The effect of corn trypsin inhibitor and inhibiting antibodies for FXIa and FXIIa on coagulation of plasma and whole blood: comment’

We read with interest the article by Hansson et al. [1] reporting that corn trypsin inhibitor (CTI), in addition to being an efficient inhibitor of factor (F) XIIa proteolytic activity, also partially inhibits FXIa amidolytic activity at concentrations used in many laboratories worldwide. Based on their data, the authors recommend using CTI for the prevention of contact pathway-initiated thrombin generation at concentrations not exceeding 20 µg mL⁻¹.

We have been using CTI for the prevention of the contact pathway in tissue factor (TF)-triggered thrombin generation and clot formation for almost two decades [2, 3]. The concentration of CTI in all our experiments has been 100 µg mL⁻¹ (8.0 µM), which is, by coincidence, almost identical to the K_i of 8.1 µM for FXIa reported by Hansson et al. [1]. Our choice of this CTI concentration has been based on the observation [2], that at lower concentrations, particularly at 20 µg mL⁻¹, CTI inhibition of contact pathway-related plasma clotting is quite limited. A similar conclusion could be drawn from the data published by Hansson and co-workers (see Fig. 1 in [1]): their data suggest that at recommended CTI concentrations (below 20 µg mL⁻¹), an efficient inhibition of the contact pathway is impossible, especially if the goal of an experiment is to analyze processes occurring in blood or plasma triggered with low TF concentrations [4–6]. It is even more important to efficiently inhibit contact pathway initiation when endogenous FXIa present in plasma of cardiovascular patients and of those with inflammation [7, 8] is quantitated. An inefficient inhibition of FXIIa would lead to the activation of FXI during a thrombin generation assay in the absence of an exogenous trigger. Conversely, inhibition of endogenous FXIa by CTI would underestimate the concentration of FXIa in a patient’s plasma.

To address these issues, we analyzed the effect of CTI on clotting of a congenital FXII-deficient plasma (George King Bio-Medical Inc., Overland Park, KS, USA) triggered with purified human FXIa. CTI (in-house preparation [9]) and FXIa (Haematologic Technologies Inc., Essex Junction, VT, USA) were added at varying concentrations to FXII-deficient plasma followed by the addition of either an APTT reagent (Trinity Biotech PLC Bray, Wicklow, Ireland) or 20 µM synthetic phospholipid vesicles [9] and recalcification. Concentrations of CTI varied between 25 µg mL⁻¹ and 300 µg mL⁻¹ and concentrations of FXIa between 50 and 500 pM. Clotting times were measured in the ST-8 apparatus (Diagnostica Stago, Parsippany, NJ, USA). Neither with the APTT reagent nor with synthetic phospholipids did the clotting times of FXII-deficient plasma become prolonged at any combination of FXIa and CTI concentrations, indicating the lack of detectable inhibition of FXIa activity by CTI.

The most likely cause of the discrepancy between the data published by Hannson et al. [1] and our results is the environment of FXIa in plasma. In the purified system, CTI has no competitors for the active site of FXIa (with the exception of a synthetic substrate which is an essential component of the assay), whereas in plasma there are several known natural substrates for FXIa, such as FIX [10–12], FXI [11], FV and FVIII [10], and a number of serine protease inhibitors [13]. All these substrates and inhibitors are competitors of CTI for the active site of FXIa and, due to the relatively low affinity of CTI for the active site of FXIa, they will minimize CTI binding and its influence on FXIa activity in plasma.

In conclusion, our data indicate that CTI, at concentrations as high as 300 µg mL⁻¹, does not inhibit FXIa activity in plasma and, presumably, in blood.