Highly parallelized microfluidic droplet cultivation and prioritization on antibiotic producers from complex natural microbial communities

To investigate the overwhelming part of the bacterial diversity still evading standard cultivation for its potential use in antibiotic synthesis, we have compiled a microscale-cultivation and screening system. We devised a strategy based on droplet-microfluidics taking advantage of the inherent miniaturization and high throughput. Single cells of natural samples were confined in 9 x 106 aqueous droplets and subjected to long-term incubation under controlled conditions. Subsequent a high-throughput screening for antimicrobial natural products was implemented, employing a whole cell reporting system using the viability of reporter strains as a probe for antimicrobial activity. Due to the described microscale cultivation a novel subset of bacterial strains was made available for the following screening for antimicrobials. We demonstrate the merits of the in-droplet cultivation by comparing the cultivation outcome in microfluidic droplets and on conventional agar plates for a bacterial community derived from soil by 16S rRNA gene amplicon sequencing. In-droplet cultivation resulted in a significantly higher bacterial diversity without the common overrepresentation of Firmicutes. Natural strains able to inhibit either a Gram-positive or a Gram-negative reporter strain were isolated from the microscale system and further cultivated. Thereby a variety of rare isolates was obtained. The natural products with antimicrobial activity were elucidated for the most promising candidate. Our method combines a new cultivation approach with a high-throughput search for antibiotic producers to increase the chances of finding new lead substances.