Atherosclerosis 176 (2004) 95 99

Atherosclerosis 176 (2004) 95 99 Platelet membrane collagen receptor glycoprotein VI polymorphism is associated with coronary thrombosis and fatal myocardial infarction in middle-aged men Elina Ollikainen a, Jussi Mikkelsson a,, Markus Perola b, Antti Penttilä c, Pekka J. Karhunen a a Medical School, University of Tampere and Research, Unit of Clinical Chemistry, Tampere University Hospital, Tampere FIN-33014, Finland b Department of Human Molecular Genetics, National Public Health Institute, Helsinki, Finland c Department of Forensic Medicine, University of Helsinki, Helsinki, Finland Received 1 October 2003; accepted 26 March 2004 Available online 9 June 2004 Abstract Glycoprotein VI is a platelet collagen receptor binding to subendothelial collagen after a rupture of an atherosclerotic plaque. The GPVI gene is polymorphic with several SNPs and the T13254C polymorphism predicting amino acid substitution (serine to proline) has been associated with the risk of MI in a preliminary study. We studied the association of the GPVI T13254C with fatal myocardial infarction (MI) and coronary artery disease among the 300 men of the Helsinki Sudden Death Study (HSDS). Genotype frequencies were 77.9% for TT, 20.7% for CT and 1.4% for CC. We found a significant association (P = 0.02) between the C-allele carriers (CT or CC) and coronary thrombosis (OR 2.5, 95% CI: 1.05 6.2). There was also a tendency (P = 0.07) for an association between the C-allele and acute myocardial infarction (AMI) (OR 2.2). The average area of complicated coronary lesions was also significantly (P = 0.01) larger in carriers compared to non-carriers of the C-allele. Our findings support previous results on the role of this GPVI polymorphism, or another linked polymorphism, as a possible predictor of the risk of coronary thrombosis. 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: GPVI; Myocardial infarction; Polymorphism; Sudden cardiac death 1. Introduction Coronary artery disease (CAD) is the single largest cause of natural death in the Western world. Approximately 50% of deaths due to CAD are sudden and unexpected. The incidence of sudden cardiac death (SCD) increases with age, but the proportion of CAD deaths that are sudden and unexpected is greater in the group of younger individuals [1]. Known risk factors for common cardiovascular diseases are mostly environmental in nature. However, genetic elements are also important risk factors especially for premature disease. In younger people the genetic factors play especially important roles in SCD [2]. Factors influencing Corresponding author. Fax: +358 3 2157511. E-mail address: Jussi.Mikkelsson@dnainternet.net (J. Mikkelsson). the development of atherosclerosis include lipoproteins, inflammatory factors, components of blood coagulation system and platelets [3]. Platelet adhesion and activation by components of the extracellular matrix are crucial in arresting bleeding after trauma but can also be harmful for tissue by occluding atherosclerotic vessels [4,5]. Collagens are major components of extracellular matrix and connective tissue. After damage to the vascular endothelium, platelets adhere to the exposed collagen fibers of the vessel wall leading to secretion of various mediators and platelet aggregation, contributing to the formation of haemostatic plug. Coronary artery thrombosis is usually initiated by rupture or erosion of the atheromatous plaque revealing extracellular matrix protein collagen, which strongly activates platelets [5,6]. Platelets have two major receptors for collagen, Integrin alpha2beta1 and glycoprotein VI [7,8]. 0021-9150/$ see front matter 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2004.03.021

96 E. Ollikainen et al. / Atherosclerosis 176 (2004) 95 99 GPVI is a member of Ig superfamily and has a key role in platelet activation through tyrosine phosphorylation and activation of phospholipase C 2 [9,10]. The T13254C polymorphism in the GPVI gene predicts the substitution of serine to proline, which might influence structure-function of the receptor [11]. The GPVI 13254CC genotype has been associated with an increased risk of MI, particularly in older individuals in a preliminary study by Croft et al. [11]. 2. Materials and methods 2.1. Prospective autopsy series of middle-aged men The original study population comprised a prospective series of 300 Caucasian men, aged 33 69 years, who were subjected to a medico-legal autopsy at the Department of Forensic Medicine, University of Helsinki, from 1991 to 1992 [12]. This autopsy series included 42% of all deaths of those <65 years of age during those years in the area of Helsinki and its surroundings. The reason for the medico-legal autopsy was unexpected sudden or violent death occurring outside a hospital, often unwitnessed. The cause of death was recent MI in 12.0% (n = 36), old MI in 11.0% (n = 33), sudden coronary death with no histological MI in 10.3% (n = 31), other cardiac causes 5.3% (n = 16), other diseases 21.3% (n = 64), and violent deaths (suicides and accidents) 40% (n = 120). The study was approved by the Ethics Committee of the Department of Forensic Medicine, University of Helsinki. 2.2. Measuring the percent stenosis in silicone rubber casts of the coronary arteries At autopsy, coronary angiography was performed by use of vulcanizing liquid-silicone rubber mixed with lead oxide as the contrast medium. This procedure does not dislodge attached thrombus from its site and has been successfully used in the routine postmortem diagnosis of thrombotic and other complications after coronary artery bypass surgery [13]. The proximal, middle, and distal stenoses of the main trunks of the left anterior descending coronary artery, left circumflex artery, and right coronary artery were measured from the cast rubber model with a mauser. The percentage of stenosis was obtained by dividing the diameter (millimetres) of the artery with the greatest stenosis by the diameter of the nearest proximal undamaged part of the cast model of the artery. 2.3. Measuring the area of atherosclerosis by computer-assisted morphometry of coronary arteries Coronary arteries were fixed in 10% buffered formalin and stained for fat by the Sudan IV staining method. The areas of the fatty streaks, elevated fibrous lesions, calcification and complicated lesions (with fissures, hematoma, or thrombosis) [14] were measured by computer-assisted morphometry. Percentage areas of these changes were also calculated. 2.4. Confirmation of MI At autopsy, the presence of MI in the series was confirmed by macroscopic and histological examination of the myocardium. Coronary thrombosis was recorded during opening of the coronary arteries after angiography. Diagnostic studies of MI were done independent of any measurements of the cast of the arteries. Of successfully genotyped cases, 32 men had died of recent MI and an additional 2 had suffered an acute, fatal, occluding coronary thrombosis without histological features of acute MI, owing to their short survival time. These 34 cases were grouped together as recent MI cases for statistical analysis. Old, nonfatal MI was diagnosed in an additional 33 cases. Among these 67 men with MI, 22 (31.9%) were associated with coronary thrombosis, whereas in the remaining 47 (68.1%) men with MI, a macroscopic thrombus could not be found. 2.5. Determination of GPVI genotype The dimorphism of the T13254C in the GPVI gene was detected by polymerase chain reaction and restriction digestion. Genomic DNA (10 30 g) extracted from frozen cardiac muscle samples taken at autopsy was used in each amplification. In the amplification of the fragment encompassing exon 5 primers 5 -ACATCCACAACAGTCCAGTG (forward) and 5 -ATCGAGAAGTCTAGGCAGAG (reverse) were used. DNA was amplified with GeneAmp 9700 PCR System (Perkin-Elmer) for 32 cycles of denaturation at 94 C for 60 s, annealing at 58 C for 60 s, and extension at 72 C for 90 s. The final extension step was at 72 C for 4 min. The 279 bp-product was then incubated at 37 C for 6 h with 10 U of MspI. Digestion resulted in fragments 120 bp, 112 bp, and 47 bp in the presence on the T-allele or 112 bp, 95 bp, 47 bp, and 25 bp in the presence of C-allele [11]. These resulting fragments were then separated by size in a 16% polyacrylamid gel and visualized by ethidium bromide staining in UV light. The electrophoresis was ran at 200 V for 5 h. Genotyping was succesfull in 280 cases. 2.6. Risk factors for CAD A spouse, relative, or close friend of the deceased could be interviewed in 147 cases. Among questions pertaining to risk factors for sudden death, questions were included that delineated past and recent smoking habits as well as previous illnesses. On the basis of these interviews, men were classified as smokers (n = 99) and non-smokers (n = 32). Ex-smokers (n = 16) were excluded from statistical analysis. Hypertension had been diagnosed before death in 50 men and diabetes mellitus in 22.

E. Ollikainen et al. / Atherosclerosis 176 (2004) 95 99 97 2.7. Controls We used men with a non-cardiac cause of death and without significant coronary artery stenosis (>50% vessel stenosis) as controls for analyses of the risk of MI. Another control group was formed from 92 healthy blood donors from the same geographical region of Helsinki. No diabetics were included, smoking habits and hypertension were recorded. 2.8. Statistical analysis The data analysis for stenosis and areas of atherosclerotic changes was based on ANCOVA, in which the possible confounding effects of age, body mass index, diabetes, hypertension and smoking (if data were available) were taken into account by including them into the model as covariates in a stepwise fashion. Dichotomous analyses with the same covariates are based on stepwise logistic regression. The computation was carried out using Statistica 5.0 for Windows. 3. Results Genotype frequencies were 77.9% for TT, 20.7% for CT and 1.4% for CC genotype in the entire study population (Table 1). The frequencies of alleles in this series were similar to those reported in a study in UK by Croft et al. [11].We could not find a direct association between C-allele carriers and sudden cardiac death (P = 0.3). We found a significant (P = 0.0085) association between the C-allele carriers and coronary thrombosis (OR 2.4, 95% CI: 1.1 5.4). Men with C-allele tended to be associated with acute MI although this association did not reach statistical significance (Table 2). Supporting the association of C-allele carriers with coronary thrombosis, we also found significant association between the C-allele carriers and the larger area of complicated coronary lesions (P = 0.01). The average area of complicated lesions were 2.6% for TT and 3.3% for TC/CC. The associations between thrombosis and complicated lesions were significant both in all men and also when analyses were limited to men with interview data (P = 0.003 and P = 0.04, respectively). Thus we chose to report the associa- Table 1 GPVI genotype frequencies in the whole study population, in autopsy controls, in victims of sudden cardiac death and in healthy controls TT TC CC Study population 218 (77.9) 58 (20.7) 4 (1.4) Autopsy controls 126 (79.7) 30 (19.0) 2 (1.3) Sudden cardiac death 82 (74.5) 26 (23.6) 2 (1.8) Healthy controls 74 (80.4%) 18 (19.6) 0 Values are n (%). Table 2 GPVI genotypes, different autopsy findings and P-values TT TC + CC P All MI 50 (74.6) 17 (25.4) 0.9*, 0.8** AMI 22 (64.7) 12 (35.3) 0.2*, 0.07** Thrombosis 14 (63.6) 8 (36.4) 0.0085*, 0.02** Autopsy Controls 126 (79.7) 32 (20.3) Healthy Controls 74 (80.4) 18 (19.6) Values are n (%). Symbol (*) shows comparisons with autopsy controls and (**) comparisons with healthy controls. Table 3 Association of GPVI genotypes with average degree of coronary stenosis (n = 263) and average areas of fatty streaks, elevated lesions, calcified lesions and complicated lesions (n = 272 for all) TT TC + CC P Stenosis 37.7 (S.D. 22.1) 35.3 (S.D. 20.6) 0.9 Fatty streaks 9.8 (S.D. 7.7) 10.0 (S.D. 7.6) 0.9 Elevated lesions 6.0 (S.D. 6.2) 5.9 (S.D. 5.9) 0.6 Calcified lesions 4.7 (S.D. 7.4) 2.9 (S.D. 4.3) 0.1 Complicated lesions 2.6 (S.D. 7.3) 3.3 (S.D. 8.1) 0.01 Values are in %. tion for the entire study population. The results for coronary thrombosis were also significant when we used healthy controls as the reference group (P = 0.02, OR: 2.5; 95% CI: 1.05 to 6.2) (Table 3). 4. Discussion We found the C-allele of the platelet collagen receptor GPVI gene T13254C polymorphism to be associated with coronary thrombosis and the area of complicated coronary lesions. Similar results have been published in a preliminary study in UK [11] with a differing study population. Our study population consisted of men suffering fatal pre-hospital MI and the UK study population consisted of older MI patients who survived to hospital. Although the study populations were so unlike, our study confirms the previous results. Our hypothesis is that since the degree of coronary narrowing is similar among men with different GPVI genotypes, the association of the 13254C allele with complicated coronary lesions and coronary thrombosis is likely to be due to increased platelet tendency to adhere to pre-existent coronary plaques in a high shear environment. The major collagen receptors on platelets are Integrin alpha2beta1 and glycoprotein VI [7,8]. The role of the glycoprotein VI as a collagen receptor was established following the identification of a patient with a mild bleeding disorder whose platelets lacked this receptor [8]. GPVI is physically associated with Fc receptor -chain, which acts as a signaling part of the complex. Binding of collagen to GPVI leads

98 E. Ollikainen et al. / Atherosclerosis 176 (2004) 95 99 to phosphorylation of this -chain which leads to activation of thyrosine kinase Syk and phospholipase C 2 and hence to platelet activation [10,15]. GPVI appears to possess more profound role in platelet thrombus formation under flow conditions than previously thought. It has a crucial role in thrombus formation on the surface of collagen under flow conditions in humans. Furthermore it has also been shown to be involved in the process of firm platelet adhesion to the surface of immobilized vwf in the absence of collagen [16]. The GPVI gene has eight exons [17 19]. Exon V SNP T13254C predicts amino acid substitution of serine to proline. The location of T13254C dimorphism relative to the protein has been reported to be in the stem region of the receptor just outside the platelet membrane [20] The T13254C polymorphism is in linkage disequilibrium with several other polymorphisms of GPVI. Since functional evidence on the T13254C polymorphism is lacking, any one of these linked polymorphisms could be the true variant associated with MI. Accordingly, in a small Japanese study [21] a polymorphism in exon VII (A21908G) of the GPVI gene associated with MI. Interestingly, in Japanese this polymorphism is not in linkage disequilibrium with the T13254C polymorphism as it is in Caucasians [11] and the T13254C was not associated with MI in Japanese. This suggests that the true variant associated with MI in the present study and in the study by Croft et al. [11] could be the exon VII polymorphism of GPVI gene. The location of the A21908G polymorphism relative to the protein is in its mucin-like domain [11]. A recent study on platelet function and the allelic differences in GPVI showed that the low frequency allele homozygotes, such as CC genotype of T13254C dimorphism, are associated with reduced binding of platelets to collagen when compared to high frequency allele homozygotes [20]. The relationship between these and our findings remains unclear since our results are mainly due to differences between high frequency allele homozygotes (TT) and heterozygotes (CT). GPVI has an important role in platelet thrombus formation through interaction with collagen. Selective inhibition of the collagen activation pathway in platelets may provide strong antithrombotic protection while preserving other platelet functions [16]. In the future therapeutic strategies against platelet activation through GPVI could provide a new method for treatment of cardiovascular diseases. The major weakness of the present study is the limited study size. Our results, even though confirmatory to a previous report, should still be interpreted with caution before larger population based studies are performed on the GPVI genetics. In conclusion we have shown that possession of C-allele of the GPVI T13254C polymorphism is associated with coronary thrombosis and an increased area of complicated coronary lesions. These results suggest that the C-allele of the 13254 polymorphism of GPVI may be a risk factor for fatal complications of CAD. Acknowledgements MP is supported by the Paavo Nurmi Foundation, JM supported by the Maud Kuistila Foundation, The Finnish Medical Society Duodecim and the Paavo Nurmi Foundation. References [1] Kannel WB, Thomas HE. Sudden coronary death: the Framingham study. Ann NY Acad Sci 1982;382:3 21. [2] Marenberg ME, Risch N, Berkman LF, Rloderus B, de Faire U. Genetic susceptibility to death from coronary heart disease in a study of twins. N Engl J Med 1994;330:1041 6. [3] Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362:801 9. [4] Weiss HJ. Platelet physiology and abnormalities of platelet function. N Engl J Med 1975;293:531 41. [5] Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes. N Engl J Med 1992;326:242 50. [6] Baumgartner HR. Platelet interaction with collagen fibrils in flowing blood. Part I. Reaction of human platelets with alpha chymotrypsindigested subendothelium. Thromb Haemost 1977;37:1 16. [7] Santoro SA, Zutter MM. The alpha 2 beta 1 integrin: a collagen receptor on platelets and other cells. Thromb Haemostasis 1995;74:813 21. [8] Moroi M, Jung SM, Okuma M, Shinmyoxu K. A patient with platelets deficient in glycoprotein VI that lack both collagen-induced aggregation and adhesion. J Clin Invest 1989;84:1440 5. [9] Clemetson JM, Polgar J, Magnenat E, Wells TM, Clemetson KJ. The platelet collagen receptor glycoprotein VI is a member of the immunoglobulin superfamily closely related to FcalphaR and the natural killer receptors. J Biol Chem 1999;274:29019 24. [10] Gibbins JM, Okuma M, Farndale R, et al. Glycoprotein VI is the collagen receptor in platelets which underlies tyrosine phosphorylation of the Fc receptor -chain. FEBS Lett 1997;413:255 9. [11] Croft SA, Samani NJ, Teare MD, et al. Novel platelet membrane glycoprotein VI dimorphism is a risk factor for myocardial infarction. Circulation 2001;104:1459 63. [12] Mikkelsson J, Perola M, Laippala P, et al. Glycoprotein IIIa PlA polymorphism associates with progression of coronary artery disease and with myocardial infarction in an autopsy series of middle-aged men who died suddenly. Arterioscler Thromb Vasc Biol 1999;19:2573 8. [13] Karhunen PJ, Männikkö A, Penttilä A, Liesto K. Diagnostic angiography in postoperative autopsies, Am J Forensic Med Pathol 1989:303 309. [14] Stary HC, Chandler AB, Dinsmore RE, et al. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis: a report from the Committee of Vascular Lesions of the Council on Arteriosclerosis, American Heart Association. Circulation 1996;92:1355 574. [15] Tsuji M, Ezumi Y, Arai M, Takayama H. A novel association of Fc receptor -chain with glycoprotein VI and their co-expression as a collagen receptor in human platelets. J Biol Chem 1997;272:23528 31. [16] Goto S, Tamura N, Handa S, Arai M, Kondama K, Takayama H. Involvement of Glycoprotein VI in platelet thrombus formation on both collagen and von Willenbrand factor surfaces under flow conditions. Circulation 2002;106:266 72. [17] Ezumi Y, Uchiyama T, Yakayama H. Molecular cloning, genomic structure, chromosomal localization, and alternative splice forms of the platelet collagen receptor glycoprotein VI. Biochem Biophys Res Commun 2000;277:27 36.

E. Ollikainen et al. / Atherosclerosis 176 (2004) 95 99 99 [18] Clemetson J, Polgar J, Magnenat E, Wells T, Clemetson K. The platelet collagen receptor glycoprotein VI is a member of the immunoglobulin superfamily closely related to Fc R and the natural killer receptors. J Biol Chem 1999;274:29019 24. [19] Miura Y, Ohnuma M, Jung S, Moroi M. Cloning and expression of the platelet-specific collagen receptor glycoprotein VI. Thromb Res 2000;98:301 9. [20] Joutsi-Korhonen L, Smethurst P, Rankin A, et al. The low frequency allele of the platelet collagen signalling receptor glycoprotein VI is associated with reduced functional responses and expression. Blood 2003;101:4372 9. [21] Takagi S, Iwai N, Baba S, et al. A GPVI polymorphism is a risk factor for myocardial infarction in Japanese. Atherosclerosis 2002;165: 397 8.