Information mining in parallelized and miniaturized cultivation systems

The continued development of microbial production hosts is necessary to satisfy the urgent need of novel drugs, food, and especially production routes for bulk and fine chemicals from sustainable resources such as second-generation feedstocks. Therefore, strain screening and characterization throughput was increased by miniaturization and highly parallelization of cultivation systems with integrated online sensors and automated at-line analytics. These systems could be already used for screening of best clones, media optimization, and characterization of influential factors, but are still not able to fully replace classical lab scale reactors. The lack of time resolved samples hinders to unravel the growth and production dynamics at this early stage and the accessibility to this information is crucial for further process development and scale up. Within this project a methodology is developed to extract scalable, physiological information from miniaturized cultivation systems.

The focus of the analysis was the growth behaviour and the dynamics of Corynebacterium glutamicum, a bacterium used for the million-ton-scale production of amino acids and capable to efficiently use sugars like glucose and arabinose present in lignocellulosic waste-streams [1]. The developed methodology allows to easily analyze growth properties of C. glutamicum on different sugar mixtures. Rather than qualitatively compare the data, the analysis tool results in quantitative and scalable growth parameters such as growth and uptakes rates, conversion yields as well as component affinities and inhibitions. This information can be used to efficiently select the right strain candidate and at the same time to predict its performance in a larger scale.