Improving the recovery of "difficult to release" periplasmically-expressed products from recombinant E. Coli

[摘要] The periplasm of E. coli has been recognized a suitable location for the production of large quantities of many industrially important soluble recombinant proteins, and offers some important advantages over both intracellular and extracellular production. The oxidative environment of the periplasm promotes correct disulphide bonding and protein folding; there is reduced risk of proteolytic attack in the periplasm; and the periplasm accounts for <5% of total cell protein, so that selective release reduces subsequent purification demands. Despite these merits, periplasmic expression systems have not yet fulfilled their true potential, largely due to the lack of reliable general methods for efficient selective release of periplasmically expressed proteins at large-scale. The classical osmotic shock procedure, the only reliable method for releasing proteins from the periplasmic space of E. coli, is expensive and time-consuming, and thus it is not feasible at large scale. The main objective of this study has been to develop a gentle chemical permeabilisation method for selective release of periplasmically-expressed proteins from recombinant E. coli. In the first experiments, the titre, location and form of anti-lysozyme Fab D1.3 were determined during fed-batch cultivation of E. coli. It was shown that the Fab produced as both soluble and insoluble forms and released into the culture medium over the course of fed-batch fermentation. Purification of Fab D1.3 was then performed using various chromatographic methods, and the most effective target Fab purification was achieved by using sequential cation exchange – Protein G affinity chromatography route with an overall yield of 83%. Pure model proteins typically secreted to the periplasm (i.e. beta-lactamase, alpha amylase and Fab D1.3 fragment) were subsequently exposed to various chemicals, and alterations in the secondary structure of the proteins in the presence of various chemicals were investigated by high throughput circular dichroism (ht-CD) system. Chemicals such as 0.1% Triton X-100, 0.05% benzalkonium chloride (BAC), 0.1% cetyltrimethylammonium bromide (CTAB), and 2 M urea remarkably changed the secondary structure of beta-lactamase and alpha-amylase. The secondary structure of Fab D1.3 was more vulnerable to the tested chemicals. The biological activity of the target proteins in the presence of various chemicals was also measured and it was revealed that changes in the secondary structure of proteins do not necessarily cause reduction in the biological activity and vice versa. Concentrations of chemicals which did not reduce the biological activity of the proteins were eventually examined in subsequent periplasmic release experiments using recombinant E. coli strains producing the same target proteins, and the performance of various chemical permeabilisers were evaluated by comparing to classical osmotic shock and mechanical cell disruption. It was demonstrated that low concentrations of chemicals such as sodium deoxycholate (DOC) and/or chelating agents, isoamyl alcohol released the periplasmic proteins as efficient as or more efficient (up to 168%) than osmotic shock treatments. It was also proved that chemicals could increase the periplasmic release efficiency when they used in combinations. For instance, 1 M EDTA in combination with detergents could increase the periplasmic release of beta-lactamase and Fab D1.3 up to 80% and 130%, respectively. Such synergetic effect for release of alpha-amylase and Fab D1.3 was also observed when 1% solvents (hexane, xylene, benzene, toluene, and isoamyl alcohol) were combined with detergents such as 0.025% DOC, 0.01% CTAB and 0.1% Triton X-100.

[发布机构] University:University of Birmingham;Department:School of Chemical Engineering

[关键词] T Technology;TP Chemical technology