Catalytic antibodies and enzyme inhibitors

Catalytic antibodies or abzymes as we have termed them have generated an avalanche of "forecasts and predictions" concerning their future applications. This, because in principle their only limitations are the designer's creativity and one's ability to access the entire immune systems repertoire. Keeping these so-called predictions in mind, we have sought out a truly formidable test for our de novo catalyst program. That being the creation of catalytic antibodies that have no enzymatic or synthetic catalyst equivalent. An antibody is reported which catalyzes a highly disfavored ring closure reaction with exquisite stereospecificity. The design and synthesis of peptidomimetic enzyme inhibitors continues to be an active area of research. Such compounds have proven useful in elucidating mechanisms of catalysis and as therapeutic agents. The discovery that the human immunodeficiency virus encodes an aspartic protease (HIV PR) vital for its propagation has brought this protein under intense scrutiny. In this regard, the development of compounds which inhibit the HIV PR has been particularly rapid. Herein we report on a new amide bond replacement, y[P02-CH2N+] which we term an "exploding transition-state analogue". Incorporation of this unit in peptide sequences and its potent, slow-binding inhibition of the HIV protease is discussed. A considerable amount of effort has been focused on the development of efficient catalysts for the synthesis or modification of complex molecules. Two methodologies have generally been applied in these endeavors; protein and synthetic ligand binding catalysts. To attack such problems, we have recruited antigen induced catalysts from the immune system.l These catalytic antibodies or abzymes as we have termed them have generated an avalanche of "forecasts and predictions" concerning their future applications. This, because in principle their only limitations are the designer's creativity and one's ability to access the entire immune systems repertoire. Keeping these so-called predictions in mind, we have sought out a truly formidable test for our de novo catalyst program. That being the creation of catalytic antibodies that have no enzymatic or synthetic catalyst equivalent2 Ring-forming reactions are important and common processes in organic chemistry. Almost twenty years ago J.E. Baldwin supplied a specific set of "favored" and "disfavored" rules for certain closings of 340 7membered rings.3 The physical basis of these guidelines was formulated based on stereochemical requirements of the transition states for the various ring closure processes. Wherein favored pathways are those in which the length and nature of the linking chain enable the terminal atoms to achieve the proper geometrys for reaction, and disfavored cases require severe distortion of bond angles. Many cases in the literature are in substantial accord with Baldwin's rules, both experimentally and theoreti~ally.39~ The skeletal structures of a number of bioactive marine natural products are found to contain 0heterocyclic rings.5 Of particular importance are the ubiquitous tetrahydropyrans, towards the synthesis of which much work has been done.6 One attractive strategy reduces to a regioselective 6-endo-tet-ring opening of an epoxide by an internal nucleophilic oxygen (scheme I). However, conventional considerations (Baldwin's rules) suggest the 5-exo-tet mode of cyclization as the preferred pathway, (scheme I). Indeed, experimental evidence confirms this wherein the tetrahydrofuran system is the exclusively observed product. A specifically tailored catalytic antibody could provide a means for obtaining the desired, yet disfavored 6-endo-tet adduct. Hence, a catalytic regio-controlled detour, and thus a possible general methodology for the re-routing of disfavored ring closure reactions. The concept on which we relied to achieve this selectivity is shown in scheme I, which depicts a scenario of an acid-catalyzed cyclization of hydroxyl epoxide 1. Based on our "bait and switch" principle7, the problem reduces to the design of a hapten which could induce a regiospecific negatively charged participatory amino acid in the binding pocket of an antibody. According to this concept, a negative charge when strategically placed adjacent to the epoxide unit should act as a selective stabilizing element