The G haplotype increases TFPIα function. (A) As in Figure 1A, a schematic representing exon 13 with flanking introns. However, the x-axis has been extended at the 3′ end to indicate rs10800453, which was shown in the INTERVAL study to be associated with increased plasma TFPI levels .⁵⁰ The alternate allele of rs10800453, A, is in complete LD with the 4 SNVs (rs6032, rs4525, rs4524, and rs6021), originally assigned as the G haplotype (marked with asterisks), as are 22 other SNVs (supplemental Table 2) of which 5 are plotted on this diagram because they have genetic coordinates within the range of the x-axis. (B-E) The functional effect of the extended G haplotype on the TFPIα pathway was tested using thrombin generation assays. Thrombin generation was initiated by 1 pM TF in citrate plasma treated with corn trypsin inhibitor (CTI) in the presence of 4 μM phospholipids (20:60:20 DOPS/DOPC/DOPE), 16 mM CaCl₂, and the presence or absence of inhibitory anti-TFPI Abs (77.5 μg/mL) (supplemental Methods). (B) Representative thrombin generation curves are shown for the median samples (based on LT ratio) of the F5-G/G and F5-ref groups. The arrows indicate the LT for each thrombin generation curve. (C-D) The effect of the G haplotype on the thrombin LT is shown in the absence (C) and presence (D) of inhibitory anti-TFPI Abs. (E) The ratio of thrombin LT obtained in the presence and absence of anti-TFPI Abs was calculated. For panels C-E, each data point represents the mean of 3 independent replicate analyses of individual samples and the height of the bar represents the median and the error bar the IQR. The green square refers to the results obtained from a plasma pool of 15 healthy individuals (supplemental Methods) as a comparison with the F5-ref group. This point was excluded from all statistical analyses. N.B. ns, nonsignificant; ∗P < .05; ∗∗P < .01.