Discovery and Synthesis of Macrobicyclic Lincosamide Antibiotics

Abstract

This dissertation presents the discovery and synthesis of macrobicyclic lincosamide antibiotics. In the first chapter, the design conception, chemical synthesis, and microbiological evaluations of the bridged macrobicyclic antibiotic cresomycin (CRM) are presented. CRM overcomes evolutionarily diverse forms of antimicrobial resistance that render modern antibiotics ineffective, and exhibits in vitro and in vivo efficacy against both Gram-positive and Gram-negative bacteria, including multidrug-resistant strains of Staphylococcus aureus, Escherichia coli, and Pseudomonas aeruginosa. We show that CRM is highly preorganized for ribosomal binding by determining its density functional theory–calculated, solution-state, solid-state, and (wild-type) ribosome-bound structures, all of which align identically within the macrobicyclic subunits. Lastly, we report two additional X-ray crystal structures of CRM in complex with bacterial ribosomes separately modified by the ribosomal RNA methyltransferases, chloramphenicol-florfenicol resistance (Cfr) and erythromycin-resistance ribosomal RNA methyltransferases (Erm), revealing concessive adjustments by the target and antibiotic that permit CRM to maintain binding where other antibiotics fail. Extant lincosamide antibiotics such as lincomycin and clindamycin undergo rapid hepatic oxidation upon first-pass metabolism at their anomeric thioethers to the comparatively inactive sulfoxides. Efforts to replace the sulfur atom with a more metabolically stable functionality have invariably resulted in antibiotic candidates with inferior antibacterial efficacy. In the second chapter, we present the chemical syntheses, MIC analyses, and ribosome-bound X-ray co-crystal structures of sulfur atomic replacement CRM analogs O-CRM (S → O), C-CRM (S → CH2), and Se-CRM (S → Se), making side-by-side comparisons with CRM. The high-resolution co-crystal structures alongside antibacterial susceptibility measurements reveal why sulfur is uniquely (and probably irreplaceably) effective in targeting the bacterial ribosome. In view of the promising antibacterial activity and pharmacokinetic profile of CRM, we hope to advance the antibiotic candidate in pre-clinical, investigational new drug (IND)–enabling studies. These experiments will require multigram-quantities of CRM, which can be retrosynthetically disconnected into two halves of equal complexity — a macrobicyclic thiolincosamine northern fragment and a fused bicyclic oxepanoproline southern fragment. In the third chapter, the multigram-scale synthesis of the northern macrobicyclic thiolincosamine fragment of CRM is presented. A key transformation in the synthetic route is the highly diastereoselective addition of a putative allenylzinc nucleophile to a versatile Ellman sulfinimine intermediate using a zinc-promoted Barbier-type propargylation protocol. The transformation proceeds with dynamic kinetic resolution of the enantiomeric pair of allenylzinc nucleophiles and use just 1.2 equivalents of the propargyl bromide precursor. In the fourth chapter, the multigram-scale synthesis of the southern oxepanoproline fragment of CRM is presented. A key transformation in the synthetic route involves an efficient one-pot procedure for the TiCl4-mediated conjugate addition of an allylsilane to an Evans N-acryloyloxazolidinone, followed by aldol addition of the resultant titanium enolate to (R)-Garner’s aldehyde, setting all the stereocenters within the target molecule in a single operation. Together, the synthetic strategies presented facilitated the preparation of >5 g-amounts of CRM, as well as other promising antibiotic candidates, for extensive in vitro and in vivo profiling. Over the course of our studies on the chemical synthesis of lincosamide antibiotics, we explored the use of azomethine ylides as suitable reactive intermediates for the preparation of highly functionalized proline derivatives. The fifth chapter describes the unexpected regioselectivity of a [3+2] dipolar cycloaddition reaction of a stabilized azomethine ylide with an electron-deficient dipolarophile, which was found to be counter to literature precedent. Lastly, the sixth chapter describes the tritium-labeling and biochemical profiling of iboxamycin, a semisynthetic lincosamide antibiotic reported by our laboratory in 2021. Hydrogen-tritium exchange is widely employed for radioisotopic labeling of molecules of biological interest but typically involves the metal-promoted exchange of sp2-hybridized carbon-hydrogen bonds, a strategy that is not directly applicable to iboxamycin, which possesses no such bonds. We show that ruthenium-induced epimerization at the α-hydrogen of the amide carbonyl (position C2′) of 2′-epi-iboxamycin in HTO of low specific activity (180 mCi/mmol) affords tritium-labeled iboxamycin (3.55 µCi) with a specific activity of 53 mCi/mmol. Iboxamycin displayed an apparent inhibition constant (Ki, app) of 41 ± 30 nM towards Escherichia coli ribosomes, binding approximately 70-fold more tightly than the antibiotic clindamycin (Ki, app = 2.7 ± 1.1 µM).