ABSTRACT
BACKGROUND: The aspartic proteinase renin plays an important physiological role in the regulation of blood pressure. It catalyses the first step in the conversion of angiotensinogen to the hormone angiotensin II. In the past, potent peptide inhibitors of renin have been developed, but none of these compounds has made it to the end of clinical trials. Our primary aim was to develop novel nonpeptide inhibitors. Based on the available structural information concerning renin-substrate interactions, we synthesized inhibitors in which the peptide portion was replaced by lipophilic moieties that interact with the large hydrophobic S1/S3-binding pocket in renin. RESULTS: Crystal structure analysis of renin-inhibitor complexes combined with computational methods were employed in the medicinal-chemistry optimisation process. Structure analysis revealed that the newly designed inhibitors bind as predicted to the S1/S3 pocket. In addition, however, these compounds interact with a hitherto unrecognised large, distinct, sub-pocket of the enzyme that extends from the S3-binding site towards the hydrophobic core of the enzyme. Binding to this S3(sp) sub-pocket was essential for high binding affinity. This unprecedented binding mode guided the drug-design process in which the mostly hydrophobic interactions within subsite S3(sp) were optimised. CONCLUSIONS: Our design approach led to compounds with high in vitro affinity and specificity for renin, favourable bioavailability and excellent oral efficacy in lowering blood pressure in primates. These renin inhibitors are therefore potential therapeutic agents for the treatment of hypertension and related cardiovascular diseases.
Subject(s)
Antihypertensive Agents/chemical synthesis , Antihypertensive Agents/pharmacology , Drug Design , Protease Inhibitors/chemical synthesis , Protease Inhibitors/pharmacology , Renin/antagonists & inhibitors , Angiotensin-Converting Enzyme Inhibitors/adverse effects , Angiotensinogen/analogs & derivatives , Animals , Antihypertensive Agents/chemistry , Antihypertensive Agents/metabolism , Binding Sites/physiology , Callithrix , Crystallography, X-Ray , Humans , Hydrogen Bonding/drug effects , Hypertension/drug therapy , Hypertension/physiopathology , Mice , Models, Molecular , Molecular Structure , Protease Inhibitors/chemistry , Protease Inhibitors/metabolism , Protein Binding/physiology , Protein Conformation , Renin/metabolism , Structure-Activity Relationship , Substrate SpecificityABSTRACT
We have developed a computer program, called NEWLEAD, for the automatic generation of candidate structures conforming to the requirements of a given pharmacophore. The treatment consists in connecting the pharmacophoric pieces with spacers assembled from small chemical entities (atoms, chains, or ring moieties). We have tested the program on several sets of input fragments, each comprising selected functional groups obtained from the bioactive conformations of reference molecules. In addition to the expected solutions, the program can generate new structures that are chemically unrelated to the reference molecules. This provides an unbiased starting point for the design of new generations of lead structures. The concept used in this approach is presented and discussed. The present possibilities of the program are illustrated by some examples. The treatment is very fast, because only a few bonds are created between building blocks already having ideal geometries. The ability to generate rapidly a variety of molecules conforming to a three-dimensional pharmacophoric model makes NEWLEAD a useful tool with wide applicability in rational drug design, including the areas of molecular mimicry and peptidomimetism.
Subject(s)
Drug Design , Software , Algorithms , Computer Simulation , HIV Protease Inhibitors/chemistry , Indomethacin/chemistry , Methotrexate/chemistry , Models, Molecular , Molecular Structure , Sugar Alcohols/chemistry , Valine/analogs & derivatives , Valine/chemistrySubject(s)
Models, Molecular , Software , Computer Graphics , Drug Design , Receptors, Drug , X-Ray DiffractionABSTRACT
Recent observations reveal deficiencies in the accepted theory rationalizing the biological activities of the beta-lactam antibiotics, since a study of strained carbapenem beta-lactams has shown that the observed antibacterial activities do not correlate either with the pyramidal character of the beta-lactam nitrogen atom or with the ease of base hydrolysis of the lactam amide bond. The contradiction can be reconciled by an analysis of the three-dimensional (3-D) features of a set of the representative active and inactive beta-lactam structures, which shows that highly specific 3-D recognition sites may exist in the enzymes in their recognition of the antibiotics. The identification of the geometrical requirements for antibacterial activity also reveals how it could be possible to restore antibiotic activities to inactive structures, up to now considered as devoid of any therapeutic interest.
