¹¹¹ . The effect of aPL antibodies on blood coagulation tests is likely related to cell-free DNA and NETs levels. In fact, in vitro assays showed that histones are able to prolong the blood coagulation tests (prothrombin, activated partial thromboplastin, and thrombin times as well as diluted Ruselĺs viper venom test), likely due to their affinity to anionic phospholipids such as phosphatidylserine, and/or their binding to thrombin, and/or to direct inhibition of a contact phase activator. However, by using thrombinography, it has been shown that histones increase the thrombin generation rate and the endogenous thrombin potential without affecting fibrin formation^111,112 .Moreover, a reduced generation of activated PC mediated by histones likely due to the binding to its gamma-carboxyglutamic acid-rich domain of PC has been demonstrated. ¹¹² Moreover, the binding of H4 histone to prothrombin improves their affinity to FXa in absence of phospholipids however, this reaction is inhibited in presence of the other elements of prothrombinase complex (FVa and phospholipids)^{. 113}

A synthesis of the pathophysiological mechanisms described in APS patients are depicted in Figures 3 and 4. Figure 3 shows the effects of histones on platelets, endothelial cells, eosinophils, and soluble factors while Figure 4 depicts a summary of the mechanisms involved in the development of the clinical manifestations of APS.