Three-dimensional models of ACE and NEP inhibitors and their use in the design of potent dual ACE/NEP inhibitors

A composite template for angiotensin converting enzyme (ACE, EC 3.4.15.1) inhibitors and a hypothetical model of the active site of neutral endopeptidase (NEP, EC 3.4.24.11) have been constructed and used to guide the design of dual ACE/NEP inhibitors. For the ACE template, a new computer program was used to flexibly superimpose potent, conformationally restricted ACE inhibitors. This program, which only considers the structures of the ligands, generated three possible templates. It was possible to evaluate the plausibility of these templates because new X-ray data is extending our knowledge of the binding of ligands to zinc metalloproteases. We have found that the available X-ray structures of inhibitors complexed to different zinc metalloproteases share certain conformational features. In each complex, the regions between the catalytic zinc and the P-1' side chain were found to have almost the same geometry. This geometry appears to be dictated by the mechanism of catalysis. Only one of the templates displays this geometry and is, therefore, proposed as a pharmacophore for ACE. To simulate NEP, we used the crystal structure of the active site of thermolysin (EC 3.4.24.4). These models of ACE and NEP predict that the conformation an inhibitor must adopt to bind to ACE differs from that required for binding to NEP. We have designed inhibitors in which conformationally restricted sections are linked by a flexible hinge, allowing the molecules to adapt to the conformation required by each enzyme. One of these inhibitors, a tricyclic alpha-thiol, 18 (CGS 28106), was found to inhibit both ACE and NEP with an IC50 of 40 and 48 nM, respectively The models predict that 18 binds to the S-1', S-2', and S-3' subsites of NEP and thermolysin and to the S-1, S-1', and S-2' subsites of ACE. The predicted mode of binding of 18 to thermolysin was experimentally verified by the determination of the X-ray crystal structure of the thermolysin/18 complex. This is the first reported three-dimensional structure of an alpha-thiol bound to a zinc metalloprotease. Except for a single NEP inhibitor, the models we propose for ACE and NEP are able to differentiate between active and inactive compounds reported in the present as well as other studies of dual ACE/NEP inhibition.