Thrombin-Activatable Fibrinolysis Inhibitor Antigen Levels and Cardiovascular Risk Factors
Abstract—Thrombin-activatable fibrinolysis inhibitor (TAFI) is a recently described fibrinolysis inhibitor that circulates in plasma as a procarboxypeptidase and is converted into an active form during coagulation. The physiological relevance of TAFI is not known, but it might be involved in pathways regulating fibrin deposition. Our aim was to determine the interindividual variability of plasma TAFI antigen values and their associations with conventional cardiovascular risk factors. Six hundred twenty-six consecutive patients (277 men) attending a metabolic ward for primary prevention were studied. TAFI antigen presented a large range of values, with a 2- to 3-fold increase between the 10th and 90th percentiles. No difference was observed between the 2 sexes. A significant correlation was observed between age and TAFI levels in women only. After adjustment for age, TAFI antigen was positively correlated in men for the waist-to-hip circumference ratio and blood pressure, whereas no significant correlation was observed in women. Stepwise multiple linear regression analysis indicated a low contribution of the parameters studied to the variability of TAFI antigen levels; the waist-to-hip circumference ratio accounted for only 2% in men, and age accounted for only 3% in women. Results were compared with those of fibrinogen and plasminogen activator inhibitor-1; cardiovascular risk factors in men and women accounted for 16% and 9.5%, respectively, of the fibrinogen variance and 36% and 32%, respectively, of the plasminogen activator inhibitor-1 variance. These observations did not attribute an important role to lifestyle characteristics in the control of TAFI antigen concentration in plasma. Because of the large interindividual variability of TAFI levels in plasma, genetic control may be involved.
- thrombin-activatable fibrinolysis inhibitor
- cardiovascular risk
- plasminogen activator inhibitor-1
- Received October 20, 1999.
- Accepted April 25, 2000.
Thrombin-activatable fibrinolysis inhibitor (TAFI) is a recently described fibrinolysis inhibitor cloned from human liver.1 It circulates as a procarboxypeptidase B zymogen, which is converted into an active form, carboxypeptidase U (EC126.96.36.199), or TAFIa, during coagulation2 after thrombin cleavage.3 Generation of TAFIa is dependent on the quantity of thrombin generated during coagulation and is drastically potentiated by thrombomodulin.4 5 6 7 Generation of TAFIa during clotting leads to a retardation of clot lysis that is mainly due to the fast elimination of carboxy-terminal lysine residues, thus decreasing plasminogen binding on partially degraded fibrin.8 9 10 11 In vitro, experiments have shown that enrichment by TAFI increased the lysis time of plasma clots in a dose-dependent manner.3 9 In a canine model of coronary thrombolysis, the time to achieve reperfusion correlates directly with levels of a TAFIa-like activity, because it can be inhibited ex vivo by a specific inhibitor of plasma carboxypeptidase B.12 Moreover, coadministration of this inhibitor and tissue plasminogen activator significantly improved tissue plasminogen activator thrombolysis.13
The importance of the hemostatic system in predisposing to or precipitating coronary heart disease has gained increasing recognition over the past several years. The published results from prospective studies are remarkably consistent for fibrinogen, indicating a highly statistically association with coronary artery disease.14 A rise in the circulating level of the fibrinolytic inhibitor, plasminogen activator inhibitor (PAI)-1, has also been shown to predict the occurrence of myocardial infarction.15 Interestingly, both of these factors are related to lifestyle.16 17 Thus, conventional cardiovascular risk factors may be involved in atherosclerosis through modifications of the plasma levels of hemostatic factors. Therefore, it is of importance to analyze the associations between hemostatic parameters and conventional cardiovascular risk factors to determine whether they could be affected by modification of lifestyle.
The physiological relevance of TAFI is not known, but it might be involved in pathways regulating the balance between fibrin formation and deposition within and in the vicinity of the vascular bed. Thus, TAFI could represent a candidate gene for thrombosis and atherogenesis. The quantity of TAFIa generated during coagulation could not be easily measured, but in small populations, its level has been shown to be positively correlated with the level of circulating TAFI antigen.18
Our aim was to determine the interindividual variability of the TAFI antigen in a large population of subjects attending an outpatient clinic for primary prevention of cardiovascular disease. We then analyzed the association of conventional cardiovascular risk factors with TAFI antigen plasma levels. Results were compared with those obtained for fibrinogen and PAI-1.
The population consisted of 626 consecutive patients (277 men and 349 women) attending a metabolic ward (Center de Détection et de Prévention de l’Athérosclérose, Marseille, France) for primary prevention of coronary disease.
Patients were submitted to a standardized examination. Height and weight were recorded, and the body mass index (BMI) was calculated as kilograms per square meter. The waist and the hip circumferences were measured, with patients in a standing position, midway between the lower rib margin and the iliac crest and at the greatest circumference, respectively; the former divided by the latter is reported as the waist-to-hip circumference ratio (WHR) and is a measure of fat distribution.
While the patient was in the sitting position, systolic blood pressure (SBP) and diastolic blood pressure (DBP) measurements were taken on the left arm by an automated device (OMRON 705 CP); the mean of 3 measurements was used for analysis. A 12-lead ECG was recorded.
Noninvasive arterial explorations were performed with a B-mode ultrasound imager (High Definition Imaging, Advanced Technologies Laboratories). Left and right common carotid arteries were examined with a 5- to 10-MHz linear probe. Artery images were obtained in the anteroposterior projection and were perpendicular to the far wall of the vessel. The measurement of the intima-media thickness (IMT) was performed by the same physician with the built-in software from 3 images of each carotid artery. The mean of these 6 measures was used for analysis.
The interview included questions on health status, personal and familial history of cardiovascular disease, socioprofessional factors, and drug use, including hormonal therapy. Patients were graded as nonsmokers, exsmokers, and current smokers, with a quantification in packs per year.
Total energy and nutrient intakes were calculated by recording 3 days of nutrition and analyzed by use of nutritional software (GENI). Alcohol consumption (kilocalories per day) was automatically calculated from self-report.
Anxiety and physical activity were assessed by self-administered questionnaires.
Blood samples were obtained from the antecubital vein, after an overnight fasting, between 8:00 and 10:00 am, collected on citrate (3.8% citrate, 0.129 mol/L), and centrifuged (2500g for 30 minutes at 4°C). Platelet-poor plasma was kept frozen at <−80°C until analysis. Antigen determination of TAFI was performed with a commercially available kit from Milan Analytica. This assay is based on affinity-purified sheep anti-TAFI IgG raised against TAFI purified from plasma. These antibodies do not recognize carboxypeptidase N and are able to recognize the proenzyme as well as the active form of TAFI. Results are expressed as a percentage of a control pooled plasma from 30 healthy volunteers. Fibrinogen concentration was determined by the Clauss thrombin clotting method. PAI-1 antigen was assayed by a commercially available kit (Asserachrom PAI-1, Stago).
Total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, apoB, glucose, creatinine, and microalbuminuria were determined by routine clinical chemical procedures.
Statistical analyses were performed by use of Statview and SPSS software. Values are expressed as mean±SD. Distribution of BMI, PAI-1 antigen, lipid parameters, and microalbuminuria values were skewed, and logarithmically transformed values were used. The Mann-Whitney test was used to determine differences in mean values between men and women. One-way ANOVA was computed to compare the hemostatic parameter levels according to the presence or absence of the cardiovascular risk factors and was substituted by the Kruskal-Wallis test for small subpopulations. Spearman correlation coefficients were calculated to study the associations between hemostatic parameters and other variables. When the influence of the hormonal status was evaluated, the interaction with age was tested. Stepwise multiple linear regression analysis was performed to evaluate independent correlates of hemostatic parameters. A value of P<0.05 was considered statistically significant.
Characteristics of the population studied are given in Table 1⇓. The percentage of individuals with risk factors are as follows, for men and women, respectively: excess weight (BMI >25), 69.8% and 68.0%; dyslipemia (total cholesterol >250 mg/dL and/or triglycerides >200 mg/dL or current treatment), 64.5% and 48.9% (P<0.001); smoking (tobacco consumption >5 cigarettes per day), 33.8% and 29.1% (P<0.001); hypertension (SBP >160 mm Hg and/or DBP >90 mm Hg or current treatment), 25.8% and 14.3% (P<0.001); sedentary lifestyle (physical activity <500 kcal/wk), 25.8% and 41.2% (P<0.001); and glucose anomaly (fasting glycemia >125 mg/dL or current treatment), 16.2% and 7.8% (P<0.001). Most patients presented multiple risk factors. The prevalence of 2 risk factors was 27.7% in men and 33.3% in women, and the prevalence of 3 risk factors was 32.0% in men and 29.1% in women.
TAFI Antigen, Clottable Fibrinogen, and PAI-1 Antigen Distribution in the Population
Table 2⇓ shows the distribution of the hemostatic parameters according to the study participants. TAFI antigen levels presented a wide range of values in men and in women, with a 2.5- to 3-fold increase between the 10th and 90th percentiles. The distribution was gaussian in men and women. However, a small population (13.4%) with low values (<60%) was identified (Figure 1⇓). Compared with the rest of the population, this small population did not exhibit differences in all the variables studied, nor did it present any difference in familial history of thrombosis or the use of blood pressure–lowering or lipid-lowering agents. In contrast to fibrinogen and PAI-1, no sex difference was observed in TAFI antigen levels. A significant correlation was observed between age and TAFI, fibrinogen, and PAI-1 levels in women (r=0.17, 0.16, and 0.15, respectively; P<0.01) but not in men. To evaluate further the association between TAFI levels and age in women, we compared levels of TAFI antigen according to quartiles of age (Figure 2⇓). It appears that the oldest women were more likely to have high levels of TAFI antigen (P<0.01). The same trend was observed for men, without reaching significance (P=0.09).
Associations Between Hemostatic Variables and Cardiovascular Risk Factors
By univariate analysis, in men, TAFI antigen was significantly and positively correlated with BMI (P<0.03), WHR (P<0.05), SBP (P<0.02), DBP (P<0.01), and fibrinogen (P<0.03), whereas in women, it was correlated with age (P<0.01), SBP (P<0.02), total cholesterol (P<0.05), apoB (P<0.01), fibrinogen (P<0.03), and IMT (P<0.01). After adjustment for age, the correlations between TAFI and other parameters disappeared in women, but TAFI remained significantly correlated with WHR and blood pressure in men (Table 3⇓). No change related to the use of blood pressure–lowering or lipid-lowering agents was observed.
After adjustment for age, in men, fibrinogen was significantly and positively correlated with BMI, WHR, SBP, DBP, smoking, and glycemia and significantly and negatively correlated with HDL cholesterol and alcohol intake, whereas in women, fibrinogen levels were correlated with BMI, DBP, and microalbuminuria (Table 2⇑). PAI-1 was strongly correlated with variables related to insulin resistance state (BMI, WHR, blood pressure, HDL cholesterol, triglycerides, and glucose) in both sexes. It was also significantly correlated with IMT in men but not in women (Table 3⇑).
Interestingly, in the whole population, the TAFI antigen levels were significantly higher in patients with a self-reported familial history of cardiovascular disease than in those without (106±35% versus 100±35% [mean±SD], respectively; P=0.05).
Influence of Hormonal Status on Hemostatic Parameter Levels
TAFI antigen was evaluated in the 198 premenopausal and 151 menopausal women according to the use of hormonal therapy. The use of oral contraceptives (n=20 women) did not affect TAFI or fibrinogen levels, whereas PAI-1 levels in women not receiving oral contraceptives (n=178) were significantly higher (data not shown). When women were studied according to menopausal status, postmenopausal women receiving hormonal substitution (n=43), compared with postmenopausal women not receiving hormonal substitution (n=108), had 6.6% lower levels of TAFI antigen (103±41% versus 111±35%, respectively; P<0.01), and the mean level was not significantly different from that of premenopausal women (data not shown). The same evolution was observed for fibrinogen and PAI-1 (data not shown). However, except for PAI-1, this difference disappeared after age adjustment.
Contribution of Environmental Factors and Age to the Variability of Hemostatic Parameters
Results of the multivariate step-by-step analysis are reported in Table 4⇓. Results showed a low contribution of the parameters studied to the variability of TAFI antigen levels. Indeed, WHR explained only 2% of the variability of the TAFI antigen values in men, and age explained only 3% of the variability in women. On the other hand, the environmental factors studied could account for ≈16% of the fibrinogen variance in men and 9.5% of the variance in women. The influence of metabolic factors on PAI-1 antigen was even more pronounced, because it could account for ≈36% of the variance in men and 32% of the variance in women. Multivariate analysis was also performed, restricted to the upper part of the distribution (TAFI antigen >60%), and did not lead to different results.
TAFI may occupy a key position in the regulation of endogenous fibrinolysis because it acts directly on fibrin to delay or inhibit plasminogen activation of the fibrin network. This presumed role and the results showing an association between TAFI level and delay to reperfusion in animals during thrombolysis could implicate TAFI in the development of thrombogenesis and atherogenesis. This argues for devoting special attention to TAFI in humans. Our results showed large interindividual variation in circulating TAFI antigen. A 2.5- to 3-fold difference was noticed between the 10th and 90th percentiles. These results agree with those of Schatteman et al,19 who reported a 3-fold difference between the minimum and maximum TAFI value by use of a functional assay based on quantitative conversion of TAFI to its active form by the thrombin-thrombomodulin complex. When the distribution of the values was analyzed, it appeared to be biphasic, with a first peak including 13.4% of the individuals with TAFI antigen values <60%. This remains unexplained, because these individuals did not differ from the others in all the variables studied, including familial history of cardiovascular disease.
The finding of a positive association between levels of fibrinogen, PAI-1, and cardiovascular risk factors is consistent with previous studies. In accordance with the Prospective Epidemiological Study of Myocardial Infarction (PRIME),20 the traditional cardiovascular risk factors accounted for ≈10% and 30% of the total variance of circulating fibrinogen and PAI-1, respectively. Whereas the same percentage was obtained in men and women for PAI-1, this was not the case for fibrinogen. Cardiovascular risk factors are implicated differently in men than in women. These data are highly consistent with those previously reported by the second Monitoring Trends and Determinants in Cardiovascular Disease (MONICA) survey.16 Indeed, in men, the relation between fibrinogen and BMI was less strong than in women, and WHR was consistently included before BMI in the stepwise procedure in men but not in women.
Another difference between men and women was the absence of any effect of smoking on fibrinogen levels in women. These results also illustrate the discrepancy between fibrinogen and PAI-1 in regard to lifestyle. Whereas fibrinogen levels are moderately affected, their contribution to circulating PAI-1 levels is strong and unique among the hemostatic parameters. This difference likely reflects the high proportion of insulin-resistant obese patients included in this population. Indeed, in contrast to fibrinogen, PAI-1 levels are dramatically increased in the insulin-resistant state.21 Moreover, the predictive effect of plasma PAI-1 for myocardial infarction has been shown to disappear after adjustment for parameters of the insulin resistance syndrome,15 which is not the case for fibrinogen.22
In contrast to fibrinogen and PAI-1 levels, a weak or no relationship was found between TAFI antigen levels and all the variables studied. The percentage of TAFI variance that could be accounted for was very low, 2% and 3% in men and women, respectively. The association between age and TAFI antigen values confirms previous results,19 but in the present study, this relationship was restricted to women. The observation that TAFI antigen levels were mainly elevated in the oldest women and that postmenopausal women exhibited higher levels of TAFI antigen raises questions about the role of estrogen in controlling circulating TAFI levels. As previously reported,23 postmenopausal women receiving estrogen replacement therapy had lower levels of PAI-1 than did postmenopausal women not receiving therapy; this difference was maintained after age adjustment.
Altogether, these observations did not attribute an important role to lifestyle in the control of TAFI antigen levels. This was different from the results observed for fibrinogen and PAI-1 levels. Thus, the plasma levels of PAI-1 and TAFI, 2 circulating fibrinolytic inhibitors, are differently regulated. Because of the large interindividual variability of TAFI antigen levels and the weak relationship with environment, it is possible that TAFI antigen levels are mainly under genetic control. This awaits further confirmation.
This work was supported by grants from the Institut National de la Santé et de la Recherche Médicale (INSERM, EPI 99/36), from the Ministère de l’Education Nationale et de la Recherche (Programme quadriennal Université de la Méditerranée), and from the Ministère de la Travail et des Affaires Sociales, Program Hospitalier de Recherche Clinique (PHRC). We thank A. Charles for statistical expertise, J. Ansaldi for excellent technical assistance, and E. Guenoun for organization of the samples.
Eaton DL, Malloy BE, Tsai SP, Henzel W, Drayna D. Isolation, molecular cloning, and partial characterization of a novel carboxypeptidase B from human plasma. J Biol Chem. 1991;266:21833–21838.
Bajzar L, Manuel R, Nesheim ME. Purification and characterization of TAFI, a thrombin-activable fibrinolysis inhibitor. J Biol Chem. 1995;270:14477–14484.
Boffa MB, Wang W, Bajzar L, Nesheim ME. Plasma and recombinant thrombin-activable fibrinolysis inhibitor (TAFI) and activated TAFI compared with respect to glycosylation, thrombin/thrombomodulin-dependent activation, thermal stability, and enzymatic properties. J Biol Chem. 1998;273:2127–2135.
Bajzar L, Morser J, Nesheim M. TAFI, or plasma procarboxypeptidase B, couples the coagulation and fibrinolytic cascades through the thrombin-thrombomodulin complex. J Biol Chem. 1996;271:16603–16608.
Kokame K, Zheng X, Sadler JE. Activation of thrombin-activable fibrinolysis inhibitor requires epidermal growth factor-like domain 3 of thrombomodulin and is inhibited competitively by protein C. J Biol Chem. 1998;273:12135–12139.
Bajzar L, Nesheim M, Morser J, Tracy PB. Both cellular and soluble forms of thrombomodulin inhibit fibrinolysis by potentiating the activation of thrombin-activable fibrinolysis inhibitor. J Biol Chem. 1998;273:2792–2798.
Redlitz A, Tan AK, Eaton DL, Plow EF. Plasma carboxypeptidases as regulators of the plasminogen system. J Clin Invest. 1995;96:2534–2538.
Bajzar L, Nesheim ME, Tracy PB. The profibrinolytic effect of activated protein C in clots formed from plasma is TAFI-dependent. Blood. 1996;88:2093–2100.
Sakharov DV, Plow EF, Rijken DC. On the mechanism of the antifibrinolytic activity of plasma carboxypeptidase B. J Biol Chem. 1997;272:14477–14482.
Wang W, Boffa MB, Bajzar L, Walker JB, Nesheim ME. A study of the mechanism of inhibition of fibrinolysis by activated thrombin-activable fibrinolysis inhibitor. J Biol Chem. 1998;273:27176–21781.
Redlitz A, Nicolini FA, Malycky JL, Topol EJ, Plow EF. Inducible carboxypeptidase activity: a role in clot lysis in vivo. Circulation. 1996;93:1328–1330.
Klement P, Liao P, Bajzar L. A novel approach to arterial thrombolysis. Blood. 1999;94:2735–2743.
Juhan-Vague I, Pyke SD, Alessi MC, Jespersen J, Haverkate F, Thompson SG. Fibrinolytic factors and the risk of myocardial infarction or sudden death in patients with angina pectoris: ECAT Study Group: European Concerted Action on Thrombosis and Disabilities. Circulation. 1996;94:2057–2063.
Krobot K, Hense HW, Cremer P, Eberle E, Keil U. Determinants of plasma fibrinogen: relation to body weight, waist-to-hip ratio, smoking, alcohol, age, and sex: results from the second MONICA Augsburg survey 1989–1990. Arterioscler Thromb. 1992;12:780–788.
Henry M, Tregouet DA, Alessi MC, Aillaud MF, Visvikis S, Siest G, Tiret L, Juhan-Vague I. Metabolic determinants are much more important than genetic polymorphisms in determining the PAI-1 activity and antigen plasma concentrations: a family study with part of the Stanislas Cohort. Arterioscler Thromb Vasc Biol. 1998;18:84–91.
Schatteman KA, Goossens FJ, Scharpe SS, Neels HM, Hendriks DF. Assay of procarboxypeptidase U, a novel determinant of the fibrinolytic cascade, in human plasma. Clin Chem. 1999;45:807–813.
Scarabin PY, Aillaud MF, Amouyel P, Evans A, Luc G, Ferrieres J, Arveiler D, Juhan-Vague I. Associations of fibrinogen, factor VII and PAI-1 with baseline findings among 10,500 male participants in a prospective study of myocardial infarction: the PRIME Study: Prospective Epidemiological Study of Myocardial Infarction. Thromb Haemost. 1998;80:749–756.
Juhan-Vague I, Morange P, Renucci JF, Alessi MC. Fibrinogen, obesity and insulin resistance. Blood Coagul Fibrinolysis. 1999;10:S25–S28.
Scarabin PY, Alhenc-Gelas M, Plu-Bureau G, Taisne P, Agher R, Aiach M. Effects of oral and transdermal estrogen/progesterone regimens on blood coagulation and fibrinolysis in postmenopausal women: a randomized controlled trial. Arterioscler Thromb Vasc Biol. 1997;17:3071–3078.