A THEORETICAL-STUDY OF THE ACTIVE-SITE COMPLEXES FORMED BETWEEN ENDOTHIAPEPSIN AND 3 POTENT INHIBITORS - PEPSTATIN-A, AND PEPTIDE ANALOGS CONTAINING DIFLUOROSTATONE AND PHOSPHOSTATINE - IMPLICATIONS FOR INHIBITOR DESIGN

The aspartic proteinases are a family of proteolytic enzymes known to be involved in several medical disorders (such as hypertension, AIDS a nd cancer) and are therefore an important therapeutic target for the r ational design of inhibitors. The crystal structures of several enzyme /inhibitor complexes have already been reported. Many of these inhibit ors contain a hydroxyl group which forms hydrogen bonds to the carboxy l groups of the active site aspartates, D32 and D215, (pepsin numberin g). In many cases it is not possible to identify unambiguously the hyd rogen bonds formed between the inhibitor and the catalytic aspartates from the crystal structures, and therefore one cannot determine which of these inhibitors are good transition state analogues. We have perfo rmed ab initio Hartree-Fock calculations on a small portion of the act ive site of endothiapepsin complexed with three potent inhibitors: pep statin A, and two peptide analogues which contain difluorostatone and phosphostatine to identify the hydrogen bonds formed between the inhib itor and the active site aspartates. Our calculations suggest that inh ibitors which contain a gem-diol group (e.g. difluorostatone-containin g inhibitors) are more likely to be good transition state analogues.