In a recent issue of Blood, Henry et al1 reported the identification of several single nucleotide polymorphisms (SNPs) in the promoter and 3′ untranslated region of thethrombin-activatable fibrinolysis inhibitor (TAFI)gene. They also investigated the 505A/G SNP in the coding region of the TAFI gene resulting in an amino acid (aa) substitution, Thr147Ala.2 Henry et al1 showed that plasma TAFI antigen (Ag) levels are strongly associated with all these SNPs that turned out to be in strong linkage disequilibrium. This association confirmed and extended preliminary reports on the association of TAFI plasma levels and SNPs in the TAFI gene promoter.3 4
We identified another SNP, 1040C/T, in the coding region of theTAFI gene by comparing published sequences (GenBank no. NM_001872 and NM_016413). This SNP also results in an aa substitution (Thr325Ile). This SNP is of particular interest because it has been shown that the presence of an Ile325 residue has a positive influence on both TAFIa activity and stability in vitro resulting in increased antifibrinolytic activity.5 On the contrary, the Thr147Ala substitution has no effect on the functional properties of TAFI in vitro.2
We determined in a group of 152 blood donors (mean age 45.8 years, SD 11.7, range 19-71 years, 94 males and 58 females) the genotype frequency and allele frequency of the 1040C/T SNP and correlated this with TAFI Ag levels.6
The 1040C/T SNP was significantly associated with TAFI Ag levels with the C/C genotype corresponding with the highest and the T/T genotype with the lowest TAFI Ag levels (Table 1). The 1040C/C genotype corresponds to Thr/Thr at position 325. Activated TAFI-Thr325 has a normal in vitro half-life of 8 minutes (at 37°C), whereas activated TAFI-Ile325 has a half-life of about 16 minutes and a 60% greater antifibrinolytic activity compared to TAFI-Thr325.5 The reduced antifibrinolytic activity of TAFI-Thr325 is very likely the result of the greater instability of this isoform. It is remarkable that the Thr325 allele, which is associated with increased TAFI Ag levels in vivo, encodes a TAFI isoform that is less stable after activation in vitro. At present, it is not known whether the substitution of Thr325Ile will make TAFI less susceptible to degradation by plasmin via an effect on the cleavage at Arg328.7
To determine if the 1040C/T SNP is in linkage disequilibrium with the SNPs studied by Henry et al,1 we also determined the genotype and allele frequencies of the −438A/G TAFIpromoter SNP and the 505A/G SNP in the coding region of theTAFI gene. The −438A/G and 505A/G SNP showed similar genotype distribution and allele distributions as described before1 (Table 1) and all 3 analyzed SNPs were in strong linkage disequilibrium (P < .001, χ2analysis). The −438G/G and the 505A/A genotype correspond to the highest TAFI Ag levels (Table 1), which is also in agreement with earlier observations.1,3 4
In conclusion, the 1040T/C SNP located in the coding region of the TAFI gene and resulting in the Thr325Ile substitution is associated with TAFI plasma levels, just like other SNPs in and around this gene. Presently, it is not known which SNP is/are responsible for this effect on TAFI levels. On the other hand, studies using recombinant proteins have demonstrated a functional effect of the Thr325Ile substitution on the stability of activated TAFI resulting in altered antifibrinolytic activity. The possibility that the same TAFI haplotype will affect levels and antifibrinolytic activity in opposite directions will hamper the interpretation of genetic association studies using this SNP.
