Advertisement

Antibodies to blood coagulation components are implicated in patients with multiple sclerosis

Open AccessPublished:March 31, 2022DOI:https://doi.org/10.1016/j.msard.2022.103775

      Highlights

      • Antibodies to coagulant factors are involved to venous thromboembolism development.
      • They can trigger signaling inflammatory pathways in neuroinflammatory diseases.
      • Antibodies to coagulant factors were detected in patients with multiple sclerosis.
      • Reactivities against FVIIa, FXII, and plasmin were the most prevalent.
      • Thrombin-IgG seropositivity was associated with worse outcomes.

      Abstract

      Background

      The strong link between innate immunity and thrombosis/coagulation has recently been investigated in the light of antibodies directed against serine proteases of the coagulation pathway. The antibodies have been proposed as contributing factors to venous thromboembolism development and as key molecules in the initiation of signaling inflammatory pathways in neuroinflammatory diseases. Preliminary studies of Multiple Sclerosis (MS) progression characteristics with the reactivity of antibodies against coagulant components are limited. Considering the development of thrombosis at the early onset of MS, our study aimed to detect antibodies against coagulant components in MS and evaluate their possible association with the clinical profile of the disease.

      Method

      A cross-sectional study was carried out to identify antibodies to factor(F)VIIa, thrombin, prothrombin, FXa, FXII, plasmin, and protein C in serum samples from 167 patients with MS and 40 healthy controls using the enzyme-linked immunosorbent assay. Statistical analysis was performed for the evaluation of the data.

      Results

      The analysis revealed a significantly higher prevalence of IgG in MS patients (n = 72, 43%) compared to HCs (n = 8, 20%, p < 0.01). Specifically, elevated anti-FVIIa (n = 19, 11.4%, mean activity p < 0.0001), anti-FXII (n = 12, 7.2%, mean activity p < 0.001) and anti-plasmin (n = 20, 12%, mean activity p < 0.01) levels were observed in patients compared to controls. Additionally, the highest scores of clinical characteristics like the expanded disability status scale and MS severity score were linked with IgG seropositivity against thrombin, whilst anti-FXII levels corresponded with the lowest disease progression.

      Conclusion

      The findings of our study illustrate the presence of antibodies against serine proteases of the coagulation cascade in MS and demonstrate the association of antibody activity with disease progression. In particular, thrombin IgG seropositivity was demonstrated to be associated with worse outcomes and a severe disease phenotype. These observations suggest the implication of antibodies in patient monitoring and prognosis, and further evaluation may elucidate inflammatory cascades in which antibodies act as key mediators.

      Graphical abstract

      Keywords

      1. Introduction

      Multiple Sclerosis (MS) is a heterogeneous and multifactorial disease of the central nervous system (CNS) (
      • Ruiz F.
      • Vigne S.
      • Pot C.
      Resolution of inflammation during multiple sclerosis.
      ), characterized by multifocal plaque formation within the CNS and the disruption of self-immune tolerance, thereby promoting a neurodegenerative and autoimmune phenotype (
      • Huang W.J.
      • Chen W.W.
      • Zhang X.
      Multiple sclerosis: pathology, diagnosis and treatments (review).
      ).
      The adaptive immune system plays a fundamental role in the development of MS. Activated T-lymphocytes are crucial for the initiation and maintenance of inflammatory responses as they induce the secretion of pro-inflammatory components and lead to the leukocyte adhesion to the endothelial cells, the disruption of blood-brain barrier (BBB) permeability, and the leukocyte invasion into the CNS (
      • Faria S.S.
      • Fernandes P.C.
      • Silva M.J.B.
      • Lima V.C.
      • Fontes W.
      • Freitas R.
      • Eterovic A.K.
      • Forget P.
      The neutrophil-to-lymphocyte ratio: a narrative review.
      ;
      • Ghasemi N.
      • Razavi S.
      • Nikzad E.
      Multiple sclerosis: pathogenesis, symptoms, diagnoses and cell-based therapy.
      ;
      • Koudriavtseva T.
      • Zannino S.
      • Filippi M.M.
      • Cortese A.
      • Piantadosi C.
      • Lapucci C.
      • Fiorelli M.
      • Giannarelli D.
      • Mandoj C.
      • Stefanile A.
      • Conti L.
      • Salvetti M.
      • Inglese M.
      Coagulation activation and cerebral hypoperfusion in relapsing-remitting multiple sclerosis.
      ).
      Recently, it has been demonstrated that components of innate immunity are essential effectors for activation and stimulation of biochemical and cellular mechanisms (
      • Koudriavtseva T.
      • Zannino S.
      • Filippi M.M.
      • Cortese A.
      • Piantadosi C.
      • Lapucci C.
      • Fiorelli M.
      • Giannarelli D.
      • Mandoj C.
      • Stefanile A.
      • Conti L.
      • Salvetti M.
      • Inglese M.
      Coagulation activation and cerebral hypoperfusion in relapsing-remitting multiple sclerosis.
      ;
      • Pavelek Z.
      • Angelucci F.
      • Souček O.
      • Krejsek J.
      • Sobíšek L.
      • Klímová B.
      • Šarláková J.
      • Halúsková S.
      • Kuča K.
      • Vališ M.
      Innate immune system and multiple sclerosis. Granulocyte numbers are reduced in patients affected by relapsing-remitting multiple sclerosis during the remission phase.
      ;
      • Yadav S.K.
      • Mindur J.E.
      • Ito K.
      • Dhib-Jalbut S.
      Advances in the immunopathogenesis of multiple sclerosis.
      ). Activated microglia together with infiltrated macrophages have been detected in MS lesions and associated with the damage of the white matter (
      • Chu F.
      • Shi M.
      • Zheng C.
      • Shen D.
      • Zhu J.
      • Zheng X.
      • Cui L.
      The roles of macrophages and microglia in multiple sclerosis and experimental autoimmune encephalomyelitis.
      ;
      • Gandhi R.
      • Laroni A.
      • Weiner H.L.
      Role of the innate immune system in the pathogenesis of multiple sclerosis.
      ;
      • Mayo L.
      • Quintana F.J.
      • Weiner H.L.
      The innate immune system in demyelinating disease.
      ).
      Inflammatory responses due to the activation of microglia and infiltrating immune cells in the CNS lead to a reciprocal relationship between coagulation and inflammation (
      • Ryu J.K.
      • Petersen M.A.
      • Murray S.G.
      • Baeten K.M.
      • Meyer-Franke A.
      • Chan J.P.
      • Vagena E.
      • Bedard C.
      • Machado M.R.
      • Coronado P.E.R.
      • Prod'homme T.
      • Charo I.F.
      • Lassmann H.
      • Degen J.L.
      • Zamvil S.S.
      • Akassoglou K.
      Blood coagulation protein fibrinogen promotes autoimmunity and demyelination via chemokine release and antigen presentation.
      ). In the coagulation-inflammation activation, the Tissue Factor (TF) transmembrane receptor is overexpressed in response to vessel injury, inflammatory cytokines, and other damaged-associated molecular patterns expression (
      • Grover S.P.
      • Mackman N.
      Tissue factor: an essential mediator of hemostasis and trigger of thrombosis.
      ;
      • Spronk H.M.H.
      • Ten Cate H.
      • Van Der Meijden P.E.J.
      Differential roles of tissue factor and phosphatidylserine in activation of coagulation.
      ) and forms the TF-FVIIa complex with the circulated factor VII. This complex activates FX followed by the generation of thrombin from its zymogen prothrombin (
      • Grover S.P.
      • Mackman N.
      Tissue factor: an essential mediator of hemostasis and trigger of thrombosis.
      ) that cleaves fibrinogen to fibrin (
      • Kamikubo Y.
      • Mendolicchio G.L.
      • Zampolli A.
      • Marchese P.
      • Rothmeier A.S.
      • Orje J.N.
      • Gale A.J.
      • Krishnaswamy S.
      • Gruber A.
      • Østergaard H.
      • Petersen L.C.
      • Ruf W.
      • Ruggeri Z.M.
      Selective factor VIII activation by the tissue factor–factor VIIa–factor Xa complex.
      ). Fibrin clots can also be formed by activating the FXI-intrinsic coagulation pathway through the FXII activity (
      • Renné T.
      • Schmaier A.H.
      • Nickel K.F.
      • Blombäck M.
      • Maas C.
      In vivo roles of factor XII.
      ). Under pathological conditions, thrombin induces the activation of pro-inflammatory signaling pathways and the migration of leukocytes into the CNS through the function of thrombin-activated protease-activated receptors (PARs) (
      • Rezaie A.R.
      Protease-activated receptor signalling by coagulation proteases in endothelial cells.
      ). On the other hand, fibrin clots are bound to integrin MAC-1 (CD11b/CD18) receptors on microglial cells activating microglia and promoting pro-inflammatory cytokine expression (
      • Davalos D.
      • Akassoglou K.
      Fibrinogen as a key regulator of inflammation in disease.
      ).
      Studies in autoimmune disorders that have common clinical characteristics with MS provide new insights into the coagulation-inflammation circuit and reveal the presence of antibodies against coagulant components that may contribute to the thromboembolism development (
      • Artim-Esen B.
      • Pericleous C.
      • Mackie I.
      • Ripoll V.M.
      • Latchman D.
      • Isenberg D.
      • Rahman A.
      • Ioannou Y.
      • Giles I.
      Anti-factor Xa antibodies in patients with antiphospholipid syndrome and their effects upon coagulation assays.
      ;
      • Lambrianides A.
      • Turner-Stokes T.
      • Pericleous C.
      • Ehsanullah J.
      • Papadimitraki E.
      • Poulton K.
      • Ioannou Y.
      • Lawrie A.
      • MacKie I.
      • Chen P.
      • Latchman D.
      • Isenberg D.
      • Rahman A.
      • Giles I.
      Interactions of human monoclonal and polyclonal antiphospholipid antibodies with serine proteases involved in hemostasis.
      ). These molecules act and antagonize the interaction between anticoagulant mediators and the catalytic region of serine proteases, preventing the protease inactivation or inhibiting the fibrin degradation and thus, lead to the venous thromboembolism (VTE). The high similarity of several coagulant serine proteases either in amino acid content or in secondary structure indicates that antibodies can cross-react with homologous coagulation components, thereby affecting the regulation of the coagulation cascade and the coagulation-inflammation interaction (
      • Lambrianides A.
      • Turner-Stokes T.
      • Pericleous C.
      • Ehsanullah J.
      • Papadimitraki E.
      • Poulton K.
      • Ioannou Y.
      • Lawrie A.
      • MacKie I.
      • Chen P.
      • Latchman D.
      • Isenberg D.
      • Rahman A.
      • Giles I.
      Interactions of human monoclonal and polyclonal antiphospholipid antibodies with serine proteases involved in hemostasis.
      ).
      Despite the in-depth research that has been conducted up to date, the etiology of MS is still elusive and no study has shed light on the development and progression of the disease. In addition, there are no previous studies to evaluate the relationship between antibodies against coagulant components and MS progression. To this date, numerous studies have investigated the above hypothesis in antiphospholipid syndrome (APS), but since both diseases share common clinical features, this should be investigated further in MS. Therefore, the purpose of this study was to detect the seroprevalence of antibodies against procoagulant, anticoagulant, and fibrinolytic molecules in patients with MS and to determine whether these antibodies contribute to the progression and disability of the disease.

      2. Materials and methods

      2.1 Ethical considerations

      All individuals provided an informed consent form that was approved by the Cyprus National Bioethics Committee (EEBK/EΠ/2016/51).

      2.2 Study participants

      In the current study, 167 participants diagnosed with MS at the Neuroimmunology Department of the Cyprus Institute of Neurology and Genetics were enrolled during routine follow-up appointments between September 2017 and January 2019. McDonald's revised criteria (
      • Polman C.H.
      • Reingold S.C.
      • Banwell B.
      • Clanet M.
      • Cohen J.A.
      • Filippi M.
      • Fujihara K.
      • Havrdova E.
      • Hutchinson M.
      • Kappos L.
      • Lublin F.D.
      • Montalban X.
      • O'Connor P.
      • Sandberg-Wollheim M.
      • Thompson A.J.
      • Waubant E.
      • Weinshenker B.
      • Wolinsky J.S.
      Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria.
      ) were satisfied by all participants. Furthermore, the following criteria were required for inclusion: ages above 18 years old; sufficient demographic and clinical data (age of onset, Expanded Disability Status Scale (EDSS) and treatments received); patients with clearly defined clinical courses [Relapsing-Remitting MS (RRMS), Secondary Progressive MS (SPMS), Primary Progressive MS (PPMS)]. The investigators were blinded to MS courses. The exclusion criteria were: lack of or inability to provide informed consent; drug abuse and alcoholism; pregnancy. Out of 167 patients with MS, 130 had RRMS, 29 SPMS, seven PPMS, and one participant was diagnosed with clinically isolated syndrome (CIS). According to an extensive review of their medical history, none of the patients had any other autoimmune disease or diagnosed with other non-autoimmune disorders not attributable to MS. Table 1 illustrates the demographic and clinical information of participants studied. In addition, 40 healthy individuals were recruited, matched for both age and gender to patients with MS.
      Table 1Demographic and clinical profile of participants.
      FeaturesMS patients(n = 167)HCs(n = 40)
      Gender

      Female

      Male


      110

      57


      21

      19
      Age in years

      Mean±SD

      Min–Max


      48 ± 13

      21–80


      45 ± 10

      24–63
      Disease course (CIS/RRMS/SPMS/PPMS)

      1/130/29/7


      N/A
      Disease Duration (years)

      Mean±SD

      Median (interquartile range)


      15.3 ± 8.9

      15 (9–21)


      N/A
      EDSS

      Median (interquartile range)

      Mild: 0–3.0 [n (%)]

      Moderate: 3.5–5.5 [n (%)]

      Severe: 6.0–9.5 [n (%)]


      3.5 (2.5–5.5)

      75 (45.0)

      57 (34.0)

      35 (21.0)


      N/A

      MSSS

      Median (interquartile range)

      Benign MS: 1–2 [n (%)]

      Aggressive MS: 7–10 [n (%)]


      3.9 (2.4–5.5)

      20 (12.0)

      20 (12.0)


      N/A
      Medication [n(%)]

      Interferon beta-1a or −1b

      Natalizumab

      Fingolimod

      Other
      Other type of treatment: Azathioprine, Dimethyl fumarate, Glatiramer acetate, Methotrexate, Mycophenolate, Rituximab, Teriflunomide.


      None


      40 (24.0)

      18 (11.0)

      26 (16.0)

      24 (14.0)

      59 (35.0)
      N/A
      MS: Multiple Sclerosis; HCs: Healthy controls; CIS: Clinically Isolated Syndrome; RRMS: Relapsing-Remitting MS; SPMS: Secondary Progressive MS; PPMS: Primary Progressive MS; SD: Standard Deviation; EDSS: Expanded Disability Status Scale; MSSS: Multiple Sclerosis Severity Score; N/A: Not Applicable.
      low asterisk Other type of treatment: Azathioprine, Dimethyl fumarate, Glatiramer acetate, Methotrexate, Mycophenolate, Rituximab, Teriflunomide.

      2.3 Blood collection

      Fresh blood samples were collected in vacutainers without any additive and allowed to clot for a period up to 3 h. Subsequently, sample centrifugation was carried out for 10 min at 1500 xg at 25 °C. Cell-free serum was finally collected and stored at −20 °C until further processing.

      2.4 Detection of IgG antibodies using ELISA

      The indirect form of in-house ELISA assays was used to assess antibody binding activity directed against seven serine proteases of the coagulation pathway. Specifically, IgG seropositivity was determined against FVIIa, Thrombin, Prothrombin (PT), FXa, FXII, plasmin, and protein C (PC). As per protocol, an antigen was diluted in specific diluent on the test side of the 96-well ELISA plate, while the control side had only diluent added and was incubated overnight at 4 °C. In each ELISA assay, serum samples were diluted at 1:25 in blocking buffer and loaded on both sides of the plates. Each serum sample was examined in duplicate for reproducibility. Positive and negative controls were used in each plate to ensure the accuracy of the test, reliability, and validation. HRP-conjugated anti-human IgG antibodies were used to detect bound anti‑serine proteases IgG antibodies, and the detection was accomplished by using tetramethylbenzidine chromogenic solution (TMB) (Life Technologies, USA). After stopping the enzyme-substrate reaction with 1 M HCl stop solution, absorbance was measured at the wavelength of 450 nm.
      Net absorbance was calculated by subtracting the control side's absorbance from the test side's. The results were determined as the percentage of the average net absorbance divided by the absorbance of a reference sample. Serum samples considered positive had antibody-binding activity above the 95th percentile of HCs (95th percentile is calculated as the two standard deviations above the mean antibody binding activity of HCs).

      2.4.1 Anti-FVIIa ELISA

      Anti-FVIIa IgG antibodies were detected according to the protocol described by
      • Artim-Esen B.
      • Pericleous C.
      • Mackie I.
      • Ripoll V.M.
      • Latchman D.
      • Isenberg D.
      • Rahman A.
      • Ioannou Y.
      • Giles I.
      Anti-factor Xa antibodies in patients with antiphospholipid syndrome and their effects upon coagulation assays.
      . Briefly, MaxiSorp 96-wells plates (ThermoFisher, Waltham, Massachusetts, USA) were coated with 1.5 μg/ml of human FVIIa. Plates were then blocked with 2% bovine serum albumin (BSA) (Sigma-Aldrish, St. Louis, USA) in phosphate-buffered saline (PBS) for 90 min at 37 °C. Serum samples were diluted 1:25 in PBS/1% BSA.

      2.4.2 Anti-thrombin ELISA

      Anti-Thr. IgG detection was performed based on
      • Lambrianides A.
      • Turner-Stokes T.
      • Pericleous C.
      • Ehsanullah J.
      • Papadimitraki E.
      • Poulton K.
      • Ioannou Y.
      • Lawrie A.
      • MacKie I.
      • Chen P.
      • Latchman D.
      • Isenberg D.
      • Rahman A.
      • Giles I.
      Interactions of human monoclonal and polyclonal antiphospholipid antibodies with serine proteases involved in hemostasis.
      . High binding capacity Costar plates (ThermoFisher, Waltham, Massachusetts, USA) were coated with 10 μg/ml of human alpha-Thr diluted in Tris-buffered saline buffer (TBS; 0.05 M Tris, 0.15 M NaCl and pH 7.5). Plates were blocked with TBS/0.3% porcine gelatin for 90 min at room temperature (RT) and then sera were diluted in TBS/0.1% gelatin.

      2.4.3 Anti-PT ELISA

      Anti-PT IgG antibodies were detected as reported previously
      • Matsuda J.
      • Sanaka T.
      • Nishizawa A.
      • Gotoh M.
      • Gohchi K.
      Two antiprothrombin antibodies against prothrombin and prothrombin-phosphatidyl serine show partial but not total identity.
      . A concentration of 10 μg/ml of human prothrombin in TBS coated MaxiSorp plates. Serum samples were diluted 1:25 in TBS/1% BSA and incubated for 1.5 h at RT.

      2.4.4 Anti-FXa ELISA

      The protocol of
      • Yang Y.H.
      • Hwang K.K.
      • FitzGerald J.
      • Grossman J.M.
      • Taylor M.
      • Hahn B.H.
      • Chen P.P.
      Antibodies against the activated coagulation factor X (FXa) in the antiphospholipid syndrome that interfere with the FXa inactivation by antithrombin.
      was used for the detection of anti-FXa IgG. Costar plates were coated with 5 μg/ml of FXa diluted in TBS and incubated overnight at 4oC. Plates were blocked using TBS/0.3% gelatin and the plates were incubated for 90 min at RT. Incubation of sera in TBS/0.3% gelatin was carried out for 1.5 h at RT.

      2.4.5 Anti-FXII ELISA

      Anti-FXII antibodies were detected according to the protocol described by
      • Jones D.W.
      • Gallimore M.J.
      • Mackie I.J.
      • Harris S.L.
      • Winter M.
      Reduced factor XII levels in patients with the antiphospholipid syndrome are associated with antibodies to factor XII.
      . MaxiSorp plates were coated with 5 μg/ml of FXII antigen diluted in carbonate-bicarbonate buffer. Overnight incubation at 4 °C was performed and the plates were then blocked with TBS/2% BSA for 60 min at RT. Serum samples 1:25 diluted in TBS/1% BSA were incubated for 1 h at RT.

      2.4.6 Anti-plasmin ELISA

      Costar plates were coated with 5 μg/ml of plasmin diluted in PBS and incubated overnight at 4 °C (
      • Yang C.D.
      • Hwang K.K.
      • Yan W.
      • Gallagher K.
      • FitzGerald J.
      • Grossman J.M.
      • Hahn B.H.
      • Chen P.P.
      Identification of anti-plasmin antibodies in the antiphospholipid syndrome that inhibit degradation of fibrin.
      ). Plates were blocked using PBS/0.25% gelatin for 60 min at RT. Sera were then diluted 1:25 in PBS/0.1% gelatin and incubated for 90 min at RT.

      2.4.7 Anti-PC ELISA

      A concentration of 5 μg/ml of PC in TBS/2.5 mM CaCl2 coated the test half of Costar plates as described by
      • Lambrianides A.
      • Turner-Stokes T.
      • Pericleous C.
      • Ehsanullah J.
      • Papadimitraki E.
      • Poulton K.
      • Ioannou Y.
      • Lawrie A.
      • MacKie I.
      • Chen P.
      • Latchman D.
      • Isenberg D.
      • Rahman A.
      • Giles I.
      Interactions of human monoclonal and polyclonal antiphospholipid antibodies with serine proteases involved in hemostasis.
      . After overnight incubation at 4 °C, the plates were blocked with 0.3% gelatin in TBS/2.5 mM CaCl2 and serum samples were then diluted 1:25 in TBS/0.1% gelatin and incubated for 60 min at RT.

      2.5 . Statistical analysis

      The analysis of the results was carried out using GraphPad Prism V.5.00 for Windows(La Jolla, California, USA). The D'Agostino-Pearson test was used to test for normality. The non-parametric Mann-Whitney U-test was applied for the age-matching analysis and the comparison of differences in the antibody distribution between the patients with MS and HCs. Kruskal–Wallis one-way ANOVA followed by Dunn's post hoc test for multiple comparisons was used to assess antibody presence in more than two groups of participants. The Fisher's exact test was performed to assess the matching of gender among the study groups as well as the significance of IgG antibody detection among the above groups. Area under the receiver operating characteristic (ROC) was measured between the variables (patients vs HCs) for each antibody, indicating the modeling accuracy. Spearman correlation matrix was performed to analyze the association between antibody responses and clinical outcomes. A logistic regression analysis was conducted to assess the detection of antibodies with regards to baseline characteristics such as gender and age. P-values less than 0.05 were statistically significant.

      3. Results

      3.1 . Patient demographic and clinical data

      This study was undertaken involving 167 MS patients, along with 40 healthy controls recruited at the same period, in which the clinical and non-clinical features of the participants were examined. No significant difference was found in the age (p = 0.3) or gender (p = 0.2) between the patients and controls. According to the data, the disease duration of the patients with MS was 15.3 ± 8.9 years. The median EDSS score was 3.5 (interquartile range: 2.5–5.5). Forty-five percent of patients had mild disease (EDSS score:0–3), 34% moderate disease (EDSS:3.5–5.5), and 21% of the MS population had severe disability (EDSS:6.0–9.5). The median value of MS severity score (MSSS) was 3.9 (interquartile range: 2.4–5.5), while 12% of patients showed aggressive disease progression (MSSS: 7–10).

      3.2 Detection of antibodies against serine proteases in patients with MS

      The IgG antibody activity to serine proteases FVIIa, thrombin, PT, FXa, FXII, plasmin, and PC was examined in patients with MS and healthy individuals and was expressed as the percentage of each sample's binding activity compared to the binding activity of the reference sample. Patients with activity values more than 2 S.D. above the mean of binding activity of HCs (95th percentile) were considered positive.
      As Fig. 1 shows, MS patients had significantly higher mean index of anti-FVIIa IgG [mean: 26.86 (standard error of the mean; SEM: 0.75)] than HCs [mean: 21.03 (SEM: 0.95); p < 0.0001), with an AUC prediction value of 0.74 [95% Confidence Interval (CI): 0.65–0.82)] (Fig.1A). Reactivity to thrombin, prothrombin, FXa, and PC was also observed in patients, but at a frequency that did not differ significantly from controls. In contrast, elevated mean index of anti-FXII activity was detected more frequently in MS group [mean: 87.87 (SEM: 0.95)] compared to HCs [mean: 72.17 (SEM: 3.57); p < 0.001], showing an AUC value of 0.70 (95% CI: 0.60–0.80) (Fig.1B). Furthermore, binding activity against plasmin differed significantly between the two study groups [MS mean index: 40.04 (SEM: 1.86); HC mean index: 29.06 (SEM: 3.02); p < 0.01], showing an AUC value of 0.64 (95% CI: 0.55–0.74)] (Fig.1C).
      Fig 1
      Fig. 1IgG activity against serine proteases, FVIIa (A), FXII (B), and plasmin (C) in MS patients and HCs. Results are expressed as the percentage of the O.D. value of each sample to the corresponding O.D of a reference sample. The evaluation was performed between each tested group and HC group. Dashed lines represent the threshold above which the samples are considered as positive, calculated as the 2 S.D. above the mean of HCs. Bars represent the mean ± SEM. Mann-Whitney test applied for the analysis (⁎⁎p < 0.01, ⁎⁎⁎p < 0.001). AUC is measured (data with statistical significance are shown). MS: multiple sclerosis; S.D.: standard deviation; SEM: standard error of mean; AUC: area under the curve.
      Overall, an increased proportion of patients with MS (n = 72, 43%) showed antibody positivity for the presence of at least one IgG antibody compared to HCs (n = 8, 20%, p < 0.001) (Fig.2).
      Fig 2
      Fig. 2Antibody-binding status of patients with MS versus HCs. The results are expressed as the percentage of participants who were found to be positive (bars shown in black) or negative (bars shown in white) for the presence of at least one IgG antibody of interest in serum samples. Fisher's exact test was used for comparison (⁎⁎⁎p < 0.001).
      Table 2 shows the number of participants with IgG seropositivity, revealing a relatively increased prevalence for almost all the studied serine proteases in the MS group when compared to HCs. Specifically, of the MS patients, an important prevalence was tested positive for the presence of IgG antibodies against FVIIa (n = 19, 11.4%), thrombin (n = 11, 6.6%), and prothrombin (n = 11, 6.6%), while seropositivity to FXa was observed in MS patients (n = 17, 10.2%), revealing a significant differentiation when compared to HCs (n = 1, 2.5%; p<0.05). Furthermore, seropositivity against plasmin (n = 20, 12%) and PC (n = 7, 4.1%) was shown in MS patients in a greater prevalence than in HCs (Table.2).
      Table 2Seropositivity status of participants to serine proteases of the coagulation cascade.
      Seropositivity
      IgG againstMS [n(%)]HCs [n (%)]Mean activity P value
      FVIIa19 (11.4)1 (2.5)<0.0001
      Thrombin11 (6.6)1 (2.5)>0.05
      Prothrombin11 (6.6)2 (5.0)>0.05
      FXa
      Statistically significant seropositivity in patient group compared to healthy controls (p = 0.04). MS: Multiple Sclerosis; HCs: Healthy controls; IgG: Immunoglobulin G.
      17 (10.2)0>0.05
      FXII12 (7.2)1 (2.5)<0.01
      Plasmin20 (12.0)2 (5.0)<0.01
      Protein C7 (4.1)1 (2.5)>0.05
      low asterisk Statistically significant seropositivity in patient group compared to healthy controls (p = 0.04). MS: Multiple Sclerosis; HCs: Healthy controls; IgG: Immunoglobulin G.

      3.3 Antibody binding and disease courses

      Across all disease courses, there was an approximately equal prevalence of IgG positivity, revealing an increasing number of participants who tested positive for at least one IgG. Specifically, 42.3% of RRMS patients (55 out of 130), 44.8% of SPMS patients (13 out of 29), and 42.8% of PPMS patients (3 out of 7) demonstrated IgG seropositivity. In both RRMS and SPMS patients, anti-plasmin IgG was the most frequently detected antibody, followed by anti-FXa and anti-FVIIa IgG in RRMS and anti-FXII IgG in SPMS patients.
      The results for each antibody tested are indicated in the Table 3, in which multiple comparisons are shown after performing the post hoc Dunn's test. CIS and PPMS are not included due to the small number of patients analyzed.
      Table 3Binding activity against coagulation serine proteases with respect to disease courses. The Kruskal–Wallis test was performed for the analysis and Dunn's test for the multiple comparisons.
      AntibodiesMean Index (±SEM)Kruskal-Wallis testDunn's multiple comparisons test
      RRMS-HCsSPMS-HCsRRMS-SPMS
      RRMSSPMSHCsP valueP value
      Anti-FVIIa26.80 (±0.82)26.91

      (±2.13)
      21.03

      (±0.95)
      < 0.0001
      p value < 0.001.
      p value < 0.01.
      ns
      Anti-thrombin68.41

      (±5.08)
      66.87

      (±11.80)
      75.15

      (±6.48)
      = 0.37nsnsns
      Anti-prothrombin8.49

      (±1.49)
      5.69

      (±2.64)
      5.18

      (±1.83)
      = 0.15nsnsns
      Anti-FXa33.79

      (±2.10)
      35.10

      (±4.77)
      27.85

      (±2.67)
      = 0.45nsnsns
      Anti-FXII88.13

      (±2.77)
      83.42

      (±4.96)
      72.17

      (±3.57)
      = 0.0023
      p value < 0.01.
      nsns
      Anti-plasmin36.96

      (±1.89)
      48.04

      (±4.92)
      29.06

      (±3.02)
      = 0.0024ns
      p value < 0.01.
      ns
      Anti-Protein C103.1

      (±8.24)
      86.46

      (±8.60)
      110.70

      (±6.67)
      = 0.15nsnsns
      Results are expressed as the percentage of the O.D. value of each sample with respect to a reference sample. HCs: Healthy controls; RRMS: Relapsing-Remitting MS; SPMS: Secondary Progressive MS; ns: not significant;.
      low asterisklow asterisk p value < 0.01.
      low asterisklow asterisklow asterisk p value < 0.001.
      Both RRMS and SPMS exhibited anti-FVIIa positivity with significantly higher mean antibody activity levels [RRMS mean: 26.80 (SEM: 0.82); p < 0.0001; SPMS mean: 26.91 (SEM: 2.13); p < 0.01, respectively] in comparison to HCs [mean: 21.03 (SEM: 0.95)] (Fig.3A). Similarly, RRMS patients had a significantly higher frequency (p < 0.01) of anti-FXII IgG activity [mean: 88.13 (SEM: 2.77), than the controls [mean: 72.17 (SEM: 3.57)] (Fig.3B); however, this difference was not observed when comparing SPMS patients and HCs or antibody activity of RRMS and SPMS patients. On the other hand, SPMS exhibited greater mean index of activity [mean: 48.04 (SEM: 4.92)] than RRMS [mean: 36.96 (SEM: 1.89)], and differentiated significantly from HCs (p < 0.01), as shown in Fig. 3C.
      Fig 3
      Fig. 3IgG activity against serine proteases FVIIa (A), FXII (B), and plasmin (C) among MS types and HCs. Results are expressed as the percentage of the O.D. value of each sample with respect to a reference sample. Dashed lines represent the threshold for positivity calculated as the 2 S.D. above the mean of HCs. Bars represent the mean ± SEM. The Kruskal–Wallis test was performed for the analysis and Dunn's test for the multiple comparisons (⁎⁎p < 0.01, ⁎⁎⁎p < 0.001) (data with statistical significance are shown).

      3.4 Correlation between the levels of IgG antibodies

      Among the patients with seropositivity status, 17 out of 72 (23.6%) showed reactivity for more than one IgG antibody against trypsin/thrombin-like serine proteases of the coagulation cascade.
      Further analysis of the correlation between antibody binding responses by Spearman correlation coefficient revealed a negative correlation between anti-FVIIa IgG and anti-prothrombin IgG levels (rs=−0.23, p < 0.01) (Fig. 4A), and this was also observed between anti-FVIIa IgG levels and anti-PC IgG activity levels (rs=−0.29, p < 0.05) (Fig. 4B). On the other hand, a monotonically increasing correlation was revealed between the anti-plasmin and anti-FXa IgG activity levels (rs=0.3, p < 0.001) (Fig. 4C), and likewise, a positive correlation was observed between anti-plasmin and anti-thrombin IgG (rs=0.23, p < 0.01) (Fig. 4D), between anti-plasmin and anti-PC IgG (rs=0.34, p < 0.01) (Fig. 4E), and between anti-PC IgG and anti-thrombin IgG activity levels (rs=0.25, p < 0.05) (Fig. 4F). These correlations are also illustrated in the heat map (Fig. 5), in addition to correlation between anti-plasmin IgG and age (rs=0.17, p < 0.05).
      Fig 4
      Fig. 4Correlation between antibody activity against coagulant components. Negative correlation was revealed between anti-FVIIa IgG levels and anti-prothrombin IgG activity levels (A), and between anti-FVIIa IgG levels and anti-PC IgG activity levels (B). Positive correlation was demonstrated between anti-plasmin IgG levels and anti-FXa IgG activity levels (C), between anti-plasmin IgG and anti-thrombin IgG activity levels (D), between anti-plasmin and anti-PC IgG (E), and between anti-PC IgG and anti-thrombin IgG activity levels (F). Values are expressed as the percentage of the O.D. value of each sample with respect to a reference sample (data with statistical significance are shown).
      Fig 5
      Fig. 5Correlation matrix between antibody responses and clinical outcomes in MS patients. All the correlation analyzes were performed using Spearman correlation coefficient. The color-coded R values range between −1 (white color) and 1 (gray color) as provided on heat map. Demographic and clinical outcomes are as follows: age, disease duration, EDSS, MSSS.
      To evaluate the association between the study IgG antibodies and the clinical outcomes in MS patients, the odds ratio (OR) was calculated (Table 4). A significant association between anti-plasmin IgG and age was observed [OR:1.04 (95% CI: 1.0–1.1)], and activity to plasmin was also associated with EDSS [OR:1.30 (95% CI: 1.0– 1.7)] (Table 4). Furthermore, anti-thrombin IgG was correlated significantly with MSSS [OR:1.50 (95% CI: 1.1–2.0)] (Table 4). Interestingly, the correlation analysis between anti-prothrombin IgG and age showed and an odds ratio less than 1 [OR:0.95 (95% CI: 0.9–1.0); p < 0.05], and likewise, a similar observation was revealed between anti-prothrombin IgG and disease duration [OR:0.93 (95% CI: 0.85–1.0); p < 0.05] (Table 4).
      Table 4Logistic regression results between IgG binding activity and demographic or clinical features of MS patients.
      Odds Ratio (95% confidence interval)
      AntibodiesAgeGenderDisease durationDisease CoursesEDSSMSSSMedication
      Anti-FVIIa0.97 (0.94–1.01)0.68 (0.26–1.86)0.97 (0.91–1.02)1.14 (0.46–2.37)0.89 (0.67–1.17)1.0 (0.80–1.25)1.04 (0.77–1.40)
      Anti-thrombin1.0 (0.95–1.05)2.22 (0.54–15.0)0.97 (0.90–1.04)1.26 (0.40–2.98)1.35 (0.98–1.90)1.47 (1.10–2.02)

      p = 0.009 (⁎⁎)
      1.22 (0.82–1.94)
      Anti-prothrombin0.95 (0.89–0.99)

      p = 0.03 (
      Statistically significant values (p < 0.05) are shown in bold. EDSS: expanded disability status scale; MSSS: multiple sclerosis severity score.
      )
      0.93 (0.27–3.67)0.93 (0.85–0.99)

      p = 0.049 (
      Statistically significant values (p < 0.05) are shown in bold. EDSS: expanded disability status scale; MSSS: multiple sclerosis severity score.
      )
      1.32 (0.42–3.08)0.74 (0.47–1.07)0.92 (0.66–1.24)1.03 (0.71–1.52)
      Anti-FXa0.99 (0.95–1.03)1.78 (0.59–6.55)1.0 (0.94–1.06)1.06 (0.39–2.31)1.22 (0.93–1.60)1.20 (0.95–1.51)0.89 (0.65–1.20)
      Anti-FXII1.0 (0.95–1.05)0.59 (0.17–2.19)1.01 (0.95–1.08)1.02 (0.33–2.40)0.95 (0.67–1.321.17 (0.83–1.65)1.23 (0.82–1.96)
      Anti-plasmin1.04 (1.0–1.08)

      p = 0.03 (
      Statistically significant values (p < 0.05) are shown in bold. EDSS: expanded disability status scale; MSSS: multiple sclerosis severity score.
      )
      0.59 (0.23–1.56)1.03 (0.98–1.08)1.45 (0.66–2.88)1.30 (1.02–1.69)

      p = 0.03 (
      Statistically significant values (p < 0.05) are shown in bold. EDSS: expanded disability status scale; MSSS: multiple sclerosis severity score.
      )
      1.10 (0.89–1.37)1.26 (0.94–1.75)
      Anti-Protein C1.02 (0.95–1.05)3.54 (0.58–68.2)0.97 (0.88–1.07)0.52 (0–2.18)1.16 (0.77–1.72)0.95 (0.71–1.25)0.79 (0.47–1.94)
      low asterisk Statistically significant values (p < 0.05) are shown in bold. EDSS: expanded disability status scale; MSSS: multiple sclerosis severity score.

      3.5 Clinical outcomes of MS patients with IgG seropositivity

      As Table 5 illustrates, patients with seropositivity of IgG against FVIIa, PT, FXII, and PC had a mild disability status, which corresponds to an EDSS score up to 3.5, whereas patients with anti-thrombin IgG positivity had the greatest EDSS median index. Likewise, an increased median MSSS score was found in the presence of anti-thrombin IgG [median: 6.5 (range:3.3–7.5)], followed by anti-FXa [median: 5.2 (range:2.8–6.9)] and anti-plasmin IgG [median: 4.5 (range: 3.1–6.1)]. On the contrary, patients with anti-FXII positivity had the lowest MSSS score [median: 3.0 (range:2.6–4.7)].
      Table 5Disease severity corresponding to the immunological profile of patients with MS found to be positive against serine proteases.
      IgGEDSS MSSS
      Median (interquartile range)Median (interquartile range)
      Anti-FVIIa2.5 (2.0–5.0)4.3 (2.4–5.2)
      Anti-thrombin5.0 (3.0–7.0)6.5 (3.3–7.5)
      Anti-prothrombin2.5 (2.0–4.0)4.0 (2.8–4.9)
      Anti-FXa4.0 (3.0–6.5)5.2 (2.8–6.9)
      Anti-FXII3.5 (2.2–4.7)3.0 (2.6–4.7)
      Anti-plasmin4.5 (3.0–6.0)4.5 (3.1–6.1)
      Anti-Protein C3.5 (3.0–5.0)3.1 (3.0–7.3)
      EDSS: expanded disability status scale; MSSS: multiple sclerosis severity score; IgG: immunoglobulin G.

      4. Discussion

      The presence of immunoglobulin G against serine proteases of the coagulation cascade was analyzed in patients with MS to test our hypothesis that such antibodies might contribute to MS progression. Compared with HC (n = 8, 20%), 72 patients or a proportion of 43% showed significantly higher levels of IgG for at least one antibody (p < 0.001). Furthermore, our findings indicated that patients had significantly elevated levels of binding activity to FVIIa (n = 19, 11.4%, mean activity p < 0.0001), FXII (n = 12, 7.5%, mean activity p < 0.001), and plasmin (n = 20, 12%, mean activity p < 0.01) than healthy individuals. Noteworthy, the anti-FVIIa antibody was found to be an important marker for patients suffering from RRMS and SPMS in our study since the binding activity levels were significantly elevated in these groups. Moreover, the levels of anti-FXII could also differentiate RRMS patients from HCs, while patients with SPMS differed significantly from HCs in binding activity to plasmin. This observation was further extended with the correlation between anti-plasmin and EDSS as well as between anti-plasmin and age, demonstrating the implication of such antibodies in disease disability over time.
      To our knowledge, this is the first study that involves antibodies against coagulant components in neurodegenerative and inflammatory diseases of the CNS. Prior studies have revealed the presence of the coagulation components in a large proportion of RRMS and SPMS patients with specificity for prothrombin and FX (
      • Göbel K.
      • Kraft P.
      • Pankratz S.
      • Gross C.C.
      • Korsukewitz C.
      • Kwiecien R.
      • Mesters R.
      • Kehrel B.E.
      • Wiendl H.
      • Kleinschnitz C.
      • Meuth S.G.
      Prothrombin and factor X are elevated in multiple sclerosis patients.
      ). Additionally, in vitro studies have shown that early symptoms of Alzheimer's disease are associated with vascular dysfunction and increased BBB permeability mediated by thrombin (
      • Festoff B.W.
      • Sajja R.K.
      • Van Dreden P.
      • Cucullo L.
      HMGB1 and thrombin mediate the blood-brain barrier dysfunction acting as biomarkers of neuroinflammation and progression to neurodegeneration in alzheimer's disease.
      ;
      • Iannucci J.
      • Renehan W.
      • Grammas P.
      Thrombin, a mediator of coagulation, inflammation, and neurotoxicity at the neurovascular interface: implications for alzheimer's disease.
      ). In contrast, the inhibition of the FXII activity by anti-FXII in an AD mouse model could reduce thrombin-mediated inflammatory responses via inhibition of the intrinsic coagulation pathway and, therefore, could regulate the brain pathology (
      • Suidan G.L.
      • Singh P.K.
      • Patel-Hett S.
      • Chen Z.L.
      • Volfson D.
      • Yamamoto-Imoto H.
      • Norris E.H.
      • Bell R.D.
      • Strickland S.
      Abnormal clotting of the intrinsic/contact pathway in alzheimer disease patients is related to cognitive ability.
      ;
      • Zamolodchikov D.
      • Renné T.
      • Strickland S.
      The Alzheimer's disease peptide β-amyloid promotes thrombin generation through activation of coagulation factor XII.
      ).
      The results of our study are consistent with other research studies, which have revealed the presence of anti-FVIIa IgG in patients diagnosed with diseases that share similar clinical features with MS. Namely, the majority of patients in an APS study were characterized as positive for the presence of the FVIIa-specific antibody (n = 22/33, 67%) (
      • Bidot C.J.
      • Jy W.
      • Horstman L.L.
      • Huisheng H.
      • Jimenez J.J.
      • Yaniz M.
      • Ahn Y.S.
      Factor VII/VIIa: a new antigen in the anti-phospholipid antibody syndrome.
      ), while elevated anti-FVIIa IgG levels have been detected in plasma samples from patients with bleeding disorders (
      • Bidot C.J.
      • Jy W.
      • Horstman L.L.
      • Huisheng H.
      • Jimenez J.J.
      • Yaniz M.
      • Ahn Y.S.
      Factor VII/VIIa: a new antigen in the anti-phospholipid antibody syndrome.
      ) and coronary heart disease (
      • Eriksson Berg M.
      • Silveira A.
      • Orth Gomér K.
      • Hamsten A.
      • Schenck Gustafsson K.
      Coagulation factor VII in middle-aged women with and without coronary heart disease.
      ). It is also worth mentioning that the same studies have characterized two distinct isotypes of anti-FVIIa antibodies with different binding properties. The first type of anti-FVIIa antibody was calcium-dependent and directed against the TF-FVIIa complex to inhibit its function. The second isotype was calcium-independent with a weak affinity for its target (
      • Kamikubo Y.I.
      • Miyamoto S.
      • Iwasa A.
      • Ishii M.
      • Okajima K.
      Purification and characterization of factor VII inhibitor found in a patient with life threatening bleeding.
      ). Even though it is unclear how these molecules play a role in coagulation-inflammation interactions, anti-FVIIa antibodies can exert either a procoagulant or anticoagulant response, depending on the FVIIa complex in which they participate, their targeted epitope, and the microenvironment in which they act (
      • Bajaj M.S.
      • Birktoft J.J.
      • Steer S.A.
      • Bajaj S.P.
      Structure and biology of tissue factor pathway inhibitor.
      ;
      • Kamikubo Y.I.
      • Miyamoto S.
      • Iwasa A.
      • Ishii M.
      • Okajima K.
      Purification and characterization of factor VII inhibitor found in a patient with life threatening bleeding.
      ).
      Similarly, IgG antibodies directed against the fibrinolytic enzyme plasmin have recently been described as procoagulant molecules with a pivotal role in the cross-talk between coagulation-inflammation (
      • Yang C.D.
      • Hwang K.K.
      • Yan W.
      • Gallagher K.
      • FitzGerald J.
      • Grossman J.M.
      • Hahn B.H.
      • Chen P.P.
      Identification of anti-plasmin antibodies in the antiphospholipid syndrome that inhibit degradation of fibrin.
      ). Fibrinolysis is a tightly regulated process involving both enhancers, such as tissue plasminogen activator (tPA), and inhibitors, such as plasminogen activator inhibitors (PAI-1 and PAI-2) and a2-antiplasmin. The latter inhibits the function of tPA and plasmin, respectively (
      • Booth N.A.
      Fibrinolysis and thrombosis. Bailliere's.
      ;
      • Galli M.
      • Barbui T.
      Antiprothrombin antibodies: detection and clinical significance in the antiphospholipid syndrome.
      ). Surprisingly, purified anti-plasmin IgG antibodies from APS and SLE patients have been shown to reduce the fibrinolytic activity of plasmin, interfering with fibrin clot degradation and causing accumulation of thrombi (
      • Yang C.D.
      • Hwang K.K.
      • Yan W.
      • Gallagher K.
      • FitzGerald J.
      • Grossman J.M.
      • Hahn B.H.
      • Chen P.P.
      Identification of anti-plasmin antibodies in the antiphospholipid syndrome that inhibit degradation of fibrin.
      ). Pathological conditions associated with the expression profile of hormones and cytokines that induce the expression of PAI-1 inhibitor can cause hypercoagulability in the blood, leading to the appearance of sclerosis in the capillaries (
      • Angelini F.
      • Pagano F.
      • Bordin A.
      • Picchio V.
      • De Falco E.
      • Chimenti I.
      Getting old through the blood: circulating molecules in aging and senescence of cardiovascular regenerative cells.
      ;
      • Eren M.
      • Boe A.E.
      • Klyachko E.A.
      • Vaughan D.E.
      Role of plasminogen activator inhibitor-1 in senescence and aging.
      ). As a result, PAI-1 expression and impaired fibrinolysis have been linked to diseases in which thrombosis and vascular sclerosis play a critical role in their progression (
      • Angelini F.
      • Pagano F.
      • Bordin A.
      • Picchio V.
      • De Falco E.
      • Chimenti I.
      Getting old through the blood: circulating molecules in aging and senescence of cardiovascular regenerative cells.
      ). Surprisingly, purified anti-plasmin IgG from APS and SLE patients could also reduce the fibrinolytic activity of plasmin, interfering with fibrin clot degradation and causing accumulation of thrombi (
      • Yang C.D.
      • Hwang K.K.
      • Yan W.
      • Gallagher K.
      • FitzGerald J.
      • Grossman J.M.
      • Hahn B.H.
      • Chen P.P.
      Identification of anti-plasmin antibodies in the antiphospholipid syndrome that inhibit degradation of fibrin.
      ).
      We have also demonstrated a mild outcome of disease progression corresponding with FXII-IgG seropositivity. Analysis of the role of anti-FXII antibodies was also conducted by
      • Matafonov A.
      • Leung P.Y.
      • Gailani A.E.
      • Grach S.L.
      • Puy C.
      • Cheng Q.
      • Sun M.F.
      • McCarty O.J.T.
      • Tucker E.I.
      • Kataoka H.
      • Renné T.
      • Morrissey J.H.
      • Gruber A.
      • Gailani D.
      Factor XII inhibition reduces thrombus formation in a primate thrombosis model.
      , who evaluated the contribution of monoclonal anti-FXII antibodies in the thrombosis-inflammation interplay. Matafonov and his colleagues found a significant reduction of thrombin generation and fibrin clot formation after a treatment with these specific antibodies, suggesting a protective role against thrombus formation (
      • Matafonov A.
      • Leung P.Y.
      • Gailani A.E.
      • Grach S.L.
      • Puy C.
      • Cheng Q.
      • Sun M.F.
      • McCarty O.J.T.
      • Tucker E.I.
      • Kataoka H.
      • Renné T.
      • Morrissey J.H.
      • Gruber A.
      • Gailani D.
      Factor XII inhibition reduces thrombus formation in a primate thrombosis model.
      ). To our knowledge, the current study is the first study analyzing antibody levels with MSSS. Therefore, further studies with a larger sample size from multiple centers are required to validate these results further in order to investigate the role of anti-FXII antibodies in the coagulation-inflammation interplay and characterize their role in MS progression. On the other hand, anti-thrombin IgG positivity correlated with the greatest disease severity score leading to speculation that anti-thrombin antibodies are implicated in the mechanisms that underlie MS pathology and being involved in disease progression.
      It is also worth mentioning that 17 of 72 patients (23.6%) who were positive for at least one antibody of interest showed multiple reactivities for more than one member of the trypsin/thrombin-like serine protease family. In accordance with our findings, other studies have also illustrated the binding of a single antibody to serine proteases of hemostasis that share similar or homologous catalytic binding sites (
      • Matsuda J.
      • Sanaka T.
      • Nishizawa A.
      • Gotoh M.
      • Gohchi K.
      Two antiprothrombin antibodies against prothrombin and prothrombin-phosphatidyl serine show partial but not total identity.
      ). Antibodies against PT have been shown to bind plasminogen and to prevent the creation of plasmin and activation of the fibrinolytic mechanism, and therefore to cause thrombosis in patients with myocardial infarction. In addition, monoclonal antibodies derived from patients with APS recognize common epitopes in plasmin and thrombin catalytic regions, and it should be further evaluated whether these interactions can provide thrombotic potency through the inhibition of protease inactivation by anticoagulant molecules or interfere with fibrinolytic mechanisms (
      • Yang C.D.
      • Hwang K.K.
      • Yan W.
      • Gallagher K.
      • FitzGerald J.
      • Grossman J.M.
      • Hahn B.H.
      • Chen P.P.
      Identification of anti-plasmin antibodies in the antiphospholipid syndrome that inhibit degradation of fibrin.
      ).
      Attempting to understand the relevance of these antibodies to MS increasingly emphasizes the importance of the link between the dysfunction of the vascular microenvironment and MS development. In patients with MS, the abnormal endothelial function may be initiated or sustained by proteins like PAR receptors that are activated by coagulant factors such as thrombin and fibrin. or platelet aggregation. Beyond the interactions of the endothelial cells mediated by PAR-mediated proinflammatory cytokine expression, coagulant components like thrombin and fibrin/fibrinogen are crucial for the function of the vascular barrier. Through PAR-1 activation, thrombin can initiate proinflammatory signaling pathways such as NF-kB or mitogen-activated protein kinases (MAPKs), which are pivotal for the maintenance of the coagulation-inflammation interplay leading to disease progression. Furthermore, patients with MS have an increased risk for venous thromboembolism along with ishaemic stroke, cardiovascular perturbation and pulmonary embolism during the early stages of the disease.
      In experimental autoimmune encephalomyelitis (EAE), thrombin and fibrinogen/fibrin deposit in the vasculature of the CNS thereby promoting proinflammatory signaling pathways. More specifically, in MS animal models, thrombin activity has been detected in the vasculature of the CNS prior to the neurological signs and accumulated gradually throughout the disease progression. Importantly, thrombin activity is strongly correlated with both microglia activity and lesion size at the peak of EAE (
      • Davalos D.
      • Baeten K.M.
      • Whitney M.A.
      • Mullins E.S.
      • Friedman B.
      • Olson E.S.
      • Ryu J.K.
      • Smirnoff D.S.
      • Petersen M.A.
      • Bedard C.
      • Degen J.L.
      • Tsien R.Y.
      • Akassoglou K.
      Early detection of thrombin activity in neuroinflammatory disease.
      ). The early thrombin activity can modify microglia motility toward the pre-demyelinated regions and promote the proliferation of microglial cells. In addition, binding of fibrin to the integrin CD11b/CD18 receptor expressed on microglial cells promotes the production of reactive oxygen species and the secretion of proinflammatory cytokines and chemokines sustaining the inflammatory cellular processes, as well as cell adhesion. Similarly, thrombin can induce the proliferation of astrocytes and microglia, as well as promote the expression of adhesion molecules on endothelial cells, enhance the increased permeability of the BBB, and the high intracellular calcium and contribute to neuroinflammation and axonal damage. Interestingly, thrombin activity is inhibited by hirudin, resulting in the prevention of PAR-mediated inflammatory responses while reducing microglial activity and ameliorating the severity of the disease. It is also worth noting that when plasma prothrombin is reduced genetically in the MS animal model, a significant reduction of thrombin is revealed in the spinal cord at the peak of the disease (
      • Davalos D.
      • Baeten K.M.
      • Whitney M.A.
      • Mullins E.S.
      • Friedman B.
      • Olson E.S.
      • Ryu J.K.
      • Smirnoff D.S.
      • Petersen M.A.
      • Bedard C.
      • Degen J.L.
      • Tsien R.Y.
      • Akassoglou K.
      Early detection of thrombin activity in neuroinflammatory disease.
      ). Depletion of fibrinogen in genetically modified animal models has also been shown to suppress microglia activation and neuroinflammation and prevent axonal injury in pre-demyelinated areas (
      • Davalos D.
      • Kyu Ryu J.
      • Merlini M.
      • Baeten K.M.
      • Le Moan N.
      • Petersen M.A.
      • Deerinck T.J.
      • Smirnoff D.S.
      • Bedard C.
      • Hakozaki H.
      • Gonias Murray S.
      • Ling J.B.
      • Lassmann H.
      • Degen J.L.
      • Ellisman M.H.
      • Akassoglou K.
      Fibrinogen-induced perivascular microglial clustering is required for the development of axonal damage in neuroinflammation.
      ).
      MS is a multifactorial disease. In addition to genetic, epigenetic, and epidemiological factors, viral infections have been linked to the activation of immunity and auto-reactivity of T- and B-cells, especially Epstein-Barr virus (EBV) (
      • Saberi A.
      • Akhondzadeh S.
      • Kazemi S.
      Infectious agents and different course of multiple sclerosis: a systematic review.
      ). Prior studies have suggested the molecular mimicry hypothesis to explain the autoimmunity in MS, whereby epitopes of foreign antigens have a similar structure to host own antigens. As a consequence, immune cells cross-react with foreign and self-antigens (
      • Meier U.C.
      • Cipian R.C.
      • Karimi A.
      • Ramasamy R.
      • Middeldorp J.M.
      Cumulative roles for epstein-barr virus, human endogenous retroviruses, and human herpes virus-6 in driving an inflammatory cascade underlying MS Pathogenesis.
      ). Recently, the hypothesis that MS is caused by EBV has been examined in a cohort comprising of more than 10 million young adults, where they detected that the risk of developing MS increased 32-fold after infection with EBV but was not increased after infection with other viruses (
      • Bjornevik K.
      • Cortese M.
      • Brian C.
      • Healy J.K.
      • Mina M.J.
      • Leng Y.
      • Elledge S.J.
      • Niebuhr D.W.
      • Scher A.I.
      • Munger, Kassandra L.A.A.
      Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis.
      ). Researchers identified for the first time that antibodies from the cerebrospinal fluid of MS patients bound to both Epstein-Barr nuclear antigen 1 (EBNA1) and the glial cell adhesion molecule (GlialCAM), an antigen which is detected in the CNS. Interestingly, in the MS mouse model following immunization with EBNA1, antibodies against EBNA1 and GlialCAM were detected and positively associated with severe disease phenotypes (
      • Lanz T.V.
      • Brewer R.C.
      • Ho P.P.
      • Moon J.S.
      • Jude K.M.
      • Fernandez D.
      • Fernandes R.A.
      • Gomez A.M.
      • Nadj G.S.
      • Bartley C.M.
      • Schubert R.D.
      • Hawes I.A.
      • Vazquez S.E.
      • Iyer M.
      • Zuchero J.B.
      • Teegen B.
      • Dunn J.E.
      • Lock C.B.
      • Kipp L.B.
      • Cotham V.C.
      • Ueberheide B.M.
      • Aftab B.T.
      • Anderson M.S.
      • DeRisi J.L.
      • Wilson M.R.
      • Bashford-Rogers R.J.M.
      • Platten M.
      • Garcia K.C.
      • Steinman L.
      • Robinson W.H.
      Clonally expanded B cells in multiple sclerosis bind EBV EBNA1 and GlialCAM.
      ). Our findings have shown the presence of antibodies to coagulant components in a high prevalence of MS patients. Thus, studies are needed to determine whether there is any potential relevance between EBV antigens and coagulation components that would facilitate the triggering of an autoimmune response by causing cross-reaction. Beyond that, they would be useful in the development of new therapeutic strategies.
      This study has some limitations. There was a significant difference in sample size between MS groups, and thus it made it insufficient to draw valid conclusions. Larger population sample sizes in future research would allow for more precise results. In addition, we did not examine whether the antibodies were consistently positive, which might also be contributing to the disease. Longitudinal evaluation should be conducted to assess the persistent positivity.

      5. Conclusion

      Overall, the IgG antibodies directed against serine proteases of the coagulation cascade illustrated an increased prevalence in patients with MS. Moreover, clinical features of the disease can also be associated with the presence of particular IgG antibodies against the coagulation serine proteases. Further studies are needed to evaluate these molecules as part of the coagulation-inflammation circuit and their clinical significance in MS progression to assess whether they can contribute to better monitoring and prognosis of the disease as well as whether they can serve as targets for novel therapeutic strategies.

      Data availability statement

      The data that support the findings of this study are available from the corresponding author upon reasonable request.

      CRediT authorship contribution statement

      Maria S. Hadjiagapiou: Methodology, Formal analysis, Data curation, Writing – original draft, Writing – review & editing. George Krashias: Writing – review & editing. Elie Deeba: Writing – original draft, Writing – review & editing. Christina Christodoulou: Writing – review & editing. Marios Pantzaris: Resources, Writing – review & editing. Anastasia Lambrianides: Conceptualization, Visualization, Data curation, Writing – original draft, Writing – review & editing.

      Declaration of Competing of Interest

      All authors have no potential conflict of interest to report.

      Acknowledgment

      This work was supported by the Cyprus Muscular Dystrophy Association (Telethon 2017). We would like to thank all patients and HCs at the Cyprus Institute of Neurology and Genetics for donating blood for this research. Our deepest gratitude to Dr. Kyriaki Michailidou and Dr. Christiana Demetriou for their assistance with the statistical analysis.

      Ethical standards

      All procedures were done according to ethical standards of the Cyprus National Bioethics Committee. All patients informed and signed an approved consent (EEBK/EΠ/2016/51).

      References

        • Angelini F.
        • Pagano F.
        • Bordin A.
        • Picchio V.
        • De Falco E.
        • Chimenti I.
        Getting old through the blood: circulating molecules in aging and senescence of cardiovascular regenerative cells.
        Front. Cardiovasc. Med. 2017; https://doi.org/10.3389/fcvm.2017.00062
        • Artim-Esen B.
        • Pericleous C.
        • Mackie I.
        • Ripoll V.M.
        • Latchman D.
        • Isenberg D.
        • Rahman A.
        • Ioannou Y.
        • Giles I.
        Anti-factor Xa antibodies in patients with antiphospholipid syndrome and their effects upon coagulation assays.
        Arthritis Res. Ther. 2015; https://doi.org/10.1186/s13075-015-0568-7
        • Bajaj M.S.
        • Birktoft J.J.
        • Steer S.A.
        • Bajaj S.P.
        Structure and biology of tissue factor pathway inhibitor.
        Thromb. Haemost. 2001; https://doi.org/10.1055/s-0037-1616518
        • Bidot C.J.
        • Jy W.
        • Horstman L.L.
        • Huisheng H.
        • Jimenez J.J.
        • Yaniz M.
        • Ahn Y.S.
        Factor VII/VIIa: a new antigen in the anti-phospholipid antibody syndrome.
        Br. J. Haematol. 2003; 120: 618-626https://doi.org/10.1046/j.1365-2141.2003.04161.x
        • Bjornevik K.
        • Cortese M.
        • Brian C.
        • Healy J.K.
        • Mina M.J.
        • Leng Y.
        • Elledge S.J.
        • Niebuhr D.W.
        • Scher A.I.
        • Munger, Kassandra L.A.A.
        Longitudinal analysis reveals high prevalence of Epstein-Barr virus associated with multiple sclerosis.
        Science. 2022; 375: 296-301https://doi.org/10.1126/science.abj8222
        • Booth N.A.
        Fibrinolysis and thrombosis. Bailliere's.
        Best Pract. Res. Clin. Haematol. 1999; https://doi.org/10.1053/beha.1999.0034
        • Chu F.
        • Shi M.
        • Zheng C.
        • Shen D.
        • Zhu J.
        • Zheng X.
        • Cui L.
        The roles of macrophages and microglia in multiple sclerosis and experimental autoimmune encephalomyelitis.
        J. Neuroimmunol. 2018; https://doi.org/10.1016/j.jneuroim.2018.02.015
        • Davalos D.
        • Akassoglou K.
        Fibrinogen as a key regulator of inflammation in disease.
        Semin. Immunopathol. 2012; https://doi.org/10.1007/s00281-011-0290-8
        • Davalos D.
        • Baeten K.M.
        • Whitney M.A.
        • Mullins E.S.
        • Friedman B.
        • Olson E.S.
        • Ryu J.K.
        • Smirnoff D.S.
        • Petersen M.A.
        • Bedard C.
        • Degen J.L.
        • Tsien R.Y.
        • Akassoglou K.
        Early detection of thrombin activity in neuroinflammatory disease.
        Ann. Neurol. 2014; https://doi.org/10.1002/ana.24078
        • Davalos D.
        • Kyu Ryu J.
        • Merlini M.
        • Baeten K.M.
        • Le Moan N.
        • Petersen M.A.
        • Deerinck T.J.
        • Smirnoff D.S.
        • Bedard C.
        • Hakozaki H.
        • Gonias Murray S.
        • Ling J.B.
        • Lassmann H.
        • Degen J.L.
        • Ellisman M.H.
        • Akassoglou K.
        Fibrinogen-induced perivascular microglial clustering is required for the development of axonal damage in neuroinflammation.
        Nat. Commun. 2012; https://doi.org/10.1038/ncomms2230
        • Eren M.
        • Boe A.E.
        • Klyachko E.A.
        • Vaughan D.E.
        Role of plasminogen activator inhibitor-1 in senescence and aging.
        Semin. Thromb. Hemost. 2014; https://doi.org/10.1055/s-0034-1387883
        • Eriksson Berg M.
        • Silveira A.
        • Orth Gomér K.
        • Hamsten A.
        • Schenck Gustafsson K.
        Coagulation factor VII in middle-aged women with and without coronary heart disease.
        Thromb. Haemost. 2001; https://doi.org/10.1055/s-0037-1615719
        • Faria S.S.
        • Fernandes P.C.
        • Silva M.J.B.
        • Lima V.C.
        • Fontes W.
        • Freitas R.
        • Eterovic A.K.
        • Forget P.
        The neutrophil-to-lymphocyte ratio: a narrative review.
        Ecancermedicalscience. 2016; https://doi.org/10.3332/ecancer.2016.702
        • Festoff B.W.
        • Sajja R.K.
        • Van Dreden P.
        • Cucullo L.
        HMGB1 and thrombin mediate the blood-brain barrier dysfunction acting as biomarkers of neuroinflammation and progression to neurodegeneration in alzheimer's disease.
        J. Neuroinflammation. 2016; https://doi.org/10.1186/s12974-016-0670-z
        • Galli M.
        • Barbui T.
        Antiprothrombin antibodies: detection and clinical significance in the antiphospholipid syndrome.
        Blood. 1999; https://doi.org/10.1182/blood.v93.7.2149
        • Gandhi R.
        • Laroni A.
        • Weiner H.L.
        Role of the innate immune system in the pathogenesis of multiple sclerosis.
        J. Neuroimmunol. 2010; https://doi.org/10.1016/j.jneuroim.2009.10.015
        • Ghasemi N.
        • Razavi S.
        • Nikzad E.
        Multiple sclerosis: pathogenesis, symptoms, diagnoses and cell-based therapy.
        Cell J. 2017; https://doi.org/10.22074/cellj.2016.4867
        • Göbel K.
        • Kraft P.
        • Pankratz S.
        • Gross C.C.
        • Korsukewitz C.
        • Kwiecien R.
        • Mesters R.
        • Kehrel B.E.
        • Wiendl H.
        • Kleinschnitz C.
        • Meuth S.G.
        Prothrombin and factor X are elevated in multiple sclerosis patients.
        Ann. Neurol. 2016; https://doi.org/10.1002/ana.24807
        • Grover S.P.
        • Mackman N.
        Tissue factor: an essential mediator of hemostasis and trigger of thrombosis.
        Arterioscler. Thromb. Vasc. Biol. 2018; https://doi.org/10.1161/ATVBAHA.117.309846
        • Huang W.J.
        • Chen W.W.
        • Zhang X.
        Multiple sclerosis: pathology, diagnosis and treatments (review).
        Exp. Ther. Med. 2017; https://doi.org/10.3892/etm.2017.4410
        • Iannucci J.
        • Renehan W.
        • Grammas P.
        Thrombin, a mediator of coagulation, inflammation, and neurotoxicity at the neurovascular interface: implications for alzheimer's disease.
        Front. Neurosci. 2020; https://doi.org/10.3389/fnins.2020.00762
        • Jones D.W.
        • Gallimore M.J.
        • Mackie I.J.
        • Harris S.L.
        • Winter M.
        Reduced factor XII levels in patients with the antiphospholipid syndrome are associated with antibodies to factor XII.
        Br. J. Haematol. 2000; https://doi.org/10.1046/j.1365-2141.2000.02251.x
        • Kamikubo Y.
        • Mendolicchio G.L.
        • Zampolli A.
        • Marchese P.
        • Rothmeier A.S.
        • Orje J.N.
        • Gale A.J.
        • Krishnaswamy S.
        • Gruber A.
        • Østergaard H.
        • Petersen L.C.
        • Ruf W.
        • Ruggeri Z.M.
        Selective factor VIII activation by the tissue factor–factor VIIa–factor Xa complex.
        Blood. 2017; https://doi.org/10.1182/blood-2017-02-767079
        • Kamikubo Y.I.
        • Miyamoto S.
        • Iwasa A.
        • Ishii M.
        • Okajima K.
        Purification and characterization of factor VII inhibitor found in a patient with life threatening bleeding.
        Thromb. Haemost. 2000; https://doi.org/10.1055/s-0037-1613758
        • Koudriavtseva T.
        • Zannino S.
        • Filippi M.M.
        • Cortese A.
        • Piantadosi C.
        • Lapucci C.
        • Fiorelli M.
        • Giannarelli D.
        • Mandoj C.
        • Stefanile A.
        • Conti L.
        • Salvetti M.
        • Inglese M.
        Coagulation activation and cerebral hypoperfusion in relapsing-remitting multiple sclerosis.
        J. Neurol. Sci. 2019; https://doi.org/10.1016/j.jns.2019.10.1161
        • Lambrianides A.
        • Turner-Stokes T.
        • Pericleous C.
        • Ehsanullah J.
        • Papadimitraki E.
        • Poulton K.
        • Ioannou Y.
        • Lawrie A.
        • MacKie I.
        • Chen P.
        • Latchman D.
        • Isenberg D.
        • Rahman A.
        • Giles I.
        Interactions of human monoclonal and polyclonal antiphospholipid antibodies with serine proteases involved in hemostasis.
        Arthritis Rheumatol. 2011; https://doi.org/10.1002/art.30525
        • Lanz T.V.
        • Brewer R.C.
        • Ho P.P.
        • Moon J.S.
        • Jude K.M.
        • Fernandez D.
        • Fernandes R.A.
        • Gomez A.M.
        • Nadj G.S.
        • Bartley C.M.
        • Schubert R.D.
        • Hawes I.A.
        • Vazquez S.E.
        • Iyer M.
        • Zuchero J.B.
        • Teegen B.
        • Dunn J.E.
        • Lock C.B.
        • Kipp L.B.
        • Cotham V.C.
        • Ueberheide B.M.
        • Aftab B.T.
        • Anderson M.S.
        • DeRisi J.L.
        • Wilson M.R.
        • Bashford-Rogers R.J.M.
        • Platten M.
        • Garcia K.C.
        • Steinman L.
        • Robinson W.H.
        Clonally expanded B cells in multiple sclerosis bind EBV EBNA1 and GlialCAM.
        Nature. 2022; https://doi.org/10.1038/s41586-022-04432-7
        • Matafonov A.
        • Leung P.Y.
        • Gailani A.E.
        • Grach S.L.
        • Puy C.
        • Cheng Q.
        • Sun M.F.
        • McCarty O.J.T.
        • Tucker E.I.
        • Kataoka H.
        • Renné T.
        • Morrissey J.H.
        • Gruber A.
        • Gailani D.
        Factor XII inhibition reduces thrombus formation in a primate thrombosis model.
        Blood. 2014; https://doi.org/10.1182/blood-2013-04-499111
        • Matsuda J.
        • Sanaka T.
        • Nishizawa A.
        • Gotoh M.
        • Gohchi K.
        Two antiprothrombin antibodies against prothrombin and prothrombin-phosphatidyl serine show partial but not total identity.
        Blood Coagul. Fibrinolysis. 2002; https://doi.org/10.1097/00001721-200212000-00005
        • Mayo L.
        • Quintana F.J.
        • Weiner H.L.
        The innate immune system in demyelinating disease.
        Immunol. Rev. 2012; https://doi.org/10.1111/j.1600-065X.2012.01135.x
        • Meier U.C.
        • Cipian R.C.
        • Karimi A.
        • Ramasamy R.
        • Middeldorp J.M.
        Cumulative roles for epstein-barr virus, human endogenous retroviruses, and human herpes virus-6 in driving an inflammatory cascade underlying MS Pathogenesis.
        Front. Immunol. 2021; https://doi.org/10.3389/fimmu.2021.757302
        • Pavelek Z.
        • Angelucci F.
        • Souček O.
        • Krejsek J.
        • Sobíšek L.
        • Klímová B.
        • Šarláková J.
        • Halúsková S.
        • Kuča K.
        • Vališ M.
        Innate immune system and multiple sclerosis. Granulocyte numbers are reduced in patients affected by relapsing-remitting multiple sclerosis during the remission phase.
        J. Clin. Med. 2020; https://doi.org/10.3390/jcm9051468
        • Polman C.H.
        • Reingold S.C.
        • Banwell B.
        • Clanet M.
        • Cohen J.A.
        • Filippi M.
        • Fujihara K.
        • Havrdova E.
        • Hutchinson M.
        • Kappos L.
        • Lublin F.D.
        • Montalban X.
        • O'Connor P.
        • Sandberg-Wollheim M.
        • Thompson A.J.
        • Waubant E.
        • Weinshenker B.
        • Wolinsky J.S.
        Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria.
        Ann. Neurol. 2011; https://doi.org/10.1002/ana.22366
        • Renné T.
        • Schmaier A.H.
        • Nickel K.F.
        • Blombäck M.
        • Maas C.
        In vivo roles of factor XII.
        Blood. 2012; https://doi.org/10.1182/blood-2012-07-292094
        • Rezaie A.R.
        Protease-activated receptor signalling by coagulation proteases in endothelial cells.
        Thromb. Haemost. 2014; https://doi.org/10.1160/TH14-02-0167
        • Ruiz F.
        • Vigne S.
        • Pot C.
        Resolution of inflammation during multiple sclerosis.
        Semin. Immunopathol. 2019; https://doi.org/10.1007/s00281-019-00765-0
        • Ryu J.K.
        • Petersen M.A.
        • Murray S.G.
        • Baeten K.M.
        • Meyer-Franke A.
        • Chan J.P.
        • Vagena E.
        • Bedard C.
        • Machado M.R.
        • Coronado P.E.R.
        • Prod'homme T.
        • Charo I.F.
        • Lassmann H.
        • Degen J.L.
        • Zamvil S.S.
        • Akassoglou K.
        Blood coagulation protein fibrinogen promotes autoimmunity and demyelination via chemokine release and antigen presentation.
        Nat. Commun. 2015; https://doi.org/10.1038/ncomms9164
        • Saberi A.
        • Akhondzadeh S.
        • Kazemi S.
        Infectious agents and different course of multiple sclerosis: a systematic review.
        Acta Neurol. Belg. 2018; https://doi.org/10.1007/s13760-018-0976-y
        • Spronk H.M.H.
        • Ten Cate H.
        • Van Der Meijden P.E.J.
        Differential roles of tissue factor and phosphatidylserine in activation of coagulation.
        Thromb. Res. 2014; https://doi.org/10.1016/j.thromres.2014.03.022
        • Suidan G.L.
        • Singh P.K.
        • Patel-Hett S.
        • Chen Z.L.
        • Volfson D.
        • Yamamoto-Imoto H.
        • Norris E.H.
        • Bell R.D.
        • Strickland S.
        Abnormal clotting of the intrinsic/contact pathway in alzheimer disease patients is related to cognitive ability.
        Blood Adv. 2018; https://doi.org/10.1182/bloodadvances.2018017798
        • Yadav S.K.
        • Mindur J.E.
        • Ito K.
        • Dhib-Jalbut S.
        Advances in the immunopathogenesis of multiple sclerosis.
        Curr. Opin. Neurol. 2015; https://doi.org/10.1097/WCO.0000000000000205
        • Yang C.D.
        • Hwang K.K.
        • Yan W.
        • Gallagher K.
        • FitzGerald J.
        • Grossman J.M.
        • Hahn B.H.
        • Chen P.P.
        Identification of anti-plasmin antibodies in the antiphospholipid syndrome that inhibit degradation of fibrin.
        J. Immunol. 2004; https://doi.org/10.4049/jimmunol.172.9.5765
        • Yang Y.H.
        • Hwang K.K.
        • FitzGerald J.
        • Grossman J.M.
        • Taylor M.
        • Hahn B.H.
        • Chen P.P.
        Antibodies against the activated coagulation factor X (FXa) in the antiphospholipid syndrome that interfere with the FXa inactivation by antithrombin.
        J. Immunol. 2006; https://doi.org/10.4049/jimmunol.177.11.8219
        • Zamolodchikov D.
        • Renné T.
        • Strickland S.
        The Alzheimer's disease peptide β-amyloid promotes thrombin generation through activation of coagulation factor XII.
        J. Thromb. Haemost. 2016; https://doi.org/10.1111/jth.13209