OUP user menu

Expression of platelet collagen receptor glycoprotein VI is associated with acute coronary syndrome

Boris Bigalke, Stephan Lindemann, Raila Ehlers, Peter Seizer, Karin Daub, Harald Langer, Tanja Schonberger, Elisabeth Kremmer, Dorothea Siegel-Axel, Andreas E. May, Meinrad Gawaz
DOI: http://dx.doi.org/10.1093/eurheartj/ehl192 2165-2169 First published online: 21 August 2006


Aims Platelet collagen receptor glycoprotein VI (GPVI) is critical for the formation of arterial thrombosis. In this observational study, we examined the platelet surface expression of GPVI in patients with symptomatic coronary artery disease (CAD).

Methods and results We evaluated a consecutive cohort of 367 patients with symptomatic CAD, who underwent coronary angiography. The surface expression of platelet activation markers (GPVI, CD62P, and CD42b) was determined by flow cytometry. Patients with acute coronary syndrome (ACS) showed a significantly enhanced GPVI expression on admission when compared with patients with stable angina pectoris (SAP) (ACS: 21.4±9.7; SAP: 18.6±7.1 mean fluorescence intensity±SD; P=0.004). The expression of GPVI correlated with CD62P (r=0.702; P=0.001). Logistic regression analysis demonstrated that on admission, elevated platelet GPVI expression was associated with ACS, independent of markers of myocardial necrosis such as troponin and creatine kinase.

Conclusion Platelet GPVI surface expression is elevated in patients with ACS and is associated with imminent acute coronary events. The determination of the platelet-specific thrombotic marker GPVI may help to identify patients at risk before myocardial ischaemia is evident.

  • Platelets
  • Glycoprotein VI
  • Collagen
  • Acute coronary syndrome


Exposure of collagen to the bloodstream occurs at the site of plaque rupture and initiates platelet activation and platelet-mediated thrombus formation.1 Previously, several platelet collagen receptors have been identified, and it has become clear by now that collagen binding to glycoprotein VI (GPVI) is the initial trigger of platelet activation and aggregation at the site of vascular injury where extracellular matrix components, including collagen, are exposed.24 GPVI-mediated platelet activation leads to firm platelet adhesion and secretion of procoagulative and proinflammatory compounds, which, in turn, aggravate and consolidate thrombus formation.2 Further, collagen binding to platelets is followed by the expression of phosphatidylserine on the platelet membrane. Phosphatidylserines provide prothrombinase activity that enhances plasmatic thrombin formation.2 Moreover, the adhering platelets stay alive for several hours or days at the site of vascular lesions and initiate de novo protein synthesis and interleukin-1 beta release through shedding of microvesicles,57 thereby inflammatory activity around a platelet thrombus.

Recently, GPVI has been described to be the major collagen receptor mediating platelet adhesion and aggregation on human atherosclerotic plaque tissues.8,9 Inhibition of GPVI results in the substantial attenuation of platelet adhesion to atherosclerotic tissue.8 Although GPVI is constitutively surface-expressed, activation of platelets leads to further release and enhanced plasma membrane expression of GPVI.2,10 Preliminary data suggest that platelet collagen receptor GPVI density may be related to myocardial infarction.11 Thus, an altered GPVI expression level on circulating platelets may indicate a prothrombotic disease state such as imminent myocardial infarction.

The present work evaluates the expression levels of platelet GPVI in a large cohort of patients with symptomatic coronary artery disease (CAD). The data imply that patients with enhanced platelet GPVI are at increased risk for acute coronary events. Determination of GPVI levels may be an early marker of acute coronary syndrome (ACS) before myocardial ischaemia is evident.


Study population and enrolment criteria

Our study population comprised 367 consecutive patients, in whom CAD was suspected and thus received coronary angiography: 240 patients with stable angina pectoris (SAP) and 127 patients with ACS/MI.

The exclusion criteria were age below 18 years, lack of informed consent, and false positive elevation of creatine kinase (CK) and CKMB. Twenty-six patients were excluded, among these, 19 refused to give their consent and the other seven patients were excluded because of exclusion criteria.


SAP: stable angina pectoris, symptoms without progression and stable intensity.

ACS: episodes of >20 min of angina pectoris with progress over time or with recurrent episodes at rest or with minimal physical strain within the last 24 h. The ECG shows at least one of the following criteria: new ST-depression of at least 0.05 mV, transient (<20 min) ST-elevation of at least 0.1 mV, or negative T of at least 0.3 mV in two or more leads of standard electrocardiogram. CK and troponin-I (Tn-I) levels may be increased, but can be normal as well. If CK and Tn-I levels are increased, we also have a STEMI or NSTEMI, depending on the ECG criteria (TN+ACS, Figure 1B).

Figure 1

Platelet collagen receptor GPVI expression is elevated in patients with ACS. (A) Platelets of patients with SAP, ACS, and healthy control subjects (combined control level for GPVI; 15.8±4.3; n=45; dotted line) were stained for the collagen receptor (GPVI) and P-selectin (CD62P) using monoclonal antibodies conjugated with either FITC or PE. Two-colour whole blood flow cytometry was performed according to the standardized protocol and the results were expressed as MFI. (B) Surface expression of platelet GPVI and P-selectin (CD62P) in patients with Tn-I- positive ACS (Tn+ACS) and with myocardial infarction without/with ST-segment elevation (NSTEMI/STEMI). P-value indicates statistical significance vs. SAP in (A). (C) Correlation of platelet GPVI and CD62P surface expression.

STEMI: episodes of >20 min of angina pectoris with the following ECG and laboratory criteria: ST-segment elevation in two or more leads with at least 0.1 mV or a two-fold elevation of CK when compared with the normal level and with a significant CKMB or an increased Tn-I level.

NSTEMI: episodes of >20 min of angina pectoris with the following ECG and laboratory criteria: no ST-segment elevation in the standard ECG. Two-fold elevation of CK when compared with the normal level and with a significant CKMB or an increased Tn-I.


First, we wanted to check the platelet GPVI expression of apparently healthy controls. The personnel of our lab were investigated (22 men and eight women, age 31.8±8.3 years, no medication, and except for four smokers, no cardiovascular risk factors). These controls did not receive coronary angiography.

We then examined a group of subjects in whom CAD was excluded. These patients (nine men and six women, age 57.4±9.4 years) were angiographied because of chest pain but did not have CAD. The patients had the following cardiovascular risk profiles: hypertension (6x), diabetes (1x), family history for CAD (1x), hypercholesterolaemia (1x), and smoking (1x). Medical treatment includes β-blockers (5x), ASS (1x), statins (1x), and ACE-inhibitors (1x).

Sample collection

Arterial blood was drawn from the sheath that was introduced into the femoral artery at the beginning of cardiac catheterization. Cardiac catheterization was started immediately after admission to the hospital in patients with ACS. Patients with SAP were investigated on schedule in the morning. In the apparently healthy controls, the blood was obtained by venipuncture of the right anticubital vein. The blood was filled into 5 mL citrate phosphate dextrose adenine vials and analysed by flow cytometry.

FACS analysis

The surface expression of the platelet receptors GPVI, CD62P, and GPIb was determined by two-colour whole blood flow cytometry, as described previously.12,13 Each measurement was performed in duplicate and mean fluorescence intensity (MFI) was used as the index of receptor expression. Fluorescein (FITC)-conjugated anti-CD62P (clone CLB-Thromb/6) and phycoerythrin (PE)-conjugated anti-CD42b (clone SZ2) monoclonal antibodies (mAb) were purchased from Immunotec, Beckman Coulter, Inc., USA. The anti-GPVI mAb 4C9 was generated and characterized, as described previously.10,14


Pairwise comparisons of ACS, Tn+, NSTEMI, and STEMI with SAP were evaluated with appropriate non-parametric tests (the Kruskall–Wallis test). For corrections of multiple testing, a Bonferroni–Holm correction was applied. A probability value of less than 0.05 was considered as statistically significant. Comparison between GPVI and clinical parameters was made by the Spearman correlation. Adjustment by possible cofounders was performed by the multifactorial analysis of covariance for the decadic logarithm of GPVI. Association of the platelet GPVI surface expression was found to be independent of medical treatment at the time of admission, classical cardiovascular risk factors, and laboratory markers. The association of GPVI after adjustment for conventional laboratory markers was assessed by the logistic regression analysis. All statistical analyses were performed using SPSS version 13.


We consecutively investigated the platelet surface expression of collagen receptor GPVI in a total of 367 patients with symptomatic CAD. In all patients, the diagnosis was verified and the severity of the disease was assessed by coronary angiography. The demographic details are given in Table 1.

View this table:
Table 1

Baseline patients'characteristics and medical treatment on admission

CharacteristicsTotal (n=367)SAP (n=240)ACS (n=127)
Age, years (±SD)67.1±10.166.9±9.467.5±11.3
Sex, no. (%)
 Female96 (26.2)59 (24.6)37 (29.1)
 Male271 (73.8)181 (75.4)90 (70.9)
Cardiovascular risk factors, no. (%)
 No risk factors31 (8.4)17 (7.1)14 (11)
 Arterial hypertension260 (70.8)176 (73.3)84 (66.1)
 Hyperlipidaemia201 (54.8)151 (62.9)50 (39.4)
 Diabetes134 (36.5)88 (36.7)46 (36.2)
 Family history of CAD88 (24.0)59 (16.1)29 (22.8)
 Smoking116 (31.6)75 (24.6)41 (32.3)
CAD, no. (%)
 One vessel83 (22.6)47 (19.6)36 (28.4)
 Two vessels117 (31.9)80 (33.3)37 (29.1)
 Three vessels167 (45.5)113 (47.1)54 (42.5)
LVEF, no. (%)
 Normal201 (54.8)135 (56.3)66 (52)
 Slightly reduced92 (25.1)61 (25.4)31 (24.4)
 Moderate47 (12.8)27 (11.2)20 (15.7)
 Low27 (7.4)17 (7.1)10 (7.9)
Medication, no. (%)
 No medication113 (30.8)52 (21.7)61 (48)
 ACE-inhibitors174 (47.4)129 (53.8)45 (35.4)
 Angiotensin receptor blockers30 (8.2)28 (11.7)2 (1.6)
 Beta-blockers216 (58.9)165 (68.8)51 (40.2)
 Statins204 (55.6)159 (66.3)45 (35.4)
 Aspirin216 (58.9)162 (67.5)54 (42.5)
 Clopidogrel75 (20.4)63 (26.3)12 (9.4)
 Vitamin K antagonist21 (5.7)15 (6.3)6 (4.7)

The platelet surface expression of GPVI was significantly enhanced in patients with ACS when compared with patients with stable CAD (SAP) [ACS (MFI±SD): 21.4±9.7; SAP: 18.6±7.1; P=0.004] (Figure 1A). Patients with acute myocardial infarction (particularly STEMI) tended to reveal the highest GPVI expression levels compared with patients with minor myocardial ischaemia (troponin positive ACS), although this did not reach statistical significance [Tn pos. ACS (n=57): 21.6±9.2; NSTEMI (n=51): 20.7±8.2; P=0.672; STEMI (n=19): 24±14.2; P=0.894] (Figure 1B). Similarly, CD62P surface expression was significantly elevated in patients with ACS compared with SAP [ACS (MFI±SD): 15.4±11.4; SAP: 13.5±8.3; P=0.036] (Figure 1A and B). Platelet GPVI correlated with CD62P surface expression levels (r=0.702; P=0.001) (Figure 1C). In order to obtain the information about normal platelet GPVI expression in apparently healthy people, we investigated our lab personnel (n=30; MFI±SD: 15.6±4.7). These controls did not receive coronary angiography. We then examined a group of subjects in whom CAD was excluded. These patients were angiographied because of chest pain but did not have CAD (n=15; MFI±SD: 16.2±2.7). The mean MFI for GPVI expression of both controls combined was 15.8±4.3 (n=45; Figure 1, dotted line).

Next, we asked whether platelet GPVI correlates with markers of myocardial ischaemia. We found that GPVI levels determined on admission did not correlate with the respective initial Tn-I or CK values (initial Tn: r=0.035, P=0.567; initial CK: r=0.069, P=0.232) (Figure 2A). In contrast, GPVI levels correlated with maximum Tn-I or CK values determined later on during hospital stay (maximum Tn: r=0.178, P=0.031; maximum CK: r=0.181, P=0.047) (Figure 2A). No correlation was found for GPVI and C-reactive protein levels.

Figure 2

Correlation of platelet GPVI with markers of myocardial injury. (A) Correlation of platelet GPVI surface expression with initial Tn-I and initial CK at the time of hospital admission. (B) Correlation of platelet GPVI surface expression on admission with maximum values of troponin and CK determined in the time course after hospital admission.

Comparison of the decadic logarithm of GPVI between ACS and SAP was adjusted by possible cofounders such as medical treatment and cardiovascular risk factors (Table 2). To test whether GPVI is associated with ACS independent of baseline conventional laboratory markers of ACS, we performed a logistic regression analysis that included Tn-I, CK, and C-reactive protein determined at the time of admission (Table 3). We found that the platelet surface expression of GPVI was independently associated with ACS (Table 3).

View this table:
Table 2

Multifactorial analysis of covariance for the logarithm of GPVI

Angiotensin receptor blockers0.844
Vitamin K antagonist0.586
Cardiovascular risk factorsArterial hypertension0.713
History of CAD0.370
Conventional laboratory markersInitial troponin I (ng/mL)0.110
Initial CK (U/L)0.484
Initial C-reactive protein (mg/dL)0.216
GroupsACS vs. SAP0.024
View this table:
Table 3

Association of GPVI after adjustment for conventional laboratory markers

ParametersP-valueOR95% CI for OR
GPVI (MFI)0.0131.0211.0211.189
Initial Tn-I (ng/mL)0.1671.9660.7545.124
Initial CK (U/L)0.0251.0071.0011.014
Initial C-reactive protein (mg/dL)0.8821.0090.8991.132


The present study is an observational study of 367 consecutive patients with SAP or ACS. We demonstrated that the platelet GPVI expression levels are significantly elevated in patients with ACS when compared with those with SAP. The platelet GPVI surface expression determined on admission is associated with ACS, independent of markers of myocardial ischaemia (troponin and CK). Enhanced GPVI levels on hospital admission are associated with the development of ACS and myocardial infarction. The findings of the present work imply that platelet GPVI may be an early biomarker for the identification of patients with ACS. Because GPVI levels may be elevated when markers of myocardial necrosis are still within the normal range in patients admitted to hospital with ACS, it is tempting to speculate that a platelet-specific thrombotic marker may help to define patients at risk before myocardial ischaemia is evident.

Evaluation of patients who present at the hospital with a complaint of chest pain or other signs or symptoms suggestive of ACS is time-consuming and problematic. Recent investigations have indicated that increases in biomarkers upstream from biomarkers of necrosis (cardiac troponins I and T), such as markers of inflammation (e.g. C-reactive protein and CD40L) or myocardial stretch (BNP), may provide an earlier assessment of overall patient risk and help in identifying patients with higher risk of an adverse event.

Platelet activation and platelet-mediated thrombus formation at the site of plaque rupture are the initial trigger events in ACS.15 At the site of plaque rupture, subendothelial extracellular matrix proteins, namely collagen, are exposed to the blood stream that are rapidly covered by a layer of activated platelets.1 GPVI is the major collagen receptor that mediates platelet adhesion and activation to fibrillar collagen.2,14,16 GPVI expression is limited to the megakaryocyte system and is, thus, platelet-specific.2

Limited data are available, to date, that document changes in GPVI expression levels on circulating platelets in humans. Previously, we found that GPVI is altered in type 2 diabetes.10 Others found that GPVI levels are enhanced in patients who suffered from myocardial infarction.11 Genetic elements seem to play an important role in the development of coronary thrombosis and fatal myocardial infarction, as demonstrated by a study on GPVI polymorphism.17 We did not investigate our cohort for GPVI polymorphisms, but GPVI polymorphisms may significantly contribute to the risk for a coronary event independently of platelet GPVI expression. In the present study, we used a specific anti-GPVI monoclonal antibody that was developed for flow cytometric analysis, as described recently.10 We found in a large consecutive cohort of patients with symptomatic CAD that the platelet surface expression of GPVI is enhanced in patients with ACS and myocardial infarction compared with those with stable CAD. Moreover, we found that the GPVI expression levels determined on admission are associated with ACS, independent of myocardial necrosis marker such as troponin or CK. Our data strengthen the concept that platelet activation occurs in ACS and is critical for disease progression. Further, we provide evidence that a platelet-specific activation marker that plays a critical pathophysiological role in ACS (GPVI/collagen interaction) may be a promising candidate for a biomarker for risk stratification in patients with ACS. Determination of elevated GPVI levels may indicate a prothrombotic state in patients developing ACS and may serve as an early biomarker of imminent myocardial ischaemia.


We thank Heike Runge, Iris Schäfer, Felix Waibel, and Andreas Schönfeld for expert technical assistance.

Conflict of interest: none declared.


View Abstract