SHORT COMMUNICATION

Lack of association between polymorphisms of thrombogenicgenes and disease susceptibility in rheumatoid arthritis

Theodoros Dimitroulas • Karen M. J. Douglas •

Jacqueline Smith • Vasilis F. Panoulas •

George D. Kitas

Received: 23 October 2011 / Accepted: 11 March 2012

� Springer-Verlag 2012

Abstract Rheumatoid arthritis (RA) is associated with

increased mortality due to cardiovascular disease (CVD).

Abnormalities in coagulation have been linked with CVD

in general and RA population. The aim of our study is to

determine whether particular single nucleotide polymor-

phisms thought to be involved in the regulation of coagu-

lation are over-represented in patients with RA compared

to controls. We compared the frequency of atherothrom-

botic polymorphisms (Factor V Leiden, fibrinogen G455A,

prothrombin G20210A and plasminogen activator inhibi-

tor 4G5G) in 322 RA patients [231 females, mean age

61.5 ± 12, median disease duration 10 years (IQR = 14)]

with 441 local controls. No significant differences were

observed in genotype or allele frequencies either between

RA and controls or between the disease subgroups studied.

Whereas these polymorphisms may be of importance at the

level of individual patients, they are unlikely to be clini-

cally important on a population basis.

Keywords Rheumatoid arthritis � Genetic

polymorphism � Atherothrombosis � Cardiovascular disease

Introduction

Rheumatoid arthritis (RA) is associated with excess risk of

cardiovascular morbidity and mortality. Increased fibrino-

lytic activity as assessed by abnormal concentrations of

fibrinogen, plasminogen activator inhibitor (PAI) and tis-

sue plasminogen activator (tPA) are markers of athero-

thrombotic events and cardiovascular events [1]. In RA

patients, the derangement of coagulation system has been

associated with future thrombotic or cardiovascular events

in an 8-year follow-up study [2].

A significant number of candidate genes and polymor-

phisms have been associated with abnormalities in coagu-

lation factors and future cardiovascular disease (CVD),

including prothrombin G2021A, Factor V Leiden [3], PAI

4G5G [4], and fibrinogen G455A [5] mutations. Few

studies have assessed the distribution of atherothrombotic

genes within the RA population [6, 7].

The aim of our study was to determine whether partic-

ular single nucleotide polymorphisms thought to be invol-

ved in the coagulation system are over-represented in

patients with RA compared to controls, or within specific

RA disease phenotypes associated with increased risk for

CVD.

Materials and methods

Study population

A cohort of 322 patients who fulfilled the 1987 ACR cri-

teria for RA [8] were recruited from routine rheumatology

outpatient clinics of the Dudley Group of Hospitals, NHS

FT, West Midlands, UK, between August 2004 and August

2006. The study had local research ethics committee and

The authors Theodoros Dimitroulas and Karen M. J. Douglas have

equally contributed to this work and therefore share first authorship.

T. Dimitroulas � K. M. J. Douglas � J. Smith �V. F. Panoulas � G. D. Kitas (&)

Department of Rheumatology, Dudley Group of Hospitals NHS

Foundation Trust, Russells Hall Hospital, Pensnett Road,

Dudley, West Midlands DY1 2HQ, UK

e-mail: gd.kitas@dgoh.nhs.uk; g.d.kitas@bham.ac.uk

123

Rheumatol Int

DOI 10.1007/s00296-012-2392-6

research and development approval and all participants

gave written informed consent according to the Declaration

of Helsinki. The control group consisted of 441 age- and

sex-matched apparently healthy individuals from the same

ethnic and geographic background, who had previously

consented to anonymously donate blood for DNA extrac-

tion for a similar local polymorphism study. As genetic

polymorphisms vary between races and populations, all the

controls had the same ethnic and geographic background

(based on the place of birth) with the patients, and were

selected on the basis of age and gender. Ethical approval

was granted to use these anonymous samples for determi-

nation of the local prevalence of the polymorphisms.

DNA extraction and assays

A 5-ml sample of blood was taken in EDTA tubes for DNA

extraction using the Whatman Bioscience DNA extraction

method [9]. Polymorphisms were assessed in all subjects

using the LightCyclerTM System (Idaho Technology Inc.

Salt Lake City, Utah, USA), as previously described [10].

Cycle conditions were as follows: denaturation of the

template DNA for 1 cycle of 95 �C for 10 min, pro-

grammed transition rate of 20 �C/s; amplification of the

target DNA for 35 cycles of 95 �C for 3 s, 56 �C for 10 s

and 72 �C for 10 s, each with a temperature transition rate

of 20 �C/s; melting curve analysis for 1 cycle of 95 �C for

0 s and 40 �C for 30 s, each with a transition temperature

rate of 20 �C/s, and then increasing to 80 �C. PAI 4G/5G

and fibrinogen G455A polymorphisms were assessed by

probes and primers commercially available (Tib Molbiol,

Berlin, Germany). The 50 and 30 primers for PAI 4G/5G

were AGC CAG ACA AGG TTG TTG ACA C and CAG

AGG ACT CTT GGT TTT CCC respectively. The sensor

probe used was 50-TGA CTC CCC CAC GTG TCC-30-FLUORESCEIN and the anchor one 50-ACT CTC TCT

GTG CCC CTG AGG GCT CT-30-LCRED640. For

fibrinogen G455A, the forward primer was 50-GAT GTG

TAT TTT TCA TAG AAT AGG GTA-30, the reverse one

50-ATT TGA CCT ACT CAC AAG GCA A-30, the sensor

probe 50-CAT TAC TAT TGA TTT TAA TAG CCC CT-

30FLUORESCEIN and the anchor one 50LCRED640-TGA

AAT AGA ATT ATG TCA TTG TCA GAA AAC-30PH.

Factor V Leiden and prothrombin G20210A polymor-

phisms were detected by commercially available kits

(Roche, Basel, Switzerland) for the LightCycler. These kits

provide all the primers, probes and reagents required for

hybridisation.

Statistical analysis

Variables are presented as mean ± SD, apart from duration

of disease, which is expressed as the median. Genotype and

allele frequencies of each polymorphism were compared

between groups using v2 tests (or Fisher’s exact test where

n \ 5). Statistical significance was set at p \ 0.05.

Results

The study included 231 females (72 %) and 91 males

(28 %) [mean age 61.9 ± 11.63 median disease duration

10 years (IQR = 14)]. Of all, 232 (72 %) were rheumatoid

factor positive and 171 (53 %) anti-CCP positive. Mean

values of erythrocyte sedimendation rate, C-reactive pro-

tein and disease activity score 28 were 29.56 ± 26.2,

18.16 ± 23.17 mg/lt and 4.24 ± 1.42, respectively.

Genotype and allele frequencies of the polymorphisms

in the RA and control populations studied are shown in

Table 1. No significant differences were found for any of

them. Similarly, no differences were found between any of

the RA disease sub-groups studied: males versus females;

rheumatoid factor and/or anti-CCP positive versus negative

patients; with versus without extra-articular disease.

Discussion

In this study, we did not identify variations in the fre-

quency of any of the atherothrombotic polymorphisms or

alleles studied between controls and RA or within sub-

groups of RA patients. To the best of our knowlegde this is

the first study investigating Factor V Leiden and pro-

thrombin G20210A polymorphisms in RA. CVD mani-

festing as acute myocardial infarction, stroke or sudden

Table 1 Frequencies of the polymorphisms in the RA and control

populations

Gene Genotype Frequency of

polymorphism (%)

Significance

Controls

n = 441

RA patients

n = 324

PAI 4G4G 139 (31.5) 103 (31.8) NS

4G5G 197 (44.7) 151 (46.6)

5G5G 105 (23.8) 70 (21.6)

Prothrombin GG 436 (98.9) 317 (97.8) NS

GA 5 (1.1) 7 (2.2)

AA 0 0

Factor V

Leiden

GG 422 (95.7) 310 (95.7) NS

GA 19 (4.3) 13 (4.0)

AA 0 1 (0.3)

Fibrinogen GG 317 (71.9) 224 (69.0) NS

GA 124 (28.1) 99 (30.7)

AA 0 1 (0.3)

PAI plasminogen activator inhibitor, RA rheumatoid arthritis

Rheumatol Int

123

death is highly prevalent and has worse outcomes in RA

[11]. Under impaired fibrinolysis, plaque rupture and/or

vasospasm may result in occlusive thrombus and sub-

sequent CVD event. Based on this hypothesis, we tried to

determine whether there was an enhanced genetic suscep-

tibility of RA patients to abnormal coagulation.

PAI 4G allele has been associated with ischaemic heart

disease in a large cohort of 467 RA patients [6], independent

of variables such as sex, age, hypertension, smoking, disease

activity and treatment. Combined carriage of TNFRII 196R

variant and fibrinogen G455A was a stronger predictor for

hypertension than each genotype separately. In another

study including 132 RA patients, the C/G genotype was

associated with increased circulating plasma levels of

PAI-1, which has been described previously as a predictor

of future CVD events in RA [7]. The same group found

higher PAI-1 expression in C/C genotype carriers than C/G

and G/G carriers, in Mexican subjects without underlying

heart disease [12].

It has been demonstrated that systemic extra-articular

and seropositive and/or erosive disease [13, 14] are major

predictors of cardiovascular mortality in RA. If a proco-

agulant genetic polymorphism was to be over-represented

in RA patients, then these would be the most likely groups

to be affected. We found no such increase in prevalence of

the studied polymorphisms in patients sero-positive for

either RF or anti-CCP antibodies. It is therefore unlikely

that differences in the frequencies of the genetic poly-

morphisms studied, on a population basis, is the explana-

tion for the coagulation abnormalities in RA patients. This

does not necessarily preclude genetic polymorphisms from

being important determinants of the levels of the relevant

coagulation factors within a certain individual.

The strengths of the present study include a large local

control group and the number of RA patients greater than

in previous studies. Without prior knowledge of the local

prevalence, an appropriate power calculation was not

possible. Instead, we based our calculations on published

ranges in other populations, assumed our sample to con-

tain 300 RA patients and then determined what differ-

ences in prevalence would be significant. We therefore

acknowledge that our sample size may have been inade-

quate to determine small differences and this is a common

criticism of many polymorphism studies [15]. However,

the determined mutation frequencies in the present study

can now be used to estimate what sample size would have

been required to find a significant difference between

groups: this ranged from 3,000 to 100,000 patients in each

arm, depending on the polymorphism concerned, inferring

that any difference may be of limited or no clinical

relevance.

In conclusion, the polymorphisms studied are not over-

represented in patients with RA and cannot therefore

explain the coagulation differences between RA and con-

trols on a population basis. The complexity of CVD in RA

and the risk factor interactions make it unlikely that genetic

epidemiology will identify genes involved in these pro-

cesses without a better understanding of environmental and

disease-related influences.

Acknowledgments Dr Theodoros Dimitroulas has been supported

by the Rheumatology Society of Northern Greece and the Hellenic

Society for Rheumatology.

References

1. Juhan-Vague I, Morange P, Christine Alessi M (1999) Fibrino-

lytic function and coronary risk. Curr Cardiol Rep 1:119–124

2. Wallberg-Jonsson S, Cederfelt M, Rantapaa Dahlqvist S (2000)

Hemostatic factors and cardiovascular disease in active rheuma-

toid arthritis: an 8 year followup study. J Rheumatol 27:71–75

3. Ye Z, Liu EH, Higgins JP, Keavney BD, Lowe GD, Collins R,

Danesh J (2006) Seven haemostatic gene polymorphisms in

coronary disease: meta-analysis of 66,155 cases and 91,307

controls. Lancet 367:651–658

4. Iacoviello L, Burzotta F, Di Castelnuovo A, Zito F, Marchioli R,

Donati M (1998) The 4G/5G polymorphism of the PAI-1-1

promoter gene and the risk of myocardial infarction: a meta-

analysis. Thromb Haemost 80:1029–1030

5. Green FR, Hamsten A, Blomback M (1993) The role of the

fibrinogen genotype in determining plasma fibrinogen levels in

the young survivors of myocardial infarction and healthy controls

from Sweden. Thromb Haemost 70:915–920

6. Arlestig L, Wallberg Jonsson S, Stegmayr B, Rantapaa-Dahlqvist

S (2007) Polymorphism of genes related to cardiovascular disease

in patients with rheumatoid arthritis. Clin Exp Rheumatol

25:866–871

7. Torres-Carrillo N, Torres-Carrillo NM, Martınez-Bonilla GE,

Vazquez-Del Mercado M, Palafox-Sanchez CA, Oregon-Romero

E, Bernard-Medina AG, Rangel-Villalobos H, Munoz-Valle JF

(2009) Plasminogen activator inhibitor-1 C/G polymorphism in

relation to plasma levels in rheumatoid arthritis. Clin Exp Med

9:223–228

8. Arnett FC, Edwarthy SM, Bloch DA, McShane DJ, Fries JF,

Cooper NS (1988) The American Rheumatism Association 1987

revised criteria for the classification of rheumatoid arthritis.

Arthritis Rheum 31:315–324

9. Subrungruang I, Mungthin M, Chavalitshewinkoon-Petmitr P,

Rangsin R, Naaglor T, Leelayoova S (2004) Evaluation of DNA

extraction and PCR methods for detection of Enterocytozoonbieneusi in stool. J Clin Microbiol 42:3490–3494

10. Panoulas VF, Douglas KM, Smith JP, Metsios GS, Elisaf MS,

Nightingale P, Kitas GD (2009) Galectin-2 (LGALS2) 3279C/T

polymorphism may be independently associated with diastolic

blood pressure in patients with rheumatoid arthritis. Clin Exp

Hypertens 31:93–104

11. Douglas KMJ, Pace AV, Treharne GJ, Saratzis A, Nightingale P,

Erb N, Banks MJ, Kitas GD (2005) Excess recurrent cardiac

events in rheumatoid arthritis patients with acute coronary syn-

drome. Ann Rheum Dis 65:348–353

12. Torres-Carrillo N, Magdalena Torres-Carrillo N, Vazquez-Del

Mercado M, Rangel-Villalobos H, Parra-Rojas I, Sanchez-

Enrıquez S, Francisco Munoz-Valle J (2008) Distribution of –844

G/A and Hind III C/G PAI-1 polymorphisms and plasma PAI-1

levels in Mexican subjects: comparison of frequencies between

populations. Clin Appl Thromb Hemost 14:220–226

Rheumatol Int

123

13. Goodson NJ, Wiles NJ, Lunt M, Barrett EM, Silman AJ,

Symmons DP (2002) Mortality in early inflammatory polyar-

thritis: cardiovascular mortality is increased in seropositive

patients. Arthritis Rheum 48:2010–2019

14. Turesson C, McClelland RL, Christianson TJ, Matteson EL

(2006) Multiple extra-articular manifestations are associated with

poor survival in patients with rheumatoid arthritis. Ann Rheum

Dis 65:1533–1534

15. Lane DA, Grant PJ (2000) Role of hemostatic gene polymor-

phisms in venous and arterial thrombotic disease. Blood

95:1517–1532

Rheumatol Int

123