(Circulation. 2001;104:140.)
© 2001 American Heart Association, Inc.
Clinical Investigation and Reports |
From the Institute for Prevention of Cardiovascular Disease, Beth Israel Deaconess Medical Center (D.F., I.L., M.M.), the Cardiovascular Division, Brigham and Womens Hospital (K.L.), the Cardiology Division, Massachusetts General Hospital, Harvard Medical School (C.J.O., J.E.M.), and the Statistics and Consulting Unit, Department of Mathematics, Boston University (R.B.D.), Boston, Mass; the Framingham Heart Study, the National Institutes of Health National Heart, Lung, and Blood Institute (M.G.L., C.J.O., P.A.S., D.L.), Framingham, Mass; and the Royal North Shore Hospital (G.H.T.), Sydney, Australia.
Correspondence to DaLi Feng, MD, Cardiology Division, Massachusetts General Hospital, Harvard Medical School, 55th Fruit Street, Bulfinch 106, Boston, MA 02114. E-mail dfeng2{at}caregroup.harvard.edu
| Abstract |
|---|
|
|
|---|
Methods and Results Glycoprotein IIIa PlA genotype, fibrinogen level, and platelet aggregability were ascertained in 1340 subjects enrolled into the Framingham Offspring Study. Platelet aggregability was evaluated by the Born method. Higher fibrinogen levels were associated with increased epinephrine-induced aggregation (P=0.002) and a trend for ADP-induced aggregation (P=0.07). The fibrinogen effect was genotype specific, however, in that the increase in platelet aggregability with higher fibrinogen was present for the PlA1/A1 genotype (P=0.0005 and P=0.03 for epinephrine- and ADP-induced aggregation, respectively) but not for the PlA2-positive genotype (P>0.90).
Conclusion Higher fibrinogen levels were associated with increased platelet aggregability. However, the association between fibrinogen and platelet aggregability was genotype specific. This interaction may be responsible for the conflicting findings regarding PlA genotype and platelet aggregability. Further study of this gene-environment interaction may provide insight into cardiovascular disease risk.
Key Words: platelets genetics glycoproteins fibrinogen
| Introduction |
|---|
|
|
|---|
Weiss and colleagues1 first reported that patients with acute coronary syndromes were more likely to carry the PlA2 variant of the GP IIIa receptor than were controls. The risk associated with PlA2 was especially high for those aged
60 years at the time of infarction. Recently Walter and colleagues 2 reported that patients with the PlA2 variant had an increased risk of coronary stent thrombosis compared with PlA1-homozygous individuals. In addition, PlA2 has been associated with restenosis after coronary stent placement.3 However, an association between the PlA2 and cardiovascular disease (CVD) has not been a consistent finding.410
Recently, we found that the presence of the PlA2 variant was associated with increased platelet aggregability as indicated by a lower epinephrine threshold concentration.11 There was also a trend toward PlA2 being associated with a decreased threshold concentration for ADP, which was directionally consistent with the results seen with epinephrine-induced aggregation.11 Similar results have been reported with thrombin receptor-activating peptide.12,13 Paradoxically, however, Goldschmidt-Clermont et al14 showed that platelets with the PlA2 variant bound less exogenous fibrinogen than did the Pl A1-homozygous platelets when stimulated with ADP. Fibrinogen mediates platelet aggregation by binding to GP IIb/IIIa, and fibrinogen levels influence platelet aggregability. 1517 In addition, vWF is a glycoprotein that mediates platelet adhesion by binding with platelet surface receptor GP Ib/IX. This receptor is particularly important in hemostasis and thrombosis at high shear rate. vWF also binds with the platelet GP IIb/IIIa receptor, mediating platelet aggregation. In the current study, we hypothesized that platelet GP IIIa PlA genotypes may interact with plasma fibrinogen in modulating platelet aggregability. We also evaluated whether plasma vWF levels were associated with platelet aggregability, and if so, whether PlA genotypes may interact with vWF levels in modulating platelet aggregability.
| Methods |
|---|
|
|
|---|
Determination of Platelet Aggregability, Fibrinogen, and vWF Levels
Blood samples were obtained in the morning to avoid circadian variation in platelet aggregability.19 Blood was drawn in 3.8% sodium citrate solution (9:1 vol/vol). Platelet-rich plasma was separated by centrifugation for 10 minutes at 160g. Platelet aggregation was measured according to the method of Born. 20 The aggregation agents tested were epinephrine and ADP in varying concentrations (0.01 to 30 µmol/L) and a fixed concentration of arachidonic acid (1.6 µmol/L). The lowest concentrations of ADP and epinephrine required to produce a biphasic response with >50% aggregation (threshold concentration) were determined. Therefore, a lower threshold concentration indicates an increase in platelet aggregability.
Blood for fibrinogen was drawn into 3.8% sodium citrate (9:1, vol/vol), and blood for vWF was collected in EDTA. Plasma was separated by centrifugation for 20 minutes at 2000g and stored at -80°C for later analysis. Fibrinogen was determined by the Clauss method. vWF antigen was measured by ELISA. The intra-assay coefficients of variation for fibrinogen and vWF in our laboratory were 2.6% and 8.8%, respectively, and the inter-assay coefficients of variation were 4.7% and 10.6%, respectively.
Genotyping
Detailed methods have been reported previously.11,21 Polymerase chain reaction results were scored without knowledge of platelet aggregation results. Genotyping was successful in 97% of the samples.
Statistical Analysis
Demographic and clinical characteristics were compared by Students t tests or
2 tests between genotype groups defined by the presence or absence of the PlA2 allele. Observed genotype frequencies were compared with the Hardy-Weinberg equilibrium prediction, also using
2 tests. Data on epinephrine and ADP threshold concentrations were log-transformed. Multiple linear regression models were used to analyze platelet aggregability, adjusting for age, sex, body mass index, diabetes, triglycerides, HDL cholesterol, and estrogen replacement status, together with the presence or absence of the PlA2 allele and fibrinogen quartile (determined separately for each sex).22,23 Generalized estimating equation algorithms were used to correct for intra-family correlation.24 Data on platelet aggregation were expressed as geometric mean±95% CI.
| Results |
|---|
|
|
|---|
|
Plasma Fibrinogen Levels and Platelet Aggregability
Increased plasma fibrinogen levels were associated with incrementally increased platelet aggregability for epinephrine-induced aggregation and a trend for ADP-induced aggregation, as indicated by decreases in the threshold concentrations for epinephrine and ADP (P=0.002 and 0.07, respectively) (Table 2), after adjusting for covariates and the presence of PlA2 allele.
|
Interaction Between Plasma Fibrinogen and Pl A Genotype on Platelet Aggregability
The threshold concentrations for epinephrine-stimulated aggregation decreased with higher fibrinogen levels among the PlA1-homozygous subjects, with mean epinephrine threshold concentrations of 1.11 (range, 0.95 to 1.30) µmol/L in the lowest fibrinogen quartile, 1.02 (0.87 to 1.19) µmol/L in the second quartile, 0.85 (0.72 to 1.00) µmol/L in the third quartile, and 0.72 (0.60 to 0.86) µmol/L in the highest fibrinogen quartile (trend P=0.0005) (Figure 1). However, the threshold concentrations were not related to fibrinogen levels among the PlA2-positive genotype, with respective mean epinephrine threshold concentrations of 0.67 (0.52 to 0.86) µmol/L, 0.81 (0.63 to 1.04) µmol/L, 0.80 (0.62 to 1.05) µmol/L, and 0.63 (0.48 to 0.83) µmol/L (trend P=0.91). The interaction term describing the fibrinogen-related alterations in epinephrine threshold concentrations between genotypes was of borderline statistical significance (P=0.06, multivariate analyses).
|
For ADP-induced aggregation, the threshold concentrations also decreased with increased fibrinogen levels in the PlA1-homozygous group, with ADP threshold concentrations of 3.20 (range, 3.02 to 3.39) µmol/L, 3.18 (3.01 to 3.37) µmol/L, 2.97 (2.80 to 3.16) µmol/L, and 2.89 (2.70 to 3.09) µmol/L from the lowest quartile to the highest quartile (trend P=0.03) (Figure 2). As seen with epinephrine-induced platelet aggregation, there was no relation found between fibrinogen level and ADP threshold concentration in the subjects with PlA2-positive genotype. The respective ADP threshold concentrations were 2.83 (2.58 to 3.10) µmol/L, 2.97 (2.71 to 3.26) µmol/L, 3.17 (2.87 to 3.49) µmol/L, and 2.88 (2.61 to 3.19) µmol/L (trend P=0.97). The interaction term assessing fibrinogen-related differences in ADP-induced aggregation between genotypes was statistically significant (P=0.05, multivariate analyses).
|
Similar results were obtained when we analyzed the relation between fibrinogen levels and platelet aggregability and the interaction between fibrinogen and PlA genotype on platelet aggregability with fibrinogen as a continuous variable (data not shown).
Platelet aggregation thresholds were modestly but statistically significantly correlated between siblings for epinephrine threshold concentration (r=0.245, P<0.0001) and for ADP threshold concentration (r=0.235, P<0.0001). We obtained very similar results for the regression interaction terms relating PlA genotype with fibrinogen quartile to aggregation thresholds whether the sibship information was included or excluded (data not shown).
There was no significant association between plasma vWF antigen levels and platelet aggregation induced by ADP or epinephrine (P=0.70 and 0.96, respectively; multivariate analyses). Furthermore, vWF did not modify the association between the PlA genotype and platelet aggregability (data not shown).
| Discussion |
|---|
|
|
|---|
Studies of twins suggest an important pathogenic role for genetic factors in the pathogenesis of CVD.25 Although a small proportion of CVD can be attributed to a single gene defect, the nature of additional contributing genetic factors and their interaction with environmental factors remains largely unknown. Because platelets play a central role in acute CVD, it is possible that inherited platelet receptor variants and their interaction with environmental factors may contribute to the onset of CVD.
GP IIIa Polymorphism, Fibrinogen, and Platelet Aggregability
Platelet GP IIb/IIIa is the most abundant platelet receptor.26 The receptor is highly polymorphic and has long been recognized as having alloantigens, which cause immune-mediated platelet destruction. We found that one variant of these polymorphisms, the PlA2 allele, was associated with heightened platelet aggregability as indicated by a lower epinephrine threshold concentration.11 Although these findings were supported by studies with thrombin receptoractivating peptide (TRAP)12,13 and ADP, 27 the association between the PlA2 and increased platelet aggregability has not been consistent. For example, Lasne et al28 found that platelets of PlA2-positive genotype needed higher concentrations of TRAP to induce platelet dense granule secretion. Goldschmidt-Clermont et al14 quantified exogenous fibrinogen binding to platelets of different genotypes. Paradoxically, these investigators found that platelets with the PlA2 allele bound less exogenous fibrinogen than did platelets that were homozygous for PlA1. However, using the same method, Meiklejohn et al 29 found that the percentage of platelets positive for fibrinogen binding was similar among the PlA2-positive platelets and the PlA1-homozygous platelets.
The reason for the difference in platelet reactivity associated with the PlA genotype among the various studies is uncertain. The difference is not caused by altered platelet receptor density.27,30 One possible explanation is that the mild platelet aggregability change produced by the genetic variation could be masked when a large fixed concentration of agonist is used.29 In addition, studies with small sample sizes may not have the power to detect differences.31 In comparison, in our large sample, we titrated the platelet agonists to obtain threshold concentration.11
In the present study, increased plasma fibrinogen levels in the physiological range were associated with increases in platelet aggregability as indicated by reductions of epinephrine and ADP threshold concentrations. Prior studies similarly reported that an increase of fibrinogen either in the physiological range in the plasma or added extrinsically was associated with a reduction of the median effective dose of ADP.15,16 These findings are consistent with fibrinogens role as ligand for the GP IIb/IIIa receptor. In addition, increased fibrinogen concentrations have also been shown to enhance ADP- and TRAP-induced platelet P-selectin expression.32 It is therefore possible that platelet reactivity associated with the PlA polymorphism may be modulated by plasma fibrinogen levels.
Plasma Fibrinogen Levels Interact With PlA Polymorphism on Platelet Aggregability
Our results show an interaction between the PlA genotype and fibrinogen levels on platelet aggregability such that the enhanced platelet aggregability with increased fibrinogen levels were observed in the PlA1/A1 genotype but not in the PlA2-positive genotype. However, the PlA2-positive genotype still had an overall higher platelet aggregability compared with the PlA1-homozygous even though the PlA2-positive genotype platelets bind a smaller 14 or a similar29 amount of exogenous fibrinogen with the GP IIb/IIIa receptor than do the PlA1/A1 platelets. Our findings partially explain the conflicting findings regarding the PlA2 allele and platelet function.1114,2729,31 Plasma fibrinogen level should be taken into account when platelet reactivity studies are interpreted because fibrinogen level is an important determinant for platelet reactivity and a modulator of platelet aggregability associated with the PlA genotype.15,16,32
The mechanism for this interaction is unknown. The PlA antigen system is not in the 2 putative RGD (Arg-Gly-Asp)-binding regions of the GP IIIa (PlA1 is a leucine and PlA2 a proline at amino acid position 33). However, according to Calvete,33 the PlA system could be brought into immediate proximity with the binding regions by 2 disulfide bonds. Because of the unique structure of proline and its propensity for inducing conformational change, the amino acid substitution could influence the affinity of GP IIb/IIIa receptor for fibrinogen. Thus, the platelet with the PlA2 variant might have an increased binding with endogenous plasma fibrinogen at basal conditions. If that is the case, fewer GP IIb/IIIa receptors will be available for exogenous fibrinogen after being stimulated as reported by Goldschmidt-Clermont et al.14 Further study of endogenous and exogenous fibrinogen binding is needed to test the hypotheses. Alternatively, platelets with the PlA2 allele may have intrinsically increased aggregability as indicated by a lower epinephrine threshold concentration and therefore may be less dependent on plasma fibrinogen levels to achieve a similar degree of platelet aggregation as measured by the Born method.1113
Strengths and Limitations of the Study
In a large number of subjects, we obtained genotype status, fibrinogen levels, and platelet aggregability data to evaluate the relation between the PlA genotype and platelet aggregability and the interaction between the genotype and fibrinogen on platelet aggregability. A detailed risk factor profile was collected and adjusted for potential confounders. However, this cross-sectional study had several limitations. First, our analysis was based on the subset of Framingham subjects in whom both genotype and phenotype data were available. However, the genotype distribution and subject characteristics were similar among subjects excluded from analysis or included in the present analysis (data not shown). Second, we used an in vitro method to evaluate the relation between the PlA polymorphism, fibrinogen levels, and platelet aggregability. The in vivo significance of our in vitro findings and their clinical relevance need to be evaluated further. Finally, we used epinephrine and ADP threshold concentrations to determine platelet aggregability. Many different methods and parameters have been used to evaluate platelet aggregability, which may influence the results. For example, in one study,17 increased fibrinogen concentrations led to increased platelet fibrinogen bonding but reduced aggregation velocity. Meade et al15 found that increased fibrinogen levels were associated with a reduced median effective dose of ADP, a finding similar to our current result, indicating increased platelet aggregability. However, they also reported that increased fibrinogen was associated with a lower maximal response when higher ADP concentration was used, suggesting decreased aggregability.
Implications of the Study
We found that increased plasma fibrinogen levels were associated with increased platelet aggregability, although this association was present for the PlA1/A1 but not the PlA2 genotype. Prospective studies are warranted to examine the combined effect of fibrinogen levels and the PlA polymorphism on incidence of CVD. If individuals with the PlA2 variant (who have increased platelet aggregability irrespective of fibrinogen levels) or PlA1 homozygotes with high fibrinogen levels have a higher incidence of CVD, they may benefit from more aggressive measures for prevention and treatment, including therapy with antiplatelet agents.
| Acknowledgments |
|---|
Received August 22, 2000; revision received April 3, 2001; accepted April 17, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Chandrakantan, D. H. McDermott, H. T. Bich Tran, M. Jurewicz, L. Gallon, R. Gaston, E. Milford, and R. Abdi Role of {beta}3 Integrin in Acute Renal Allograft Rejection in Humans Clin. J. Am. Soc. Nephrol., November 1, 2007; 2(6): 1268 - 1273. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Mahmud, J. J. Cavendish, S. Tsimikas, L. Ang, C. Nguyen, G. Bromberg-Marin, G. Schnyder, S. Keramati, V. Palakodeti, W. F. Penny, et al. Elevated Plasma Fibrinogen Level Predicts Suboptimal Response to Therapy With Both Single- and Double-Bolus Eptifibatide During Percutaneous Coronary Intervention J. Am. Coll. Cardiol., June 5, 2007; 49(22): 2163 - 2171. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. N. Bongers, M. P.M. de Maat, M.-L. P.J. van Goor, V. Bhagwanbali, H. H.D.M. van Vliet, E. B. Gomez Garcia, D. W.J. Dippel, and F. W.G. Leebeek High von Willebrand Factor Levels Increase the Risk of First Ischemic Stroke: Influence of ADAMTS13, Inflammation, and Genetic Variability Stroke, November 1, 2006; 37(11): 2672 - 2677. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. V. Vijayan and P. F. Bray Molecular Mechanisms of Prothrombotic Risk Due to Genetic Variations in Platelet Genes: Enhanced Outside-In Signaling Through the Pro33 Variant of Integrin {beta}3. Experimental Biology and Medicine, May 1, 2006; 231(5): 505 - 513. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. L. Yee, C. W. Sun, A. L. Bergeron, J.-f. Dong, and P. F. Bray Aggregometry detects platelet hyperreactivity in healthy individuals Blood, October 15, 2005; 106(8): 2723 - 2729. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Holmberg, M.-L. Persson, M. Uhlen, and J. Odeberg Pyrosequencing Analysis of Thrombosis-Associated Risk Markers Clin. Chem., August 1, 2005; 51(8): 1549 - 1552. [Full Text] [PDF] |
||||
![]() |
J. Mikkelsson, M. Perola, and P. J. Karhunen Genetics of Platelet Glycoprotein Receptors: Risk of Thrombotic Events and Pharmacogenetic Implications Clinical and Applied Thrombosis/Hemostasis, April 1, 2005; 11(2): 113 - 125. [Abstract] [PDF] |
||||
![]() |
S. Tsiara, M. Elisaf, I. A. Jagroop, and D. P. Mikhailidis Platelets as Predictors of Vascular Risk: Is There a Practical Index of Platelet Activity? Clinical and Applied Thrombosis/Hemostasis, July 1, 2003; 9(3): 177 - 190. [Abstract] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2001 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |