Figures
Abstract
Background and Aims
The aims of this study were to investigate (1) if P2Y12 polymorphisms defining the P2Y12 H2 allele are associated with any other SNPs that may explain the previously reported association with increased ADP induced platelet activation and association with peripheral arterial disease and coronary artery disease and (2) if such variants are associated with acute myocardial infarction (AMI) or classical risk factors for AMI.
Methods and Results
The P2Y13 Met-158-Thr polymorphism was found to be in linkage disequilibrium (LD) with the P2Y12 H2 haplotype (all examined SNPs: D′ = 1.0, r2 = 0.936–1.0), defining a novel P2Y12 H2/P2Y13 Thr-158 haplotype. Genotyping of an AMI case control population (n = 1244 cases, 2488 controls) revealed no association of the P2Y13 Thr-158 allele with AMI (OR = 0.96, 95% C.I. 0.82–1.12, P = 0.63). Also, no differences between the genotype frequencies of P2Y13 Met-158-Met and Met-158-Thr/Thr-158-Thr were seen in AMI case-control subpopulations (early onset AMI OR = 1.06, 95% C.I. 0.85–1.31, P = 0.62); family history of AMI (OR = 0.98, 95% C.I. 0.78–1.22, P = 0.83) nor in early onset AMIs with family history of AMI (OR = 1.0, 95% C.I. 0.74–1.36, P = 1.0). Genotyping of the P2Y13 Met-158-Thr polymorphism in a population based sample (n = 6055) revealed no association with cardiovascular risk factors. In addition, the P2Y13 Met-158-Thr polymorphism was genotyped in a diabetes case-control population, and associations were found neither with DM nor with any examined DM risk factors.
Citation: Amisten S, Braun OÖ, Johansson L, Ridderstråle M, Melander O, Erlinge D (2008) The P2Y13 Met-158-Thr Polymorphism, Which Is in Linkage Disequilibrium with the P2Y12 Locus, Is Not Associated with Acute Myocardial Infarction. PLoS ONE 3(1): e1462. https://doi.org/10.1371/journal.pone.0001462
Academic Editor: Florian Kronenberg, Innsbruck Medical University, Austria
Received: October 9, 2007; Accepted: December 22, 2007; Published: January 23, 2008
Copyright: © 2008 Amisten et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The study has been supported by the Swedish Scientific Research Council, the Swedish Heart and Lung Foundation Lund University Hospital funds and the Vascular Wall program (Lund Medical faculty).
Competing interests: The authors have declared that no competing interests exist.
Introduction
Three human adenosine diphosphate (ADP) receptors have been cloned: P2Y1, P2Y12 and P2Y13 [1]–[4]. On platelets, P2Y1 and P2Y12 mediate ADP-induced platelet activation and aggregation [5]. In red blood cells, activation of P2Y13 by the adenosine triphosphate (ATP) metabolite ADP activates a negative feedback loop that inhibits ATP release from erythrocytes [6]. Rare mutations in the P2Y12 gene that disrupt P2Y12 receptor function result in compromised ADP-induced platelet activation and increased bleeding times [1], [7], [8].
The clinical importance of the P2Y12 receptor as a mediator of platelet activation has become evident in several large-scale clinical studies and inhibition of the P2Y12 receptor with clopidogrel is one of the cornerstones in treatment and prevention of acute coronary syndromes [9]. Even greater P2Y12 inhibition by prasugrel was recently shown to be even more effective in preventing ischemic events than the standard regimen of clopidogrel [10].
A group of single nucleotide polymorphisms (SNPs) in the P2Y12 gene, forming the so called P2Y12 H2 haplotype [11], have been associated with increased platelet responsiveness to ADP and increased risk of peripheral arterial disease (PAD) [11]–[13]. Recently, Cavallari et al showed an association of the P2Y12 H2 haplotype with coronary artery disease (CAD) [14]. It has also been proposed that this haplotype may account for variations in response to clopidogrel. However, several studies have failed to confirm any association between platelet function and the H2 haplotype [15]–[17].
The group of polymorphisms that make up the P2Y12 H2 haplotype are all synonymous polymorphisms that do not change the amino acid sequence of the P2Y12 protein, and no mechanistic explanation to the reported increased platelet reactivity to ADP associated with this haplotype has been presented [11]. However, a non-synonymous polymorphism, Met-158-Thr, in the neighboring P2Y13 gene, located only 8 kb away from P2Y12, could be in linkage disequilibrium with the P2Y12 H2 haplotype. We hypothesized that the P2Y13 Met-158-Thr polymorphism of the P2Y13 receptor could account for the reported effects of the P2Y12 H2 haplotype since the receptors share the same ligand, ADP. The P2Y13 receptor has been found on red blood cells and inflammatory cells, both cell types known to interact with platelets [4], [6]. The first objective of this study was to examine possible linkage disequilibrium (LD) between the P2Y12 H2 haplotype and the P2Y13 Met-158-Thr polymorphism. After showing that this was the case, we aimed at investigating if the P2Y13 Met-158-Thr polymorphism is associated with acute myocardial infarction (AMI) or diabetes mellitus, two diseases strongly associated with peripheral arterial disease [18], [19]. Our hypothesis that the P2Y12 H2 haplotype and SNPs in LD with the P2Y12 H2 haplotype would be associated with AMI was strengthened further by a report linking the P2Y12 H2 haplotype with coronary artery disease (CAD) [14], since myocardial infarction is the major complication of CAD [20]. In order to do so, the Met-158-Thr polymorphism was genotyped in more than 10,000 individuals divided in three study populations: two sub-populations of the Malmö Diet and Cancer study (an AMI case-control population and a large population with cardiovascular risk factor data [21], [22]) and a diabetes mellitus case-control population with data on several DM risk factors [23].
Materials and Methods
Malmö diet and cancer population (MDC)
The study population is made up of 28098 randomly selected men (born 1923–1945) and women (born 1923–1950) living in the Swedish city of Malmö (population 250 000). Overall participation rate in the study was 41%.
A baseline examination was performed between 1991–1996, including assessment of dietary habits, a questionnaire on socio-economic, demographic and lifestyle factors, heredity, medication and previous and current diseases. Blood samples were taken and DNA, lymphocytes, granulocytes, erythrocytes and plasma/serum were stored in a biological bank [21], [22].
AMI Case control population
On 31 December 2000 the study population was matched with the Swedish National Board of Health and Welfare's National Patient Registry and Cause of Death Registry. AMI cases (first AMI) were identified using the diagnosis criteria defined by the International Classification of Diseases, 9th and 10th and Revision, Clinical Modification (ICD 9 and 10); ICD 9 codes 410 in the Swedish Patient Registry or 410-414 in the Swedish Cause of death Registry; ICD 10 codes I21 in the Swedish Patient Registry and I21-I25 in the Swedish Cause of Death Registry.
Two gender- and age (±1 year) -matched AMI-free controls from the MDC population were assigned to each AMI case, resulting in a case-control material consisting of 1244 AMI cases and 2488 AMI-free controls. The myocardial infarction group was then further subdivided into early onset AMIs (EO, n = 622), age at first AMI event <62.8 years (median age of all first event AMI cases) and late onset AMIs (LO, n = 622, age at first AMI event >62.8 years). Family history AMIs (FH, n = 611) were defined as AMI cases where at least one blood related first degree family member had suffered an AMI, and non familial AMIs (n = 633) as cases without any first degree family history of AMI. 319 cases had both early onset and family history AMI (table 1). DNA was available from all cases and controls (n = 3732).
Cardiovascular group population
Of the MDC, 6103 individuals were randomly selected into a “Cardiovascular cohort” (MDC-CV), a sample thus being representative of MDC, in whom cardiovascular risk factors were measured, including systolic blood pressure, smoking status and anthropometric data and, in the majority (n = 5540), fasting plasma analyses of glucose, lipids and C-reactive protein (CRP). DNA for genotyping was obtained from 6055 of the 6101 selected individuals.
Diabetes mellitus case-control population
The diabetes mellitus case control material has been described elsewhere [23]. Briefly, all study subjects originate from the Botnia region in Western Finland and the Helsinki area and age- and gender matched controls were assigned to all type 2 diabetes cases. The case group is composed of 307 unrelated randomly selected individuals with type 2 diabetes (146 males and 161 females, mean age 61 (55–67) years, mean BMI 28.7 (26.0–31.7)). The control group consisted of 307 unrelated individuals with confirmed normal oral glucose tolerance and without a family history of diabetes (146 males and 161 females, mean age 60 (53–67) years, BMI 26.4 (24.1–29.2)).
Extent of H2 haplotype linkage disequilibrium and genotyping of case-control and cardiovascular group populations
Using HapMap and the Human Genome assembly build 36.2, SNPs in or within 1000 base pairs (bp) upstream of the known genes located in the 3q24-25 region (P2Y12 locus) were identified. By means of DNA sequencing using BigDye v. 3.1 (Applied Biosystems, CA, USA) in 20 individuals, selected SNPs were probed for linkage disequilibrium with the known P2Y12 H2 haplotype SNPs using one of the P2Y12 H2 SNPs, rs2046934, as a marker of the P2Y12 H2 haplotype [11]. Single nucleotide polymorphisms (SNPs) that displayed high degrees of LD with the P2Y12 H1/H2 haplotype SNPs were selected for genotyping in a randomly selected sub-population of the DM case-control population (n = 295) using TaqMan or Sequenom and a haplotype map was constructed using the Haploview software [24].
Genotyping of the AMI (n = 3732) and DM (n = 614) case control populations and the cardiovascular group population (n = 6055) was performed using Sequenom (Sequenom Inc., CA, USA) or TaqMan ABI 7900 according to the manufacturers' instructions. Two different persons who were unaware of the phenotypic status of the study participants read all genotypes. For genotyping primers and probes, see table 2.
Statistical analysis
In the AMI case control population, conditional logistic regression was used to calculate odds ratios and p values. The Cardiovascular group population was subjected to ANOVA and t-tests for continuous normally distributed variables, in case of non-normality Kruskal-Wallis test or Mann-Whitney test was used. Chi-2 test was used to test for significant differences in dichotomous variables. In the DM case control population, variables were log transformed for normal distribution. P-values were calculated using the GLM-ANCOVA using sex and age as covariates. Adjustment for multiple testing was not done. Statistical analyses were performed with SPSS.
Power calculations
For myocardial infarction.
Accepting a significance level of 0.05, 1244 AMI cases and 2488 controls have a power of 95% to detect a genotype relative risk of 1.20 for the P2Y13 Met-158-Thr polymorphism. Thus, it is unlikely that our result is a false negative finding.
For diabetes.
Accepting a significance level of 0.05, 307 diabetes cases and 307 controls have a power of 35% to detect a genotype relative risk of 1.20 for the P2Y13 Met-158-Thr polymorphism. Accepting a significance level of 0.05, 532 diabetes cases and 5522 controls in the cardiovascular group population have a power of 79% to detect a genotype relative risk of 1.20 for the P2Y13 Met-158-Thr polymorphism. By analyzing both these two diabetes materials, it is unlikely that our result that the P2Y13 Met-158-Thr polymorphism is not associated with diabetes is a false negative finding. Powercalculations were performed using the program CaTS [25].
Results
Extent of P2Y12 H2 haplotype linkage disequilibrium
The pilot linkage disequilibrium analysis of SNPs in the P2Y12 locus revealed three SNPs (rs1466684, rs38211667, rs11922647) that showed signs of LD with the known P2Y12 H2 SNPs. Genotyping in a larger population (n = 295) confirmed complete LD of the three SNPs with the P2Y12 H2 haplotype SNPs (D′ = 1.0, r2 = 0.936–1.0, figure 1) [11]. Two of the three SNPs were located in or in close proximity to the P2Y12 gene: rs38211667 in the non-coding region of P2Y12 exon 2 and rs11922647 within 1000 bp of transcription start of transcript variant 2 of the P2Y12 gene. The third SNP (rs1466684) was found to be a non-synonymous SNP of the P2Y13 gene, causing a Met-Thr amino acid substitution at position 158 of the P2Y13 receptor. The complete LD of the P2Y12 H2 haplotype with the P2Y13 Thr-158 variant thus defines a novel P2Y12 H2/P2Y13 Thr-158 haplotype. The P2Y13 Met-158-Thr polymorphism was selected as reference SNP for the P2Y12 H2/P2Y13 Thr-158 haplotype in subsequent genotyping.
All examined SNPs displayed a very high degree of linkage disequilibrium (D′ = 1.0; r2 = 0.936–1.0 (see figure for each individual r2 value).
Genotyping of the P2Y12 H2/P2Y13 Thr-158 haplotype in the AMI and DM case control populations
In the AMI case control population, containing 3273 individuals, 92.8% of those eligible for the case–control study were genotyped successfully (table 3), representing 1134 pairs in total (n = 1134 cases and one to two control subjects matching every case (n = 2139). Genotype frequencies were in accordance with Hardy–Weinberg Equilibrium.
No association of the P2Y13 Met-158-Thr polymorphism (Thr-158-Met and Thr-158-Thr vs. Met-158-Met) was found with AMI (OR = 0.96, 95% C.I. 0.82–1.12, P = 0.63). Also, no differences were seen in the AMI case subpopulations (EO OR = 1.06, 95% C.I. 0.85–1.31, P = 0.62; FH OR = 0.98, 95% C.I. 0.78–1.22, P = 0.83; EO+FH OR = 1.0, 95% C.I. 0.74–1.36, P = 1.0).
In the diabetes mellitus (DM) case control study 576 individuals (93.8%) were genotyped successfully. No associations of the P2Y13 polymorphism (Thr-158-Met and Thr-158-Thr vs. Met-158-Met) were found with diabetes mellitus or any examined DM risk factor (table 4).
Genotyping of P2Y13 Met-158-Thr in the cardiovascular group population
5846 individuals (96.5%) of 6055 in the cardiovascular group (CVG) were genotyped successfully. No association was found for the P2Y13 Met-158-Thr polymorphism (Thr-158-Met and Thr-158-Thr vs. Met-158-Met) regarding any of the examined cardiovascular risk factors, including systolic blood pressure, diastolic blood pressure, BMI, waist circ., diabetes, total cholesterol, triglycerides, HDL, LDL, CRP, smoking or alcohol intake (table 5). Furthermore, no stronger association with the above mentioned risk factors was seen in the homozygous P2Y13 Thr-158 group compared to the heterozygous Met-158-Thr carriers (data not shown).
Discussion
In this study we show that the P2Y12 H2 haplotype [11] is in complete linkage disequilibrium with the non-synonymous Met-158-Thr polymorphism in the P2Y13 gene, defining a P2Y12 H2/P2Y13 Thr-158 haplotype. Based on the observed LD between the studied P2Y12 and P2Y13 polymorphisms, we assume that all disease and risk factor associations made with the P2Y13 Met-158-Thr polymorphisms are also valid for the P2Y12 H1/H2 haplotypes. We hypothesized that this finding could provide a potential mechanistic explanation to the previously observed clinical associations of the P2Y12 H2 haplotype with CAD, PAD or platelet function [11], [12], [14], since the receptors share the same ligand. However, no associations of the P2Y13 Met-158-Thr polymorphism with AMI or DM were found in our large material. Indeed, no associations with any of the investigated cardiovascular or diabetes mellitus risk factors were observed. This was unexpected, since strong associations have been reported between CAD, PAD, AMI and diabetes [18]–[20]. All studies involving the P2Y12 H2 haplotype are listed in table 6.
The concentration of extracellular nucleotides in the blood is tightly regulated by ectonucleotidases on leukocyte and endothelial cells to prevent excessive ADP accumulation and subsequent platelet activation [26]. Red blood cells contain millimolar amounts of ATP and are therefore a major source of nucleotides in the blood [6], [27]. This ATP pool could potentially be an important contributor to the regulation of platelet activation. Recently, it was shown that extracellular ADP activates P2Y13 expressed on red blood cells, resulting in a subsequent decreased release of nucleotides from the red blood cells in a classic negative feedback manner [6]. It is possible that this negative feedback loop might be important in the regulation of nucleotide-induced platelet activation in vivo. Thus, a non-synonymous polymorphism leading to a structurally and functionally altered P2Y13 could potentially alter the nucleotide concentrations in the blood stream, thereby affecting platelet activation in vivo. Indeed, in silico analysis using the polymorphism phenotyping prediction software PolyPhen [28] indicated that the P2Y13 Met-158-Thr amino acid substitution could possible affect the function of the P2Y13 receptor.
In 2002 Fontana et al reported the P2Y12 H2 haplotype to be associated with a gain of function in terms of ADP induced platelet aggregation. The polymorphisms in the H2 haplotype are either located in intronic regions of the gene or were silent, i.e. causing no alterations of the P2Y12 receptor protein. The possibility remained that the polymorphisms could potentially be coupled to mRNA processing or translation events, thereby altering P2Y12 receptor protein expression. However, no such data has been presented.
In a subsequent study, Fontana et al also reported an association between the P2Y12 H2 haplotype and PAD [12]. PAD causes an increased atherosclerotic burden throughout the whole cardiovasculature and patients with PAD have a marked increase in coronary artery disease [29]. The progression of chronic atherosclerotic lesions is mainly driven by an inflammatory reaction, with recruitment of inflammatory cells and subsequent reactive changes in the vessel wall [21]. In acute thrombotic complications of atherosclerosis, such as myocardial infarction, platelets constitute the major role. A gain of function polymorphism leading to increased platelet reactivity would likely be more prominent when studied in a setting were platelet activation is a main pathogenic factor, such as AMI. However, to our surprise, genotyping in several thousand individuals revealed no association with AMI.
The early onset or family history AMI case-control subpopulations are believed to contain a stronger genetic component of AMI. The lack of association also in these populations emphasizes further that the P2Y12 H2/P2Y13 Thr-158 haplotype is not associated with cardiovascular disease.
It has been proposed that the P2Y12 H2 haplotype might be involved in the variation in response to clopidogrel treatment. However, subsequent studies have not been able to confirm that variations in response after a high loading dose of clopidogrel are associated with the haplotype. The failure to confirm this hypothesis agrees well with our study and supports the lack of association of the P2Y12 H2/P2Y13 Thr-158 haplotype with cardiovascular disease.
In conclusion, we found that the P2Y13 Met-158-Thr polymorphism was in complete LD with the P2Y12 H2 haplotype, defining a novel P2Y12 H2/P2Y13 Thr-158 haplotype. Genotyping of more than 10 000 individuals in three separate study populations revealed no associations with AMI, DM or related risk factors. Therefore, it seems very unlikely that the examined polymorphisms of the P2Y12 and P2Y13 genes contribute to the pathogenesis of cardiovascular disease or diabetes mellitus.
Author Contributions
Conceived and designed the experiments: OM DE SA. Performed the experiments: LJ SA. Analyzed the data: OM LJ DE SA OB MR. Contributed reagents/materials/analysis tools: OM DE MR. Wrote the paper: OM LJ DE SA OB MR.
References
- 1. Hollopeter G, Jantzen HM, Vincent D, Li G, England L, et al. (2001) Identification of the platelet ADP receptor targeted by antithrombotic drugs. Nature 409: 202–207.
- 2. Ayyanathan K, Webbs TE, Sandhu AK, Athwal RS, Barnard EA, et al. (1996) Cloning and chromosomal localization of the human P2Y1 purinoceptor. Biochem Biophys Res Commun 218: 783–788.
- 3. Zhang FL, Luo L, Gustafson E, Lachowicz J, Smith M, et al. (2001) ADP is the cognate ligand for the orphan G protein-coupled receptor SP1999. J Biol Chem 276: 8608–8615.
- 4. Communi D, Gonzalez NS, Detheux M, Brezillon S, Lannoy V, et al. (2001) Identification of a novel human ADP receptor coupled to G(i). J Biol Chem 276: 41479–41485.
- 5. Murugappa S, Kunapuli SP (2006) The role of ADP receptors in platelet function. Front Biosci 11: 1977–1986.
- 6. Wang L, Olivecrona G, Gotberg M, Olsson ML, Winzell MS, et al. (2005) ADP acting on P2Y13 receptors is a negative feedback pathway for ATP release from human red blood cells. Circ Res 96: 189–196.
- 7. Cattaneo M, Zighetti ML, Lombardi R, Martinez C, Lecchi A, et al. (2003) Molecular bases of defective signal transduction in the platelet P2Y12 receptor of a patient with congenital bleeding. Proc Natl Acad Sci U S A 100: 1978–1983.
- 8. Remijn JA, MJ IJ, Strunk AL, Abbes AP, Engel H, et al. (2007) Novel molecular defect in the platelet ADP receptor P2Y12 of a patient with haemorrhagic diathesis. Clin Chem Lab Med 45: 187–189.
- 9. Bertrand ME, Simoons ML, Fox KA, Wallentin LC, Hamm CW, et al. (2002) Management of acute coronary syndromes in patients presenting without persistent ST-segment elevation. Eur Heart J 23: 1809–1840.
- 10. Wiviott SD, Braunwald E, McCabe CH, Montalescot G, Ruzyllo W, et al. (2007) Prasugrel versus Clopidogrel in Patients with Acute Coronary Syndromes. N Engl J Med.
- 11. Fontana P, Dupont A, Gandrille S, Bachelot-Loza C, Reny JL, et al. (2003) Adenosine diphosphate-induced platelet aggregation is associated with P2Y12 gene sequence variations in healthy subjects. Circulation 108: 989–995.
- 12. Fontana P, Gaussem P, Aiach M, Fiessinger JN, Emmerich J, et al. (2003) P2Y12 H2 haplotype is associated with peripheral arterial disease: a case-control study. Circulation 108: 2971–2973.
- 13. Ziegler S, Schillinger M, Funk M, Felber K, Exner M, et al. (2005) Association of a functional polymorphism in the clopidogrel target receptor gene, P2Y12, and the risk for ischemic cerebrovascular events in patients with peripheral artery disease. Stroke 36: 1394–1399.
- 14. Cavallari U, Trabetti E, Malerba G, Biscuola M, Girelli D, et al. (2007) Gene sequence variations of the platelet P2Y12 receptor are associated with coronary artery disease. BMC Med Genet 8: 59.
- 15. Cuisset T, Frere C, Quilici J, Morange PE, Saut N, et al. (2007) Role of the T744C polymorphism of the P2Y12 gene on platelet response to a 600-mg loading dose of clopidogrel in 597 patients with non-ST-segment elevation acute coronary syndrome. Thromb Res.
- 16. Schettert IT, Pereira AC, Lopes NH, Hueb WA, Krieger JE (2006) Association between platelet P2Y12 haplotype and risk of cardiovascular events in chronic coronary disease. Thromb Res 118: 679–683.
- 17. von Beckerath N, von Beckerath O, Koch W, Eichinger M, Schomig A, et al. (2005) P2Y12 gene H2 haplotype is not associated with increased adenosine diphosphate-induced platelet aggregation after initiation of clopidogrel therapy with a high loading dose. Blood Coagul Fibrinolysis 16: 199–204.
- 18. Lee DK, Nguyen T, Lynch KR, Cheng R, Vanti WB, et al. (2001) Discovery and mapping of ten novel G protein-coupled receptor genes. Gene 275: 83–91.
- 19. Criqui MH, Langer RD, Fronek A, Feigelson HS, Klauber MR, et al. (1992) Mortality over a period of 10 years in patients with peripheral arterial disease. N Engl J Med 326: 381–386.
- 20. McCullough PA (2007) Coronary artery disease. Clin J Am Soc Nephrol 2: 611–616.
- 21. Amisten S, Melander O, Wihlborg AK, Berglund G, Erlinge D (2006) Increased risk of acute myocardial infarction and elevated levels of C-reactive protein in carriersof the Thr-87 variant of the ATP receptor P2Y11. Eur Heart J.
- 22. Manjer J, Carlsson S, Elmstahl S, Gullberg B, Janzon L, et al. (2001) The Malmo Diet and Cancer Study: representativity, cancer incidence and mortality in participants and non-participants. Eur J Cancer Prev 10: 489–499.
- 23. Orho-Melander M, Klannemark M, Svensson MK, Ridderstrale M, Lindgren CM, et al. (2002) Variants in the calpain-10 gene predispose to insulin resistance and elevated free fatty acid levels. Diabetes 51: 2658–2664.
- 24. Barrett JC, Fry B, Maller J, Daly MJ (2005) Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 21: 263–265.
- 25. Skol AD, Scott LJ, Abecasis GR, Boehnke M (2006) Joint analysis is more efficient than replication-based analysis for two-stage genome-wide association studies. Nat Genet 38: 209–213.
- 26. Coade SB, Pearson JD (1989) Metabolism of adenine nucleotides in human blood. Circ Res 65: 531–537.
- 27. Bergfeld GR, Forrester T (1992) Release of ATP from human erythrocytes in response to a brief period of hypoxia and hypercapnia. Cardiovasc Res 26: 40–47.
- 28. Ramensky V, Bork P, Sunyaev S (2002) Human non-synonymous SNPs: server and survey. Nucleic Acids Res 30: 3894–3900.
- 29. Leng GC, Lee AJ, Fowkes FG, Whiteman M, Dunbar J, et al. (1996) Incidence, natural history and cardiovascular events in symptomatic and asymptomatic peripheral arterial disease in the general population. Int J Epidemiol 25: 1172–1181.
- 30. Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, Ramirez C, Cavallari U, et al. (2005) Lack of association between the P2Y12 receptor gene polymorphism and platelet response to clopidogrel in patients with coronary artery disease. Thromb Res 116: 491–497.
- 31. Hetherington SL, Singh RK, Lodwick D, Thompson JR, Goodall AH, et al. (2005) Dimorphism in the P2Y1 ADP receptor gene is associated with increased platelet activation response to ADP. Arterioscler Thromb Vasc Biol 25: 252–257.