Preservation of Beneficial TAFI Functions by an Activated Protein C Variant with Normal Cytoprotective Functions but Minimal Anticoagulant Activity.

Abstract

Activated Thrombin Activatable Fibrinolysis Inhibitor (TAFIa), a carboxypeptidase B-like enzyme, both inhibits fibrinolysis and exerts anti-inflammatory effects via removal of C-terminal Arg residues of bradykinin (BK) and the complement anaphylatoxins, C3a and C5a. Proteolytic activation of TAFI is caused by thrombin or thrombin-thrombomodulin; thus, TAFI activation can be down-regulated by the anticoagulant, activated protein C (APC). Recombinant APC reduces mortality in severe sepsis patients (PROWESS trial), but it increases risk of serious bleeding. This risk is presumably due to APC's anticoagulant activity and possibly exacerbated by profibrinolytic effects derived from down-regulation of thrombin-mediated activation of TAFI. In addition, APC therapy might ironically aggravate sepsis by reducing TAFIa generation because TAFIa inactivates BK, C3a and C5a, well known pathological mediators in sepsis. Recently we made a non-anticoagulant APC variant, 5A-APC (RR229/230AA and KKK191–193AAA), that retains normal in vitro cytoprotective effects and an ability to reduce mortality in murine sepsis models (Kerschen et al, J Exp Med, 2007). This current study compares the effects of this variant to wild type (wt) APC on profibrinolytic and TAFIa's anti-inflammatory activities in plasma-based assays. To determine the effects of APC on TAFIa-mediated inhibition of fibrinolysis, clot lysis was studied in a plasma system of thrombin-induced clot formation and tPA-mediated fibrinolysis. Compared to wt-APC (IC50=0.36 nM APC), almost 10-times more 5A-APC was required for 50% inhibition of clot lysis (IC50=3.1 nM 5A-APC) and 20-times more 5A-APC was needed to inhibit completely TAFI activation under the conditions employed. The effects of wt-APC and 5A-APC on TAFIa's anti-inflammatory activities were determined by following proteolytic inactivation of BK in plasma during tissue factor-induced clotting. In the absence of added thrombomodulin, angiotensin-converting enzyme, carboxypeptidase N, and TAFIa each inactivated approximately 10–15% of BK under the study conditions. But in the presence of thrombomodulin, BK was fully inactivated in plasma and converted to des-Arg9-BK (> 98%), an effect mainly attributable to TAFIa. Analysis of the time course of BK inactivation in the presence of thrombomodulin indicated that TAFIa rapidly converted BK to des-Arg9-BK with only a minor contribution from carboxypeptidase N. Under conditions where TAFIa fully converted BK to des-Arg9-BK, wt-APC, but not 5A-APC, dose-dependently diminished BK inactivation and des-Arg9-BK generation. At the highest tested concentrations of APC (40 nM), wt-APC completely inhibited TAFIa-dependent BK inactivation whereas non-anticoagulant 5A-APC permitted normal TAFI activation and showed no significant inhibition of TAFIa-dependent BK inactivation. Thus, 5A-APC left intact the TAFIa-mediated anti-inflammatory mechanism for inactivation of BK in plasma and, presumably, TAFIa's ability to inactivate the C3a and C5a anaphylatoxins. These studies show that genetic engineering can selectively alter APC's multiple activities to provide an APC variant that retains this enzyme's beneficial anti-inflammatory and cytoprotective effects but that diminishes bleeding risk due to reduction in APC's anticoagulant and profibrinolytic activities. Thus, these data support the hypothesis that preservation of TAFI activation in plasma by normally cytoprotective 5A-APC variant would preserve TAFIa-dependent anti-inflammatory effects that are lost with the use of wt-APC.