Search

Using the crystal structure of an inhibitor complexed with the serine protease thrombin (PDB code ) and the functional group definitions contained within the Catalyst software, a representation of the enzyme's active site was produced (structure-based pharmacophore model). A training set of 16 homologous non-peptide inhibitors whose conformations had been generated in continuum solvent (MacroModel) and clustered into conformational families (XCluster) was regressed against this pharmacophore so as to obtain a 3D-QSAR model. To test the robustness of the resulting QSAR model, the synthesis of a series of non-peptide thrombin inhibitors based on arylsuphonyl derivatives of an aminophenol ring linked to a pyridyl-based S1 binding group was undertaken. These compounds served as a test set (20-24). The crystal structure for the novel symmetrical disulfonyl compound 24, in complex with thrombin, has been solved. Its calculated binding mode is in general agreement with the crystallographically observed one, and the predicted K(i) value is in close accord with the experimental value.

The binding of the zymogenic form of urokinase-type plasminogen activator (pro-uPA) to its specific cellular receptor, uPAR, leads to a large potentiation of plasmin generation. This is dependent on the concurrent cellular binding of plasminogen, and is completely abrogated by the plasminogen lysine-binding site ligand, 6-aminohexanoic acid. Previous data have provided circumstantial evidence for the formation of specific complexes to mediate the kinetically favorable reciprocal interactions between the protease and zymogen components [Ellis, V., and Dano, K. (1993) J. Biol. Chem. 268, 4806-4813]. To further investigate the formation of these putative complexes, we have studied the effect of various lysine-binding site ligands on the binding and activation of plasminogen on U937 cells. Lysine-binding site ligands resembling internal lysine residues, such as Nalpha-acetyl-L-lysine methyl ester, were found to specifically inhibit uPAR-mediated cell-surface plasminogen activation at concentrations up to 40-fold lower than those inhibiting the cellular binding of 125I-labeled plasminogen (IC50s 300 microM vs 8.5 mM). By contrast, 6-aminohexanoic acid, resembling a C-terminal lysine residue, did not display this disparity (IC50s 25 vs 30 microM). These lysine analogues were also found to compete a non-active-site interaction between uPA and plasminogen, detected by surface plasmon resonance (Kd 50 nM), at concentrations correlating with their effect on cell-surface plasminogen activation, suggesting that this interaction is part of the kinetic mechanism. Consistent with this, synthetic peptides corresponding to the sequence uPA149-158 (GQKTLRPRFK) and uPA149-157 (GQKTLRPRF) specifically abolished the amplification of cell-surface plasminogen activation. These data demonstrate that a novel non-active-site interaction between uPA and plasminogen is necessary for the assembly and efficiency of cell-surface plasminogen activation complexes.

The affinity of the hirudin49-64 segment for exosite 1 of thrombin has been used previously to enhance the potency of simple competitive inhibitors [DiMaio, J., Gibbs, B., Munn, D., Lefebvre, J. , Ni, F., Konishi, Y. (1990) J. Biol. Chem. 265, 21698-21703., and Maraganore, J. M., Bourdon, P., Jablonski, J., Ramachandran, K. L., and Fenton, J. W., II (1990) Biochemistry 29, 7095-7087.]. Using a similar approach, we have enhanced the activity of two active site directed thrombin inhibitors by attaching this segment via a novel reverse oriented linker to each of two tripeptide boronate inhibitors. At P1, compound 1 contains an arginine-like, isothiouronium, side chain, while compound 2 contains an uncharged, bromopropyl residue. Inhibition of human alpha-thrombin by compound 1 shows slow, tight-binding competitive kinetics (final Ki of 2.2 pM, k1 of 3.51 x 10(7) M-1 s-1, and k-1 of 1.81 x 10(-)4 s-1). The addition of hirugen peptide (20 microM) competes for exosite 1 binding and restores the k1 and k-1 to that of the analogous tripeptide, 0.29 x 10(7) M-1 s-1 and 0.13 x 10(-)4 s-1, respectively. Compound 1 has enhanced specificity for thrombin over trypsin with KiTry/KiThr of approximately 900 compared to the analogous tripeptide, with KiTry/KiThr of approximately 4. Compound 2 acts as a competitive inhibitor (KiThr of 0.6 nM) and is highly selective with no effect on trypsin. Crystallographic analysis of complexes of human alpha-thrombin with compound 1 (1.8 A) and compound 2 (1.85 A) shows a covalent bond between the boron of the inhibitor and Ser195 (bond lengths B-O of 1.55 and 1.61 A, respectively). The isothiouronium group of compound 1 forms bidentate interactions with Asp189. The P2 and P3 residues of the inhibitors form interactions with the S2 and S3 sites of thrombin similar to other D-Phe-Pro based inhibitors [Bode, W., Turk, D., and Karshikov, A. (1992) Protein Sci. 1, 426-471.]. The linker exits the active site cleft of thrombin forming no interactions, while the binding of Hir49-64 segment to exosite 1 is similar to that previously described for hirudin [Rydel, T. J., Tulinsky, A., and Bode, W. (1991) J. Mol. Biol. 221, 583-601.]. Because of the similarity of binding at each of these sites to that of the analogous peptides added alone, this approach may be used to improve the inhibitory activity of all types of active site directed thrombin inhibitors and may also be applicable to the design of inhibitors of other proteases.

Z-D-Phe-Pro-boroMpg-OPin (9a)1,2 has been shown previously to be a highly specific inhibitor of thrombin in spite of lacking an arginine-like guanidino group at the P1 site. A range of compounds have been synthesized based upon this lead compound, varying the neutral side chain at the P1 site. Of the 20 examples based upon the structures at P2 and P3 of Z-D-X-Pro (X being Phe or beta,beta-diphenylalanine), all were found to be effective inhibitors of thrombin (Ki's between 10 and 100 nM). Furthermore all exhibited a high specificity toward thrombin having values for a Ki(trypsin)/Ki(thrombin) ratio of between 10- and 100-fold. High ratio values were found for a number of the compounds tested against a range of serine proteinases (plasmin, factor Xa, kallikrein, urokinase, protein Ca, chymotrypsin, elastase, and cathepsin G). As far as potency toward thrombin, compounds containing the methoxypropyl group at P1 were favored over those with a methoxy grouping on a shorter alkyl chain (8) or without the methoxy group (1-5). The compounds display potent anticoagulant activity with values for 18 in thrombin time of 0.63 microM and in activated partial thromboplastin time of 2.0 microM. 11B NMR has been used to confirm interaction of the boron atom with the active site. From the high specificity shown with all the compounds we propose that the compounds, constitute a new class of thrombin inhibitors.

In a search for inhibitors of 17 alpha-hydroxylase-C17,20-lyase and testosterone-5 alpha-reductase, target enzymes in the development of drugs to treat hormone-dependent prostatic cancer, we have identified certain compounds chemically derived by the hydrolysis of decafluoroazobenzene (4) as novel inhibitors for these two enzymes. Hydrolysis of 4 gave the known 4-hydroxynonafluoroazobenzene (1) and the novel 2-hydroxynonafluoroazobenzene (2). By AlI3 demethylation of 4,4'-dimethoxyoctafluoroazobenzene (5) or by hydrolysis of 4 under phase-transfer conditions 4,4'-dihydroxyoctafluoroazobenzene (3) was obtained. Compounds 1 and 2 were inhibitors of the hydroxylase (IC50 values, respectively, 30 and 63 microM) and of the lyase (IC50 values 33 and 16 microM) steps on the pathway of androgen biosynthesis. The 2-hydroxy compound 2 underwent spontaneous conversion into octafluorodibenz[b,f][1,4,5]oxadiazepine (6) which had IC50 values, respectively, of 50 and 15 microM for the hydroxylase and lyase steps and which contributed to the observed activity of 2. Effective inhibitors of the 5 alpha-reductase were 1 (Ki 10 microM) and 3 (Ki4 microM): the activities of 1 and 3 were markedly pH dependent, with respective IC50 values of 14 and 5 microM at pH 7.4 and of 2 and 0.8 microM at pH 6.6.