Structure-activity relationships in chemically modified coumermycin.

ABSTRACT Coumermycin A1 was early recognized as a potentially useful antibiotic because of its very good spectrum of activity against gram-positive organisms (particularly against staphylococci), in addition to appreciable activity against a variety of gram-negative bacteria. However, in spite of its excellent in vitro activity, the insoluble character of coumermycin A1 made parenteral administration unsatisfactory. Poor oral absorption was also observed. Because of these shortcomings, coumermycin A1 was chemically modified either by an acyl interchange reaction or by direct condensation with 3amino-4-hydroxy-8-methyl-7-[3-O-(5-methyl-2-pyrrolylcarbonyl)noviosyloxy] coumarin (PNC-amine), which was produced by chemical degradation of coumermycin A1. A large number of semisynthetic derivatives were so prepared, and interesting structure-activity patterns will be discussed. A number of the new semisynthetic derivatives possessed markedly improved oral absorbability in mice. This improvement in some cases was >25 fold over the parent compound and more than compensated for the lower antibacterial potency possessed by most of the derivatives. Further, some of the more active compounds showed a significant improvement in the in vivo antistaphylococcal therapeutic effect, and marked improvement in oral efficacy in experimental D. pneumoniae and Streptococcus pyogenes mouse infections. Compounds demonstrating the broadest spectrum of antibacterial activity were members of the 3-substituted-4-hydroxybenzoic acid class. Among the aliphatic derivatives, 3-isobutyramido-4-hydroxy-8-methyl-7{3-O-(5-methyl-2-pyrrolylcarbonyl)noviosyloxy]coumarin (designated as BLC43) was by far the most interesting. Characterized by its overall trend toward improved in vitro activity, and its significant in vivo activity, this compound was selected for extensive pre-clinical evaluation. The results of these investigations will be discussed.