The Role of Lysine Residues 42, 43, and 44 in the Activation Peptide Region of TAFI in Its Activation by the Thrombin-Thrombomodulin Complex.

Abstract 3187

Poster Board III-124

Thrombin-activatable fibrinolysis inhibitor (TAFI) is a 60 kDa plasma protein that is activated to the enzyme TAFIa, by a single cleavage at Arg92 by thrombin, plasmin or trypsin. TAFIa is a carboxypeptidase B-like enzyme that attenuates fibrinolysis. Thrombomodulin (TM) is a cofactor which increases the overall efficiency of thrombin-mediated TAFI activation by 1250-fold. Thus, the thrombin-TM complex is believed to be the physiological TAFI activator. The minimal structure of TM required for efficient TAFI activation contains the EGF-like domains 3 through 6. New structure models have postulated that the C-loop of TM EGF-like domain 3 has a negatively charged molecular surface that could interact with several positively charged surface patches on TAFI. One positively charged surface patch of TAFI consists of the three consecutive lysine residues at positions 42, 43, and 44, which are unique to the TAFI activation peptide as no corresponding residues exist in rattus, bovine or human tissue procarboxypeptidases A and B. More interestingly, all three lysine residues are conserved in human, rattus, murine and canine TAFI, but not for bovine TAFI which only has a single lysine residue at position 42. We previously reported that when the three lysine residues are substituted by alanine residues (K42/43/44A), compared to the wild-type, the catalytic efficiencies for TAFI activation by thrombin-TM complex decreased 8-fold. In order to identify which residue(s) are key for TAFI activation by the thrombin-TM complex, combinations of mutations of the three lysine residues were constructed and expressed. TAFI wild-type or mutants were activated by thrombin for 10 minutes in the absence or presence of TM at varying levels. At this point, the levels of TAFIa formed were measured by adding the synthetic substrate AAFR containing PPAck and measuring the absorbance change at 349nm. The rates were used to determine the kinetic parameters of TAFI activation. The non-linear regression analysis with the NONLIN module of SYSTAT returned best fit values along with their asymptotic standard errors (A.S.E) for the kinetic parameters of TAFI activation (k_cat, K_m, and K_d). The value of K_d (the dissociation constant for the thrombin-TM interaction) is assumed to be the same for wild-type TAFI and the mutants, because all reactions have this interaction in common. The regression analysis yielded K_d = 22.4 ± 1.3 nM for this interaction. This value agrees favourably with a value of 22 nM measured directly and reported previously. The k_cat values (1/sec) ranged from 1.06 ± 0.18 (K44A) to 1.19 ± 0.18 (K43A). The value for wild-type TAFI was 1.50 ± 0.63 (1/sec). K_m values ranged from 1.14 ± 0.73 μM (WT) to 3.01 ± 2.17 μM (K42A). The k_cat / K_m ratios (1/sec/μM), which provides the best indication of overall catalytic efficiency, ranged from 1.43 ± 0.27 (WT) to 0.43 ± 0.17 (K42A). When the three lysine residues are individually substituted by alanine residues (K42A, K43A, and K44A), compared to the wild-type, their catalytic efficiencies (k_cat / K_m) for TAFI activation by the thrombin-TM complex decreased 3.3-fold for K42A, 1.83-fold for K43A, and 1.96-fold for K44A. When Lys43 and Lys44 are substituted by alanine residues simultaneously (K43/44A), its catalytic efficiency decreased 3.3-fold. Together, our data show that each of these lysine residues on the activation peptide of TAFI may contribute partially to the interactions of TAFI with the thrombin-TM complex that are needed for efficient activation. In addition, the effects of the mutations may be additive.