The disclosure of the novel antibiotic platensimycin (1, ) by the Merck group of Wang, Soisson and Singh et al. has attracted considerable attention from within the scientific community as manifested by a series of publications describing its chemical synthesis. The intense interest in this natural product can be attributed to its unique molecular architecture and extraordinary antibacterial activity against Gram positive bacteria, including multiple drug-resistant strains such as methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus faecalis (VREF). Platensimycin exhibits its impressive antibacterial activity through a novel mode of action that involves selective interference with the catalytic machinery responsible for the biosynthesis of bacterial fatty acids by selective inhibition of the elongation condensing enzyme FabF. X-ray crystallographic analysis of platensimycin in complex with a mutant version of the FabF enzyme [ecFABF(C163Q)] designed to mimic the acyl enzyme intermediate reveals two distinct binding domains (): (I) the highly polar benzoic acid unit which docks to the malonate pocket of the condensing enzyme, and (II) the lipophilic cage that is positioned in the mouth of the enzyme. Within binding domain I, the carboxylate of the benzoic acid interacts with the active site histidine residues (H303 and H340), an interaction that is likely to have some ionic character. Moreover, the aromatic ring of the benzoic acid is involved in an edge-to-face π-interaction with F400, which is oriented in the position it is believed to occupy in the acyl-enzyme intermediate. Within binding domain II, the hydrophilic ketolide motif, which is partially exposed to solvent, has significant van der Waals interactions with the protein surface, covering some 122 Å of the protein surface. The cage also makes hydrogen bonds involving its ether and carbonyl oxygen atoms with T270 and A309, respectively. The amide linker, connecting both binding domains, is involved in two hydrogen bonds to the T307 side chain and the amide backbone at T270.
As shown in , the molecular structure of platensimycin (1) comprises two structurally distinct domains, a substituted benzoic acid moiety and a lipophilic cage, that are connected through an amide linker. In order to determine the importance of the polar aromatic subunit and its substitution pattern for biological activity, we aimed to synthesize members of analog series I (). Within this series, the tetracyclic cage domain remains intact whilst increasingly complex aromatic surrogates (compounds 6, 18 to 15) probe the specific interactions of this domain with the targeted enzyme through single, double and triple functional group deletions.
Synthesis and biological evaluation of platensimycin analogs
The first member of analog series I (), platensic acid (6), is a key intermediate in our asymmetric synthesis of platensimycin and was obtained enantiomerically enriched from an (S,S)-pseudoephedrine amide, as described in our previous communication. Access to aniline derivatives 15–18 relied on late-stage divergence from this carboxylic acid (). Thus, HATU-mediated amide coupling of 6 with amines 19, 22 and 25 (see ), as well as aniline itself, gave the corresponding amides 20, 23, 26 and 18 in 82, 90, 83 and 86% yield, respectively. As required, MOM-protecting group removal (aq. HCl, THF, 45 °C) followed by deprotection of the ester functionality using fluoride (TASF, DMF, 40 °C) or basic conditions (aq. LiOH, THF, 45 °C), revealed platensimycin derivatives 15–18 in excellent yields, as indicated in .
He graduated from Kyoto University in 1970
A potential antibiotic and a structural analogue of platensimycinwas synthesized. The focal step of the synthesis is a highly diastereo-and enantioselective Diels-Alder cycloaddition (A + B → D).