An efficient platform for antibiotic discovery

About This Grant

The goal of this project is to develop a novel discovery platform by accessing uncultured bacteria, and efficiently mining them for antibiotics. The vast majority of bacteria on the planet do not grow in the lab, which presents a critical barrier for basic and applied microbiology. We developed methods to culture environmental bacteria in situ in diffusion chambers, which increases recovery of cells up to 40%. However, the majority of bacterial Phyla remain uncultured. These include soil Acidobacteria, Verrucomicrobia, and Gemmatimonadetes that contain biosynthetic gene clusters (BGCs) only distantly related to known ones. There is a great opportunity to tap into this enormous diversity of hidden chemistry. The challenge is formidable, without access to uncultured bacteria, not clear how to study them. The majority of soil bacteria live in microaggregates (ma), ~100 µ particles. We reasoned that ma contain self-assembled communities of symbionts, with helper bacteria enabling the growth of uncultured. In preliminary studies, sequencing individual ma produces a network, indicating likely helper species for uncultured bacteria. These include Rhizobacteria, and we found that their extracts stimulate the growth of Acidobacteria. We also developed a general method for isolating uncultured species by staining soil bacteria with a membrane die and incubating in growth medium; dividing cells dilute the dye, while uncultured do not. Uncultured bacteria can then be isolated by FACS and paired with cultivable bacteria to find helpers. Focusing on specific groups, we developed a capturing method for Acidobacteria with fluorescently labeled phage receptor proteins. Extracts from Rhizobacteria enable cultivation of captured soil uncultured bacteria. Screening a set of Acidobacteria and Verrucomicrobia confirmed that they produce antimicrobials. Identifying various helper species will lead to the isolation of a set of growth factors for growing attractive uncultured phyla. Uncultured bacteria will then be screened for novel antibiotics using a droplet microfluidics approach we developed, which pairs bacteria with reporters and operates at a rate of 106 droplets/hr. Hits will be purified and characterized in vitro and in vivo. Promising leads will validate the discovery platform. Preliminary studies establish feasibility for the discovery platform, and confirm that previously inaccessible phyla are a promising source of antibiotics. This project will result in an efficient antibiotic discovery platform and novel antibiotics from uncultured microorganisms.