Abstract

Multi-functional enzymes such as thrombin acquire specificity from exosite-dependent interactions. Peptide inhibition studies of protein-protein interactions are useful to identify exosite functions. We developed a novel general approach for exosite analysis using photocrosslinking of exosite inhibitory peptides that also carry a fluorescent tag and reporter. Here we demonstrate the method’s feasibility in studying the thrombin-hirugen (enzyme-peptide) complex. Thrombin contains two positively-charged exosites, named exosites I and II. Hirugen, the C-terminal residues 54 to 65 of hirudin, binds exosite I of thrombin. A series of desulfo-hirugen analogs were synthesized with either an additional fluorescein-(benzoyl) phenylalanine-(abbreviated Fl-bF-), or Fl-bF-mercaptopropionic acid or Fl-bF-lactic acid moiety conjugated to the N-terminal end of the peptide; i.e. a fluorescent label followed by a photocrosslinker and a cleavable linker in succession were conjugated to the N-terminus of hirugen analogs. Each hirugen analog was bound and photocrosslinked to thrombin, and the resulting covalent complex was purified from the reactants to homogeneity. These thrombin-hirugen adducts hydrolyzed small oligopeptide substrates (CBS 34-47 or S2366) with an efficiency similar to that of native thrombin, showing that the active site was unaffected by labeling. However, these adducts were significantly inhibited in assays of fibrinogen clotting and of thrombomodulin-mediated protein C activation, implying that exosite I was blocked, causing reduced accessibility for the substrate, fibrinogen, or for the cofactor, thrombomodulin. Exosite II mediates thrombin binding to heparin. The thrombin-hirugen adduct bound to and eluted from heparin-Sepharose like native thrombin, suggesting that exosite II was unaffected by the photocrosslinked hirugen analog. Treatment of the thrombin-(Fl-bF-mercaptopropionic acid) hirugen adduct that contains a thioester bond with the mild nucleophile hydroxylamine released the amino acids C-terminal to the thioester bond from the adduct, effectively releasing hirugen residues 54–65 from the complex while retaining the fluorescein reporter near exosite I (Fl-thrombin). This treatment did not affect the amidolytic activity of Fl-thrombin. Additionally, Fl-thrombin could be inhibited by unlabeled hirugen suggesting that exosite I was uncovered by release of the peptide by hydroxylamine treatment. In summary, our results show that exosites of clotting factors, e.g., thrombin, can be specifically targeted and labeled with fluorescent reporters. This novel technology may have broad applicability for studies of protein-protein interactions that regulate coagulation.

Author notes

Corresponding author