Thrombin, a serine protease, plays a central role in the initiation and propagation of thrombotic events. An extensive search for new thrombin inhibitors was performed, using an unconventional approach. Screening of small basic molecules for binding in the recognition pocket of thrombin led to the discovery of (aminoiminomethyl)piperidine (amidinopiperidine) as a weak, but intrinsically selective, thrombin inhibitor. Elaboration of this molecule provided compounds which inhibit thrombin with Ki's in the range of 20-50 nM and with selectivities of 1000-4000 against trypsin. These inhibitor compounds show a new and unexpected binding mode to thrombin. Modification of the central building block and then of one of the hydrophobic substituents led to the discovery of a new family of thrombin inhibitors which has reverted to the former binding mode to thrombin. This last class of compounds shows inhibitory activities in the picomolar range, low toxicity, and a short plasma half life which favors its use for an intravenous application. From this series of thrombin inhibitors, 19f(Ro 46-6240) was selected for clinical development as an antithrombotic agent for intravenous administration.

The prediction of a ligand binding constant requires generating three-dimensional structures of the complex concerned and reliably scoring these structures. Here, the scoring problem is investigated by examining benzamidine-like inhibitors of trypsin, a system for which errors in the structures are small. Precise and consistent binding free energies for the inhibitors are determined experimentally for this test system. To examine possible improvement of scoring methods, we test the suitability of continuum electrostatics to account for solvation effects and use an ideal-gas entropy correction to account for the changes in the degrees of freedom of the ligand. The small observed root-mean-square deviation of 0.55 kcal/mol of the calculated relative to the experimental values indicates that the essentials of the binding process have been captured. Even though all six ligands make the same salt bridge and H-bonds to the protein, the electrostatic contribution varies among the ligands by as much as 2 kcal/mol. Moreover, although the ligands are rigid and similar in size, the entropic terms also significantly affect the relative binding affinities (by up to 2.7 kcal/mol). The present approach to solvation and entropy may allow the ranking of the ligands to be considerably improved at a cost that makes the method applicable to the optimization of lead compounds or to the screening of small collections of ligands.

In a systematic fluorine scan of a rigid inhibitor to map the fluorophilicity/fluorophobicity of the active site in thrombin, one or more F substituents were introduced into the benzyl ring reaching into the D pocket. The 4-fluorobenzyl inhibitor showed a five to tenfold higher affinity than ligands with other fluorination patterns. X-ray crystal-structure analysis of the protein-ligand complex revealed favorable C-F...H-C(alpha)-C=O and C-F...C=O interactions of the 4-F substituent of the inhibitor with the backbone H-C(alpha)-C=O unit of Asn98. The importance of these interactions was further corroborated by the analysis of small-molecule X-ray crystal-structure searches in the Protein Data Base (PDB) and the Cambridge Structural Database (CSD). In the C--F...C=O interactions that are observed for both aromatic and aliphatic C-F units and a variety of carbonyl and carboxyl derivatives, the F atom approaches the C=O C atom preferentially along the pseudotrigonal axis of the carbonyl system. Similar orientational preferences are also seen in the dipolar interactions C--F.C[triple chemical bond]N, C-F.C-F, and C-F...NO(2), in which the F atoms interact at sub-van der Waals distances with the electrophilic centers.

Exposure of type III collagen coats on plastic cover slips In parallel-plate perfuslon chambers to flowing nonanticoagulated human blood resulted In deposition of platelets and fibrin. Blood was drawn directly from an antecubltal vein by an occluslve roller pump over the collagen coats In chambers having flow silts of different dimensions, so that wall shear rates of 100,650, and 2600 s~1 were obtained at 10 ml/mln. Coagulation was minimally activated during the passage of blood from the vein to the chamber as shown by fibrlnopeptlde A levels of 3.7 ng/ml after 5-mlnute perfuslons. The surface coverage with platelets increased from 18% at 100 s" 1 to 59% at 2600 s \ and the corresponding thrombus volumes Increased from 2 to 22 fim'/fim', respectively. This contrasted with the coverage with fibrin on collagen, which decreased from 28% at 100 s 1 to 9% at 2600 s~1. Fibrin deposits on the thrombi covered 6% of the surface irrespective of the shear rate, Indicating that some of the deposited platelets accelerated the deposition of fibrin. The type III collagen preparation did not activate factor XII and did not possess tissue factor activity, Indicating that the surface Itself was not procoagulant However, a correlation between deposited leukocytes and surface coverage with fibrin was observed ( T he component of the vessel wall that most potently triggers the formation and growth of thrombi is considered to be collagen.1 -2 So far, 13 species of collagen have been identified, and many have been localized in the vessel wall.3 ' 4 One of these is type III collagen, which is synthesized by endothelial cells 6 and is present in subendothelium. Type III collagen binds platelets and builds up thrombi when exposed to flowing citrated blood. 87 Furthermore, in fibrillar form It triggers platelet aggregation when added to stirring platelet-rich plasma.8 However, controversy exists as to whether collagens are able to initiate coagulation through activation of coagulation factor XII 9 1 0 1 1 and thereby amplify its thrombogenic potential. Activation of factor XII leads to generation of thrombin and fibrin, which respectively reinforce growth and stability of thrombi.12 - 18From F. Hoffmann-La Roche Ltd, Pharma Research/CVD, Basle, Switzerland.Helene Sage is at the Department of Biological Structure, University of Washington, Seattle, Washington.Kjell S. Sakariassen's present address is at the Biotechnology Center of Oslo, Gausdalsallein 21, Post-box 1125, Bllndem 0316, University of Oslo, Oslo 3, Norway., Address for correspondence: Prof. Hans R. Baumgartner, F. Hoffmann-La Roche Ltd, Pharma Research/CVD, Qrenzacherstrasse 124, CH-4002 Basle, Switzerland. Received May 3,1989; revision accepted November 17,1989. The interactions between nonanticoagulated blood and type III collagen have not yet been studied. Since type III collagen is one of the most abundant collagens in subendothelium, 17 and as such may become exposed to blood at the sites of vascular injury, we decided to expose surfaces coated with fibrillar human t...

SummaryPlatelet adhesion to natural and artificial surfaces and adhesion-induced aggregation were investigated in vitro using an annular perfusion chamber. The surfaces were exposed to anticoagulated blood under identical flow conditions (~ arterial shear rates). The initial attachment of platelets (contact) appeared less surface specific than spreading and release. Fibrillar collagen was the most powerful inducer of platelet degranulation whereas elastin, microfibrils and epon were virtually inactive. Fibrillar collagen caused release also in the absence of spreading. Surface coverage with platelets did not exceed 25 % unless spreading occurred. Perfusion with platelet-free plasma or platelet-poor blood did not remove adhering platelets. However, platelets were translocated from mural thrombi to the surface by such perfusion. In addition, platelets which detached from mural thrombi adhered more readily to elastin or microfibrils than platelets from the circulating blood. The initial attachment of platelets to subendothelium was inhibited in von Willebrand’s disease, the Bernard-Soulier syndrome and at high concentrations of dipyridamole; spreading was inhibited in storage pool disease of rats, at low temperature (20° C), with EDTA (3 mM) and Prostaglandin E1 (1 μM); and adhesion-induced aggregation was inhibited in thrombasthenia, storage pool disease and after ingestion of sulfinpyrazone or Aspirin.It is concluded that the initial attachment (contact) of platelets, spreading and surface-induced release of platelet constituents are at least partially independent phenomena, the latter two being highly surface specific. At flow conditions which cause the disappearance of platelet thrombi, platelet adhesion appears as an irreversible process.