Affiliation: Eshelman School of Pharmacy, Division of Chemical Biology and Medicinal Chemistry
Over 2 million people in the United States, and millions more worldwide, are affected by antibiotic resistant bacterial infections. The discovery of new antibiotics has been stalled by the lack of incentives from pharmaceutical companies and academia from both an economic and regulatory standpoint. The advent of next-generation DNA sequencing and gene annotation through homology models has provided a vast pool of biosynthetic clusters from which new and old bioactive molecules can be predicted. This research focuses on the characterization of the enzymes TclM, TmlU, and HolE and the visible light activated Oglycosylation of thioglycosides. We have characterized the enzyme TclM in the thiocillin biosynthetic pathway from Bacillus cereus ATCC 14579 as a pyridine synthase by ligating a leader peptide recognition sequence to a semi-synthetic core substrate through native chemical ligation and subsequent conversion of cysteine side chains to dehydroalanines by elimination of a trialkylated tetrahydrothiophene. We have additionally characterized the enzymes TmlU and HolE from the thiomarinol biosynthetic pathway from Pseduoalteromonas spp. SANK7339 and found them to be responsible for ligating the components pseudomonic acid C-coenzyme A thioester and holothin, respectively, to create the hybrid antibiotic. Structure-activity-relationship studies were done on the dithiolopyrrolone holothin and found group promiscuity at the endocyclic and exocyclic amides but bioactivity was lost when the disulfide was oxidized to the thiosulfinate moiety. Further reactivity studies showed the ability of dioxoholomycin to oxidize small molecule thiols and redox-sensitive cysteines in the proteins YwlE and AtAPSK. We have also designed a visible light mediated O-glycosylation of thioglycosides by utilizing blue LEDs as a light source in the presence of an Ir (III) catalyst. It was found that 1o, 2o, and 3o alcohols react and serve as acceptors. Mechanistic studies showed the need for a bromine-trihalogenated carbon bond for the initiation step, but that subsequent propagation relies on the p-methoxythiophenol disulfide, formed through dimerization from the activated starting material.