ABSTRACT
The lead compound 5-bromoindolyl-3-acetohydroxamic acid (10) was recently identified as a potent inhibitor of bacterial peptide deformylases (PDFs). The synthesis and associated activities of new variants were investigated at position 5 to optimize the fit at the S1' subsite and at position 1 to improve both potency and antibacterial activity. A morphomimetic series, termed "reverse-indole" was synthesized. The indole derivatives remain selective in vitro inhibitors of PDF2 over PDF1. Bromide is the best group at position 5 and cannot be replaced by bulkier substituents. In this series, an N-benzyl group at position 1 in 19 e improves the potency relative to 10. In the case of PDF1, and unlike PDF2, potency is increased as the alkyl chain becomes longer and more ramified. These data support the results of NMR footprinting experiments that were performed with (15)N-labeled Ni-PDF and the corresponding 3-acetic acid derivatives. Most of the compounds have antibacterial activities toward B. subtilis, but are inefficient toward E. coli owing to active removal by the major efflux pumps. Among the reverse-indole derivatives, 23 c, which is the exact mirror image of 19 e, shows strong potency in vitro against PDF2, but little against PDF1, although this compound displays significant antibacterial activity toward an efflux-minus mutant of E. coli. All the compounds were assessed with major pathogenic bacteria, but most of them are inefficient antibacterial agents. The reverse-indole compounds 23 a and 23 c have potency against S. pneumoniae that is similar to that of actinonin.
