Bernd Jilma, MD, Department of Clinical Pharmacology, or Petra Jilma, MD, Transfusion Medicine, Währinger Gürtel 18–20, A-1090 Wien, Austria. E-mail: Bernd.Jilma@univie.ac.at
Summary. Platelet polymorphisms (Kozak, VNTR and HPA-2) within glycoprotein (GP)Ibα may be associated with an increased risk of arterial thrombosis. However, the functional role of these polymorphisms has not been clarified. Their influence on platelet plug formation under high shear rates was, therefore, examined in 233 healthy individuals. Collagen–adrenaline-induced closure time was shorter in carriers of the C/D versus C/C VNTR allele and in homozygotes with the (−5)T/T versus (−5)C/T Kozak genotype as determined by novel polymerase chain reaction methods. The HPA-2 genotype had no effects, and the density of GPIbα molecules was not influenced by GPIbα genotypes.
The glycoprotein (Gp) Ib/IX complex is the major receptor for von Willebrand factor (VWF), which mediates shear-stress-induced platelet adhesion and activation. Three genetically inherited polymorphisms of GpIbα may affect its receptor function: a length polymorphism that results from a variable number of tandem repeats (VNTR) in the macroglycopeptide region of GpIbα; the (−5)C/T dimorphism in the Kozak region; and, third, a 524C/T dimorphism, leading to the HPA-2 antigens, which may elicit alloimmune response.
Despite extensive studies in patients, only one small study in 40 subjects has examined the influence of these GpIbα polymorphisms on platelet function (Cadroy et al, 2001). Hence, we evaluated the role of GpIbα polymorphisms in platelet plug formation under pathophysiologically relevant high shear conditions.
Materials and methods
The study was approved by the Ethics Committee of the Vienna University. Written informed consent was obtained from all subjects.
Healthy individuals (n = 233; M/F = 134/99; aged 33 ± 10 years) without any bleeding history, who had not regularly or recently (within the last 5 d) had an intake of cyclooxygenase inhibitors were studied. High-shear-induced (5000/s to 6000/s) platelet plug formation was tested using a platelet function analyser (PFA-100; Dade Behring, Marburg, Germany) to determine the collagen–adrenaline closure time (CEPI-CT). Single determinations were performed in one batch of cartridges, and individual day-to-day variability of CEPI-CT averaged 9% (Jilma, 2001). Collagen adenosine diphosphate (CADP)-induced platelet plug formation was not measured, because we have previously shown a very good correlation between CADP-CT and CEPI-CT (r = 0·83) when values were repeatedly determined on 4 subsequent days. von Willebrand factor antigen level (VWF-Ag) was measured with a fully automated simultaneous thermal analyser (STA) using the VWF-Liatest (Diagnostica Stago, Paris, France), because VWF-Ag levels influence CEPI-CT values. Surface expression of GpIb (CD42b) was measured by flow cytometry from whole citrated blood as described previously (Jilma-Stohlawetz et al, 2001).
Genotyping of GPIbα polymorphisms. Each genotype was determined in a separate laboratory. VNTR-polymorphism: 2·5 µl of genomic DNA were added to a 25-µl reaction mixture containing 10·5 µl polymerase chain reaction (PCR)-grade water, 2·5 µl 0·5 mol/l GC-RICH solution, 2 µl dNTP (10 m µmol/l each), 1 µl 5 µmol/l of each sense primer (5′-4158-ACACTTCACATGGAATCCAT-4177–3′) and antisense primer (5′-4546-GGGTCATTTCTGGAGCTCTC-4527–3′), 5 µl 5 × GC-RICH PCR reaction buffer and 0·5 µl (1 U) GC-RICH PCR system enzyme mix (Boehringer Mannheim, Germany). After initial denaturation at 95°C for 3 min, amplification was performed in a DNA thermocycler (Hybaid, Ashford, Middlesex, UK) for 35 cycles (denaturation at 95°C for 30 s, annealing at 58°C for 30 s and extension at 72°C for 45 s). The PCR products were analysed by 2·0% agarose gel (Roth, Karlsruhe, Germany) electrophoresis using Tris-borate/EDTA buffer and visualized by ethidium bromide staining. A DNA ladder of 50 bp (50–800 bp) was used as molecular marker (Gibco BRL, Karlsruhe, Germany).
For the detection of the (−5)T/C polymorphism in the Kozak sequence of the GPIb-α gene, a mutagenic-separated PCR (MS-PCR) was established. MS-PCR is a single-tube PCR-based technique with allele-specific primers differing in length by 7 bp (base-pairs). Deliberate base mismatches in the primers minimize cross-reactions of the allele-specific PCR products during PCR amplification. The alleles are easily discernible by electrophoresis on high resolution gels. Primer sequences were: sense GP1bWT-F 5′-GCTCCCTTGCCCACGGGT-3′ (nucleotide position 3067–3064; 5 pmol/25 µl), GP1nMT-F 5′-ACTCAAGTATCCCTTGCCCACAAGC-3′ (position 3040–3064, 10 pmol/25 µl), and antisense 5′-GGTTGTGTCTTTCGGCAGG-3′ (position 3218–3200, 10 pmol/25 µl). PCRs were performed in a Biometra Thermocycler (Göttingen, Germany) using 25 µl reaction volumes containing 0·625 U of AmpliTaq Gold (Applied Biosystems, Roche), 2 mmol/l MgCl2, 2·5 µl 10 × PCR buffer (Applied Biosystems), 100 µmol of each dNTP (Amersham Pharmacia Biotech, Uppsala, Sweden), primers and approximately 25 ng DNA. A 10 min denaturation at 95°C was followed by 39 cycles at 95°C for 45 s, 54°C for 1 min and 72°C for 1 min, and a final extension step of 10 min at 72°C. PCR products [179 bp (mutant) and 172 bp (wild type)] were analysed by gel electrophoresis on Spreadex EL 400 S-50 N-Polyacrylamide minigels (Elchrom Scientific, Zürich, Switzerland) at 200 V for 2 h (Fig 1). Gels were stained with Sybr Green (1:10 000; Molecular Probes, Eugene, OR, USA). Reagent controls without DNA served as negative controls.
HPA-2 dimorphism within the GPIBα gene was determined by sequence-specific PCR as described previously (Sperr et al, 1998).
Statistical analysis. Data is expressed as means and 95% confidence intervals (CI). Comparisons were made by the Kruskal–Wallis anova and Mann–Whitney U-test. A two-tailed P-value < 0·05 was considered significant. Non-closure in the PFA-100 test system was assigned at 301 s for conservative statistical comparisons. Subjects with abnormally low VWF levels (< 60%) were excluded from statistical analysis, as were individuals with VNTR genotypes of very low frequency although these individuals are presented in Table I.
Table I. Influence of the VNTR, HPA-2 and Kozak polymorphisms on platelet plug formation under high shear stress.
P < 0·05 C/D vs C/C and T/T vs C/T.
P = 0·08 B/C vs C/D.
Data is given as the mean ± 95% CI, median and range.
CEPI-CT, collagen–adrenaline-induced closure times; VWF-Ag, von Willebrand factor antigen levels; MFI, mean fluorescence intensity.
Interestingly, all seven Kozak C/C individuals were VNTR C/C, and heterozygous carriers of the VNTR D-allele were more frequently carriers of the Kozak T/T genotype (15%) than of the C/T genotype (2%; P = 0·007 χ2 test).
Platelet plug formation under high shear rates, similar to those in arterial stenosis, was about 15–20% higher in individuals carrying the Kozak T/T or the VNTR C/D genotype than in those subjects with other genotypes (Table I). Individuals with the Kozak T/T genotype had 20% shorter CEPI-CT than those with the C/T genotype (P = 0·02; Table I), even when VNTR C/D genotypes were excluded. Carriers of the VNTR C/D genotype had the shortest CEPI-CT (i.e. haemostatically most competent platelets; Table I, P = 0·047). As these polymorphisms were in linkage disequilibrium, we grouped the T/T individuals into two populations, those with the C/D genotype and others. The T/T individuals who were C/D had lower mean CEPI-CT values than the others (147 s vs 165 s). Changes in avidity and affinity determine the function of integrins. The C and D VNTR size polymorphisms are shorter than A and B. It may be speculated that differences in size may alter the affinity for VWF and hence affect platelet function, although experimental proof is currently not available.
Our data could be clinically applied to target therapies for prothrombotic patients at highest risk. More importantly, the efficacy of anti-VWF therapy (under development) could be affected by GPIb polymorphisms, which, therefore, should be determined during all phases of clinical drug development.
However, HPA-2 polymorphisms did not affect platelet function (Table I). Flow cytometric analysis showed that all GPIb genotypes expressed a similar density of platelet surface GPIb (Table I), and thus confirms previous observations (Corral et al, 2000; Santoso et al, 2002).
Our findings are in contrast to a small study (Cadroy et al, 2001), which showed increased platelet deposition in vitro at physiological shear rates of 650/s in eight Kozak C/T individuals as compared to 32 T/T subjects. However, no difference was observed at the pathophysiologically relevant shear rate of 2600/s, where GpIb polymorphisms particularly should play a major role in platelet adhesion.
In summary, we have established a new mutagenic separated PCR to detect the (−5)T/C polymorphism in the Kozak sequence of the GPIBα gene and demonstrated its feasibility. Importantly, subjects with Kozak T/T and VNTR C/D genotypes had enhanced platelet plug formation under high shear rates. However, the pathogenic role of the different GpIbα polymorphisms in arterial thrombosis can only be determined in prospective clinical studies.