Identification of ITGA2B and ITGB3 Single-Nucleotide Polymorphisms and Their Influences on the Platelet Function

The aim of the study was to investigate ITGA2B and ITGB3 genetic polymorphisms and to evaluate the variability in the platelet function in healthy Chinese subjects. The genetic sequence of the entire coding region of the ITGA2B and ITGB3 genes was investigated. Adenosine diphosphate-induced platelet aggregation, glycoprotein IIb/IIIa content, bleeding time, and coagulation indexes were detected. Thirteen variants in the ITGA2B locus and 29 variants in the ITGB3 locus were identified in the Chinese population. The rs1009312 and rs2015049 were associated with the mean platelet volume. The rs70940817 was significantly correlated with the prothrombin time. The rs70940817 and rs112188890 were related with the activated partial thromboplastin time, and ITGB3 rs4642 was correlated with the thrombin time and fibrinogen. The minor alleles of rs56197296 and rs5919 were associated with decreased ADP-induced platelet aggregation, and rs55827077 was related with decreased GPIIb/IIIa per platelet. The rs1009312, rs2015049, rs3760364, rs567581451, rs7208170, and rs117052258 were related with bleeding time. Further studies are needed to explore the clinical importance of ITGA2B and ITGB3 SNPs in the platelet function.


Introduction
Platelet aggregation plays a central role in the pathogenesis of acute thrombosis in coronary heart disease, stroke, and peripheral arterial disease. The cellular events leading to platelet aggregation are mediated by the binding of fibrinogen to the glycoprotein (GP)IIb/IIIa receptor of platelets as a final common pathway. GPIIb/IIIa is a platelet-specific, surface membrane receptor and also called alpha IIb beta 3 (aIIb 3) in the integrin nomenclature and thus plays a primary role in both platelet adhesion and thrombus formation at the vascular injury site [1]. A large interindividual number variability for GPIIb/IIIa receptors expressed on the platelet surface is commonly observed [2,3]. Moreover, a defect in the GPIIb/IIIa complex or a qualitative abnormality of this complex is seen in Glanzmann's thrombasthenia patients with impaired platelet aggregation and increased bleeding [4].
The ITGA2B gene encodes the aIIb subunit (GPIIb), whereas the ITGB3 gene encodes 3 (GPIIIa). The ITGA2B spanning 17 kb has 30 exons, whereas the ITGB3 spanning 46 kb has 15 exons; they are closely located on chromosome 17q21.32 without evidence for coordinated expression [5]. A few studies have linked single-nucleotide polymorphisms (SNPs) in ITGA2B and ITGB3 with increased or decreased platelet responses to various agonists and the risk of acute coronary syndrome and atherosclerosis [6][7][8].
A common polymorphism in the ITGB3, known as the human platelet antigen 1 (HPA-1b, PlA2, or rs5918), arises from a single-nucleotide change at position 1565 in Exon2 of ITGB3, resulting in a leucine (PlA1) to proline (PlA2) substitution at residue 33 [9]. The SNP has been extensively studied as an inherited risk factor for acute coronary syndrome and for its effect on the platelet function [6][7][8]. The HPA-3 (rs5910) polymorphism results from a thymine to guanine base change that leads to the replacement of isoleucine (HPA-3a) by serine (HPA-3b) at codon 843 of GPIIb (ITGA2B). This polymorphism may potentially influence the activity of the GPIIb/IIIa complex and associate with thrombosis [10].

BioMed Research International
According to the aforementioned studies, GPIIb/IIIa was believed to be a platelet-platelet contact receptor playing an important role in platelet aggregation, and its SNPs were associated with platelet hyperreactivity and have an effect on the pharmacodynamics of antiplatelet drugs. However, limited information was available on other genetic polymorphisms of ITGA2B and ITGB3 in Asian populations or their association with the platelet function [11,12]. Moreover, the prevalence of the PlA2 allele (rs5918) is dependent on ethnicity, with a frequency of approximately 15/100 in Caucasian populations [13] falling to less than 1/100 in Oriental populations [12,14].
In this study, regions of the exons of ITGA2B and ITGB3 genes were sequenced in 86 unrelated healthy Chinese. In addition, the association between genetic variants of ITGA2B and ITGB3 exons and the adenosine diphosphate-(ADP-) induced platelet aggregation, GPIIb/IIIa content, bleeding time, and coagulation indexes was investigated in 55 of the 86 subjects.

Study Design.
The research was conducted in compliance with the Declaration of Helsinki. The study protocol was approved by the Ethical Review Board of the Peking University First Hospital. All subjects gave their written informed consent prior to participation in the study. Healthy native Chinese subjects ( = 86) between 18 and 45 years of age with a body mass index (BMI) of 19-24 kg/m 2 were included in the study, and their genotypes were unknown.
All the subjects were considered healthy on the basis of their medical history, physical examination, vital signs (blood pressure, pulse rate, and temperature), safety laboratory tests (blood chemistry, hematological tests, and urinalysis), and 12lead electrocardiography.
The ADP-induced platelet aggregation (transmission, % max), GPIIb/IIIa content (GPIIb/IIIa per platelet), bleeding time, and coagulation indexes, including the mean platelet volume (MPV), platelet count (PLT), prothrombin time (PT), activated partial thromboplastin time (aPTT), fibrinogen (FIB), and thrombin time (TT) were detected in 55 of the 86 subjects. None of the donors had taken any medication for 2 weeks before blood collection.

ADP-Induced Platelet
Aggregation. Blood was collected in 3.8% sodium citrate tubes, and platelet-rich plasma (PRP) was obtained by centrifuging blood at 1000 revolutions per minute (rpm) for 10 minutes at room temperature. The PRP was collected in a fresh tube and platelet was counted by Platelet Counter PL100 (Sysmex, Kobe, Japan). Platelet-poor plasma (PPP) was obtained by centrifuging the remaining blood at 3000 rpm for 10 minutes at room temperature, and platelet numbering PRP was adjusted to 250 × 10 9 /L by PPP. Detection was completed within 1 hour after sampling.
Platelet aggregation was determined by the turbidimetric method using 20 umol/L ADP as the agonist. After zero setting with the PPP, assays were performed in platelet-rich plasma in a Chrono-Log aggregometer. Platelet aggregation was quantified as the maximum change in light transmission occurring within 5 minutes of addition of agonist.

GPIIb/IIIa Content.
GPIIb/IIIa on the platelet surface was evaluated in the PRP as the maximal binding of a 125 Ilabeled GPIIb/IIIa integrin antiplatelet antibody F(ab) 2 [15]. The PRP was fixed with equal volume of anticoagulantfixative solution (10 mmol/L EDTA-Na 2 , 0.2% glutaraldehyde, and 0.02% sodium azide dissolved in PBS, pH 7.4). After stored at 4 ∘ C, overnight, platelet was washed by Tyrode's solution twice and resuspended in Tyrode solution, containing 0.35% bovine serum albumin (concentration adjusted to 1 × 10 11 platelets/L), and sodium azide was added at final concentration of 0.02% (V/V) to preserve the platelet at 4 ∘ C for a week-long before assay. For detection, the platelet suspension (100 L) was added to a 0.5 mL Eppendorf tube and then incubated with 100 L of antibody (containing 20,000 cpm of labeled monoclonal antibody (McAb) and 0.8 g of unlabeled McAb) at 37 ∘ C for 1 hour, after washed with 0.35% bovine serum albumin/Tyrode's solution for three times. The radioactivity count of the precipitate in the tube was measured. Assuming that one monoclonal antibody would only bind one GP molecule on the platelet membrane, the number of GP molecules on the platelet membrane was calculated as number of GP molecules/platelet = (binding rate × amount of antibody (g) × Avogadro's number)/(molecular weight of antibody × platelet count).

Bleeding
Time. The skin bleeding time was carried out using the Simplate-II device (General Diagnostics, NJ, USA). A sphygmomanometer cuff was inflated on the patient's arm to 40 mmHg. The arm was supported at the level of the heart, and a muscular area on the volar aspect distal to the antecubital area was identified and swabbed with alcohol. The Simplate-II device was used to perform the incision perpendicular to the antecubital fossa. Blood was blotted with sterile filter paper every 30 seconds until blood no longer stained the paper. The time from incision to stopping of bleeding was recorded.

Genetic Analysis.
Genomic DNA was extracted from peripheral whole blood samples of each subject using a DNA Purification Kit (Wizard; Promega, WI, USA). After the polymerase chain reaction (PCR) products were obtained, all samples were directly sequenced to determine the SNPs in ITGA2B and ITGB3 (Life Technologies Biotechnology Co., Ltd., China).
Primers required for PCR amplification and sequencing were designed according to the wild-type ITGA2B and ITGB3 sequences reported in the GenBank (NG 008331.1 and NG 008332.1, resp.). Tables 1 and 2 describe the primer pairs used to amplify the promoters, the 5 -and 3 -untranslated regions (UTRs), the entire coding regions and the intronexon junctions of the ITGA2B and ITGB3 genes.
2.6. Data Analysis. The Variant Reporter v1.1 (Life Technologies Biotechnology Co., Ltd.) software suite was used for the initial analysis of the sequence, including base calling, fragment assembly, SNP, and sequence insertions/deletions detection. Polymorphisms of ITGA2B and ITGB3 genes were named according to the genomic reference sequences   Tables 3 and 4.

Linkage Disequilibrium Analysis.
Using the detected polymorphisms greater than 0.01 in frequency [MA(F) > 0.02], linkage disequilibrium (LD) was analyzed for | | and 2 values (Figure 1). For ITGA2B, rs5911 was completely linked with rs850730 ( 2 = 1.00), and rs5910 was strongly linked with rs850730 and rs5911 ( 2 = 0.95). For ITGB3, complete LD was observed between rs4642 and rs4634 ( 2 = 1.00). Other LD results are shown in Figure 1. A new SNP ITGB3-09 was found for which MA(F) in this study was 0.402. The Hardy-Weinberg equilibrium values of rs1009312 and rs2015049 were <0.001. The possible reason is that the sample in this study may have a Chinese population of other ethnic groups besides the Han population. Those two SNPs were not included in the LD analysis.  (Figures 1 and 2). The relationship between GPIIb/IIIa content and APTT was not significant ( = 0.242, = 0.075), while the association between ADPinduced platelet aggregation and APTT was moderately significant ( = 0.267, = 0.049) ( Figure S1). The bleeding time had no association with other parameters. It had a negative association trend with MPV but was not significant ( = −0.244, = 0.073). The PLT was associated with PT ( = −0.415, = 0.002) and TT ( = −0.363, = 0.007).  The FIB was negatively correlated with TT ( = −0.409, = 0.002). and rs5919 were lower than those of heterozygous mutations and wild type. Moreover, the GPIIb/IIIa per platelet was associated with rs55827077. A trend of GPIIb/IIIa per platelet was observed among rs70940817 GG (45), GA (9), and AA (1) carriers (43807.7 ± 14157.5, 51086.8 ± 19368.0, 61277, resp., = 0.093), although it did not reach a significant level. The bleeding time was significantly related with rs3760364, rs7208170, rs567581451, and rs117052258 (Table 4).

Impact of ITGA2B and ITGB3 SNPs on ADP-Induced
In the present study, SNPs related with PLT were not found. ITGB3 rs1009312, which is completely linked with rs2015049, was significantly associated with the MPV  (Table 4). ITGB3 rs70940817 was correlated with the PT and APTT ( = 0.002 and = 0.003, resp.). The coagulation indexes and their values are summarized in Table 5.

Discussion
In this study, 13 variants in the ITGA2B locus and 29 variants in the ITGB3 locus in the Chinese population were observed. Two of the 29 variants located in ITGB3 were novel SNPs with MA(F) > 0.02. Variants in ITGA2B and ITGB3 genes displayed significant interethnic differences in the global populations ( Table 6). The C allele frequency of SNP rs5918 (T/C), located in the ITGB3 gene, was only inhomogeneously distributed at 0.7% in the Chinese population, while it is common in whites and Africans (allele frequency 13.7% versus 12.8%). The following alleles of SNPs, rs7208170A, rs55827077C, rs11871251A, rs1009312G, and rs2015049G, were found to be more common in the Han Chinese and Africans (37%∼50%) than in the European ancestry (10%∼20%). In the ITGA2B gene, the rs5911 was completely linked with rs850730 and strongly linked with rs5910 in the Chinese population in this study, and all the three SNPs were common in whites and Asians. However, the rs5911 was not found in Africans. According to the ethnicity difference described in the preceding text, resequencing of the ITGA2B and ITGB3 genes in the Chinese population was believed to be very important to reveal the function of SNPs.
Variation in the MPV or PLT can have a profound impact on differences in the platelet function between individuals [16,17], and these traits have a strong genetic component [18][19][20][21]. However, limited information is available for Asians. In the present study, the A alleles of rs1009312 and rs2015049, located in ITGB3, were positively associated with the MPV ( = 0.029). The MPV had a trend in rs7208170 alleles but did not reach a significant level ( = 0.051). The difference between PLT and SNPs in ITGA2B or ITGB3 was not significant.
The APTT and PT are clinical tests commonly used to indicate coagulation factor deficiencies [22,23], activated coagulation, and risk of venous thromboembolism [24,25].
To date, two genome-wide association studies (GWAS) of APTT and PT conducted in the European ancestry have been reported. One study identified genome-wide significant associations of APTT and variants at F12 (MIM 610619), KNG1 (MIM 612358), and HRG (MIM 142640) [26]. In the other research, the GWAS for APTT and PT was conducted and replicated genome-wide significant associations at KNG1, HRG, F11, F12, and ABO for APTT and identified significant associations at the F7 and PROCR/EDEM2 regions for PT. Eight genetic loci accounted for ∼29% of the variance in APTT, and two loci accounted for ∼14% of the variance in PT [27]. In this study, the association of APTT, PT, FIB, and TT with ITGA2B and ITGB3SNPs was investigated. The A allele of ITGB3 rs70940817 was found to significantly correlate with the elevated APTT and PT ( = 0.003 and = 0.002, resp.). The T allele of ITGB3 rs112188890 was related with APTT ( = 0.029), and the G allele of ITGB3 rs4642 was associated with TT and FIB ( = 0.015 and = 0.029, resp.).
The ADP-induced platelet aggregation test was often used to identify the platelet function and the efficacy of antiplatelet drugs. Jones et al. detected 1327 SNPs and investigated their correlation with ADP or collagen-related peptide-induced platelet aggregation in 500 healthy northern European subjects. This identified 17 novel associations with the platelet function ( < 0.005) accounting for approximately 46% of the variation in response [28]. In this study, the minor allele of rs56197296 and rs5919 was found to be associated with the decreased ADP-induced platelet aggregation ( = 0.014 and = 0.015, resp.). GPIIb/IIIa is the central receptor of platelet aggregation. The variations of GPIIb/IIIa amount expressed on a platelet surface might affect the platelet-aggregating activity. Many studies reported that the rs5918 correlated with the GPIIb/IIIa receptor expression [29]; however, many conflict studies have also been published [3,30]. O'Halloran et al. investigated whether three polymorphisms of the GPIIIa promoter (−468G/A, −425A/C, and −400C/A) influenced the RNA expression and receptor density in the platelets of patients with cardiovascular disease [3]. They found a threefold variation between the subjects in the number of GPIIb/IIIa receptors expressed per platelet, although no association between the receptor density and the PlA2 or the three promoter polymorphisms was demonstrated. In the present study, rs55827077, which was located at the promoter region of the GPIIIs gene at position −145, was related with GPIIb/IIIa per platelet.
The bleeding time is a test that evaluates the platelet function in vivo. A prolonged bleeding time can result from platelet abnormality, Von Willebrand factor deficiency, or vascular disorders such as Ehlers-Danlos syndrome. A disruption of platelet aggregation results in a prolongation of the bleeding time [31]. In this study, the bleeding time was related with rs3760364 of ITGA2B and rs7208170, rs567581451, and rs117052258 of ITGB3 ( < 0.05). No previous study reporting on the effect of SNPs on the bleeding time was found. The mechanism of the relationship needs to be further examined. However, a negative trend between the bleeding time and MPV ( = −0.244, = 0.073) was observed. Moreover, rs1009312 and rs2015049 were found to significantly associate  with both bleeding time and MPV. Further, the minor alleles of rs7208170 and rs567581451, which were significantly related with the bleeding time, had a trend with the MPV among genotypes, although they did not reach a significant level ( = 0.051 and = 0.054, resp.). However, no relationship was observed among SNPs, bleeding time, and MPV in rs3760364 and rs117052258. In this study, a positive correlation of the level of ADPinduced aggregation and GPIIb/IIIa content was detected in healthy volunteers. This correlation is consistent with a previous report. Yakushkin et al. investigated the relationship between the number of GPIIb/IIIa and the level of ADPinduced aggregation in a group of 35 healthy volunteers and found positive and significant correlations between the level of platelet aggregation induced by different ADP doses (from 1.25 to 20 mmol/L) and the number of GPIIb/IIIa [32].
A strong positive correlation was observed between GPIIb/IIIa per platelet and PLT, PT, and TT ( < 0.01). Although the TT was strongly correlated with GPIIb/IIIa per platelet ( = 0.545, < 0.001), the relationship of SNPs/TT and SNPs/GP (IIb/IIIa) content is not linked. For example, ITGB3 rs4642 was significantly related with the TT ( = 0.015), while the value of GPIIb/IIIa per platelet had no trend among rs4642 genotypes ( = 0.789). The other coagulation indexes were the same. Therefore, the relationship between SNPs and GPIIb/IIIa content cannot be deduced from the correlation of coagulation indexes and SNPs.
Some of the significant correlations between SNPs and platelet function parameters in the present study were not reported previously, which may be due to the reason that SNPs with low MA(F) in the European ancestry were not selected as candidate SNPs in previous studies. For example, the GWAS-genotyping platforms lack the coverage of low frequency/rare variants [33]. Moreover, LD might be another reason, if different genetic patterns (haplotypes) are present in different populations [3].
A possible limitation of the current study is the limited number of subjects. After stratification, the number of subjects in each genotypic group was small, thereby limiting further haplotype association analysis and accession of small effects of an SNP with a small minor allele frequency.

Conclusion
In summary, as ethnicity difference might limit the interpretation of the function of SNPs, resequencing the ITGA2B and ITGB3 genes and investigating their functions in the Chinese population are very important. In the present study, nine SNPs were found to associate with indexes of platelet and coagulation haemostasis. Newer studies are needed, particularly, to further assess the clinical importance of the abovediscussed SNPs in disease susceptibility and antiplatelet drugs pharmacodynamics. Further studies should pay more attention to the roles of ITGA2B and ITGB3 SNPs in ethnic variations.