Actinonin, a Naturally Occurring Antibacterial Agent, Is a Potent Deformylase Inhibitor Dawn Z. Chen, Dinesh V. Patel, Corinne J. Hackbarth, Wen Wang, Geoffrey Dreyer, Dennis C. Young, Peter S. Margolis, Charlotte Wu, Zi-Jie Ni, Joaquim Trias, Richard J. White, and Zhengyu Yuan* Versicor, Inc., 34790 Ardentech Court, Fremont, California 94555 ReceiVed September 27, 1999; ReVised Manuscript ReceiVed December 2, 1999 ABSTRACT: Peptide deformylase (PDF) is essential in prokaryotes and absent in mammalian cells, thus making it an attractive target for the discovery of novel antibiotics. We have identified actinonin, a naturally occurring antibacterial agent, as a potent PDF inhibitor. The dissociation constant for this compound was 0.3 × 10 -9 M against Ni-PDF from Escherichia coli; the PDF from Staphylococcus aureus gave a similar value. Microbiological evaluation revealed that actinonin is a bacteriostatic agent with activity against Gram-positive and fastidious Gram-negative microorganisms. The PDF gene, def, was placed under control of P BAD in E. coli tolC, permitting regulation of PDF expression levels in the cell by varying the external arabinose concentration. The susceptibility of this strain to actinonin increases with decreased levels of PDF expression, indicating that actinonin inhibits bacterial growth by targeting this enzyme. Actinonin provides an excellent starting point from which to derive a more potent PDF inhibitor that has a broader spectrum of antibacterial activity. In eubacteria, protein synthesis is initiated with N- formylmethionine. In most cases, the newly synthesized polypeptide is converted to mature protein through the sequential removal of the N-formyl group and methionine by peptide deformylase and methionine amino peptidase, respectively (Scheme 1; 1, 2). In some cases, the formyl group is removed but the initiator methionine is retained. In Escherichia coli, PDF 1 is encoded by the def gene, homologues of which are present in all of the bacterial genomes sequenced to date. In those cases that have been studied, the PDF enzymes share many common properties (3). PDF activity was first described in 1968 (1), but efforts to purify it were hampered due to instability. Subsequent studies have shown that the instability is due to the oxidation of a ferrous ion at the enzyme’s active site (4). Although several classes of antibiotics target protein synthesis (e.g., tetracyclines, aminoglycosides, macrolides, and oxazolidinones), there is no report of an antibacterial that inhibits post-translational protein modification. PDF is a potentially attractive target for antibacterial drug design because (a) the gene associated with this activity is essential to bacterial growth in vitro (5-7), (b) it is present in all eubacteria examined and therefore can lead to broad spectrum activity, (c) the methionine formylation and deformylation cycle is not involved in eukaryotic cytoplasmic protein synthesis, which provides a sound basis for selective toxicity, and (d) the enzyme’s active center is very similar to several well-studied metallo hydrolases, including thermolysin and matrilysin, which thus provide prototypes for inhibitor design (8-12). PDF belongs to a new class of metallo hydrolases that utilize an Fe 2+ ion as the catalytic metal ion (4, 13, 14). The ferrous ion in PDF is very unstable and can be quickly and irreversibly oxidized into ferric ion, resulting in an inactive enzyme (15). Interestingly, the ferrous ion can be replaced with a nickel ion, resulting in little loss of enzyme activity and much greater stability. On the other hand, substitution with zinc results in a more than 10 5 -fold loss of activity (13). The three-dimensional structures of several forms of PDF have been reported in the literature (8-11, 16). The catalytic metal ion of PDF is tetrahedrally coordinated with two histidines from the conserved zinc hydrolase sequence, HEXXH, and a conserved cysteine from an EGCLS motif. The fourth position in the tetrahedron is occupied by a water molecule that presumably hydrolyzes the amide bond. Several transition state and/or substrate analogue-based inhibitors of PDF have been designed and recently reported (11, 17-19). Unfortunately, none of these inhibitors exhibit antibacterial activity, presumably due to a lack of potency against PDF and/or an inability to penetrate the bacterial cell. In this study, we report that actinonin, a naturally occurring antibacterial agent (20), is a very potent reversible PDF inhibitor and present evidence that actinonin inhibits bacterial growth through the inhibition of PDF activity. * To whom correspondence should be addressed. Phone: (510) 739- 3026. Fax: (510) 739-3003. E-mail: zyuan@versicor.com. 1 Abbreviations: PDF, peptide deformylase; MAP, methionine amino peptidase; fMAS, N-formylmethionine-alanine-serine; FDH, formate dehydrogenase; NAD + , nicotinamide adenine dinucleotide; NADH, nicotinamide adenine dinucleotide, reduced form; IPTG, isopropyl thiogalactopyranoside; LB medium, Luria-Bertani medium; TSB, tryptic soy broth; PDB, Protein Data Bank. Scheme 1: Reaction Catalyzed by PDF and MAP 1256 Biochemistry 2000, 39, 1256-1262 10.1021/bi992245y CCC: $19.00 © 2000 American Chemical Society Published on Web 01/19/2000