Determinants of thrombin specificity

Thrombin (IIa) (EC 3.4.4.13) is a serine protease most widely known for its role in the blood coagulation system, where it interacts with several macromolecules to maintain haemostasis. IIa acts as a procoagulant enzyme in the conversion of fibrinogen to fibrin in the final stages of blood coagulation, but also has regulatory activities that both amplify and attenuate the haemostatic response. The enzyme has remarkable specificity with respect to the limited number of peptide bonds that it cleaves within a variety of substrates. This narrow specificity can be attributed to four molecular mechanisms: the active site demonstrates a high selectivity for the P3-P3′ residues; two insertion loops at the surface of IIa partially occlude macromolecular substrates from the active site; two exosites bind and modulate the activity of substrates, inhibitors and modulators; and TM and Na+ can associate with IIa to overturn the enzyme’s specificity from procoagulant to anticoagulant. IIa has two exosites that mediate interactions between the enzyme and its substrates and cofactors. The binding of ligands to the exosites alters the functions of the protease, for example when the cofactor TM binds to both exosites I and II, it converts the enzyme from a procoagulant to an anticoagulant factor. It is unknown whether ligand binding to a IIa exosite will alter the substrate specificity of the enzyme and thus contribute to the changed substrate repertoire of the enzyme upon engagement with cofactors. The study initially examined whether binding of ligands to exosites I and II altered the activity of the enzyme against fluorogenic peptide substrates. The efficiency of cleavage of peptide substrates by IIa did change when TM or hirugen was present, indicating that exosite I occupancy changed the active site of the protease. The presence of heparin, did not change the activity of the enzyme, indicating that exosite II occupancy had little effect on active site function. Investigation of the effects of exosite I occupancy by hirugen on IIa specificity using phage display substrate libraries revealed that the ligand only changed the specificity of the enzyme to a small degree. Occupancy of both exosites by TM induced greater changes to the specificity of the enzyme, with the prime side showing broader changes in amino acid frequencies. Thus exosite I ligands do affect the activity and specificity of IIa, but not greatly enough to explain the altered substrate profile of the enzyme when complexed with TM. Kinetic and structural analyses have shown that the binding of a particular substrate residue at a protease subsite can have either a positive or negative influence on the binding of particular residues at other subsites. This phenomenon has been termed subsite cooperativity and has been observed in a wide range or proteases, often between non-adjacent subsites. We have used high-throughput phage display methods to elucidate the substrate specificity of IIa. The Magnum Opus software program was used to analyse this data, identifying five statistically significant amino acid motifs. These motifs indicated that cooperativity was occurring between the S3-S1; S2-S1′; S1, S1′ and S3′; and S1, S1′ and S5′ subsites of IIa. Interestingly, all five motifs occur in a range of known physiological IIa substrates such as PAR-1, carboxypeptidase B2, coagulation factors V, VIII, XIII and fibrinogen A and B. In order to investigate the contributions of substrate secondary structure and subsite cooperativity to cleavage susceptibility, the B1 IgG-binding domain of Streptococcal protein G (pG) was chosen as a model substrate. A series of mutants were synthesised that contained the P2-P1′ LRS IIa cleavage sequence within loop, alpha-helix and beta-strand regions. IIa was able to cleave the loop mutant and also, surprisingly, the alpha-helix pG mutant. A second series of mutants were synthesised with 1, 2 and 3 extra Gly residues added to the pG loop mutant, as well as one mutant with a Gly residue removed. Although there was a greater amount of pG cleavage as the loop length increased, this increase was not linear. Interestingly, the pG deletion mutant showed the greatest amount of cleavage, due to the sequence corresponding to one of the identified amino acid motifs. This suggests that the subsite cooperative effects were substantial enough to allow cleavage within a loop that should otherwise be less susceptible to cleavage. A third series of mutants were synthesised, each containing within a loop region one of the amino acid motifs identified by Magnum Opus. Cleavage assays performed on these pG mutants showed that the greatest amount of cleavage by IIa occurred with the motifs that exhibited cooperativity between the S2-S1′, S3-S1, and S1, S1′ and S3′ subsites. These results show that the predominant cooperative effects occur between subsites closest to the scissile peptide bond. Two X-ray crystal structures of a catalytically inactive form of IIa complexed with two different fluorescence-quenched peptide substrates were solved to determine the key interactions between the substrate and protease at the active site. There was a strong interaction between the P1 Arg residue of the substrate and the S1 specificity pocket. Interestingly, the Lys(Dnp) group of peptide substrate formed a ring-stacking interaction with the Trp60D residue of the IIa 60-loop that partially occludes the active site. These interactions were utilised in the design of two novel IIa peptide inhibitors. Subsequent competitive inhibitory assays showed that these peptides, as well as Lys(Dnp) alone were able to selectively inhibit IIa cleavage of both peptide and protein substrates. In addition, these peptide inhibitors did not inhibit trypsin or chymotrypsin, indicating that these inhibitors were able to distinguish between proteases of the same family. The selective nature of these peptide inhibitors is important to ensure that non-specific interactions, causing poor oral availability, stability and toxicology, do not occur.