The mRNA encoding TAFI is alternatively spliced in different cell types and produces intracellular forms of the protein lacking TAFIa activity.

TAFI (thrombin-activatable fibrinolysis inhibitor) is a pro-carboxypeptidase, encoded by the CPB2 gene in humans that links the coagulation cascade to fibrinolysis and inflammation. The liver is the main source for plasma TAFI, and TAFI expression has been documented in platelets and monocyte-derived macrophages. A recent study reported an alternatively spliced CPB2 mRNA variant lacking exon 7 (∆7) in HepG2 cells and liver. Another study identified a CPB2 mRNA variant lacking exon 7 and a 52 bp deletion in exon 11 (∆7+11) in human hippocampus. We have examined alternative splicing of CPB2 mRNA in various cell types by RT-PCR and have assessed the functional properties of TAFI variants encoded by these transcripts by recombinant expression in mammalian cells. We identified the Δ7 exon skipping event in liver, Dami megakaryoblasts, THP-1-derived macrophages, peripheral blood mononuclear cells, platelets, testis, cerebellum, and SH-SY5Y neuroblastoma cells. The Δ11 alternative splicing event was notably absent in liver cells. We also detected a novel exon Δ7+8 skipping event in liver and megakaryocytes. Of note, we detected non-alternatively spliced CPB2 transcripts in brain tissues, suggesting the expression of full-length TAFI in brain. Experiments using cultured mammalian cells transfected with wild-type CPB2-, ∆7-, ∆7+11-, and ∆11-cDNA revealed that alternatively spliced TAFI is stored inside the cells, cannot be activated by thrombin-thrombomodulin, and does not have TAFIa activity. The alternative splicing events clearly do not give rise to a secreted protein with basic carboxypeptidase activity, but the intracellular forms may possess novel functions related to intracellular proteolysis.

Regulation of the gene encoding human thrombin-activatable fibrinolysis inhibitor by estrogen and progesterone.

Thrombin-activatable fibrinolysis inhibitor (TAFI) is a pro-carboxypeptidase B-like pro-enzyme that upon activation attenuates fibrinolysis and inflammatory processes. There is a large inter-individual variability in plasma TAFI levels within the population which is primarily due to epigenetic factors. Novel associations between plasma TAFI levels and sex steroids have triggered interest in determining the role of TAFI as a mediator of the cardioprotective effects of estrogens and progestins, or as a mediator of the increased thrombotic risk that accompanies use of oral contraceptives or hormone replacement therapy (HRT). In this study, we measured the effect of sex steroids on hepatic expression of CPB2, the human gene encoding TAFI. Using human hepatocellular carcinoma cells cultured in the presence of progesterone and 17ß-estradiol, we demonstrated that the level of TAFI protein is decreased by those sex steroids. These changes in protein expression were paralleled by decreases in CPB2 mRNA abundance and promoter activity. We did not find evidence of estrogen or progesterone receptor binding sites in the CPB2 promoter region, suggesting that the genomic effects of progesterone and 17ß-estradiol are mediated indirectly, without receptor binding to the CPB2 promoter. Indeed, we found that the effect of estrogen was independent of the estrogen receptor and was mediated through a novel signaling pathway dependent on phosphatidylinositol 3-kinase and Akt that did not involve GPR30. Our findings provide the molecular explanation for the ability of female sex steroids to decrease plasma TAFI concentrations.

Regulation of the mouse gene encoding TAFI by TNFα: role of NFκB binding site.

Thrombin-activable fibrinolysis inhibitor (TAFI) is a plasma pro-carboxypeptidase, encoded by the gene CPB2, with roles in both inhibition of fibrinolysis and inflammation. In mice, plasma TAFI levels and hepatic CPB2 mRNA expression were found to increase within 24h after intra-peritoneal lipopolysaccharide (LPS) injection. On the other hand, plasma TAFI in humans decrease in experimental endotoxemia and sepsis and we have previously demonstrated that CPB2 mRNA abundance in human hepatoma cells is decreased by inflammatory cytokines. Here, we have evaluated the effects of TNFα on mouse CPB2 expression. Treatment of primary mouse hepatocytes or the mouse hepatic cell line FL83B with TNFα for 12-48h resulted in increases in CPB2 mRNA abundance of up to 2-fold; mouse TAFI protein levels secreted from FL83B cells increased 2.7-fold after 48h treatment with TNFα. When FL83B cells were transfected with reporter plasmids containing the mouse CPB2 5'-flanking region, treatment with TNFα for 24 and 48h resulted in a 1.5-fold increased mouse CPB2 promoter activity. Mutation of a putative NFκB site not conserved in the human gene ablated the increased promoter activity observed following TNFα treatment. This site binds NFκB as assessed by gel mobility shift assays, and TNFα treatment increases the translocation of NFκB from the cytoplasm to the nucleus of mouse hepatocytes. These results demonstrate that the unique NFκB site in the mouse CPB2 promoter is functional and mediates the upregulation of mouse CPB2 expression by TNFα via increase in NFκB translocation to the nucleus.