Polymorphisms in the prothrombin gene and their association with plasma prothrombin levels

*" Thromb Haeraost 2001; 85: 1066-70 © 2001 Schattauer GmbH, Stuttgart

Pofymorphisms in the Prothrombin Gene and Association

with Plasma Prothrombsn Levefs

H. Ceelie

1

, R. M. Bertina

1

, A. van Hylckama Vlieg

1

·

2

, F. R. Rosendaal

1

-

2

, H. L Vos

1

'Haemostasis and Thrombosis Research Centre, Dept. of Haematology, 2Dept. of Clinical Epidemiology, Leiden Universify Medical Centre, Leiden, The Netherlands

Key words

Prothrombin, G20210A, A1991 IG, venous thrombosis

Summary

To find genetic causes of high plasma prothrombin levels, an est-ablished prothrombotic risk factor, we searched for sequence variations in the prothrombin gene. We selected subjects with the 20210-GG ge-notype (since the 20210-A allele is already known to be associated with high levels) and elevated prothrombin levels (> 130 U/dl) from the Leiden Thrombophilia Study (LETS). No mutations were found in the l kb promoter region of the prothrombin gene in seven individuals with an isolated high prothrombin level. Comparison of the allelic frequen-cies of four different polymorphisms in the prothrombin gene in healthy volunteers and in the control subjects among the selected LETS indi-viduals indicated a higher frequency of the 19911-G allele in the latter group (allele frequency 52 vs. 78%, respectively). Homozygous car-riers of the 19911-G allele had 8 U/dl higher plasma prothrombin levels than 19911-AA carriers. This difference in prothrombin levels did not affect the tkombotic risk in 20210-GG carriers. In heterozygous 20210-A carriers the odds ratio increased from 1.6 (95% CI: 0.6-4.3) in subjects with 19911-Ato 4.7 (1.6-14.0) in subjects with 199lf-Gonthe other prothrombin allele.

of the 202ΙΟ-Λ allele in tlic general populalion is aboul 2.0fr ulthough it seems to bo more coiiinion in southern Europe (15;.

Poortei al. also shovved lliai ihe presencc of ;he 20210-A allele was associated \\iih elevaied pla.sma prolhrombin levels (4). This observati-on has been cobservati-onl'irmed in scveral sitidies. both in svmplomaiic (16-20) and in asymplomaiic carricrs of ihis allele (19). In a l'ainiK basod Mudy it was shovui b\ linkaue analysis ihal ihe (!202IOA mulalion was associated \\ nli an cf fect on bolh plasma prolhrombin levels and on the risk of venous thrombosis 1211.

An elev aied prothrombin le\ cl ilself is also a risk factor for D VT (4). Subjects in ilie hiehesi quaitilc of plasma prolhrombin levels had a more than iwol'old increased risk of DV'T compared 1o thosc in the lowest quarlile. lnleresimi!h. only aboul 14Vr of them carried the ·. 20210-A allele (4). indicalin« the prosence ol'additional causes of high..· prothrombin levels. possibK senclic in oripin. This idea is furlher sup-·. portedby a slndy ofSirnioni et al. i 19), in which il was shown mateven.;! among carriers öl the 20210-G (\\ild-l\pe) allele. many more palients ·

with previous vcnous ihroinboembolism (28'/() than non-symptomatic,1.;

individuals ιΤ',ί) had prolhrombin antigen levels abovc the cutn

valueof UM'/dl.

In the prescnt\tudv vvc looked for sequence variations in thrombin gene associalcd \\iih elevaled plasma prothrombin levelsl subjects w i l h Ihe 2()2ΙΟ-(ϊ(ϊ genolvpe. Whenever such an associalion was found. vve aNo inve.sliuaied the relation u ilh ihrombotic risk.

\f'i ·

Introduction

Prothrombin is a key component in the blood coagulation pathway. It is the precursor of the serine protease thrombin, which exhibits pro-coagulant, anticoagulant, äs well äs anti-fibrinolytic a^tivities [for a

review, see Mann [1]). The prothrombin gene is located on chromo-some 11 at position 11p 11-q 12 (2). It has a length of 21 kb and includes

14 exons and non-coding regulatory sequences (3).

Recently, a G to A transition at nucleotide position 20210 in the 3' untranslated region of the prothrombin gene was described, which is associated with a threefold-increased risk of deep vein thrombosis (DVT) (4). The importance of the 20210-A allele äs a risk factor for venous thrombosis was confirmed in several independent studies (5-12). In these studies, the 20210-A allele was found in 4-8% of patients with a history of venous thrombosis. The relative risk for venous throm-bosis in heterozygous carriers ranged from 2 to 7, Two studies reported only a weak effect of the 20210-A allele (13,14). The overall prevalence

Correspondence to: Dr H. Ceelie, Haemostasis and Thrombosis Research Centre, Dept. of Haematology, Leiden University Medical Centre, D2-19, PO Box 9600, 2300 RC, Leiden, The Netherlands - Phone: +31 71 5262076, Fax: +31 71 5266755, E-mail: hceelie@LUMC.nl

Subjects, Materials and Methods

Subjects

To estiniiite iho Ircqucnc) of piothroinhin gene poljmorphisms in the nor-mal populalion. M he<)lih\ voluniccrN not usin» oral comruccptives, wert recruitedfroiii hospual [vrsomicl. Carriersof thc 20210 A allele wcrccxcludcd (n = 3). From iw -.iibieci 110 pliwna sample \^as u\ailahlc.

A second üioup of suhiects came Irom a populalion-bascd cas*· control stud>. ilie Leiden l'romhophilia Sludy (LLTS'i. which has been descri-bed in detail belore ι22ι. liiiellv, consccuiivc patienls v\ith a tlrsl episodq.of .

DVT were ".clecie.l Irom ihrce aniieoaiiulation clinics in thc Nclherlands. AU patients weiv \oiiiu'i:r lhan 70 vcars öl agc and had not lieen diagnoscd wilh

cancer. Thi· paiienK were askcd 10 lind iheir o\\n hcalüiy conlrol subjcct; if, they were unahlc 10 do so. partners öl oihcr patienls scrvcd äs controls. Controls were matched Ιοί ace ι ·/ liv e vearsi and sc\ wiih Ihe cascs. The sludy

inclu-ded 474 paik-nis and 474 «.-oniiOls. l'roin thc>c 948 individuals, wc sclccicdall , subjects wnh ihe .'0.110 (i(i senoiyr*· and prothrombin levels 2130 UAB ; (28 patient« and i!i omnolM. Onlv 7 -uhiecis |2 paiieiiisand 5 controls)hadan isolated eleviiied piotluoinhin leiel. i ι·. all. or all bin onc, of'lheothercoagulfr ( tion factors nicasun-d il:V. Ι-ΛII. l VIII. RX. K-X. I-XI, f;XU. FXin,PWffllS, ; ProteinC, Τ.\1·Ί. :ind aniilhnmihim »cic <130'<.

Since pnuhioiiihiii IcvcN inercasc Ihe risk of ihromhosis, association hctuivn polvmorphiims and plasma prothrombin levels restricted to ilie ainirol subjccis of ihe l.l-.'l S. cxcludin«; 20210-A cameß*·;

Ceehe et al.: Polymoφhisms m the Prothrombin üene and Elevated Plasma Levels

individuals using coumarins at the time of venepuncture. DNA was unavailable for two individuals, and 460 samples were studied.

The effect of polymorphisms in the prothrombin gene on the risk of throm-bosis was studied by coraparing the prevalence in patients and controls, m some of the analyses limited to those without the 2021OA allele.

Finally, we performed an analysis restricted to 20210-A carriers (29 pati-ents, 11 controls) to assess the additional effect of the 19911-G alleie on the risk of thrombosis and plasma prothrombin levels (excluding two patients using coumarins at the time of venepuncture for the latter analysis).

A third group of 22 homozygous carriers of the 20210-A allele was used to analyse the haplotype of the prothrombin 20210-A allele. Subjects came from Australia (n = 1), Austria (n = 2), France (n = 5), Germany (n = 3), Italy (n = l), the Netherlands (n = 5), Spain (n = 4) and from the USA (n = 1).

\

\ Blood Collection and Laboratory Analysis

\

Blood was collected into tubes containing 106 mmoW trisodium citrate. Plasma was prepared by centrifugation for 10 min at 2,000 g at room tempera-ture and stored at -70° C. Genomic DNA was isolated from leukocytes using Standard methods and stored at 4° C.

In the healthy volunteers, prothrombin activity was measured with a chromogenic method using S-2238 (Chromogenix, Milan, Italy) äs a Substrate and Ecarin (Sigma Chemical Co, St Louis, MO) äs an activator on an ACL-200 following the instructions of the manufacturers. In the participants of the , Leiden Thrombophilia Study coagulation factor levels had been measured äs described previously (23-32). Prothrombin activity had been measured with a method comparable to that described above (4).

Sequencing ofthe Promoter Region ofthe Prothrombin Gene

To identify changes in the promoter region of the prothrombin gene in the seven subjects with isolated high prothrombin levels, a 947 bp fragment (nt -921 to 26 relative to the transcription Start site, numbering according to Bancroft et al. [33]) was ^mplified by Standard PCR (see below) using the following primers: 5'-CGG AAA GCT TOT GTT CCT GCT CTT TGT CCC T-3' Ad 5'-ATG GAA GCT TGT CAG CTC CTG GGT CAC TGA G-3'. Fragments were sequenced using the same primers and an ABI PRISM BigDye Terminator Cycle Sequencing Kit (Perkin Elmer-Cetus). Two additional inter-nal primers were used: 5'-CCA TGG ACA TTC CAT TCC TAA TCT CC-3' and 5'-CTG TGT GCC TCA GTT TCC TCA TC-3'.

Genotyps Analysis

For the determmation of genotypes of the prothrombin gene, four biallelic polymorphisms were used (Table 1) Three of them, the polymorphisms m intron 4 (T3728C), intron 5 (C4125G) and intron 13 (A19911G) have been previously described (34, 35). The polymorphism in intron 11 (T9832C) was identified in this study by direct sequencing of randomly selected healthy mdi-virluals. To study the haplotype of homozygous 20210-A carriers, an additional polymorphism in exon 10 (G8845A) was used (35). All polymorphisms were detected by PCR and allele-specific restriction analysis. PCR was performed according to Standard procedures, using genomic DNA (0.1-1 μg), 200 mM of

each dNTP (Pharmacia, Uppsala, Sweden), PCR buffer (67 mM Tris-HCl (pH 8.8), 6.7 mM MgCl2,6.7 μΜ EDTA, 16.6 mM (NH4)2S04 and 10 mM

ß-mer-capto-ethanol), 100-200 ng of each pnmer, 0.1 mg/ml bovine serum albumin (Pharmacia), 10% dimethylsulfoxide and l U Amplitaq DNA polymerase (Perkin Elmer-Cetus) in a final volume of 50 μΐ. For the detection of the

poly-morphism in intron 11 the PCR buffer did not contam bovine serum albumin and dimethylsulfoxide. Thermal cyclmg was carried out by incubation at 92° C for 2 min, followed by 33 cycles of denaturation at 92° C for 30 s, annealing for 30 s and extension at 72° C for l min and a final step of 4 min at 72° C. For restriction analysis 10 ml of product was digested with 2.5 to 5 U of enzyme (New England Biolabs, Beverly, MA) according to the manufacturer's instructions.

To determine haplotypes in individuals heterozygous for both the 20210-A and 19911-G allele, allele specific PCRs for the 19911-A and 19911-G allele were performed (Table l B). The two PCRs were done äs mentioned above,

with the exceptions that 7.5% DMSO and 100 ng of each primer were used and that thermal cycling was performed m 36 cycles with an extension time of 45 s. The 20210 genotype was subsequently detected by restriction analysis ofthe al-lele-specific PCR products with Hindül, äs described previously (4).

Statistical Analysis

Odds ratios were calculated äs estimates of the relative risk of thrombosis in the Standard unmatched fashion. Ninety-five percent confidence intervals (95% CI) were constructed according to Woolf (36). Haplotype frequencies, used to calculate linkage disequilibnum coefficients, were estimated according to Ott (37). The Standard disequilibnum coefficient (38) (Δ) and Lewontm's D' (39)

were used äs measures for linkage between nt 4125 and nt 19911.

Table l A: Location of single nucleotide polymorphisms in the prothrombin gene used in this study, primer properties, and

restriction enzymes used for detection. B: Primers for the allele specific detection of the 19911 polymorphism

Polymorphism (location) A) T3728C (intron 4) C4125G (intron 5) G8845A (exon 10) T9832C (intron 11) A19911G (intron 13) B) 1991 1G 19911A G20210A primers (5'-3') TCTGCCTGGCCTAGTGGGATGCATG ACCCAGACCCTCAGCACAGTTAC GAGTGAGGGGTCGGCCTTC CCTCCTCACGGTGGGGTCT CTTAGACCTGGGATTGTTAC GAACATCCTATGTGTCCCGG CCTTCAGTGACTACATTCACCCTG CCCTTGTATCCAGCCTGGAGC AGGTAAGCTTCTCTAAAGCCCAGG CATTTGATACCAGCGGTTGTTAAAGGG GTATCTAGAAACAGTTGCCTGGCtGg GTATCTAGAAACAGTTGCCTGGCtGa ATAGCACTGGGAGCATTGAaGC Nucleotide numbering nt3474nt 3952 nt 4038 nt 4243 - nt8656- nt9006- Π19527-nt 10147nt 19822 nt 19999 nt 26460 nt 26807 -3498 3930 4056 4225 8675 8987 9550 -10127 -19845 -19973 -26485 - 26786

1 Tm Product size Restriction

•ThrombHaemost2001;85: 1066-70

,_' ta.V.B'i« i

KÄS

Results

Sequence Variations in the Prothrombin Promoter Region

We screened seven individuals with an isolated high prothrombin

level for mutations in the promoter region of the prothrombin gene by

sequencing of almost l kb upstream of the transcription Start site. This

region contains all presently known transcription regulatory elements.

Compared with the nucleotide sequence of Bancroft et al. (33), two

variations, an insertion of an A at position -22 and an insertion of a G

at position -646 were found in a homozygous state in all seven

subjects and in control DNA, suggesting that they represent errors in

the original sequence. The insertion at -646 has previously been

described by Zivelin et al. (35). No other sequence variations in the

Pro-moter region were observed.

Allele Frequencies of Prothrombin Gene Polymorphisms in Selected

Individuais and Healthy Controls

All control subjects from the LETS study with plasma prothrombin

levels > 130 U/dl and with the 20210-GG genotype (n = 18) were

geno-typed for four polymorphisms in the prothrombin gene (T3728C,

C4125G and T9832C and A19911G) äs were 61 healthy volunteers

with the 20210-GG genotype. The allele frequencies are summarised in

Table 2. For the polymorphisms at nt 3728, 4125 and 9832 no major

differences were found. However, äs to the A19911G polymorphism,

we found that the 19911-G allele was more frequent among the LETS

controls with plasma prothrombin levels >130 U/dl (allele frequency

78%) than in the group of healthy volunteers (allele frequency 52%).

Moreover, it was striking that 19911-AA carriers were absent from the

first group. These results suggested an association between the

nucleo-tide at 19911 and plasma prothrombin levels. The 19911 genotypes of

the seven subjects with isolated elevated plasma prothombin levels (see

previous paragraph) were: AG (n = 3; all controls) and

19911-GG (n = 4; 2 patients and 2 controls).

The A19911G Polymorphism and Prothrombin Level

Further support for the supposition that the 19911 polymorphism is

associated with plasma prothrombin levels came from the observation

that in the healthy volunteers mean values for plasma prothrombin

levels for 19911-AA (n = 13), AG (n = 33) and GG (n = 14) carriers

were 94.4 U/dl (95% CI, 86.9-101.8), 101.8 U/dl (97.Q-105.6) and

108.6 U/dl (102.2-113.8), respectively. Despite the relatively small

number of subjects this indicated an allele dosage effect on

prothrom-bin levels, with the 19911-G allele associated with higher expression

levels. Subsequently, we determined the 19911 genotype in all

individ-uals from the LETS study. The C4125G polymorphism, which can be

detected using the same restriction enzyme (Mnl I), was analysed in the

same PCR. Mean values and confidence intervals for plasma

prothrom-bin levels for 19911-AA and -GG carriers were now clearly different

(Table 3). The 19911-G allele is associated with a 4 U/dl higher

pro-thrombin level than the 19911-A allele leading to an 8 U/dl difference

between AA and GG carriers.

The data in Table 3 suggest that there is a genotype-phenotype rela-~

tionship for the C4125G polymorphism äs well. However, when the

C4125G and A1991 IG polymorphisms were combined (Table 4), it

became clear that the association with prothrombin levels was

restrict-ed to the 19911 allele. For instance, 19911-GG carriers all had similar

prothrombin levels, independent of their genotype at nt 4125. The

apparent association between the polymorphic site at nt 4125 and

pro-Table 2 Allele frequencies for four polymorphisms m the prothrombin gene

in healthy volunteers

1

and controls with prothrombin levels > 130 U/dl selected

from the LETS

1

Polymorphism T3728C C4125G T9832C A19911G Healthy volunteers (n=61) C: 30% G: 40% C: 40% A: 48%

LETS controls with Prothrombin Z 130 U/dl (n=18) C: 25% G: 36% C:31% A: 22%

1 20210-A carriers were excluded in both cases

Table 3 Prothrombin gene polymorphisms C4125G and A19911G and

plasma prothrombin levels in control subjects from the LETS

1

Polymorphism C4125G CG CG GG A19911G AA AG GG n Mean prothrombin 95% CI levels(U/dl) 211 202 47 113 223 124 101.6 104.0 106.7 99.0 103.1 107.2 99.7-103.6 102.0-106.0 102.3-111.0 96.7-101.3 101.2-105.0 104.5-109.8 1 20210-A carriers were excluded

Table 4 Combined genotypes of the prolhrombin gcne and prolhrombin

levels in control subjects from the LETS

1

Genotypes (19911/4125) AA/CC AA/CG AA/GG AG/CC AG/CG AG/GG GG/CC GG/CG GG/GG Total n "<12 x-1 0 78 143 2 21 58 45 460 Mean prothrombin levels(U/dl) 98.9 114.0 104.2 102.4 111.0 107.1 107.8 106.5 103.2 95% CI 96.5-101.2 100.7-107.7 100.2-104.7 100.8-113.3 103.7-111.8 102.0-111.0 102-105 20210-A carriers were excluded

thrombin levels is theivforc ihe resull of the e.\Kling linkage

disequiü-brium between thi.s MIC and ihe polymorphism nt nt 19911. (Δ = -0.66

and D'= -0.98).

The A19911G Gene Pohmorphiun und ihe Rilk oj Deep Vein

Thrombosis

Table 5 illustrales lhai ihe prcsenre of the 199l l-G allele in

20210-GG carriers was not a^ocialed with an increased risk of deep vein

thrombosis, despite ils aisociaiion \viih higher prothrombin levels.

Neither heterozygou.·« nor liomo/jaous carrier.·* of the allele showedan

increased risk of venoiii thrombosis. Apparenily. the moderate in·

crease in prothrombin levels does nol iniluence ihe thrombotic risk.

We investigated whclher the 19911 -G allele may cause an additional,

risk of DVT in 202ΙΟ-Λ carriers. ll has pre\iously been shown Hut.'

20210-A allele is in simna linkaye dKequilibrium with 199ll-A(35),

Ceehe et al.: Polymorphisms m the Prothrombin Gene and Elevated Plasma Levels

; Table 5 The 19911-G allele of the prothrombin gene and the risk of venous

t" thrombosis

19911 genotypn ' AA AG GG Patients "97 ' 236 109 Controls Tf3~ 224 124 OR (95% Cl)

r"

1.2(0.9-1.7) 1.0(0.7-1.5) 'Reference group

20210-A carriers were excluded

t Tableo Effect of the 19911-G allele on the thrombotic risk of 20210-A

carriers in the LETS

Genptype 2021 0-GG •S 2021 0-AG 19911-AA 2021 0-AG 19911 -AG Patients 442 11 18 Controls 463 7 4 OR (95% Cl)

r

1 6 (0.6-4.3) 4.7(1.6-14.0) *Peference group

We could confirm this observation in 22 homozygous camers of the

20210-A allele, that were genotyped for five prothrombin gene

poly-morphisms. All subjects were homozygous for a prothrombin allele

with the haplotype: 3728T4125C-8845G-9832T-19911A-20210A (44

alleles).

Eighteen out of forty 20210-AG carriers from the LETS were

homo-zygous for the A allele, so these individuals carried one

19911-A/20210-A allele and one 19911-A/20210-G allele. The other

twenty-two 20210-AG carriers had the 19911-AG genotype. In these

individu-als haplotypes, were determined by using a PCR specific for the 1991

Ι-Α or -G allele, followed by restriction enzyme analysis to identify the

nucleotide at position 20210. It appeared that all 22 individuals carried

one 19911-A/20210-A and one 19911-G/20210-G allele.

Interestingly, among 19911-A/20210-A camers 62% of the patients

but oaly 36% of the controls carried the 19911-G/20210-G allele äs the

second allele (Table 6). Further analysis showed that, when using

20210-GG-carriers äs the reference group, 20210-AG subjects with the

19911-AG genotype, i.e. with a 19911-G (associated with high levels)

on the 20210-G allele, had an almost three times higher risk of DVT,

compared to those with the 19911-AA genotype (and thus with a

19911-A on the 20210-G allele): OR 4.7 (95% Cl: 1.6-14.0) and 1.6

(0.6-4.3), respectively, suggesting that the 19911-G/20210-G allele

contributes to the thrombotic risk of 19911-A/20210-A carriers.

How-ever, it should be noted that the confidence intervals of the estimates of

risk of DVT showed a large overlap. Within the group of 20210-AG

camers mean values of plasma prothrombin levels for 19911-AA

(n = 17) and 19911-AG (n = 21) carriers did not differ: 131.0 U/dl (95%

Cl 121.1-140.8) and 132.7 U/dl (125.3-139.3), respectively.

Discussion

In this study, we have searched for additional genetic causes of high

plasma prothrombin levels, an established risk factor for venous

throm-bosis (4). Our investigations have been focussed on the prothrombin

gene itself. The approach was first to look in control subjects from the

Leiden Thrombophilia Study (LETS) with the 20210-GG genotype and

with an isolated elevated plasma prothrombin level (>130 U/dl) for

sequence variations in l kb of the promoter region of the prothrombin

gene. We did not observe any relevant Variation. We conclude that the

high plasma levels of prothrombin in these subjects are not the result of

genetic differences in the promoter region of the prothrombin gene.

Subsequently, we determined the allele frequencies of four

pro-thrombin gene polymorphisms in the 20210-GG LETS controls with

plasma prothrombin levels >130 U/dl and compared the observed

al-lele frequencies with those observed in a group of healthy volunteers. By

this approach we found an association between the 19911-G allele and

prathrombin levels. The effect of this allele is small when compared to

that of the 20210-A allele. This might explain why the 19911-G allele

is not a risk factor for venous thrombosis in 20210-GG carriers.

Inter-estingly, it was a risk factor when present together with the 20210-A

allele. The risk associated with the 20210-A allele was the highest when

the other prothrombin allele carried a 19911-G. Using a combination of

allele-specific oligonucleotides and restriction analysis, it was

con-firmed that the 19911-G was indeed on the 20210-G allele and not on

the 20210-A allele, so recombination resulting in the combination

19911-G/20210-A could be excluded. The confidence intervals of the

estimates of risk of DVT show a large overlap and we can not exclude

the possibility that the differences in risk between carriers and

non-car-riers of the 19911 -G allele within this group are a result of chance. One

would expect to find slightly higher plasma prothrombin levels in

20210-AG camers with the 19911-G allele äs the second allele,

com-pared to those with 19911-A allele äs the second allele; this is

present-ly not the case, but to demonstrate a difference of 4 U/dl many more

individuals will be needed.

There is no obvious explanation for the association of the 19911-G

allele with increased plasma prothrombin levels. The 19911

polymor-phism is located approximately in the middle of intron 13, the last and

relatively small intron of the prothrombin gene. It has been shown that

sequences in the last intron of a gene can influence 3'-end formation

(40). However, with regard to the region of the 19911 polymorphism,

no regulatory sequences have been identified so far. It is therefore by no

means certain that this polymorphism itself influences prothrombin

expression. Possibly, it is a marker for another Variation in (part of)

these prothrombin alleles. Further detailed haplotype analysis of the

19911-G alleles can help with the identification of the sequence

Varia-tion responsible for the elevated plasma prothrombin. So far, we

identi-fied four different haplotypes of the prothrombin gene with a G at

posi-tion 19911 in 20210-A carriers (data not shown). However, we were

not able to detect significant differences in expression level between

these alleles. In conclusion, our research has shown that there are at

least two DNA variations in the prothrombin gene that are associated

with increased prothrombin levels. Coexistence of the two changes

(G20210A and A1991 IG) might increase the chance of thrombosis.

Acknowledgements

This study was supported by grant no. 97 001 from the Trombosestichting

Nederland. The LETS study was onginally supported by a grant from the

Netherlands Heart Foundation (89-063).

We like to thank J. Thom, P.A. Kyrie, I. Juhan-Vague, K.H. Reuner, A.

Siegemund, F. Bernardi, J. Danneberg, J. van der Meer, J.M. Soria and R.

Wang who kindly provided DNA samples from homozygous 20210-A carriers

and we like to thank H. de Ronde and C. Spaargaren for their technical

assistance.

References

1. Mann KG. Biochemistry and physiology of blood coagulation. Thromb

Haemostl999;82:165-74.

π·!

3. Degen SJ, Davie EW. Nucleotide sequence of the gene for human prothrombm. Biochemistty 1987; 26: 6165-77.

4. Poort SR, Rosendaai FR, Reitsma PH, Bertina RM. A common genetic Variation in the 3 '-untranslated region of the prothrombm gene is associated with elevated plasma prothrombm levels and an increase m venous throm-bosis Blood 1996; 88:3698-703.

5. Arruda VR, Annichmo-Bizzacchi JM, Gonfalves MS, Costa FF. Preva-lence of the prothrombin gene variant (nt20210A) m venous thrombosis and arterial disease. Thromb Haemost 1997; 78:1430-3.

6. Brown K, Luddington R, Williamson D, Baker P, Baglin T. Risk of venous thromboembolism associated with a G to A transition at position 20210 in the 3'-untranslated region of the prothrombin gene. Br J Haematol 1997; 98: 907-9.

7. Corral J, Gonzalez-Conejero R, Lozano ML, Rivera J, Heras I, Vicente V. The venous thrombosis risk factor 20210 A allele of the prothrombin gene is not a major risk factor for arterial thrombotic disease. Br J Haematol 1997; 99: 304-7.

8. Cumming AM, Keeney S, Salden A, Bhavnani M, Shwe KH, Hay CR. The prothrombin gene G20210A variant: prevalence m a U.K. anticoagulant clinic population. Br J Haematol 1997; 98:353-5.

9. Hillarp A, Zoller B, Svensson PJ, Dahlbäck B. The 20210 A allele of the

prothrombin gene is a common risk factor among Swedish outpatients with verified deep venous thrombosis. Thromb Haemost 1998; 78: 990-2. 10. Leroyer C, Mercier B, Oger E, Chenu E, Abgrail JF, Fe'rec C, Mottier D.

Prevalence of 20210 A allele of the prothrombin gene in venous throm-boembolism patients. Thromb Haemost 1998; 80: 49-51.

ll.Margaglione M, Brancaccio V, Giuliani N, D'Andrea G, Cappucci G, lannaccone L, Vecchione G, Grandone E, Di Mmno G. Increased risk for venous thrombosis in carriers of the prothrombin G—>A2(I2I° gene variant.

Ann Intern Med 1998; 129: 89-93.

12. Martinelli I, Sacchi E, Landi G, Taioli E, Duca F, Mannucci PM. High risk of cerebral-vein thrombosis in carriers of a prothrombin-gene mutation and in users of oral contraceptives. N Engl J Med 1998; 338:1793-7. 13. Ridker PM, Hennekens CH, Miletich JP. G20210A mutation in

prothrom-bin gene and risk of myocardial infarction, stroke, and venous thrombosis in a large cohort of US men. Circulation 1999; 99: 999-1004.

14. Tosetto A, Missiaglia E, Frezzato M, Rodeghiero F. The VITA project: pro-thrombin G20210A mutation and venous thromboembolism in the general population. Thromb Haemost 1999; 82:1395-8.

15. Rosendaai FR, Doggen CJM, Zivelin A, Arruda VR, Aiach M,.Äscovick DS, Hillarp A, Watzke HH, Bernardi F, Cumming AM, Preston FE, Reitsma PH. Geographie distribution of the 20210 G to A prothrombin variant. Thromb Haemost 1998; 79: 706-8.

16. Ferräresi P, Marchetti G, Legnam C, Cavallari E, Castoldi E, Mascoli F, Ardissino D, Palareti G, Bernardi F. 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. Arterioscler Thromb Vase Biol 1997; 17:2418-22.

17. Kyrie PA, Mannhalter C, Be'gum S, Stumpften A, Hirschl M, Weltermann A, Stain M, Brenner B, Speiser W, Pabinger I, Lechner K, Eichinger S. Clinical studies and thrombin generation in patients homozygous or hetero-zygous for the G20210A mutation in the prothrombin gene. Arterioscler Thromb Vase Biol 1998; 18:1287-91.

18. Makns M, Preston FE, Beauchamp NJ, Cooper PC, Daly ME, Hampton KK, Bayhss P, Peake IR, Miller GJ. Co-inheritance of the 2021OA allele of the prothrombin gene increases the risk of thrombosis in subjects with fa-milial thrombophilia. Thromb Haemost 1997; 78:1426-9.

19. Simioni P, Tormene D, Manfrin D, Gavasso S, Luni S, Stocco D, Girolami A. Prothrombin antigen levels in symptomatic and ίΜΐΐιρίοηκιΐιι:

carriers of the 20210A prothrombin variant. Br J Haemalol 1998. K).'· 1045-50.

20. Corral J, Zuazu-Jausoro I, Rivera 3, Gonzalez-Conejero R, Fer.vi K \ iivn-te V. Clinical and analytical relevance of the combination o)

20210A/A and factor V Leiden: results from a large family. BrJ 1999;105:560-3.

1070

21. Sona JM, Almasy L, Souto JC, Tirado I, Borell M, Mateo J, Slifer S, Stone W, Blangero J, Fontcuberta I Linkage analysis demonstrates that the pro-thrombin G20210A mutation jointly influences plasma propro-thrombin levels and risk of thrombosis. Blood 2000; 95:2780-5.

22. Koster T, Rosendaai FR, De Ronde H, Briet E, Vandenbroucke JP, Bertina RM. Venous thrombosis due to poor anticoagulant response to activated protein C: Leiden Thrombophilia Study. Lancet 1993; 342:1503-6. 23. Koster T, Rosendaai FR, Reitsma PH, van der Velden PA, Briet E,

Vandenbroucke JP. Factor VII and fibrinogen levels äs risk factors for

venous thrombosis. A case-control study of plasma levels and DNA poly-morphisms - the Leiden Thrombophilia Study (LETS). Thromb Haemost 1994; 71: 719-22.

24. Koster T, Rosendaai FR, Briet E, Vandenbroucke JP. John Hageman's factor and deep-vein thrombosis: Leiden Thrombophilia Study. Br J Haematol 1994; 87:422-4.

25. Koster T, Blann AD, Briet E, Vandenbroucke JP, Rosendaai FR. Role of clotting factor VIII m effect of von Willebrand factor on occurrence of deep-vein thrombosis. Lancet 1995; 345:152-5.

26. Koster T, Rosendaai FR, Briet E, van der Meer FJ, Colly LP, Trienekens PH, Poort SR, Reitsma PH, Vandenbroucke JP. Protein C deficiency in a controlled series of unselected outpatients: an infrequent but clear risk factor for venous thrombosis (Leiden Thrombophilia Study). Blood 1995; 85:2756-61.

27. De Visser MCH, Poort SR, Vos HL, Rosendaai FR, Bertina RM. Factor X Promoter polymorphisms, factor X levels and the risk of venous thrombo-sis. Thromb Haemost 1999; suppl: 508 (abstract).

28. Rosendaai FR, Grant PJ, Ariens R.A.S., Poort SR, Bertina RM. Factor XIIIVal34Leu, factor XIII antigen and activity levels and risfc of venous thrombosis. Thromb Haemost 1999; suppl: 508 (abstract).

29. Meijers JCM, Tekelenburg WLH, Bouma BN, Bertina RM, Rosendaai FR. High levels of coagulation factor XI äs a risk factor for venous thrombosis. N Engl J Med 2000; 342:696-701.

30. Kamphuisen PW, Rosendaai FR, Eikenboom JCJ, Bös R, Bertina RM. Factor V antigen levels and venous thrombosis: Risk profile, interaction with factor V Leiden and relation with factor Vfll antigen levels. Artenoscler Thromb Vase Biol 2000; 20:1382-6.

31. van Tilburg NH, Rosendaai FR, Bertina RM. Thrombin acdvatablefibrino-lysis inhibitor and the risk for deep vein thrombosis. Blood 2000; 95: 2855-9.

32. van Hylckama Vlieg A, Van der Linden IK, Bertina RM, Rosendaai FR. High levels of factor TX increase the risk of venous thrombosis. Blood 2000; 95:3678-82.

33. Bancroft JD, Schaefer LA, Degen SJ. Characterization of the Aiu-rich 5'-flanking region of t(!f u;:;-:.ri,.. !; vi.:·.:1 .·".ι'uVis: ρι·ηι·: idi:mific;i[i(in

of a positive cisactii; · ··.·;'..·.' ,i ·., ·.:·.· .'·.·. i11..·: ftvillc expre.v>ion.

Genel990;95:253-(·'

34. Iwahana H, Yoshimoto K, Itakura M. Highly polymorphic region of the human prothrombin (F2) gene. Hum Genet 1992; 89:123-4.

35. Zivelin A, Rosenberg N, Faier S, Kornbrot N, Peretz H, Mannhalter C, Horellou MH, Seligsohn U. A single genetic origin for the common pro-thrombotic G20210A polymorphism in the prothrombin gene. Blood 1998; 92:1119-24.

36. Woolf B. On estimating the relation between Wood gronp and diseese. Am > J Hum Genet 1955; 19:251-3.

37. Ott J. Analysis of Human Genetic Linkage. Third Edition, Baltimore and London: The John Hopkins University Press, 1999:1-382., ,

38. Chakravarti A, Buetow KH, Antonarakis SE, Waber PG, Boefan Cö,

Ka?a7ian HH. Nonuniform rivonibinaiion wilhin the human beu-j**ia LYiii· i-lusiiT. Am J Hum (ienei 1984:36: 1239-58.

39. Ιλ.·ιim B. Risi-li N. \ ^omparison öl Imkage disequilibrium manre» (et linc-tt'uli.'mappim1. (n'mmncs 1995:29:311-22.

40 V-ic 1). Chi-iiü J. MiU|Uiii l.l·.. S«|Uu-nees wilhin the last imron fi"l^"j"

K\,\ ν-jnd lorniiiiion m cultured colN. Mol Ccll Biol 1993; l* WM* ·