The C807T/G873A polymorphism in the platelet glycoprotein Ia gene and the risk of acute coronary syndrome in the Italian population
Professor Valerio De Stefano, Istituto Semeiotica Medica, Università Cattolica, Largo Gemelli 8, 00168 Rome, Italy. E-mail: email@example.com
Membrane glycoprotein (GP) Ia/IIa mediates platelet adhesion to collagen. The linked C807T/G873A polymorphisms in the GP Ia gene are correlated with a variable expression of the platelet surface receptor, the 807 TT/873 AA genotype being associated with a higher receptor density. Our study aimed to evaluate the possible role of the GP Ia C807T/G873A polymorphism as a risk factor for acute coronary syndrome in the Italian population. We investigated 157 patients with acute coronary syndrome (117 with myocardial infarction and 40 with severe unstable angina) as the first manifestation of coronary disease occurring before 65 years of age, compared with 312 healthy controls. All individuals were of Italian ancestry and were genotyped for the GP Ia C807T/G873A polymorphism. Complete linkage between the 807 and 873 sites was found in all samples. The 807 TT genotype was present in 12·7% of cases and in 4·8% of controls; the odds ratio for acute coronary syndrome was 2·9 (95% CI 1·4–5·8) for the 807 TT genotype compared with C-allele carriers and 0·6 (95% CI 0·4–0·9) for the 807 CC genotype compared with T-allele carriers. For the TT genotype, compared with CC homozygotes, the increase in risk was 3·4-fold in patients with at least one risk factor (smoking, hypercholesterolaemia, diabetes, systemic hypertension) and 4·1-fold in patients with angiographically diagnosed two- or three-vessel disease. We conclude that the GP Ia 807 TT (873 AA) genotype is associated with an increased risk of acute coronary syndrome in the Italian population; conversely, the GP Ia 807 CC (873 GG) genotype seems to represent a protective factor.
Platelet thrombi play a crucial role in the development of acute coronary syndromes, as demonstrated by histopathological findings (Fuster et al, 1992) and by clinical observations showing the efficacy of antiplatelet therapies for coronary artery disease (Antiplatelet Trialists' Collaboration, 1994). Genetic platelet variations may confer a potent risk of coronary thrombosis. If so, detection of these variants may be used to direct therapeutic decisions (Nurden, 1995).
The glycoprotein (GP) Ia/IIa is a heterodimeric platelet membrane complex that mediates platelet adhesion to collagen (Nieuwenhuis et al, 1985; Kunicki et al, 1988). Recently, two linked silent GPIa dimorphisms, 807 C→T (Phe224) and 873 G→A (Thr246) were correlated with a variable expression of the platelet surface receptor: the genotype 807 TT (873AA) was associated with a higher receptor density and the genotype 807 CC (873 GG) with a lower density, whereas heterozygous individuals expressed intermediate receptor levels (Kunicki et al, 1993, 1997; Corral et al, 1999).
Given the importance of the GP Ia/IIa in primary haemostasis, it is reasonable to suggest that, in certain circumstances, inherited differences in the platelet surface concentration of this receptor may contribute to an increased risk of either thrombosis or bleeding. The scarce data presently available on the association between these GP Ia/IIa dimorphisms and the risk of coronary artery disease have provided conflicting results (Corral et al, 1999; Croft et al, 1999; Moshfegh et al, 1999; Santoso et al, 1999).
In the present study, we evaluated the prevalence of the GPIa C807T/G873A dimorphism in a population of Italian patients with acute coronary syndrome (ACS) as the first manifestation of ischaemic heart disease before the age of 65 years, compared with healthy individuals.
Patients and methods
Subjects We performed a 1:2 case–control study. All patients and controls were of Italian ancestry from Central and Southern Italy. There were 157 patients (131 men and 26 women) with acute myocardial infarction (MI, n = 117) or severe unstable angina (UA, n = 40) as the first manifestation of ischaemic heart disease occurring before the age of 65 years (mean age 53 years, median 54, range 24–65) consecutively admitted to the Cardiology Department of the Catholic University Hospital (Policlinico A. Gemelli, Rome). Patients with MI satisfied the World Health Organization criteria on the basis of symptoms, enzyme elevations and electrocardiographic changes. Patients with UA fell into Braunwald's class IIIB (Braunwald, 1989). At the time of enrolment, a detailed history was taken to assess cardiovascular risk factors. Hypercholesterolaemia, diabetes mellitus and systemic hypertension were considered to be present if drugs for these conditions had been prescribed before admission or if these conditions were diagnosed during hospitalization. Smokers were defined as current smokers of ≥ 1 cigarette/d. Patients were defined as having a family history of ischaemic heart disease in the case of a documented episode of ACS having occurred before 60 years of age in at least one first-degree relative.
Of the 157 patients, 105 underwent coronary angiography for clinical reasons. Significant vessel disease was defined as a lumen diameter stenosis ≥ 50% in ≥ 1 major coronary artery.
The 312 control subjects were healthy individuals without a history of arterial or venous thromboembolic events recruited among blood donors (n = 117), members of the hospital staff (n = 100) and out-patients of the Department of Dermatology (n = 95) of the Policlinico A. Gemelli. Clinical history and physical examination were carried out to rule out hypertension, diabetes or dyslipidaemia. The mean age was 47 years (median 49, range 24–93).
DNA genotyping Genomic DNA was extracted from peripheral blood leucocytes according to standard procedures. Genotyping was conducted using allele-specific polymerase chain reaction (PCR) according to Dinauer et al (1999). In brief, the 807T and 873G amplicons were multiplexed in one reaction and the 807C and 873A amplicons were multiplexed in a second reaction. Genomic DNA (100 ng) was added to a 50-μl reaction mixture containing 1X Perkin-Elmer PCR buffer II (10 mmol/l Tris-HCl, pH 8·3, 50 mmol/l KCl), 2·0 mmol/l MgCl2, 0·2 mmol/l dNTP (Pharmacia), 125 ng of Intron G reverse primer (5′-GATTTAACTTTCCCAGCTGCCTTC-3′), 125 ng of Exon 8 reverse primer (5′-CTCAGTATATTGTCATGGTTG CATTG-3′) and 0·5 U of Taq polymerase. Allele-specific primers in the first reaction were 125 ng of 807C forward primer (5′-GTGGGGACCTCACAAACACATGC-3′) and 187 ng of 873A forward primer (5′-GGTGGGCGAC GAAGTGCTAGA-3′). Allele-specific primers in the second reaction were 187 ng of 807T forward primer (5′-ATGGTGGGGACCTCACAAACACATAT-3′) and 125 ng of 873G forward primer (5′-GGTGGGCGACGAAGTGCTAGG 3′). Reactions were amplified in a Perkin-Elmer 9600 thermal cycler using 35 cycles, as follows: denaturation at 94°C for 20 s, primer annealing at 64°C for 30 s and extension at 72°C for 30 s.
The PCR products were analysed on 2% agarose gels and visualized using ethidium bromide staining.
Statistical methods Differences between patients and controls were evaluated using Fisher's exact test or the chi-square test when appropriate. The odds ratios compared with controls or the relative risks among different groups of patients were calculated by simple cross-tabulation, with 95% confidence intervals (CI).
A smoking habit was present in 91 patients (58%), systemic hypertension in 73 (46%), hypercholesterolaemia in 67 (43%) and diabetes mellitus in 25 (16%) (Table I). In 142 cases (90%) at least one risk factor was present: there was only one risk factor in 51 cases, two risk factors in 62 cases, and three or four risk factors in 29 cases. In 56 cases (36%) a family history of ischaemic heart disease was documented.
Table I. Genotype distribution among the 157 investigated patients and among the patient subgroups with different risk factors for acute coronary syndrome.
|All||20 (12·7)||77 (49·0)||60 (38·2)||157 (100)|
|Smoking habit||12 (13·2)||42 (46·1)||37 (40·7)||91 (58)|
|Systemic hypertension||11 (15·1)||33 (45·2)||29 (39·7)||73 (46)|
|Hypercholesterolaemia||10 (15·5)||31 (46·3)||26 (38·9)||67 (43)|
|Diabetes mellitus||2 (8·0)||16 (64·0)||7 (28·0)||25 (16)|
|Family history of ischaemic heart disease||6 (10·8)||25 (44·6)||25 (44·6)||56 (36)|
In all cases and controls the 807T genotype was associated with the 873A genotype, confirming the complete linkage disequilibrium between 807C/873G and 807T/873A, in agreement with previous reports (Reiner et al, 1998; Moshfegh et al, 1999). We therefore analysed our data according to the 807 genotype distribution only.
The GP Ia genotype distribution in patients versus controls was 12·7% versus 4·8% for 807 TT (P = 0·004), 49% versus 45·5% for 807 CT (P = 0·49) and 38·2% versus 49·7% for 807 CC (P = 0·02). The distribution of the different genotypes among patients stratified according to the presence of risk factors is shown in Table I, with no significant difference in distribution among the different risk factor groups (P = 0·85) or between the latter and the whole patient population (P = 0·93). There was no significant difference in genotype distribution between male or female patients (P = 053 for the TT genotype, P = 0·66 for the CC genotype), between patients above or below 50 years of age (P = 0·81 for the TT genotype, P = 0·86 for the CC genotype), or between patients with or without a family history of ischaemic heart disease (P = 0·62 for the TT genotype, P = 0·22 for the CC genotype). A significant difference in genotype distribution was found between patients with a diagnosis of MI and those with a diagnosis of UA, the TT genotype being present in 17·1% of the patients with MI and in none of the patients with UA (P = 0·002), and the CC genotype being present in 33·3% of the patients with MI and in 52·5% of the patients with UA (P = 0·03).
The odds ratio for ACS among TT homozygotes compared with C-allele carriers (CT heterozygotes plus CC homozygotes) was 2·9 (95% CI 1·4–5·8), and among TT homozygotes compared with CC homozygotes was 3·4 (95% CI 1·6–7·2) (Table II).
Table II. Odds ratio (OR) values in the different subgroups of patients analysed.
|All patients||2·9 (1·4–5·8)||3·4 (1·6–7·2)||0·6 (0·4–0·9)|
|Patients without RF||3·0 (0·6–14·7)||4·1 (0·7–23·2)||0·5 (0·2–1·5)|
|Patients with at least one RF||2·9 (1·4–5·9)||3·4 (1·6–7·2)||0·6 (0·4–0·9)|
|Patients with zero-/one-vessel disease||1·7 (0·5–5·4)||2·1 (0·6–6·8)||0·7 (0·4–1·2)|
|Patients with two-/three-vessel disease||3·4 (1·3–8·4)||4·1 (1·5–11·0)||0·6 (0·3–1·0)|
The 807T allele frequency was 0·37 in the patient group and 0·27 in the control group (P = 0·002), yielding an allelic odds ratio for ACS of 1·6 (95% CI 1·2–2·1). Conversely, the frequency of the 807C allele was 0·63 in the patient group and 0·73 in the control group, with an allelic odds ratio for ACS of 0·6 (95% CI 0·5–0·8), suggesting that carriership of the GP Ia 807C allele may be a protective factor. Accordingly, the odds ratio for ACS associated with the CC genotype compared with T-allele carriers (CT heterozygotes plus TT homozygotes) was 0·6 (95% CI 0·4–0·9) (Table II).
Subgroup analyses according to the presence or absence of smoking, dyslipidaemia, hypertension or diabetes revealed that the TT genotype compared with the CC genotype was associated with a 3·4-fold increase in risk for ACS among the patients with at least one risk factor, compared with controls (Table II). The increase in risk was of similar magnitude (4·1-fold) among the patients with no risk factors; however, in this latter group the increase in risk did not reach statistical significance, probably as a result of the low number of cases. The CC genotype compared with all other genotypes was associated with a significant reduction in risk only in the patient group with ≥ 1 risk factor (Table II). However, direct comparison between cases with or without risk factors showed a similar relative risk (data not shown).
On the basis of coronary angiography, patients with two- (n = 33) or three- (n = 22) vessel disease were compared with those with single (n = 38) or non-significant (n = 12) vessel disease. With respect to the control group, the 55 patients with two- or three-vessel disease had significantly increased odds ratios for ACS associated with the TT genotype compared with both CC homozygotes (4·1-fold) and C-allele carriers (3·4-fold). In this group of patients, the CC genotype was associated with a lower odds ratio than controls (0·6) (Table II). In the group of 50 patients with single or non-significant vessel disease, no significant association was found between the TT or CC genotype and the risk for ACS. Direct comparison of patients with two- or three-vessel disease against patients with single or non-significant vessel disease showed a similar relative risk (data not shown).
GP Ia/IIa is a major collagen receptor on the platelet surface and variations in its density are related to differences in platelet adhesion to collagen (Kunicki et al, 1993); receptor density is also related to the linked GP Ia C807T/G873A polymorphism, the 807 TT (873 AA) genotype being associated with a higher density (Kunicki et al, 1997; Corral et al, 1999). The role of the GP Ia 807T allele as a possible determinant of arterial occlusive disease has been recently investigated, with conflicting results (Carlsson et al, 1999; Corral et al, 1999; Croft et al, 1999; Moshfegh et al, 1999; Santoso et al, 1999). In a study of Swiss individuals by Moshfegh et al (1999), the prevalence of the 807 TT genotype in 177 patients with myocardial infarction compared with 89 healthy controls was 16·4% and 5·6%, respectively, with a 3·3-fold increase in the risk of MI (Table III). In contrast, Croft et al (1999) failed to find an increased risk for MI associated with the GP Ia C807T polymorphism in a population from the United Kingdom, the prevalence of the TT genotype being 17% among cases and 19% among controls (Table III). Similar results were reported by Corral et al (1999) in a Spanish population of patients with heart artery disease, the prevalence of the TT genotype being 9·9% among cases and 17·8% among controls (Table III). In a study of 2237 German male patients who underwent coronary angiography, the prevalence of the TT and CC genotypes was similar in subjects with MI (14% TT and 32% CC) or without MI (14% TT and 35% CC). However, because of the lack of a healthy control group, no conclusion can be drawn from this latter report about the prevalence of the 807T allele in the general population and the associated risk of MI (Santoso et al, 1999).
Table III. Genotype prevalences of the C807T polymorphism of the GP Ia gene in Caucasian controls and in patients reported in different studies.
|Switzerland (Moshfegh et al, 1999)||177/89||16·4*||5·6||Not reported||41·0|
|UK (Croft et al, 1999)||546/547||17·0*||19·0||37·0*||33·0|
|Germany (Carlsson et al, 1999)||227/184||Not reported||15·8||30·4‡||37·5|
|USA (Reiner et al, 1998)||0/146||Not investigated||11·6||Not investigated||44·5|
|USA (Dinauer et al, 1999)||331/2957||16·0¶||15·0||38·0¶||36·0|
|USA (Hispanic) (Dinauer et al, 1999)||0/92||Not investigated||27·0||Not investigated||26·0|
|Spain (Corral et al, 1999)||0/284||Not compared||15·5||Not compared||45·4|
|Spain (Corral et al, 1999)||101/101|| 9·9†||17·8||44·6†||39·6|
|Spain (Corral et al, 1999)||104/104||18·3‡||13·5||35·6‡||41·4|
|Spain (Corral et al, 1999)||97/97||16·5¶||15·5||35·1¶||35·1|
|Italy (Sacchi et al, 2000)||70/125||27·0§||10·0||29·0§||38·0|
|Italy (present study)||157/312||12·7||4·8||38·2||49·7|
One reason for these conflicting results may be the different distribution of the GP Ia C807T polymorphism among various racial groups (Dinauer et al, 1999). Indeed, among healthy Caucasian individuals enrolled in seven other studies, the prevalence of the TT and CC genotype ranged from 5·6% to 27% and from 26% to 45% respectively (Table III). In our study, the prevalence of the TT genotype in the patient group (12·7%) was similar to that observed in the previous studies (9·9% to 17·8%), whereas the prevalence of the TT genotype in the control group was 4·8%, which is similar to that observed in the Swiss study (Moshfegh et al, 1999) but lower than that observed in other studies carried out on Caucasian populations (Table III). A recent preliminary investigation conducted in Italy reported a prevalence of the TT genotype of 10% among healthy controls (Sacchi et al, 2000), not very different from our findings; using that control group, patients with ischaemic stroke before the age of 50 years showed a significant 3·8-fold increase in risk associated with the TT genotype (Table III); on the other hand, no significant difference in risk was found between 177 patients with MI and 147 controls, whose genotype distribution was not reported in extenso (Sacchi et al, 2000).
In conclusion, the GP Ia 807 TT genotype in the Italian population seems to be associated with a 3·4-fold increase in the risk of ACS compared with the 807 CC genotype, whereas carriership of the 807 CC genotype compared with the other genotypes approximately halves such risk. The TT genotype was over-represented among patients with MI compared with patients with UA; whether such a genotype is associated with the more severe forms of ischaemic heart disease deserves further investigation. The contemporary presence of traditional cardiovascular risk factors did not enhance the risk, nor was the TT genotype over-represented among patients with a family history of ischaemic heart disease. Future investigations in larger cohorts of patients selected according to the presence or the absence of such criteria will be necessary to further define the role of the GP Ia genotype in the pathogenesis of acute coronary syndromes.
We are indebted to Dr Martin J. Hessner for providing the methodology for genotyping the C807T/G873A GP 1a polymorphism and for his valuable technical advice.