Approaches to rejuvenate antibiotic discovery: new sources, new tools

Abstract

Owing to the widespread use, overuse, and abuse of antibiotics, pathogenic bacteria have evolved extensive antibiotic resistance in recent decades. Over that same period, there has been a dramatic decline in the discovery of novel antibiotics. The failure of target-based screening using synthetic chemical libraries to discover novel classes of antibacterial drugs has renewed interest in returning to the best validated source of antibiotics i.e., natural products. However, it is self-evident that screening the same producer organisms with the same approaches used in the past will result in the rediscovery of already-known antibiotics. Therefore, new sources of natural products should be screened to access novel scaffolds and innovative discovery tools should be developed to more effectively screen established and novel sources. The work in this thesis has contributed to both of these aims. In the first instance, this study has validated clinically-isolated fungi as a novel source of antibiotics. Of 3,560 isolates screened, 12 were shown to produce apparently novel antibiotics with selective antibacterial activity that encompasses the ESKAPE pathogens. This work also describes the generation and validation of pathway-specific whole-cell biosensors for the detection of protein biosynthesis inhibitors (PBIs) that show unprecedented specificity and sensitivity. One these biosensors was engineered to contain multiple common PBI resistance genes, thereby achieving in-built dereplication for the deselection of known PBIs. Screening of >7,400 microbial extracts from the national cancer institute (NCI) natural products library for novel PBIs identified 99 putative hits; re-screening using the dereplicator strain recovered only two hits, underscoring both the utility of the whole-cell biosensor for identification of PBIs in a screening context, and the power of the dereplication function for removing known compounds from further consideration.