Discovery of a Broad-spectrum, Fluorinated Lincosamide Antibiotic Through Chemical Synthesis

Abstract

This dissertation presents an investigation of next-generation lincosamide antibiotics which led to the discovery of a fluorinated, macrobicyclic antibiotic candidate (BT-33) with best-in-class activity metrics in antibacterial potency, metabolic stability, and in vivo efficacy against multidrug-resistant bacterial pathogens. I report the discovery of BT-33 using a modular synthetic route designed to access substituted macrobicyclic thiolincosamines. Adaptations of this discovery synthesis delivered >60 macrobicyclic lincosamide analogs, more than half of which possessed antibacterial activity. The structure-activity relationships elucidated in this investigation showed that each structural feature within the macrobicyclic residue of BT-33 is essential to its potent antibacterial activity and enhanced metabolic stability. These conclusions were supported by conformational analysis and X-ray crystallography which revealed the 11-membered macrobicyclic residue of BT-33 to be rigidified in a conformation optimal for binding to the bacterial ribosome. Further investigation of this chemical series provided analogs which were unaffected by mechanisms of lincosamide resistance (namely erm and cfr methyltransferases), likely by engaging an additional binding pocket in the nascent peptide exit tunnel. I present a distinct, scalable synthesis of BT-33 that enabled its preparation on >5-g scale, an amount which was required to support an early preclinical profiling campaign. This new synthesis relied on a highly diastereoselective coupling reaction between a sulfinimine intermediate and a putative allenylzinc reagent formed in situ. Macrocyclization of the coupled product by C–S bond formation (rather than by C=C bond formation, as was used in the discovery route) followed by semireduction of the resultant cyclic alkyne intermediate set the key stereochemical features within BT-33 in an operationally simple and high yielding reaction sequence. This route was adapted to synthesize another structurally distinct, fluorinated antibiotic (11-fluoromethyl-cresomycin, a lead candidate in its own right) and synthetic analogs of BT-33 which were not accessible by the discovery route. I detail the results of an early preclinical profiling campaign to evaluate oxepanoprolinamide (OPP) antibiotics (including BT-33 and 11-fluoromethyl-cresomycin) for use in the treatment of community-acquired bacterial pneumonia. In vitro assays determined that the lead OPP molecules exhibit a highly favorable microbiological profile, acceptable physicochemical properties, and no apparent liabilities for off-target toxicity. In vivo assessments showed that BT-33 was efficacious in multiple models of infection using multidrug-resistant bacterial strains. Considering this highly promising profile, we are optimistic that one or more molecules from this class will be advanced further in preclinical development and may one day reach the clinic. Finally, I report an investigation of new lincosamide derivatives inspired by the rigid oxepanoproline residue of the prototypical OPP antibiotic, iboxamycin. Rational, structure-based design guided the synthesis and assessment of >40 lincosamide analogs with substituted, heterobicyclic amino acid residues. Ultimately, this investigation confirmed the strict steric and physicochemical requirements for a lincosamide derivative to effectively overcome antibiotic resistance in bacteria and supported the conclusion that the oxepanoprolinamide is a privileged scaffold.