Investigating determinants of periplasmic protein solubility in Escherichia coli through expression optimisation and N-terminal proteomics

Escherichia coli (E. coli) is the organism of choice for recombinant protein production in both industrial and academic settings. Over the years, E. coli has revolutionised bioprocessing due to its well-characterised genome, fast growth kinetics and low cost, providing time and optimisation-based advantages. Most therapeutic proteins such as human insulin, monoclonal antibody fragments and pharmaceutically relevant proteins are produced in E. coli. Despite all of the aforementioned examples of periplasmic protein production, E. coli can present complications which incur additional solubilisation steps and costs in small and large-scale bioprocesses. Whilst protein production in the E. coli periplasm is well studied, key knowledge gaps remain regarding protein insolubility due to translocation.
In this research, substrate binding proteins from E. coli (SapA) and Serratia marcescens (YppA), produced and localised in the periplasm were used as a test set to understand protein solubility. Various expression optimisation approaches as well as proteomics methods were utilised to characterise factors affecting the solubility of the two target proteins.
For EcSapA, a wide variety of expression optimisation methods were applied: with particular emphasis on protein primary sequencing studying the effects of Lys to Arg ratio along with manipulating various plasmid strain combinations to solubilise SapA. Membrane protein production strains were shown to withstand soluble production of EcSapA, which enabled further exploration of the properties of production strain and the predictive structural basis of the protein.
For SmYppA, unlabelled and labelled proteomics approaches were utilised to reveal its temperature dependant solubility. This research has produced a novel proteomics dataset for studying the effects of temperature, induction and media on the solubility of a periplasmic substrate binding protein in E. coli. The HUNTER N-terminal proteomics method yielded high quality peptide data providing extensive information on the host proteome as well as detailed insights into the state and quality of the recombinant SmYppA. The data presented in this thesis highlights secretion defects along with evidence of mis-cleavages revealing novel signal recognition sequences for the signal peptidase. This work demonstrates the applicability of labelling-based mass spectrometry methods for understanding the complex solubility phenotype.