The Heterozygous 20210 G/A Prothrombin Genotype Is Associated With Early Venous Thrombosis in Inherited Thrombophilias and Is Not Increased in Frequency in Artery Disease
Abstract A genetic variation in the 3′-untranslated region of the prothrombin mRNA (20210 G/A) has recently been reported to be associated with elevated plasma prothrombin levels and with an increased incidence of venous thrombosis. We determined the frequency of this mutation, the detection of which was improved by allele-specific amplification of exon 14 and by denaturing gradients (denaturing gradient gel electrophoresis), in cohorts of patients affected by venous thrombosis (n=132) or by coronary or cerebrovascular diseases (n=195) and in normal subjects from various populations. An overlapping frequency of the heterozygous genotype (4%) was found in normal subjects from Italy and Cyprus, and no carrier was detected in 40 subjects of Indian or Somali origin. The 20210 GA heterozygous genotype was not increased in frequency in patients with arterial disease. In contrast, the GA genotype was associated (P=.007) with venous thrombosis both in simple heterozygotes (16%) with a family history of thrombosis as well as in double heterozygotes (14%) for other known thrombophilic defects. A synergic interaction between the prothrombin 20210 GA genotype and the factor V Leiden mutation, both potentially affecting the prothrombinase complex, was suggested by the early onset of thrombosis (median age 22 years) in doubly heterozygous patients. The association of the 20210 A allele with higher prothrombin levels was confirmed in the Italian population. However, the prothrombin assay does not allow an efficient preselection of patients for the DNA analysis.
- Received June 3, 1997.
- Accepted July 22, 1997.
An increased risk of venous thrombosis has been found to be associated with several hereditary abnormalities of the anticoagulant pathway1 and with the FV Leiden mutation (Arg506Gln), leading to activated protein C resistance. Several studies have also established that the risk increases in carriers of combined defects,2–5 which has further supported a role for the coagulation system in the development of venous thrombosis.
An association between high plasma factor VII or fibrinogen levels, the initiation factor, and the final substrate of the coagulation cascade, respectively, and subsequent incidence of ischemic cardiovascular events has been found.6–8 Moreover, protein C and protein S deficiencies have been implicated in ischemic stroke.9,10 However, the role of deficiencies of coagulation inhibitors in artery disease has not been not firmly established, and the factor V Leiden mutation did not appear to be associated with arterial thrombosis.11,12
Thrombin, a multifunctional serine protease that acts on blood clotting proteins and platelets, links the opposing coagulation and anticoagulant pathways,13–15 and, in addition, is a potent activator of many cellular responses,16,17 which makes it a key mediator in the pathological development of thrombotic vascular occlusions.18 Thrombophilic defects may act directly through increased activation of thrombin or a decrease in its inactivation.19
Very recently, the A allele of a genetic variation (20210 G/A) in the 3′-untranslated region of the prothrombin mRNA has been found to be associated with an increase in venous thrombosis.20 This variation was also associated with elevated plasma prothrombin levels indicating, as previously observed for other hemostatic genes,21–24 the presence of a genetic component in determining protein levels in plasma.
We investigated the interaction of this gene variation with other inherited thrombophilic defects, which may determine a particularly high risk for venous thrombosis. The frequency of this marker was also investigated in cohorts of Italian vascular patients. In addition, the frequency of the 20210 A allele was estimated in other populations, including Greek Cypriots, among whom the FV Leiden mutation is found frequently.25
Patients and Control Subjects
The following groups of patients and subjects were selected in a tertiary referral center and studied: (1) 62 patients without any known thrombophilic defects who had experienced a documented thrombotic event involving the venous system and also had a family history of venous thrombosis; (2) 70 patients who had experienced documented venous thrombosis and in whom an inherited thombophilic defect had been previously diagnosed; and (3) 68 asymptomatic relatives of carriers of thrombophilic defects.
Consecutive patients with arterial disease were also studied: (1) 90 patients with angina pectoris and objective evidence of myocardial ischemia who had undergone successful intracoronary stent implantation; among these patients 44 had had a previous acute myocardial infarction; and (2) 105 patients with cerebrovascular disease who had undergone carotid endoarteriectomy; 40 of them had previously experienced a stroke event confirmed by computed tomography.
Two groups of Italian subjects with no clinical manifestations, who were matched for age and geographical origin with the patients with venous and arterial diseases, respectively, were examined as control subjects. Subjects with different ethnic origins, 80 Greek Cypriots, 40 Somali, and 41 Indians, were studied. Informed consent to participate in the study was obtained from all the subjects examined.
Genomic DNA (100 ng), extracted from peripheral blood leukocytes, was used as a template for PCR amplification of exon 14. Primers were derived from the prothrombin gene sequence of Degen and Davie.26 Three methodological approaches were used to detect the 20210 G/A variation (Figure⇓).
Amplification with forward primer 5′-TGGGAAATATG GCTTCTACA-3′ (nucleotides 20035 to 20054) and reverse primer 5′-CACTGGGAGCATTGAAGCT-3′ (nucleotides 20229 to 20211) was performed in 30 cycles, each consisting of 20 seconds of denaturation at 94°C, 30 seconds of annealing at 52°C, and 30 seconds of extension at 72°C. The amplified fragment was analyzed by DGGE on a 6% polyacrylamide gel containing 15 to 45% denaturing agent (100% denaturant = 7 mol/L urea and 40% formamide in Tris-Acetate-EDTA buffer). DNA was stained with ethidium bromide.
Because the reverse primer (5′-CACTGGGAGCATT GAAGCT-3′) contained a mutagenic nucleotide (underlined), the amplified fragment from the allele A contains a HindIII site (AAGCTT) yielding two restriction fragments (176 and 19 bp in length). The amplified fragment from allele G (195 bp in length) lacks the restriction site.
Two allele-specific (bold letters) oligonucleotides 5′-CTGG GAGCATTGAAGCTC-3′ (nucleotides 20227 to 20210) and 5′-CTGGGAGCATTGAAGCTT-3′ (nucleotides 20227 to 20210) were used as reverse primers with the above-mentioned forward primer. The presence of an additional mismatch in the ASA primers both increased their selectivity and enabled us to confirm the fidelity of the ASA by restriction analysis. For selective amplification to occur the annealing temperature was increased (56°C) and the concentration of dNTPs was lowered (from 0.2 to 0.07 mmol/L).
Laboratory Thrombophilia Diagnosis
Diagnosis of antithrombin III, protein C, and protein S deficiencies had been established using conventional functional and immunological tests. The presence of the FV Leiden mutation was confirmed by DNA analysis.27
Prothrombin Level Assay
Prothrombin activity levels were measured by chromogenic assay using S-2238 as substrate and Echis carinatus venom as prothrombin activator.28 Patients receiving oral anticoagulant treatment were excluded.
Differences between groups were assessed by the two sample proportion test. Odds ratios, using the approximation of Woolf,29 were also calculated.
For the detection of the G-to-A transition in a large number of patients and control subjects, different methodological approaches (see “Methods”) were used. The original detection by restriction20 was compared with ASA-PCR and denaturing gradient (DGGE) analysis (Figure⇑). The inspection of the 3′-untranslated region of the prothrombin gene by DGGE analysis, potentially able to reveal the presence of additional mutations, did not show evidence of any sequence variation except for the 20210 G/A mutation.
The genotyping of patients and control subjects is summarized in Table 1⇓. The 20210 AG genotype was detected in 20 (9 women and 11 men) of 132 patients with venous thrombosis, and its frequency was significantly higher (P=.007) than in normal Italian control subjects (group B, 4 of 106). A significant increase in frequency (10 of 70, P=.025) was also observed in patients with a known thrombophilic defect. Among them, 7 were carriers of the FV Leiden mutation, 2 of protein C deficiency, and 1 of antithrombin III deficiency.
In 68 relatives of these patients who were asymptomatic carriers of thrombophilic defects, the frequency of the GA genotype (2 of 68) did not differ significantly from that in control group B (4 of 106). The two asymptomatic 20210 GA heterozygous subjects carried FV Leiden and protein C deficiency, respectively. The median age of the asymptomatic carriers of thrombophilic defects (47.5 years) was comparable with that of symptomatic patients with venous thrombosis (45.5 years).
The frequency of carriers of the 20210 A allele in patients with cerebrovascular (2 of 105) and coronary artery (3 of 90) diseases was similar to that observed among matched healthy control subjects (group A) (odds ratio 0.77; 95% confidence interval 0.14 to 4.3; NS).
We extended our study to other populations with different geographical and ethnic background and particularly to the Greek community of Cyprus, where the FV Leiden mutation is particularly frequent. An overlapping frequency (0.018) of the A allele was found in Cyprus and Italy, whereas no carrier was detected in 40 subjects of Indian or Somali origin, which indicated that in these populations the A allele, if present, should be rare.
Homozygotes (20210 AA) were absent in all groups of patients and control subjects examined.
Clinical Features of Heterozygous (20210 GA) Patients With Venous Thrombosis
Of the heterozygous patients, 19 had experienced deep vein thrombosis and 2 of them also had pulmonary embolism. One patient suffered from repeated episodes of superficial thrombophlebitis. The following risk/trigger factors were found to be associated with the occurrence of the first thrombotic event: use of oral contraceptives in 7 of 9 thrombotic women, trauma and/or immobilization caused by plaster cast in 4 patients, surgery in 2, and puerperium in 1. No risk/trigger factor could be detected in the remaining 6 patients.
The median age at the first thrombotic event as well as the percentage of patients who experienced more than one thrombotic episode clearly differed in patients carrying the single 20210 GA and in those who were double heterozygotes (Table 2⇓). Among the latter, both patients who suffered pulmonary embolism were detected.
Prothrombin Activity Levels
Prothrombin levels were studied in patients with venous thrombosis characterized for the 20210 G/A genotype. Individuals with the heterozygous genotype had a higher mean prothrombin level (1.23±0.14 U/mL, P=.0002) than those with the 20210 GG homozygous genotype (1.08±0.15 U/mL). Table 3⇓ shows the distribution of the 20210 genotypes over four different categories of prothrombin activity.
For the detection of the prothrombin gene mutation, we devised experimental protocols potentially able to screen a large number of patients (ASA-PCR). Because this method is fast and cheap and, in addition, the experimental design also allows to confirm its fidelity, it was preferred in the following studies of patients and control subjects.
The association of the GA genotype with venous thrombosis was supported by the significantly increased frequency of this genotype in patients compared with normal control subjects. In these patients the 20210 A allele was found to be associated with higher prothrombin levels, thus confirming in the Italian population the contribution of this genetic component in determining protein levels in plasma. Although 65% of patients with the 20210 GA genotype were in the highest category of prothrombin activity (Table 3⇑), these assays do not allow an efficient preselection of patients for the DNA analysis.
A significant increase in the frequency of the GA genotype was also observed in patients doubly heterozygous for another known thrombophilic defect and not in their asymptomatic relatives, carriers of thrombophilic defects, who showed a genotype frequency similar to that found in normal control subjects. In Italian patients, the combinations of the 20210 A allele with other inherited thrombophilias seem to reflect their prevalence in the population, the doubly heterozygous condition with FV Leiden being the most represented.
The doubly heterozygous patients experienced the first thrombosis at a lower median age (22 years) than patients simply heterozygous for 20210 GA (30.5 years) or for other thrombophilic defects (28.5 years). Moreover, repeated thrombotic events were observed in the majority of double heterozygotes. A synergic interaction between increased prothrombin levels and delayed inactivation of FV Leiden,30 substrate, and cofactor in the prothrombinase complex, respectively, is clearly suggested by the finding of seven patients doubly heterozygous for FV Leiden and prothrombin 20210 A, who experienced thrombosis at a very young median age.
Taken together, these data suggest that the prothrombin gene component may interact with other inherited conditions, thus increasing the thrombotic risk, as previously observed for other combined defects.2–5
The crucial position of thrombin in the hemostatic system and its multiple functions make increased levels of this protein, associated with the 20210 A allele,20 a potential risk factor in arterial disease. However, the prevalence data for the 20210 A allele among patients with cerebrovascular or coronary artery disease, who included several patients with stroke or myocardial infarction, suggested that the variation in the prothrombin gene and its associated change in the hemostatic function should contribute less to the arterial disease than to venous thrombosis. More extended studies in patients selected by homogeneous criteria will be required to determine the role, if any, of this gene variation in arterial occlusion.
Because previous studies25,31 have shown in the Greek community of Cyprus a high frequency of the FV Leiden, the other frequent prothrombotic mutation, we investigated the frequency of the prothrombin gene alleles in Cypriots. A parallel high frequency of the the 20210A allele in this population would strengthen the hypothesis that positive selection of carriers might have played a role in maintaining these thrombophilic mutations in the population. However, the frequency of the 20210 A allele (0.018), similar to that of the Italian population and slightly higher than that reported for the Dutch population,20 does not support this hypothesis and suggests the presence of a distribution pattern different from that of the Leiden mutation.25
Selected Abbreviations and Acronyms
|DGGE||=||denaturing gradient gel electrophoresis|
|PCR||=||polymerase chain reaction|
This work was supported by Telethon Italy, Ministero Universita Ricerca Scientifica e Tecnologica, and Ricerche Biomediche 1996 (to F. Mascoli).
Koeleman B, Reitsma PH, Allaart CF, Bertina RM. Activated protein C resistance as an additional risk factor for thrombosis in protein C deficient families. Blood.. 1994;84:1031–1035.
Zoller B, Berntsdotter A, Garcia de Frutos PG, Dahlback B. Resistance to activated protein C as an additional genetic risk factor in hereditary deficiency of protein S. Blood. 1995;85:3518–3523.
Green F, Humphries S. Genetic determinant of arterial thrombosis. Baillière’s ClinHaematol. 1994;7:675–692.
Anzola GP, Magoni M, Ascari E, Maffi V. Early prognostic factors in ischemic stroke: the role of protein C and protein S. Stroke.. 1993;24:1496–1500.
Koller H, Stoll G, Sitzer M, Burk M, Schottler B, Freund HJ. Deficiency of both protein C and protein S in a family with ischemic strokes in young adults. Neurology. 1994;44:1238–1240.
Catto AJ, Carter AM, Ireland H. Factor V Leiden gene mutation and thrombin generation in relation to the development of acute stroke. Arterioscler Thromb Vasc Biol.. 1995;15:783–785.
Bertina RM, van Tilburg NH, de Fuw NJ, Haverkate F. Thrombin, a link between coagulation activation and fibrinolysis. Ann NY Acad Sci.1992;667:239–248.
Jackson CM. Physiology and biochemistry of prothrombin. In: Bloom AL, Forbes CD, Thomas DP, Tuddenham EGD, eds. Haemostasis and Thrombosis. Edinburg, Scotland: Churchill Livingstone; 1994:397.
Dang QD, Vindigni A, di Cera E. An allosteric switch controls the procoagulant and anticoagulant activities of thrombin. Proc Natl Acad Sci U S A.. 1995;92:5977–5981.
Walz D, Fenton JW II, Shuman M. Bioregulatory functions of thrombin. Ann NY Acad Sci. 1986;485:1–449.
Bar-Shavit R, Benezra M, Sabbah V, Dejana E, Vlodavsky I, Wilner GD. Functional domains in thrombin outside the catalytic site. In: Berliner RG, ed: Thrombin: Structure and Function. New York, NY: Plenum Press; 1992:315–350.
Poort SR, Rosendaal FR, Reitsma PH, Bertina RM. A common genetic variation in the 3′-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood.. 1996;88:3698–3703.
Spek CA, Koster T, Rosendaal FR, Bertina RM, Reitsma PH. Genotypic variation in the promoter region of the protein C gene is associated with plasma protein C levels and thrombotic risk. Arterioscler Thromb Vasc Biol.. 1995;15:214–218.
Bernardi F, Marchetti G, Pinotti M, Arcieri P, Baroncini C, Papaccini M, Zepponi E, Ursicino N, Chiarotti F, Mariani G. FVII gene polymorphisms contribute about one third of the FVII level variation in plasma. Arterioscler Thromb. Vasc Biol.. 1996;16:72–76.
Woolf B. On estimating the relation between blood group and disease. Am J Hum Genet.. 1955;19:251.
Kalafatis M, Haley PE, Lu D, Bertina RM, Long GL, Mann KG. Proteolytic events that regulate factor V activity in whole plasma from normal and activated protein C (APC)-resistant individuals during clotting: an insight into the APC-resistance assay. Blood.. 1996;87:4695–4707.