Stabilization of the E* form turns thrombin into an anticoagulant

Abstract

Previous studies have shown that deletion of nine residues in the autolysis loop of thrombin produces a mutant with an anticoagulant propensity of potential clinical relevance, but the molecular origin of the effect has remained unresolved. The x-ray crystal structure of this mutant solved in the free form at 1.55 A resolution reveals an inactive conformation that is practically identical (root mean square deviation of 0.154 A) to the recently identified E* form. The side chain of Trp(215) collapses into the active site by shifting > 10 A from its position in the active E form, and the oxyanion hole is disrupted by a flip of the Glu(192)-Gly(193) peptide bond. This finding confirms the existence of the inactive form E* in essentially the same incarnation as first identified in the structure of the thrombin mutant D102N. In addition, it demonstrates that the anticoagulant profile often caused by a mutation of the thrombin scaffold finds its likely molecular origin in the stabilization of the inactive E* form that is selectively shifted to the active E form upon thrombomodulin and protein C binding.

FIGURE 1.

Functional properties of the thrombin mutant Δ146–149e. Shown are the values of s = k_cat/K_m for the hydrolysis of chromogenic substrates H-d-Phe-Gly-Arg-p-nitroanilide (FGR), H-d-Phe-Pro-Phe-p-nitroanilide (FPF), H-d-Phe-Pro-Lys-p-nitroanilide (FPK), and FPR, fibrinogen (FpA), PAR1, PAR3, PAR4, protein C (PC), and protein C (PC+TM) or FPR (S+TM) in the presence of 100 nm thrombomodulin and 5 mm CaCl₂ for wild-type (s_wt) relative to the thrombin mutant Δ146–149e (s_mut). All substrates, except PAR4, experience a loss of activity that equals 2.30 log units (solid line) with a standard deviation of 0.09 log units (broken lines). This supports perturbation of the E*-E equilibrium in favor of the inactive form E* (see Equations 1 and 3 in the text). The larger loss for PAR4 (>4 S.D. compared with the other substrates) is likely due to direct interaction of the substrate with residues of the autolysis loop (55) that are missing in the Δ146–149e mutant. In the presence of thrombomodulin, the mutant experiences only a 2-fold drop in activity toward protein C compared with wild type. However, thrombomodulin binding alone does not restore activity toward the chromogenic substrate FPR. Experimental conditions are as follows: 5 mm Tris, 0.1% PEG8000, 145 mm NaCl, pH 7.4, at 37 °C. The values of s_wt are as follows: 0.52 ± 0.05 μm⁻¹ s⁻¹ H-d-Phe-Gly-Arg-p-nitroanilide (FGR), 0.28 ± 0.03 μm⁻¹ s⁻¹ H-d-Phe-Pro-Gly-p-nitroanilide (FGF), 4.2 ± 0.2 μm⁻¹s⁻¹ H-d-Phe-Pro-Gly-p-nitroanilide (FGK), 37 ± 1 μm⁻¹ s⁻¹ H-d-Phe-Pro-Arg-p-nitroanilide (FPR), 17 ± 1 μm⁻¹ s⁻¹ fibrinogen (FpA), 39 ± 1 μm⁻¹s⁻¹ PAR1, 0.35 ± 0.02 μm⁻¹ s⁻¹ PAR3, 0.34 ± 0.01 μm⁻¹ s⁻¹ PAR4, 59 ± 3 m⁻¹ s⁻¹ PC, 0.22 ± 0.01 μm⁻¹s⁻¹ PC+TM, 64 ± 2 μm⁻¹s⁻¹ S+TM.

FIGURE 2.

Argatroban binding to thrombin wild type (left) and Δ146–149e (right) measured by isothermal titration calorimetry. The top panel shows the heat exchanged in each individual titration for the thrombin sample (bottom trace) and the argatroban buffer control (top trace). The bottom panel is the integration of the data to yield the overall heat exchanged as a function of the ligand/protein molar ratio. Experimental conditions are 5 mm Tris, 0.1% polyethylene glycol, 145 mm NaCl, pH 7.4, 37 °C. The enzyme and argatroban concentrations are as follows: 13.44 and 140 μm (thrombin wild type); 52.5 and 777 μm (Δ146–149e). Titration curves were fit using the Origin software of the iTC200, with best fit parameter values as follows: thrombin wild type, K = 1.0 ± 0.1 10⁸ m⁻¹, ΔG = −11.3 ± 0.1 kcal/mol, ΔH = −15.2 ± 0.1 kcal/mol, and TΔS = −3.9 ± 0.1 kcal/mol; Δ146–149e, K = 7.4 ± 0.4 10⁵ m⁻¹, ΔG = −8.3 ± 0.1 kcal/mol, ΔH = −13.8 ± 0.1 kcal/mol, and TΔS = −5.5 ± 0.1 kcal/mol. The value of the stoichiometric constant N was 1.01 ± 0.01 for thrombin wild type and the Δ146–149e mutant.

FIGURE 3.

Ribbon representation of the structure of the thrombin mutant Δ146–149e (gold) overlaid with the structure of thrombin in the E conformation (35) (cyan). The newly formed peptide bond between Lys¹⁴⁵ and Gly¹⁵⁰ is indicated in red in the shortened autolysis loop of Δ146–149e (see also Fig. 4), and the loop in the E conformation is not visible between residues Trp¹⁴⁸ and Lys^149e. The 215–217 β-strand in the mutant collapses into the primary specificity pocket (red open arrowheads), with the side chain of Trp²¹⁵ (stick model) repositioned into the active site (residues of the catalytic triad His⁵⁷, Asp¹⁰², and Ser¹⁹⁵ shown as stick models) in hydrophobic interaction with Trp^60d, Tyr^60a, Leu⁹⁹, and His⁵⁷. This represents a drastic change (r.m.s.d. 0.384 Å) from the conformation of E where the side chain of Trp²¹⁵ is positioned 10.5 Å away and leaves the active site accessible to substrate. The conformation of Δ146–149e is remarkably similar (r.m.s.d. 0.154 Å) to that of E* determined recently (6, 7).

FIGURE 4.

Left, details of the collapse of Trp²¹⁵ into the active site and disruption of the oxyanion hole in the thrombin mutant Δ146–149e (CPK, yellow) are shown. The conformation of the same residues in the E form is shown by comparison (CPK, cyan). The peptide bond between Glu¹⁹² and Gly¹⁹³ is flipped in the Δ146–149e mutant (red open arrowhead), as seen in the E* form (6, 7, 9), causing disruption of the oxyanion hole contributed by the N atoms of Gly¹⁹³ and Ser¹⁹⁵. The 2F_o − F_c electron density map (green mesh) is contoured at 2.0σ. Right, deletion of residues ¹⁴⁶ETWTANVGK^149e in the autolysis loop of the Δ146–149e mutant results in a new peptide bond connection between Lys¹⁴⁵ and Gly¹⁵⁰ (CPK, yellow). The autolysis loop is rarely seen in its entirety in thrombin structures, and considerable disorder remains in the mutant Δ146–149e where the sequence ¹⁴⁴LKGQ¹⁵¹ must be contoured at 0.5σ in the 2F_o − F_c electron density map (green mesh).