Robot-assisted phenotypic screenings and utilization of biosensors to support physiological studies and cell-factory design

Biotechnology employs tailor-made cell factories for the production of numerous products such as platform chemicals, proteins and healthcare products. The development of cell factories with a desired phenotype such as productivity, growth behavior, robustness against industrial stresses or toxic compounds present in the growth media, can be a challenging undertaken and often demands the analysis of hundreds to thousands of different variants. For instance, the use of next-generation feedstocks generated for example from lignocellulosic biomass (LCB) requires the identification of strains capable of converting various carbon sources but at the same time possessing a high tolerance against LCB-derived inhibitors.

High-throughput screenings (HTS) facilitate the analysis of large libraries or strain collections. However, it can still be laborious and time-consuming when performed manually. Even when a fully automation of the entire workflow is not feasible, a semi-automation reduces the time needed to be spend on routine tasks such as colony picking, replica plating or pipetting. Furthermore, genetically encoded biosensors have emerged as a powerful analytical tool to assess internal states in a non-invasive and high-throughput manner, thereby advancing the analysis of strain libraries.

In the PhD thesis presented here, robot-supported HTS approaches were developed to support the analysis and characterization of libraries and strain collections. Besides growth-based assays dedicated for the selection of strains for fermentations on next generation feedstocks, the recognition scope of the HTS approaches established here were further expanded by implementing genetically encoded biosensors. Besides their applicability for HTS, genetically encoded biosensors can be employed for bioprocessing. Given the presence of various sugars in next-generation feedstocks, a sugar-responsive biosensor was used to control mCherry production, induced by a non-metabolizable sugar present in a next-generation feedstock.

Chapter 1:
This chapter is focusing on introducing relevant topics in relation to the presented PhD thesis. Within this chapter, special focus is laid on the implementation of robots in order to support wet-lab work especially in connection with high-throughput applications and screenings. Furthermore, different types of biosensors are introduced including transcription factor-based biosensors and fluorescent reporter proteins. Moreover, alternative feedstocks generated from renewable sources are presented and challenges related to their application for bioprocess development introduced.

Chapter 2:
This chapter describes the development of an agar plate based HTS approach. Towards a semi-automated HTS workflow, colony picking (colony picking robot QPix) and replica plating (replica plating robot Singer ROTOR) was combined to support the handling and arraying of a transposon insertion library (Tn5-library) of E. coli. The established workflow provided the basis to be further combined with HTS compatible analytical tools to assess internal states. For this, the genetically encoded biosensor mCherryEA, a pH-responsive derivate of mCherry, was applied and the internal pH visualized by the use of a Gel-documentation imaging system. The applicability of the novel combinatorial workflow presented here was demonstrated by the identification of an E. coli mutant strain possessing an improved pH-homeostatic capacity.

Chapter 3:
Fluorescence-based biosensors are a common tool to support phenotypic screening approaches. Imaging represents a common method to assess fluorescence signals of colonies. However, for each applied fluorescent protein the respective imaging device needs to be set-up with an appropriate filter-set. In this chapter, we demonstrate that a standard microplate reader can be used to measure the fluorescence signals derived from arrayed colonies. The monochromatic technology enables a brought flexibility with respect to fluorescence signals, as demonstrated by the analysis of fluorescence signals derived from different fluorescence-based biosensors in arrayed colonies.

Chapter 4:
Within this chapter, robot-supported agar plate based screenings were set-up and conducted to screen a wild-type strain collection of 32 Actinomycetes isolates with respect to their carbon source flexibility (12 different carbon sources) and tolerance against LCB-derived inhibitors (five different inhibitors). Promising candidates were selected for micro-scale fermentations on the second generation feedstock Spent sulfite liquor – ultra filtrated (SSL-UF), originating from the pulp and paper industry.

Chapter 5:
Transcription factors and their cognate promoters responding to sugars have been applied to control gene expression in bacteria. Within this chapter, an arabinose biosensor plasmid was constructed to enable L-arabinose dependent control of mCherry production in Corynebacterium glutamicum. This system was then utilized to demonstrate, that the next generation feedstock SSL-UF can be used as both, a source of carbon for growth and as inducer-containing substrate to control mCherry production.

Chapter 6:
This chapter is dedicated to an overall conclusion and future perspectives of the presented PhD thesis.