Molecular mapping of the thrombin-heparin cofactor II complex

We used 55 Ala-scanned recombinant thrombin molecules to define residues important for inhibition by the serine protease inhibitor (serpin) heparin cofactor II (HCII) in the absence and presence off glycosaminoglycans. We verified the importance of numerous basic residues in anion-binding exosite-1 (exosite-1) and found 4 additional residues, Gln²⁴, Lys⁶⁵, His⁶⁶, and Tyr⁷¹ (using the thrombin numbering system), that were resistant to HCII inhibition with and without glycosaminoglycans. Inhibition rate constants for these exosite-1 (Q24A, K65A, H66A, Y71A) thrombin mutants (0.02-0.38 × 10⁸ M^-1 min^-1 for HCII-heparin when compared with 2.36 × 108 M^-1 min^-1 with wild-type thrombin and 0.03-0.53 × 10⁸ M^-1 min^-1 for HCII-denaatan sulfate when compared with 5.23 × 10⁸ M^-1 min^-1 with wild-type thrombin) confirmed that the structural integrity of thrombin exosite-1 is critical for optimal HCII-thrombin interactions in the presence of glycosaminoglycans. However, our results are also consistent for HCII-glycosaminoglycan-thrombin ternary complex formation. Ten residues surrounding the active site of thrombin were implicated in HCII interactions. Four mutants (Asp⁵¹, Lys ⁵², Lys¹⁴⁵/ Thr¹⁴⁷/Trp¹⁴⁸, Asp ²³⁴) showed normal increased rates of inhibition by HCII-glycosaminoglycans, whereas four mutants (Trp⁵⁰, Glu ²⁰², Glu²²⁹, Arg²³³) remained resistant to inhibition by HCII with glycosaminoglycans. Using 11 exosite-2 thrombin mutants with 20 different mutated residues, we saw no major perturbations of HCII-glycosaminoglycan inhibition reactions. Collectively, our results support a "double bridge" mechanism for HCII inhibition of thrombin in the presence of glycosaminoglycans, which relies in part on ternary complex formation but is primarily dominated by an allosteric process involving contact of the "hirudin-like" domain of HCII with thrombin exosite-1.