The results of the present study provide evidence that the variant –2345 2G/1G and –1690 A/G polymorphisms of TAFI are associated with the risk of developing ACI in contrast with the –438 A/G and +1583 A/T polymorphisms. To the best of our knowledge, this result is the first to suggest an association of the –2345 2G/1G and –1690 A/G polymorphisms of TAFI with ACI in a Chinese population. Further stratification by gender revealed that the –438 A/G polymorphism of TAFI is associated with ACI in male patients. In addition, haplotype analysis demonstrated that four haplotypes are significantly associated with ACI.
TAFI is a 58-kDa plasma glycoprotein secreted by hepatocytes as an inactive form . Upon activation by thrombin, plasmin, or the thrombin-thrombomodulin complex, TAFI is transformed into a carboxypeptidase B-like enzyme (TAFIa) [2, 25]. TAFIa plays an important role in regulating the balance between coagulation and fibrinolysis by inhibiting fibrinolysis. Previous studies have indicated that high levels of TAFI in circulating plasma increase the risk of cardiovascular death during the acute phase of ischemic stroke . In addition, recent studies have reported that TAFI deficient mice are also susceptible to intracerebral thrombosis and ischemic stroke . These lines of evidence lead to the hypothesis that TAFI is of pathogenic significance to ACI.
Several studies have analyzed the relationship between TAFI polymorphisms and cardiovascular disease, with different and even contradictory conclusions drawn from various ethnical populations. While one study found that the 147Thr allele of the Ala147Thr polymorphism protected against MI , in another study, the 147Thr allele was associated with a higher risk of angina . TAFI polymorphisms C+1542G and Thr325Ile were also found to be related to the type of ACS  and thrombotic microangiopathies . Regarding cerebrovascular diseases, Akatsu et al. did not find any association between the Ala147Thr and Thr325Ile polymorphisms of TAFI and cerebral infarction . However, a prospective cohort study in Europe demonstrated that the Thr325Ile polymorphism of TAFI is associated with the incidence of stroke and the age at onset of first stroke in patients . Ladenvall et al. did not find any association between a series of tagging SNPs of the TAFI gene, but an increased risk was found between TAFI polymorphisms and stroke subtypes . In the present study, we show, for the first time, that the −2345 2G/1G and −1690 A/G polymorphisms of TAFI were associated with ACI in a Chinese population. As demonstrated by Henry et al., the four polymorphisms investigated in the present study were in strong linkage disequilibrium with each other and with the previously described Ala147Thr polymorphism . It is most likely that these polymorphisms function synergistically to contribute to the disease phenotype.
It has been widely recognized that plasma TAFI levels are strongly genetically controlled [17, 31]. TAFI levels are under the control of several SNPs located in the regulatory and coding regions of the TAFI gene [17, 18, 25, 32]. A transethnic study using fine mapping of quantitative trait nucleotides (QTNs) underlying TAFI antigen levels identified that the −2345 2G/1G SNP located in the 5' region might be one of the QTNs  and that −2345 1G alleles were independently associated with increased TAFI levels. The −1690 A/G polymorphism is located two bases downstream from a potential binding site for the HFH3 transcription factor, and the −438 A/G polymorphism is located one base downstream from a potential binding site for the VBP transcription factor . These polymorphisms may alter the binding of the corresponding transcription factors to the promoter, thus affecting promoter activity and contributing to varied TAFI expression. In the 3’-UTR, there are three transcription termination sites . The interaction of cis-acting elements within the 3'-UTR of TAFI with IL-1 and IL-6 significantly increased the stability and frequency of mRNA transcripts . In addition, polymorphisms in the 3’-UTR may alter TAFI mRNA abundance by influencing mRNA processing or stability. The haplotype carrying the +1583 T allele reduces the stability of TAFI transcripts, suggesting the +1583 A/T allele is also related to TAFI antigen levels [14, 15]. Data from Africans and Europeans converged to identify the 1583 T/A polymorphism as one of the QTNs contributing to elevated TAFI levels. In sum, polymorphisms both in the promoter and 3’-UTR regions of TAFI can affect the binding of transcription factors or mRNA stability, thereby influencing gene expression.
The stratified analysis by gender revealed that among males, those with the variant −438 A allele and AA genotype had a significantly higher risk of ACI, while among women, no statistical significance was found. TAFI levels in patients with polycystic ovary syndrome (PCOS) were significantly higher than in control groups [36, 37]. It is known that PCOS patients express high levels of androgen [36, 37]. The androgen may affect TAFI production by some unknown mechanism. Therefore, it is conceivable that the −438 AA allele that is associated with higher androgen levels could also increase TAFI levels in male ACI patients. However, more functional studies on the relationship between sex hormones and TAFI levels should be carried out.
In genes with multiple susceptibility alleles, particularly when the LD between polymorphisms is weak, haplotype-based association studies have advantages over analyses based on individual polymorphisms . In the present study, haplotype analysis identified four haplotypes significantly associated with ACI. H1(1G/G/G/T) and H4(2G/G/G/T) were associated as protective haplotypes, while H2(2G/A/G/A) and H3(2G/A/G/T) were risk haplotypes. The haplotype carrying the G allele of −1690 is a protective haplotype, while the A allele of −1690 is a risk haplotype. Ladenvall et al. found that TAFI genotypes and haplotypes showed significant associations with TAFI levels. While no haplotype was found to have a significant association with overall ischemic stroke, subtype analysis revealed an association between the H2B haplotype and cryptogenic stroke and an association between the H1B haplotype and small-vessel disease . As these results illustrate, haplotype analysis is helpful for deciphering the relationship between multiple SNPs and disease susceptibility. Our findings suggest that the haplotype of TAFI could be genetic a marker for ACI.
There are several limitations to this study. The small number of participants in this study was insufficient and therefore may have led to non-representative results. This limitation may have suppressed a true relationship due to a type II statistical error. A sample size with sufficient statistical power is critical to analyze the genetic associations of causal genes with complex diseases susceptibility. The sample size for detecting associations between disease and SNP markers is known to be highly affected by disease prevalence, disease allele frequency, linkage disequilibrium, inheritance models (e.g., additive, dominant, and multiplicative models), and effect size of the genetic variants (e.g., odds ratio, relative risk, etc.). Particular attention must be paid to sample sizes when more than one variable is studied simultaneously, such as multiple imperfectly correlated traits, intergenic interactions or gene-environment interactions . Additionally, the comparatively small sample size, together with other environmental risks may mask the genuine differences in allele frequencies between cases and controls. Hong et al. suggested that a lower sample size for testing more common SNPs with stronger effect sizes and increased LD between marker allele and disease allele might contribute to achieve adequate statistical power . Therefore, our results should be interpreted with caution. In addition, selection bias in the patient or control populations cannot be entirely excluded. The other clinical characteristics of the study group, such as hypertension, diabetes or hypercholesterolemia, might have complicated the associations between TAFI polymorphisms and ACI. Other functional polymorphisms might also influence TAFI expression, and their combined effects must be studied to improve the prediction of the occurrence, severity, and outcomes of ACI. Furthermore, the levels of TAFI were not evaluated. However, the primary purpose of our study was to establish a genetic reference for future studies. Thus, this investigation focused on assessing the association of different genetic polymorphisms of TAFI with the risk of developing ACI. Larger patient and control cohorts will be needed to confirm the association of TAFI gene polymorphisms with ACI in other populations.