Rational Design of Protein C Activators

Abstract

In addition to its procoagulant and proinflammatory functions mediated by cleavage of fibrinogen and PAR1, the trypsin-like protease thrombin activates the anticoagulant protein C in a reaction that requires the cofactor thrombomodulin and the endothelial protein C receptor. Once in the circulation, activated protein C functions as an anticoagulant, anti-inflammatory and regenerative factor. Hence, availability of a protein C activator would afford a therapeutic for patients suffering from thrombotic disorders and a diagnostic tool for monitoring the level of protein C in plasma. Here, we present a fusion protein where thrombin and the EGF456 domain of thrombomodulin are connected through a peptide linker. The fusion protein recapitulates the functional and structural properties of the thrombin-thrombomodulin complex, prolongs the clotting time by generating pharmacological quantities of activated protein C and effectively diagnoses protein C deficiency in human plasma. Notably, these functions do not require exogenous thrombomodulin, unlike other anticoagulant thrombin derivatives engineered to date. These features make the fusion protein an innovative step toward the development of protein C activators of clinical and diagnostic relevance.

Figure 1

Protein C activation (A,B) and clotting of fibrinogen (C,D) by wild-type thrombin and FPs. (A) Shown are progress curves of D-DRR-pNA hydrolysis by aPC generated with thrombin alone (0.1 nM, light gray), thrombin in complex with 10 nM TM (green), 500 nM TM456 (black) and FP31 (orange). PC concentration was 50 nM. Experimental conditions are 5 mM Tris, 0.1% PEG, 145 mM NaCl, and 5 mM CaCl₂, pH 7.4 at 37 °C. Progress curves were analyzed as described earlier to retrieve the values of specificity (k_cat/K_M) reported in Table 1. In the absence of TM, the activity of thrombin is negligible. Addition of TM, TM456 or fusing TM456 to thrombin rescues the activity of the enzyme. (B) Shown is the change in the specificity constant (s = k_cat/K_M) due to mutation, expressed as log s_mut/s_wt, under experimental conditions of 5 mM Tris, 0.1% PEG, 145 mM NaCl, and 5 mM CaCl₂, pH 7.4 at 37 °C. Each FP was tested at a concentration of 0.1 nM. The values of specificity extracted from analysis of the progress curves are also reported in Table 1 and are the average of 3 independent determinations. (C) As fibrinogen is converted to fibrin the light is scattered through the fibers and the signal can be recorded at 350 nm. Clotting curves were obtained for thrombin alone (0.3 nM, light gray), in complex with 10 nM TM (green), 500 nM TM456 (black) and FP31 (orange). The presence of TM456 either added to the solution or fused to the enzyme prolonged but did not abolish the ability of thrombin to clot fibrinogen. (D) A similar effect was obtained with FPs but the magnitude of such effect was dependent on the length of the peptide linker. FP31 and FP41 were more effective than 500 nM exogenous TM456 whereas FP69 showed barely any effect. The other constructs displayed an intermediate behavior. The bell-shaped dependence on the length unequivocally proves that the interaction between EGF456 and thrombin is intramolecular (concentration-independent) and not intermolecular. From this analysis, maximum prolongation of the clotting time was achieved with a linker 31 residues long. Each FP was tested at a concentration of 0.3 nM and the values reported in Fig. 2Dare the average of three independent determinations.

Figure 2. Optimization of the fusion proteins by site directed-mutagenesis.

Shown is the change in the specificity constant (s = k_cat/K_M) for fibrinogen (gray), extracellular fragment of PAR1 (red) and PC activation (blue) due to mutation, expressed as log s_mut/s_wt. The values of specificity for thrombin wild-type are 17 μM⁻¹s⁻¹ for fibrinogen, 30 μM⁻¹ s⁻¹ for PAR1 and 0.12 mM⁻¹s⁻¹ for PC. Fusing TM456 to thrombin with a linker of 31 residues (FP31) reduces 60- and 100-fold the ability of thrombin to cleave fibrinogen and PAR1 respectively but stimulates PC activation. Substitution of W215 alone (FP31_W215A) or W215/E217 (FP31_WE) with alanine further decreases the activity toward fibrinogen and PAR1. Yet PC activation is retained. Data are reported in Table 1 and are the average of three independent determinations.

Figure 3. Biochemical and structural validation of the FP.

Binding of thrombin wild-type (WT) and mutants W215A and WE to TM456 monitored by SPR. (A) Shown are the sensograms of the interaction between thrombin WE (1000–1.56 nM) and immobilized TM456. (B) Plot of response units (RU) as a function of the concentration of thrombin WT (solid circles), W215A (solid triangles) and WE (solid squares). Solid lines were drawn according to a simple binding equation with best-fit parameters listed in Table 2. (C) Crystal structure of FP31_WE showing a surface representation of WE (pale green) and EGF456 (gold) docked onto exosite I. Also shown are PPACK (yellow sticks) bound to the active site and a 7-unit sugar chain (gold sticks) covalently attached to N364 in EGF4. Only the initial GGGS sequence of the linker (cyan sticks) attached to the C-terminus of WE is visible in the density map, with the rest of the linker being completely disordered. The Cα-Cα distance between Ser in this linker and E346 in the N-terminus of EGF4 is 57 Å and long enough to accommodate the rest of the linker. The structure of thrombin bound to EGF456 of TM reported previously as 1DX5 is shown in cartoon representation (salmon) for comparison. The rmsd between the two structures is only 0.29 Å.

Figure 4. Effect of FPs on the aPTT.

(A) In human plasma, FP31_W215A (empty circles) and FP31_WE (filled circles) increase aPTT linearly in a dose-dependent manner featuring an anticoagulant profile. (B) Unlike aPC (25 nM), the anticoagulant effect of FPs (25 nM) requires PC since no prolongation of the aPTT time was observed with PC depleted plasma (right panel). Due to the lack of soluble TM in normal and PC deficient plasma, thrombin WE (25 nM) features slightly reduced aPTT. aPTT values are shown as the ratios of the test to normal. Reference values are 39 ± 2 s for normal plasma and 42 ± 5 for PC depleted plasma. The dashed red line identify the baseline. Each measurement is the average of four individual determinations.

Figure 5. Detection and quantification of the level of protein C in human plasma.

(A) Standard calibration curves were generated by replacing Protac with the FPs and following manufacturer’s instructions. Protac, FP31_W215A and FP31_WE efficiently convert PC to aPC, which is detected using S2366 as chromogenic substrate (Abs at 405 nm). (B) Randomized healthy (n = 6) and abnormal control plasma samples (n = 6) were tested side by side with Protac. As expected, depletion of PC from plasma results in a lower signal at 405 nm. Under these experimental conditions, FP31_W215A and FP31_WE performed better than Protac providing an improved N/A ratio, i.e. 4.5 for FP31_W215A, 4.2 for FP31_WE and 3.6 for Protac.