Introduction

Ischemic stroke is the rapid loss of brain functions due to a disturbance in the blood supply to the brain. It is a leading cause of disability and the second most common cause of death in adults around the world1. In China, with a population of 1.4 billion, the annual stroke death toll is approximately 1.6 million; thus, stroke has exceeded heart disease in becoming the major cause of death2. Aspirin, as a golden standard of antiplatelet therapy, is widely prescribed to treat ischemic stroke patients. However, recent studies have revealed that in certain cases, the platelet function was inadequately inhibited, thereby leading to thrombotic events despite therapy with the standard dose of aspirin. This phenomenon was called aspirin resistance and has an incidence of 5%–65% in patients with ischemic stroke3. Today, the underlying mechanism of aspirin resistance is still largely unknown, but it has been argued that genetic factors may be an important factor4. A number of studies have focused on the correlation between aspirin resistance and genetic polymorphisms in cyclooxygenase (COX) but have yielded conflicting results5,6. There may be other genetic factors accounting for the inter-individual differences in aspirin response.

The principal mechanism of the antithrombotic effect of aspirin is the irreversible inhibition of COX-1, which catalyzes the conversion of arachidonic acid (ARA) to prostaglandins G2 and H2, thus preventing the production of thromboxane A2 (TXA2). The binding of TXA2 to its specific receptor (TXA2 receptor, encoded by TBXA2R) could induce platelet activation and aggregation7. When the COX-dependent pathway is strongly inhibited by aspirin, maximal aggregation will then require the platelet activators in the COX-independent pathway. The platelet plasma membrane contains multiple activators that are responsible for activating the platelet physiological response. Lipoprotein-associated phospholipase A2 (Lp-PLA2, encoded by PLA2G7) is synthesized in macrophages and transported into the circulation by HDL (high-density lipoprotein) and LDL (low-density lipoprotein) cholesterol particles8. Lp-PLA2 is a strong platelet activator that has been reported to be a predictive factor for stroke and transient ischemic attack (TIA), and it might be associated with CHD (coronary heart disease) risk9. Platelet endothelial aggregation receptor-1 (PEAR1) is an epidermal growth factor repeat-containing transmembrane receptor that participates in platelet contact-induced activation10. PEAR1 genetic polymorphisms have been associated with greater platelet aggregability in platelets functioning under native conditions. Therefore, genetic polymorphisms of TBXA2R, PLA2G7, and PEAR1 may alter the aspirin response by affecting platelet function.

In its pharmacokinetic pathway, aspirin is quickly transformed into the inactive metabolite salicylic acid by carboxylesterase and is partly excreted by P-glycoprotein (P-gp, encoded by MDR1) in the gastric mucous membrane, plasma and red blood cells. Kugai et al reported that P-gp is involved in the pathogenesis of aspirin-induced intestinal epithelial injury11. It has been speculated that MDR1 genetic polymorphisms may also contribute to inter-individual differences in aspirin response by influencing the absorption of aspirin.

Hence, the present study was embarked upon with the aim to study the association of TBXA2R, PLA2G7, PEAR1, and MDR1 genetic polymorphisms with aspirin response and platelet activity in Chinese ischemic stroke patients. Six single-nucleotide polymorphisms (SNPs), which are commonly found in Asians (with a minor allele frequency higher than 0.1) and have been associated with the expression or function of a gene or protein, were chosen from candidate genes. To assess platelet reactivity during aspirin therapy, the ADP/AA-induced optical platelet aggregation method was used, and the synthesis of platelet TXB2, a chemically stable and inactive product of TXA2 hydrolysis, was assessed.

Materials and methods

Patients

In total, 283 ischemic stroke patients were enrolled from the Guangdong Provincial Hospital of Chinese Medicine from September 2012 to April 2014. Ischemic stroke was defined as a focal neurological deficit persisting for more than 24 h with evidence of cerebral infarction on neuroimaging. All patients who were ≥18 years old and who had taken 100 mg of aspirin (Bayer Healthcare Company Ltd, Beijing, China) for the previous 7 d were eligible for enrollment. Neurological severity was evaluated using the National Institutes of Health Stroke Scale (NIHSS). Exclusion criteria included current or past history neoplasm, bleeding disorders, abnormal renal function (creatinine >2.5 mg/dL), platelet count of <150 000/μL or >450 000/μL, and ingestion of clopidogrel, ticlopidine, dipyridamole, other nonsteroidal anti-inflammatory drugs, platelet glycoprotein IIb/IIIa (GPIIb/IIIa) inhibitors or fibrinolytics within the 30 d before the test. All participants submitted informed written consent before enrollment. The demographic data and relevant characteristics, such as age, gender, medical problems and lipid profile, of the patients were obtained from their medical records.

Optical platelet aggregation determination

Blood samples were drawn after the administration of the last dose of aspirin. Two tubes of whole blood, anticoagulated with 3.8% sodium citrate (4.5 mL each) were collected from each patient for platelet analysis. Turbidimetric platelet aggregation was performed in platelet-rich plasma with a platelet count adjusted to 250×103/mm3. Platelets were stimulated with 0.5 mg/mL (1.6 mmol/L) arachidonic acid (AA) and 5 and 20 μmol/L adenosine diphosphate (ADP). Aggregation was performed with a LBY-NJ4A automatic platelet aggregation analyzer (Precil Inc, Beijing, China). The extent of aggregation was defined as the maximal amount of light transmission within 6 min of the addition of the agonist, with platelet-poor plasma used as a reference. Aspirin resistance was defined as a mean aggregation of ≥70% with 10 μmol/L ADP and a mean aggregation of ≥20% with 0.5 mg/mL AA. Aspirin semi-resistance was defined as meeting one but not both of the above criteria12. In general, the analysis in this study combined patients with aspirin resistance and those with aspirin semi-resistance into an aspirin-insensitive group.

Serum TXB2 (s-TXB2) concentration quantitation

TXB2 concentrations were determined in serum samples using a commercially available ELISA kit (Thromboxane B2 EIA Kit; Cayman Chemical, San Antonio, TX, USA) following the manufacturer's instructions, and all serum samples were assayed in duplicate. The mean inter- and intra-assay coefficient of variation (CV, %) of the serum TXB2 (s-TXB2) concentration quantitation were 4.952% (1.758%–8.102%) and 6.729% (3.391%–13.89%), respectively.

Individual SNP genotyping

Total genomic DNA was extracted from the peripheral leukocytes according to a previously described method13. The six SNPs selected were TBXA2R (rs1131882), PLA2G7 (rs1051931, rs7756935), PEAR1 (rs12566888, rs12041331), and MDR1 C3435T. Genotyping was performed using an Agena Bioscience MassARRAY® system (Agena Bioscience, San Diego, CA, USA). The PCR and extension primers and MassARRAY genetic analysis spectrogram for each SNP are provided in the Supplementary data.

Statistical analysis

Statistical analysis was performed using SPSS (Statistical Package for the Social Sciences) statistical software (version 21.0). The recorded clinical data, when normally distributed in the analyzed group of patients, are presented as the mean and standard deviation (SD), and non-normally distributed data are presented as the median and interquartile range. The chi-squared test was used to compare the observed allele and genotype frequencies with the Hardy-Weinberg equilibrium prediction. The linkage disequilibrium (LD) was measured by an online calculator (http://www.oege.org/software/cubex/). Comparisons between continuous variables were performed using unpaired Student's t-tests because these variables were normally distributed, as determined by the Shapiro-Wilk test. Comparisons of genetic polymorphisms in the aspirin-sensitive and aspirin-insensitive groups were performed using the chi-squared or Fisher's exact tests, as appropriate, and are described with the odds ratio with the 95% confidence interval (CI). To account for multiple testing, the Benjamini-Hochberg false discovery rate (FDR) correction was applied. All probability values are 2-sided, and a P value <0.05 was considered statistically significant. The statistical power of the sample size was calculated using PASS (Power Analysis and Sample Size) software (version 11.0.7; PASS, NCSS, LLC).

Results

Characteristics of the patients

The demographic and clinical characteristics of the participants are summarized in Table 1. The platelet aggregation results indicate that among the 283 stroke patients, 33 patients (11.66%) were aspirin-insensitive patients and 250 patients (88.34%) were aspirin-sensitive patients. Gender, hypertension, smoking, diabetes, coronary artery disease, TC (total cholesterol), TG (triglyceride), HDL, LDL and the platelet count were not related to the aspirin response (P>0.05).

Table 1 The general characteristics and laboratory parameters of aspirin sensitive and insensitive patients.
Full size table

Allelic and genotype frequencies

Table 2 lists the allele and genotype frequencies of the variants examined in the study population. All genotypes of the examined SNPs were in Hardy-Weinberg equilibrium (P>0.05). PLA2G7 rs1051931 was in perfect linkage disequilibrium with rs7756935 (r2=1). PEAR1 rs12041331 was in tight linkage disequilibrium with rs12566888 (r2=0.83).

Table 2 Alleles and genotypes frequencies for 6 SNPs in the investigated patients.
Full size table

Associations of genetic polymorphisms with platelet aggregation and TXB2 synthesis

As shown in Table 3, 4, 5, MDR1 3435 TT genotype carriers had significantly less AA/ADP-induced platelet aggregation than did 3435 CT/CC genotype carriers (P=0.002 and P<0.001, respectively), and similarly, they tended to have a lower s-TXB2 concentration although the observed difference did not remain when the FDR correction was applied (adjusted P>0.05). For TBXA2R rs1131882, the AA-induced platelet aggregation in individuals with the rs1131882 CC genotype was higher than that in carriers of the other genotypes (P=0.014), and the ADP-induced aggregation and s-TXB2 concentration tended to be higher in the rs1131882 CC genotype group; however, the differences were not statistically significant after adjustment. With the combined effects of PLA2G7 rs1051931 and rs7756935, carriers of the AA-CC haplotype had considerably more ADP-induced platelet aggregation than did noncarriers (P=0.022), however, no association was observed between the PLA2G7 genetic polymorphisms and AA-induced platelet aggregation/s-TXB2 concentration. Neither the s-TXB2 concentration nor the AA/ADP-induced platelet aggregation correlated with the PEAR1 genetic polymorphisms (rs12041331 and rs12566888).

Table 3 Associations of MDR1, TBXA2R, PLA2G7, and PEAR1 polymorphisms with the serum TXB2 concentration in ischemic stroke patients.
Full size table
Table 4 Associations of MDR1, TBXA2R, PLA2G7, and PEAR1 polymorphisms with the maximal ADP-induced aggregation in ischemic stroke patients.
Full size table
Table 5 Associations of MDR1, TBXA2R, PLA2G7, and PEAR1 polymorphisms with the maximal AA-induced aggregation in ischemic stroke patients.
Full size table

Associations of genetic polymorphisms with aspirin resistance

The genotypic distribution of polymorphisms of the candidate genes in aspirin-sensitive and aspirin-insensitive patients are listed in Table 6. Significant differences were observed in the following SNPs: MDR1 C3435T, TBXA2R (rs1131882), and PLA2G7 (rs1051931-rs7756935) (P=0.021, P=0.028 and P= 0.023, respectively). For MDR1 C3435T, the proportion of TT genotype carriers was significantly lower in the aspirin-insensitive group than that in the sensitive group (6.1% vs 23.6%), and the risk of aspirin resistance was significantly lower in patients with the MDR1 3435 TT genotype than in the CT+CC genotype carriers (odds ratio=0.421, 95% CI: 0.233–0.759). The proportion of TBXA2R rs1131882 CC genotype carriers was significantly higher in the aspirin-insensitive group than that in the sensitive group (81.8% vs 62.4%), and the carriers of the CC genotype had a significantly higher risk of aspirin resistance compared with the risk of the T allele carriers (odds ratio=2.712, 95% CI: 1.080–6.810). Moreover, the proportion of PLA2G7 rs1051931 AA-rs7756935 CC carriers was significantly higher in the aspirin-insensitive group than in the sensitive group (9.1% vs 1.2%), and the risk of aspirin resistance in rs1051931 AA-rs7756935 CC genotypes carriers was considerably higher than the risk in the carriers of the other genotypes (odds ratio=8.233, 95% CI: 1.590–42.638).

Table 6 Associations of MDR1, TBXA2R, PLA2G7 and PEAR1 polymorphisms with aspirin resistance in ischemic stroke patients.
Full size table

However, neither of the two SNPs (rs12566888, rs12041331) of PEAR1 was correlated with the aspirin response (P>0.05).

Discussion

The role of aspirin in preventing ischemic stroke has been well documented14,15. However, some patients are not responsive to aspirin and can thus still suffer from ischemic stroke and cardiovascular events. The definition of aspirin resistance is debated; hence, the reported incidence rate varies broadly from 5% to 65% depending on the assay used for identification and the population studied16,17. In this study, the prevalence of aspirin insensitivity was 11.66%, which was evaluated by the classical gold standard, optical platelet aggregation. In addition, the level of platelet TXB2 synthesis was assessed, which is an alternative method of evaluating platelet reactivity and aspirin response18.

The thromboxane A2 produced by aggregating platelets is a potent platelet activator and vasoconstrictor, and its action is mediated by the thromboxane A2 receptor (TBXA2R). The association between TBXA2R polymorphism and platelet activity was analyzed previously in healthy volunteers; the analysis indicated that the TBXA2R rs1131882C allele might be correlated with higher platelet activity19. Consistent with this finding, the present study found that the rs1131882 CC genotype was associated with higher platelet aggregation in response to ADP/AA and was correlated with an increased level of platelet TXB2 synthesis. In addition, the rs1131882 CC genotype was a risk factor for aspirin resistance in Chinese ischemic stroke patients. The results are the first to show the association of the TBXA2R rs1131882 with platelet reactivity during aspirin therapy in subjects with ischemic stroke. TBXA2R rs1131882 is located in the common coding portion of isoforms alpha and beta, and it represents a synonymous substitution. As such, TBXA2R rs1131882 may affect the transcription and/or translation efficiency of both isoforms of the TBXA2R gene by being in linkage disequilibrium with one or several SNPs in the promoter region of the gene in the intronic silencer region or in the enhancer region. Further study is required to assess our speculations and reveal the underlying mechanisms.

Lipoprotein-associated phospholipase A2 (Lp-PLA2) is related to lipoprotein metabolism and inflammatory pathways, and it is a potent mediator of hypersensitivity and inflammatory reactions. Lipoprotein-associated phospholipase A2 is a platelet activator that has been reported to be a predictive factor for stroke and TIA9,20. Lp-PLA2 is encoded by the PLA2G7 gene, which is located on chromosome 6q21-p12. One non-synonymous polymorphism, A379F (rs1051931) in exon 11 of PLA2G7, has been associated with Lp-PLA2 enzymatic activity, and another non-synonymous SNP (rs7756935), which was in perfect linkage disequilibrium with rs1051931, was shown to be associated with Lp-PLA2 activity. In the present study, we found that the rs1051931A-rs7756935C haplotype was associated with higher platelet aggregation in response to ADP and was correlated with an increased risk of aspirin resistance in Chinese ischemic stroke patients. The current results were supported by the findings of Kruse et al. which indicated a decrease in the affinity of 379Valine (rs1051931AA genotype) in Lp-PLA2 recombinant protein for its substrate (platelet-activating factor, PAF), resulting in two-fold lower Lp-PLA2 activity in vitro21. In addition, Grallert et al demonstrated that rs7756935 was associated with Lp-PLA2 activity or mass22, and a subsequent study reported that the rs7756935C allele was associated with the risk of coronary heart disease (CHD)23. To date, this study is the first to report an association between aspirin resistance and PLA2G7 polymorphisms in Chinese ischemic stroke patients.

Platelet endothelial aggregation receptor-1 (PEAR1) is a recently identified platelet transmembrane protein that is activated by platelet contact. During platelet aggregation by various agonists, the membrane expression of PEAR1 and its tyrosine phosphorylation (phosphorylated at Tyr-925 and Ser-953/1029) increase. The polymorphisms of PEAR1, which contribute to altered platelet function, are deemed an important factor in aspirin resistance24. A recent GWAS study identified that a variant in intron 1 of the PEAR1 gene (rs12566888) was associated with ADP and epinephrine-induced aggregation, and the rs12566888T allele was associated with a decrease in aggregation response25. Another common variant (rs12041331) in intron 1 was in tight linkage disequilibrium with rs12566888, and it has been reported that the G allele was associated with greater platelet aggregation in the presence and absence of aspirin treatment in African Americans. In addition, the researchers found that the PEAR1 protein expression was greatest for the GG homozygote, intermediate for the GA heterozygote and least for the AA homozygote26. However, in our study, no association was observed between platelet activity during aspirin therapy and rs12566888/rs12041331. It was speculated that the allele frequencies of PEAR1 SNPs might have significant inter-ethnic differences, and a possible limitation in this study is the limited sample size. Therefore, further studies are required to clarify the precise mechanism involved in our findings.

In a recent analysis, MDR1 C3435T was reported to be associated with clinical outcome of aspirin treatment evaluated by the modified Rankin Scale score in ischemic stroke patients. The authors speculated that MDR1 C3435T might be involved in the mechanism responsible for aspirin resistance27. Our study evaluated a different ischemic stroke population from China using both the optical platelet aggregation method and assessment of platelet TXB2 synthesis. We found that the homozygous mutant (TT genotype) was associated with lower platelet aggregation in response to ADP/AA and was correlated with a decreased level of platelet TXB2 synthesis. Our data suggest that the 3435 TT genotype of MDR1 is a protective factor against aspirin resistance in Chinese ischemic stroke patients. The current results were supported by the functional findings of Wang et al, who showed that the 3435TT genotype resulted in lower expression and function of P-gp28, thus decreasing the efflux and increasing the bioavailability of aspirin.

In summary, we have shown that MDR1 C3435T, TBXA2R (rs1131882) and PLA2G7 (rs1051931–rs7756935) may be associated with platelet activity during aspirin therapy. The MDR1 3435 TT genotype is a protective factor, while the TBXA2R rs1131882 CC, PLA2G7 rs1051931 AA-rs7756935 CC genotypes are risk factors for aspirin resistance. To date, our study is the first to report the associations of MDR1, TBXA2R, and PLA2G7 polymorphisms with platelet reactivity in Chinese ischemic stroke patients receiving aspirin therapy. These results may be helpful for aspirin treatment in ischemic stroke. However, due to the relatively small sample size, a further study with a larger sample size is needed to provide sufficient power for drawing firm conclusion about our findings.

Author contribution

Ling-ling PENG, Jia-li LI, Jing JIN, Min HUANG, and Ye-feng CAI designed the research; Ling-ling PENG, Yuan-qi ZHAO, Zi-yi ZHOU, Min ZHAO, Xin-meng CHEN and Lin-yan CHEN performed the research; Ling-ling PENG and Jia-li LI analyzed the data; Ling-ling PENG and Jia-li LI wrote the paper.