New Prothrombin Mutation (Arg596Trp, Prothrombin Padua 2) Associated With Venous ThromboembolismSignificance
Objective—Two different prothrombin variants, p.Arg596Leu and p.Arg596Gln, conferring antithrombin resistance to patients with venous thromboembolism have been recently reported. Here, we describe a novel substitution affecting Arg596 of prothrombin molecule (Arginine596 to Tryptophan or p.Arg596Trp or Arg221aTrp in the chymotrypsinogen numbering system or prothrombin Padua 2) in 2 Italian families with venous thromboembolism.
Approach and Results—Prothrombin Padua 2 has been characterized either in plasma of carriers or using Arg596Trp recombinant prothrombin. Routine coagulation tests, thrombin generation, and antithrombin resistance tests were performed, as well as measurement of the levels of thrombin–antithrombin complexes. All carriers were heterozygotes and presented with a mild reduction of the prothrombin activity. Thrombin generation in carriers showed only a markedly prolonged decay. This finding was confirmed in plasma reconstituted with Arg596Trp recombinant prothrombin mimicking a homozygous condition, which showed longer decay and higher endogenous thrombin potential in thrombin generation than wild-type recombinant prothrombin reconstituted plasma. Patient’s plasma as well as Arg596Trp recombinant prothrombin showed a clear thrombin resistance to antithrombin inactivation. These findings were supported by the assessment of thrombin–antithrombin complexes formation, which was strongly reduced for Arg596Trp recombinant prothrombin as compared with wild-type recombinant prothrombin. In a series of 400 unrelated consecutive patients with venous thromboembolism, 2 carriers of prothrombin Padua 2 were found (estimated prevalence of 0.5%).
Conclusions—Our study showed that prothrombin Padua 2 induces antithrombin resistance and is associated with an increased risk of venous thromboembolism. Codon 596 (CGG) of prothrombin is a hot spot for mutations, which constitute a new and relatively frequent cause of inherited thrombophilia.
Numerous gene mutations in various molecules have been found in members of families with inherited thrombophilia, but many mutations remain unidentified.1 Prothrombin, the precursor of thrombin, is a serine protease that plays a key role in hemostasis and thrombosis. The prothrombin G20210A is a risk factor for venous thromboembolism (VTE).2 Recently, the prothrombin Yukuhashi mutation (c.1787G>T, p.Arg596Leu) was identified in a Japanese family.3 The mutant recombinant prothrombin showed moderately lower clotting activity than the wild type but a substantial resistance to the inhibitory effect of antithrombin, which resulted in an increased risk of thrombosis in affected family members. Another mutation was found in prothrombin Belgrade, prothrombin Amrita, and, recently, in a Japanese family, characterized by an Arginine to Glutamine substitution at the same position of the prothrombin molecule (c.1787G>A, p.Arg596Gln).4–6 Here, we studied a novel gain-of-function F2 mutation affecting the same amino acid residue, c.1786C>T (p.Arg596Trp; Arg221aTrp in the chymotrypsinogen numbering system7), named prothrombin Padua 2 by the name of the City it was discovered, inducing antithrombin resistance in 2 probands and several affected members from 2 Italian thrombophilic families.
The proband from Family 1 is a 47-year-old white man from North East of Italy who was admitted in January 2014 to our Hospital with objectively documented unprovoked occlusive right leg femoral–popliteal deep vein thrombosis (DVT) and pulmonary embolism. He had experienced the first episode of superficial thrombophlebitis to the right saphena magna vein at the age of 38 and subsequent recurrences. Thrombophlebitis episodes were successfully treated with low–molecular weight heparin for 30 to 45 days. Screening for thrombophilia (commonly known defects) and occult cancer were negative at that time. In 2012, at the age of 44, he developed provoked (knee trauma and immobilization) left distal DVT and superficial thrombophlebitis that were treated with therapeutic doses of low–molecular weight heparin for 3 months. Extensive coagulation screening was performed, and only a mild reduction of prothrombin activity in a prothrombin time (PT)–based clotting assay was observed.
During hospitalization for the current episode of VTE, associated systemic diseases were ruled out. Therapeutic doses of low–molecular weight heparin and warfarin were administered. No recurrent thrombosis has been observed over 19-months follow-up period under warfarin treatment.
Several family members experienced VTE: his mother developed recurrent DVT of both legs during chemotherapy for non-Hodgkin lymphoma at 60 year of age; his uncle on the maternal side developed recurrent DVT after the age of 45 and was anticoagulated until the age of 78 when he died of soft tissue sarcoma. The proband’s cousin is a 55-year-old woman who developed at the age of 37 massive pulmonary embolism after 1 month administration of a third-generation contraceptive pill (ethinyl estradiol plus gestoden). She received 1-year warfarin and had no recurrences. The patient had 2 pregnancies at 27 and 31 years, respectively, without complications.
Coagulation screening of the proband, performed before starting warfarin, confirmed a mild reduction of prothrombin activity (54.0%) in a PT-based functional assay.
The proband from Family 2 is a 29-year-old white woman from North East of Italy admitted to our Hospital in December 2012 with tachycardia (120/min) and sudden dyspnea while she was under estroprogestinic therapy for endometriosis (etonogestrel/ethinyl estradiol vaginal ring, delivers 150 μg/15 μg per day, 1 year treatment). Chest angio–computed tomographic scan showed numerous endoluminal filling defects of upper and lower lobes of both lungs, consistent with a diagnosis of massive pulmonary embolism. Compression ultrasonography showed femoral-popliteal DVT of the right leg. Interestingly enough, 2 years earlier, she had been investigated for common thrombophilic conditions before estroprogestinics administration, and no known defect had been detected at that time. She was initially treated with standard heparin followed by low–molecular weight heparin and warfarin (international normalized ratio 2.0–3.0). She stopped warfarin in January 2014, and no recurrent thrombosis has occurred until now. No family members experienced thrombotic episodes. We are aware of the presence of severe Hemophilia A in several members of the family on the maternal side (where the prothrombin defect comes from). Unfortunately, this part of the family was not available for study. Extensive thrombophilia screening was repeated in the proband before starting warfarin, during the acute phase of VTE. A slightly reduced prothrombin activity (65.5%) in a PT-based clotting assay was detected.
Materials and Methods
Materials and Methods are available in the online-only Data Supplement.
Plasma samples from both probands and family members examined (Figure 1A and 1B) showed normal PT, activated partial thromboplastin time (Table 1), and factor VIII, IX, X, and XI activities. Antigen and activity levels of protein C and protein S were within normal limits, as well as fibrinogen levels and antithrombin and plasminogen activities. Factor V Leiden and prothrombin G20210A polymorphisms were not detected in family members. Tests for lupus anticoagulant, anticardiolipin, and anti-beta2-glycoprotein I antibodies were negative.
Prothrombin Activity and Antigen in Family Members
In Family 1, the proband (III-2), his mother (II-2), and sister (III-3) were found to have reduced PT-based prothrombin functional activity (FII:Act), whereas the maternal cousin (III-7) exhibited a PT-based FII:Act slightly below the normal range (Table 1). Prothrombin antigen (FII:Ag) level in the proband was normal, whereas in the mother, sister, and maternal cousin, FII:Ag levels were slightly reduced (Table 1). Three out of 4 affected family members presented with a mild reduction of ecarin clotting time (ECT)–based FII:Act (Table 1). Mean values of FII:Ag and PT-based and ECT-based FII:Act in affected family members were 70.5%, 61.3%, and 70.7% of normal, respectively. The proband’s daughter (IV-1, nonaffected) presented with FII:Ag and FII:Act levels within the normal values. The proband’s mother and maternal cousin have experienced DVT and pulmonary embolism.
In Family 2, the proband (III-1), her sister (III-2), and mother (II-2) showed slightly reduced or borderline PT-based FII:Act with normal FII:Ag levels (Table 1). Two out of 3 affected family members presented with a mild reduction of ECT-based FII:Act (Table 1). Mean values of FII:Ag and PT-based and ECT-based FII:Act in affected family members were 106.3%, 70.8%, and 69.0% of normal, respectively. The proband’s father (II-1, nonaffected) had normal FII:Ag and FII:Act levels.
F2 Gene Analysis
Data from direct DNA sequencing revealed that both probands were heterozygous for a novel missense mutation in exon 14 of F2, c.1786C>T, resulting in the substitution of an arginine with tryptophan at residue 596 (p.Arg596Trp or Arg221aTrp in the chymotrypsinogen numbering system) in the prothrombin molecule (Figure 2A).
Sequencing of exon 14 was also performed on all available relatives and demonstrated that in Family 1, the proband’s mother (II-2), sister (III-3), and maternal cousin (III-7) were heterozygous for the p.Arg596Trp mutation. In the proband’s daughter, the mutation was absent.
In Family 2, the same mutation was present in heterozygous form in the proband’s mother (II-2) and sister (III-2). The proband’s father (II-1) had no mutation.
Sequencing of exon 14 in 100 healthy subjects and 100 patients with documented DVT failed to detect the c.1786C>T mutation. In a different series of 400 unrelated consecutive patients with provoked and unprovoked VTE, we found 2 carriers of prothrombin Padua 2 (prevalence of 0.5%).
Restriction Enzyme Analysis
Digestion of a normal PCR-amplified exon 14 with HpaII resulted in DNA fragments of 280 and 187 bp. Because the p.Arg596Trp mutation abolished the HpaII restriction site in exon 14, heterozygous patients showed a 467 bp band together with the 280 and 187 bp bands (Figure 2B).
Recombinant Prothrombin Padua 2 Expression
SDS-PAGE gel and silver staining of purified recombinant prothrombins showed a major band with an apparent molecular weight of 72 kDa, which had a similar mobility to that of normal prothrombin from human plasma (data not shown).
Prothrombin activities of wild-type recombinant prothrombin (WT rFII) and Arg596Trp recombinant prothrombin (Arg596Trp rFII), assessed in reconstituted plasma (final concentration 90 μg/mL; see Materials and Methods in the online-only Data Supplement) using both PT-based and ECT-based assays, are reported in Table 2. Both methods showed an activity of Arg596Trp rFII ≈8× lower than WT rFII (specific activity of Arg596Trp rFII as compared with WT rFII, 12.0% and 12.6% with the PT-based and the ECT-based activity assay, respectively).
Thrombin Generation Assay
Thrombin generation (TG) curves obtained from plasma samples with 5 pmol/L tissue factor (TF) concentration showed a significant prolongation of the decay in all subjects affected by the mutation as compared with healthy family members and pooled normal plasma. Healthy subjects in Family 1 (IV-1) and 2 (II-1) presented a decay of 18.3 and 23.7 minutes, respectively. These values were similar to that detected for pooled normal plasma (18.7 minutes). Members of the 2 families in which the mutation has been identified showed values of decay from 33.0 to 40.0 minutes. No significant difference was found for the other TG parameters (Figure 3A and 3B).
TG curves obtained from plasma samples with 1 pmol/L TF concentration showed a prolongation of the decay as seen for 5 pmol/L TF concentration. In addition, at 1 pmol/L TF, 6 out of 7 heterozygous carriers presented with a lower peak height, and all carriers showed prolonged lag time as compared with pooled normal plasma (data not shown). Finally, no significant increase of endogenous thrombin potential (ETP) values was seen in heterozygous carriers of the mutation as compared with pooled normal plasma at 1 pmol/L TF (data not shown).
TG was measured in WT rFII or Arg596Trp rFII reconstituted plasma by triggering coagulation with 5 pmol/L TF. Arg596Trp rFII-reconstituted plasma showed reduced peak, prolonged lag time and decay, and increased ETP in comparison to WT rFII reconstituted plasma. The heterozygous condition, obtained by reconstituting prothrombin-deficient plasma with 50% WT rFII and 50% Arg596Trp rFII, exhibited values of lag time, ETP, peak, and decay that were intermediate between those obtained for the homozygous WT rFII and the homozygous Arg596Trp rFII (Figure 4).
Antithrombin Resistance Analysis
Test for detection of antithrombin resistance was performed on plasma samples from all prothrombin Padua 2 mutation carriers, including those asymptomatic, and unaffected relatives of both families, in the absence and presence of heparin. As regards Family 1, after 30 minutes of inactivation using antithrombin without heparin, the only unaffected available family member (the proband’s daughter, IV-1) showed a relative residual thrombin activity (RRTA) of 29.0±0.1%, whereas all mutation carriers investigated had values from 34.7±0.1% to 42.7±1.1% (III-2, II-2, III-3 and III-7) (Figure 5A).
In Family 2, the proband’s father (II-1), a noncarrier for the mutation, showed an RRTA of 23.8±0.2%, whereas the proband (III-1) and her affected relatives studied (II-2 and III-2) had values from 36.1±0.1% to 45.1±0.9% (Figure 5B). The analysis performed in the absence of heparin did not reveal a significant difference in terms of RRTA between subjects with and without prothrombin Padua 2 mutation. On the contrary, after 1 minute of inactivation using antithrombin with heparin, subjects with prothrombin Padua 2 mutation of both families showed values of RRTA significantly higher (≈90%) than those obtained from subjects without the mutation (≈50%). This difference further increased after 5 minutes incubation of plasma samples with antithrombin and heparin (Figure 5C and 5D).
Data reported in Figure 6A indicate that homozygous Arg596Trp rFII was not inhibited even after 30 minutes incubation with a large excess of antithrombin (100 mU/mL), whereas WT rFII showed an RRTA of 17.4±0.1%. Heterozygous Arg596Trp rFII showed higher RRTA (73.0±0.4%) as compared with WT rFII but lower than homozygous Arg596Trp rFII (94.9±1.5%).
Test for detection of antithrombin resistance was performed also in the presence of 5 U/mL of heparin. As shown in Figure 6B, after 2 minutes inactivation using antithrombin plus heparin, RRTA of WT rFII, heterozygous Arg596Trp rFII, and homozygous Arg596Trp rFII were 18.6±0.3%, 44.0±0.4%, and 86.5±0.8%, respectively.
Formation of Thrombin–Antithrombin Complexes
Thrombin–antithrombin complexes levels, measured in plasma of subjects heterozygous for the mutation, were within the normal range (data not shown).
In the absence of heparin, thrombin–antithrombin complexes concentration between WT rFII and antithrombin increased over time and reached a plateau after 30 minutes of incubation, whereas in the same incubation period, Arg596Trp rFII showed a slight increase of binding with antithrombin. The presence of heparin accelerated the formation of complexes with antithrombin for both recombinant proteins, but the concentration of complexes, measured for 5 minutes, were much lower for Arg596Trp rFII (Figure 7A and 7B).
Cleavage of Fibrinogen by Recombinant (Pro)thrombins With a Clotting Method
Clotting times for WT rFII, heterozygous, and homozygous Arg596Trp rFII were 18.6±0.5, 34.2±1.1, and 210.2±56.5 sec, respectively. According to these data, in this isolated model, Arg596Trp rFII seems to poorly cleave fibrinogen.
A new missense mutation (c.1786C>T) at the last exon of F2, which results in an Arginine to Tryptophan substitution at residue 596 (p.Arg596Trp or Arg221aTrp in the chymotrypsinogen numbering system or prothrombin Padua 2), has been found in 2 unrelated Italian families. We assumed that prothrombin Padua 2 behaved similarly to the 2 prothrombin variants with amino acid substitutions at the same position recently described.3–6 Here, we characterized prothrombin Padua 2 either in patients’ plasma or after expression of the recombinant molecule.
PT and activated partial thromboplastin time were normal in all family members. In addition, not all affected members investigated showed reduced FII:Act using a PT-based clotting assay, some of them presenting with activity values at the lower limit of the normal range. This fact has clinical implication because the presence of prothrombin Padua 2 (and possibly other similar prothrombin variants) cannot be suspected on the basis of a prolongation of global tests routinely performed. However, in the presence of normal PT and activated partial thromboplastin time, isolated reduction of functional prothrombin activity in a patient with thrombotic manifestations may prompt further investigation to exclude these prothrombin mutants. As for other global tests such as standard TG in plasma with 5 pmol/L TF, it has to be noted that all parameters were normal in carriers, except for decay values. All carriers of both families exhibited a decay markedly prolonged as compared with normal subjects, indicating that antithrombin present in plasma samples is unable to completely inhibit (abnormal) thrombin. Interestingly enough, TG performed in reconstituted plasma with Arg596Trp rFII, mimicking a homozygous condition, showed a reduced thrombin peak and a slightly prolonged lag time, consistent with slightly reduced procoagulant activity and a prolonged decay and an increased ETP, consistent with increased TG and lack of thrombin inhibition by antithrombin. Naturally occurring homozygous carriers of these prothrombin variants have never been reported to date, and it is unknown if this condition can be compatible with life. As the matter of fact, the final clinical effect is unpredictable being hypercoagulable on the basis of the markedly increased ETP values and hypocoagulable because of the impairment of fibrinogen cleavage (in vitro models of homozygotes).
The resistance to thrombin inactivation by antithrombin is the main effect of this novel F2 mutation, which is in agreement with some of the findings of the TG. In the absence of heparin, RRTA measured in plasma of probands and their affected relatives after addition of antithrombin were only slightly higher than in members without the mutation. In contrast, in the presence of heparin, all carriers of the mutation showed a clear antithrombin resistance, particularly during the first 5 minutes of the inactivation reaction.
These findings are reinforced by the study of thrombin–antithrombin complexe formation in the absence and presence of heparin, which was severely compromised for Arg596Trp rFII as compared with WT rFII.
Collectively, these experiments suggest that prothrombin Padua 2 can cause antithrombin resistance as previously shown for other similar mutants.3–6 Some authors3,8 have emphasized the importance of Arg596 (Arg221a), located in the sodium-binding site of thrombin, in stabilizing the interaction between thrombin and antithrombin. In particular, the side chain of Arg596 (Arg221a) in thrombin forms 2 hydrogen bonds with the side chain of Asn265 in antithrombin. The exact mechanisms by which these new prothrombin variants, including prothrombin Padua 2, cause thrombotic manifestations are not fully elucidated. It is peculiar that in the presence of a reduction of the procoagulant function (reduced FII activity in plasma and impaired fibrinogen cleavage) as found in carriers of mutations at Arg596 (Arg221a), clinical manifestations are exclusively prothrombotic at least in the heterozygotes. These clinical findings clearly emphasize the importance of antithrombin as a natural inhibitor of thrombin. Thus, it is not surprising that these inherited defects of (pro)thrombin which result in the resistance to inhibition by antithrombin present with similar clinical manifestations reported in several inherited heterozygous antithrombin defects.
How frequent are these new F2 mutations and particularly prothrombin Padua 2? With a quite selective and arbitrary screening strategy, based only on the personal history of provoked or unprovoked VTE and the presence of an isolated mild reduction of the FII:Act with normal PT followed by gene sequencing, we identified 2 carriers of prothrombin Padua 2 out of 400 VTE patients with an approximated prevalence of 5 carriers in 1000 patients (0.5%). Whether this is an over or underestimation because of some selection or referral bias needs to be confirmed by studies in different cohorts and in different populations, and our estimate must be considered preliminary.
Interestingly enough, in a genome-wide linkage scan in thrombophilic families, performed to identify novel susceptibility regions for VTE (GENUT study), ten Kate et al reported a missense mutation in F2 characterized by the replacement of an Arginine by Tryptophan (the authors did not indicate the amino acid position).9 Whether this mutation corresponds to prothrombin Padua 2 remains to be defined.
In conclusion, substitutions at Arg596 of the prothrombin molecule, such as prothrombin Padua 2, constitute a new and relatively frequent cause of inherited thrombophilia.
We are indebted to Mariangela Fadin, Patrizia Zerbinati, Graziella Saggiorato, and Francesca Sartorello for their technical assistance.
Sources of Funding
This study was supported by Grant no 60A07-2893/13 to P. Simioni and by Grant no BULA_PRGR_P14_01 (Giovani Studiosi, Bando 2013, Contributo alla Ricerca) to C. Bulato from the Italian Ministry of University and Research (MIUR).
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.115.306914/-/DC1.
- Nonstandard Abbreviations and Acronyms
- Arg596Trp rFII
- Arg596Trp recombinant prothrombin
- deep vein thrombosis
- ecarin clotting time
- endogenous thrombin potential
- prothrombin functional activity
- prothrombin antigen
- protein C
- prothrombin time
- relative residual thrombin activity
- thrombin generation
- venous thromboembolism
- WT rFII
- wild-type recombinant prothrombin
- Received November 22, 2015.
- Accepted March 14, 2016.
- © 2016 American Heart Association, Inc.
- Poort SR,
- Rosendaal FR,
- Reitsma PH,
- Bertina RM
- Miyawaki Y,
- Suzuki A,
- Fujita J,
- Maki A,
- Okuyama E,
- Murata M,
- Takagi A,
- Murate T,
- Kunishima S,
- Sakai M,
- Okamoto K,
- Matsushita T,
- Naoe T,
- Saito H,
- Kojima T
- Djordjevic V,
- Kovac M,
- Miljic P,
- Murata M,
- Takagi A,
- Pruner I,
- Francuski D,
- Kojima T,
- Radojkovic D
- Sivasundar S,
- Oommen AT,
- Prakash O,
- Baskaran S,
- Biswas R,
- Nair S,
- Mohan CG,
- Biswas L
- Bode W,
- Turk D,
- Karshikov A
- ten Kate MK,
- van Schouwenburg IM,
- van der Steege G,
- Mulder R,
- Wijmenga C,
- Mulder AB,
- Kluin-Nelemans HC,
- Johnson DJD,
- Huntington JA,
- Ott J,
- van der Meer J
We studied a novel substitution affecting Arg596 of prothrombin (p.Arg596Trp or prothrombin Padua 2) in 2 Italian families. This new prothrombin variant resulted in antithrombin resistance and increased risk of venous thromboembolism in carriers. In a series of 400 unrelated consecutive patients with venous thromboembolism, we found 2 carriers of prothrombin Padua 2 (estimated prevalence of 0.5%). Interestingly, some authors reported in a genome-wide linkage scan in thrombophilic families, performed to identify novel susceptibility regions for venous thromboembolism (GENUT study), a missense mutation in F2 characterized by the replacement of an Arginine by Tryptophan (the amino acid position was not indicated). Whether this mutation corresponds to prothrombin Padua 2 remains to be defined. Substitutions at Arg596 of the prothrombin molecule, such as prothrombin Padua 2, constitute a new and relatively frequent cause of inherited thrombophilia.