Joyce Malfliet

Impaired fibrinolysis as a risk factor for Budd-Chiari syndrome

In Budd-Chiari syndrome (BCS), thrombosis develops in the hepatic veins or inferior vena cava. To study the relationship between hypofibrinolysis and BCS, we measured plasma levels of fibrinolysis proteins in 101 BCS patients and 101 healthy controls and performed a plasma-based clot lysis assay. In BCS patients, plasminogen activator inhibitor 1 (PAI-1) levels were significantly higher than in controls (median, 6.3 vs 1.4 IU/mL, P < .001). Thrombin-activatable fibrinolysis inhibitor and plasmin inhibitor levels were lower than in controls (13.8 vs 16.9 microg/mL and 0.91 vs 1.02 U/L, both P < .001). Median plasma clot lysis time (CLT) was 73.9 minutes in cases and 73.0 minutes in controls (P = .329). A subgroup of cases displayed clearly elevated CLTs. A CLT above the 90th or 95th percentile of controls was associated with an increased risk of BCS, with odds ratios of 2.4 (95% confidence interval, 1.1-5.5) and 3.4 (95% confidence interval, 1.2-9.7), respectively. In controls, only PAI-1 activity was significantly associated with CLT. Analysis of single nucleotide polymorphisms of fibrinolysis proteins revealed no significant differences between cases and controls. This case-control study provides the first evidence that an impaired fibrinolytic potential, at least partially caused by elevated PAI-1 levels, is related to the presence of BCS.

Compaction of fibrin clots reveals the antifibrinolytic effect of factor XIII.

Essentials Factor XIIIa inhibits fibrinolysis by forming fibrin‐fibrin and fibrin‐inhibitor cross‐links. Conflicting studies about magnitude and mechanisms of inhibition have been reported. Factor XIIIa most strongly inhibits lysis of mechanically compacted or retracted plasma clots. Cross‐links of α2‐antiplasmin to fibrin prevent the inhibitor from being expelled from the clot.

Increased N-terminal cleavage of alpha-2-antiplasmin in patients with liver cirrhosis

The activity of alpha‐2‐antiplasmin (α2AP), the main fibrinolytic inhibitor, is modified by N‐ and C‐terminal proteolytic cleavages. C‐terminal cleavage converts plasminogen‐binding α2AP (PB‐α2AP) into a non‐plasminogen‐binding derivative. N‐terminal cleavage by antiplasmin‐cleaving enzyme (APCE), a soluble, circulating derivative of fibroblast activation protein (FAP), turns native Met‐α2AP into Asn‐α2AP, which is more quickly crosslinked into fibrin.

Biological variation in tPA-induced plasma clot lysis time

Hypofibrinolysis is a risk factor for venous and arterial thrombosis, and can be assessed by using a turbidimetric tPA-induced clot lysis time (CLT) assay. Biological variation in clot lysis time may affect the interpretation and usefulness of CLT as a risk factor for thrombosis. Sufficient information about assay variation and biological variation in CLT is not yet available. Thus, this study aimed to determine the analytical, within-subject and between-subject variation in CLT. We collected blood samples from 40 healthy individuals throughout a period of one year (average 11.8 visits) and determined the CLT of each plasma sample in duplicate. The mean (± SD) CLT was 83.8 (± 11.1) minutes. The coefficients of variation for total variation, analytical variation, within-subject variation and between-subject variation were 13.4%, 2.6%, 8.2% and 10.2%, respectively. One measurement can estimate the CLT that does not deviate more than 20% from its true value. The contribution of analytical variation to the within-subject variation was 5.0%, the index of individuality was 0.84 and the reference change value was 23.8%. The CLT was longer in the morning compared to the afternoon and was slightly longer in older individuals (> 40 years) compared to younger (≤40 years) individuals. There was no seasonal variation in CLT and no association with air pollution. CLT correlated weakly with fibrinogen, C-reactive protein, prothrombin time and thrombin generation. This study provides insight into the biological variation of CLT, which can be used in future studies testing CLT as a potential risk factor for thrombosis.

The effect of ethanol and its metabolism on fibrinolysis

The role of ethanol metabolism in possible haemostatic cardioprotective effects has not yet been determined. To this end, we investigated the effect of a moderate dose of ethanol (35 g) and its metabolism, on haemostatic variables over 14 hours (h). Eighteen Caucasian males participated in a placebo-controlled, randomised, cross-over study. Blood was collected prior to alcohol consumption, and at 10 time points for 14 h. Blood ethanol peaked at 1 h and was cleared after 8 h following ethanol consumption, significantly increasing plasma acetate (p=0.0028). Ethanol did not influence the coagulation factors significantly. PAI-1act increased (p<0.0001) and tPAact (p=0.047) decreased following alcohol consumption, reaching maximum (0.69 to 22.2 IU/ml) and minimum (0.88 to 0.33 IU/ml) levels at 5 h, respectively. Significantly increased plasma clot lysis times (46.8 to 67.6 minutes) and reduced global fibrinolytic capacity of whole blood, measured as D-dimer production during incubation of blood clots (2.26 to 0.29 μg/ml), were found at 5 h. Except for PAI-1act (borderline significance; p=0.05), there was no significant difference in the fibrinolytic markers between the two groups the following morning. Moderate ethanol consumption resulted in a significant temporary fibrinolysis inhibition. Any protective effects of moderate ethanol consumption on cardiovascular disease do not appear to be due to improvement in fibrinolytic potential within the first 14 h following consumption. The use of global fibrinolytic assays is recommended for determining the true effect of ethanol on fibrinolysis.

Binding of carboxypeptidase N to fibrinogen and fibrin

The ultimate step in the blood coagulation cascade is the formation of fibrin. Several proteins are known to bind to fibrin and may thereby change clot properties or clot function. Our previous studies identified carboxypeptidase N (CPN) as a novel plasma clot component. CPN cleaves C-terminal lysine and arginine residues from several proteins. The activity of CPN is increased upon its proteolysis by several proteases. The aim of this study is to investigate the presence of CPN in a plasma clot in more detail. Plasma clots were formed by adding thrombin, CaCl(2) and aprotinin to citrated plasma. Unbound proteins were washed away and non-covalently bound proteins were extracted and analyzed with 2D gel electrophoresis and mass spectrometry. The identification of CPN as a fibrin clot-bound protein was verified using Western blotting. Clot-bound CPN consisted of the same molecular forms as CPN in plasma and its content was approximately 30 ng/ml plasma clot. Using surface plasmon resonance we showed that CPN can bind to fibrinogen as well as to fibrin. In conclusion, CPN binds to fibrinogen and is present in a fibrin clot prepared from plasma. Because CPN binds to a fibrin clot, there could be a possible role for CPN as a fibrinolysis inhibitor.

Identification and characterization of α1‐antitrypsin in fibrin clots

Preliminary studies indicated that α1‐antitrypsin (A1AT) is the most abundant protein that is non‐covalently bound to fibrin clots prepared from plasma. The aim of this study was to identify and characterize fibrin(ogen)‐bound A1AT.

Oxidized high-density lipoprotein reduces blood clot firmness.

Low high-density lipoprotein (HDL) levels are a risk factor for cardiovascular disease. The best established anti-atherogenic property of HDL is its role in reverse cholesterol transport. Unlike HDL, low-density lipoprotein (LDL) has atherogenic properties and these are enhanced when LDL is oxidatively modified. Oxidized lipids can be transferred from oxidized LDL (oxLDL) to HDL and then be transported to the liver. However, other properties of circulating oxidized HDL (oxHDL) are not extensively investigated. Therefore we studied the role of native HDL (nHDL) and oxHDL in hemostasis. We collected citrated blood from healthy volunteers (n=5) and examined the effects of 4 mg/ml added nHDL and oxHDL on tissue factor-induced clot formation and clot firmness using thromboelastography (ROTEM). The effects of nLDL and oxLDL were measured as a control. Lipoproteins were isolated from plasma with density-gradient ultracentrifugation and oxidized with copper sulphate. The effect on fibrin degradation was examined by thromboelastography in the presence of cytochalasin D (2.5 μg/ml) and by measuring fibrin degradation products in the serum after thromboelastography using an ELISA. The ROTEM parameters clotting time, clot formation time and α-angle, were not affected by the addition of nHDL or oxHDL. The clot firmness, measured as amplitude, reduced during the thromboelastography by the addition of oxHDL. The amplitude (±SD) at 30 minutes, at 45 minutes, and at 60 minutes decreased from 100 (±0)%, 96 (±1.8)% and 91 (±2.9)% to 63 (±24.8)%, 51 (±24.3)% and 45 (±23.5)%, respectively (p<0.05) by the addition of oxHDL but did not change by the addition of nHDL. Fibrin breakdown, as measured by clot firmness in the presence of cytochalasin D was not affected by added oxHDL. In addition, the amount of fibrin degradation products after thromboelastography was comparable between control blood and blood with added oxHDL. No effects on clot firmness were observed with nLDL and oxLDL. In conclusion, oxHDL, but not nHDL, reduces blood clot firmness in a time-dependent manner without interfering with the clot formation. This effect does not reflect fibrinolysis but most likely involves platelets. The precise mechanism by which oxHDL reduces clot firmness has yet to be elucidated.

Circulating fibroblast activation protein activity and antigen levels correlate strongly when measured in liver disease and coronary heart disease

Background and aim Circulating fibroblast activation protein (cFAP) is a constitutively active enzyme expressed by activated fibroblasts that has both dipeptidyl peptidase and endopeptidase activities. We aimed to assess the correlation between cFAP activity and antigen levels and to compare variations in levels. Methods In plasma of 465 control individuals, 368 patients with coronary heart disease (CHD) and 102 hepatitis C virus (HCV) infected patients with severe liver disease before and after liver transplant, cFAP activity levels were measured with a newly developed cFAP activity assay. In the same samples, cFAP antigen levels were measured using a commercially available cFAP ELISA. Correlation analyses between activity and antigen levels were performed by calculating Pearson’s correlation coefficient (ρ). Additionally, normal ranges, determinants and differences between cohorts and between anticoagulants were investigated. Results cFAP activity and antigen levels significantly correlated in controls (ρ: 0.660, p<0.001) and in CHD patients (ρ: 0.709, p<0.001). cFAP activity and antigen levels in the HCV cohort were significantly lower in the samples taken after liver transplantation (p<0.001) and normalized toward levels of healthy individuals. Furthermore, cFAP activity and antigen levels were higher in men and significantly associated with body mass index. Also, cFAP activity and antigen levels were higher in EDTA plasma as compared to the levels in citrated plasma from the same healthy individuals. Conclusions For analyzing cFAP levels, either activity levels or antigen levels can be measured to investigate differences between individuals. However, it is of importance that blood samples are collected in the same anticoagulant.

Plasma levels of plasminogen activator inhibitor-1 and bleeding phenotype in patients with von Willebrand disease

von Willebrand disease (VWD) is the most common inherited bleeding disorder. In VWD patients, large variations in bleeding tendency are observed, which cannot be completely explained by the variation in von Willebrand factor levels or activities. Thus, there must be additional factors, for instance, changes in fibrinolysis that have an effect on the variation in bleeding tendency in VWD patients.