S75 Microscale assay development for advanced microbial phenotyping in higher throughput
Tuesday, July 22, 2014: 1:00 PM
Regency Ballroom A, Second Floor (St. Louis Hyatt Regency at the Arch)
Andreas Radek, Simon Unthan, Marco Oldiges and Stephan Noack, Institute of Bio- and Geosciences, IBG-1: Biotechnology, Systems Biotechnology, Research Center Juelich, Juelich, Germany
For industrial biotechnology it is highly important to select the most promising production hosts and medium compositions in early stages of process development. However, the opposing demands of detailed process insight during high throughput are generally not fulfilled by classical cultivation systems. Today, this gap is therefore often bridged by use of high quality MTP-cultivation devices, enabling the non-invasive monitoring of online parameters while providing sufficient oxygen transfer. However, these devices still depend on manual work to inoculate, sample or induce cultivations, which limits the overall speed and throughput of microbial phenotyping.

To address this issue we constructed a mini pilot plant (MPP), by embedding a BioLector in a robotic environment to automate complete workflows for microbial phenotyping. The connected liquid handling platform enables triggered addition of media components and supplements to individual cultivation experiments. To gain a deeper process understanding, cultivation samples are processed automatically to provide supernatants for subsequent quantitative analysis with various fully automated assays in MTP-scale.

With this MPP at hand we evaluated 22 novel L-Lysine producing Corynebacterium glutamicum strains in different media for growth as well as L-Lysine yield and productivity within two weeks. Concentrations of glucose and L-Lysine in cultivation samples were automatically measured on the MPP by an enzymatic or biochemical assay, respectively. As a result novel producer strains were identified showing significantly increased L-Lysine titers compared to the reference producer DM1933. This exemplary finding demonstrates the potential of our MPP approach and shows how future microbial phenotyping will benefit from robotic automation.