Comparison Of Plasma Kallikrein Inhibition By The Endogenous C1-Inhibitor Versus DX-2930, a Monoclonal Antibody Inhibitor

Dysregulated plasma kallikrein proteolytic activity leads to edematous attacks in hereditary angioedema (HAE) and has been associated with inflammation and thrombosis. Plasma kallikrein (pKal) is a serine protease that circulates as prekallikrein, a zymogen, which, together with factor XII (FXII) and high molecular weight kininogen (HMWK), constitutes the contact system. Activation of the contact system following assembly of FXII, HMWK, and prekallikrein on a negatively charged surface promotes inflammation via the generation of bradykinin and triggers intrinsic pathway coagulation via formation of activated coagulation factor XIa. Normal hemostasis appears not to be mediated by the contact system as individuals deficient in contact system proteins are not at risk for bleeding. However, the contact system has been shown to be pathologically activated by agents that include misfolded proteins, platelet polyphosphate, and implanted devices. Therefore, pharmacologic modulation of the contact system may attenuate thrombosis and inflammation without disrupting normal hemostasis. C1 inhibitor (C1-INH) is a serpin and a key endogenous, protein-based, inhibitor of pKal activity. HAE is caused by autosomal dominant mutations in the C1-INH gene resulting in functional protein levels that are approximately 30% or less than normal (16-33 mg/dL or 1.6-3.3 μM). Prekallikrein is present in plasma at a concentration of approximately 500 nM and it has been estimated that only 30-110 nM is converted to active pKal during an HAE attack. This study investigates the requirement for super-stoichiometric amounts of endogenous C1-INH to adequately regulate pKal activity. In vitro enzyme inhibition kinetics experiments with purified proteins show that the need for high concentrations of C1-INH is likely due to its relatively slow association rate constant (1.7 x 10⁴ M^-1s^-1). In contrast, DX-2930, a human monoclonal antibody inhibitor of pKal being developed for prophylactic treatment of HAE, potently inhibited pKal (Ki = 125 pM) with a faster association rate constant (3.4 x 10⁶ M^-1s^-1). Contact activation was observed in human plasma activated by the addition of ellagic acid and monitored using a pKal-selective synthetic peptide substrate. Consistent with the data obtained using purified proteins, the apparent IC₅₀ observed upon adding exogenous C1-INH to normal human plasma was approximately 100-fold higher than that of DX-2930. Using a Western blot assay to monitor endogenous HMWK cleavage in activated plasma we similarly observed that stoichiometric additions of DX-2930 were sufficient to prevent HMWK proteolysis by active pKal; whereas significantly higher concentrations of C1-INH (e.g. 1 µM) were required to block HMWK proteolysis. Active pKal can bind endothelial cells via interactions between the non-catalytic domain of pKal with HMWK, which binds receptors (urokinase receptor, cytokeratin 1, and the globular C1q receptor) present on endothelial cells. Cell bound pKal is likely to be a physiologically relevant form of the enzyme and may provide an explanation for attack localization in HAE. In this study, active pKal was assembled in vitro on cultured human umbilical vein endothelial cells (HUVEC) and binding of a range of concentrations of either biotinylated C1-INH or biotinylated DX-2930 was observed using streptavidin-horseradish peroxidase as detection. The data obtained demonstrates that C1-INH bound cell-associated pKal with > 200-fold less potency than DX-2930. Regarding protease inhibition specificity, while DX-2930 did not inhibit any of 20 tested serine proteases at a concentration of 1 µM, C1-INH is known to inhibit multiple serine proteases. This study demonstrates that effective regulation of pKal activity requires high concentrations of C1-INH (≥ 1 µM), which are necessary to drive the kinetics of this second order, irreversible interaction. These high inhibitory concentrations of C1-INH match the normal range and provide a potential kinetic mechanism for why HAE attacks can occur at C1-INH levels that exceed expected levels of activated pKal. Furthermore, the broad specificity of C1-INH towards other proteases that could be activated during disease could sufficiently deplete C1-INH levels and thereby lead to dysregulated pKal activity.