Novel Tat-based systems for recombinant protein production and surface display in bacteria

The bacterium Escherichia coli is extensively used to produce proteins for therapeutic use. Example products include therapeutic antibodies, insulin and many others, with a total market of around $100 billion per year. A preferred approach is to 'export' the protein of interest to the periplasmic space between the 2 cell membranes, because purification of the protein is often easier and more cost-effective. Traditionally, protein export is achieved by the 'Sec' pathway, which transports the recombinant protein across the inner membrane in an unfolded form. However, some protein products either cannot refold after transport, or fold too quickly before they are transported, and thus cannot be produced. The Tat protein export pathway offers an alternative approach to the Sec pathway because it transports the protein substrate in a pre-folded form, which bypasses difficult steps involved in unfolding the protein before transport, and refolding it afterwards. We have recently shown for the first time that Tat can export very large quantities of protein.

The natural E. coli Tat system exports proteins to the periplasm, just like the Sec system, but we have recently created an E. coli strain that expresses a Tat system (termed TatAdCd) from another bacterium, Bacillus subtilis, in place of the native E. coli Tat system. This strain has unique abilities. It efficiently transports the protein product to the periplasm by the Tat pathway, but then releases the product into the culture medium because it has large holes in its outer membrane. The cells are otherwise robust and can be cultured without problems. This project will use this strain for two purposes.

First, we will develop the system for the small- and large-scale production of therapeutic proteins. The strain has enormous potential for this purpose because protein products can be purified directly from the culture medium, bypassing the problematic steps of extraction from the periplasmic space and greatly simplifying downstream processing. This part of the project will also involve use of another new technology, involving expression of a thiol oxidase, in order to export correctly folded disulphide-bonded proteins. Many proteins contain disulphide bonds, and the the thiol oxidase enzyme helps the substrate protein to form disulphide bonds before export. This will help the Tat system to export correctly folded 'high-quality' proteins.

Secondly, we will develop a novel system for 'surface display' of proteins and isolation of interacting partner proteins. Many therapeutic proteins are directed against specific protein targets, for example those on cancer cells, and a key method for isolating new therapeutic proteins is to express a massive library of different proteins on the bacterial cell surface and then add labeled target protein. Cells expressing a protein that binds to the target are isolated and the protein of interest is cloned for further testing as a therapeutic protein. Current surface display methods have real limitations, including a requirement that the library of proteins is first transported across the cell membrane by the Sec pathway. We propose to develop a new system using the TatAdCd-expressing cells. In this method, the library of proteins will be exported by the Tat pathway and tethered to the outside of the inner membrane. Because the outer membranes of these cells contain large holes, added labeled target protein will be able to bind to the exposed proteins and interacting partners will be readily identified. This technique should result in the identification of important new therapeutic proteins that have been missed using current techniques.

This project exploits an E. coli strain that possesses unique properties for protein production purposes. The twin-arginine translocation (Tat) pathway normally exports folded proteins to the periplasm, but we have shown that expression of a Bacillus subtilis Tat system (TatAdCd) in an E. coli tat null mutant leads to efficient Tat-dependent export to the periplasm, followed by release into the culture medium. The strain has a leaky outer membrane but is otherwise intact, viable and suitable for shake-flask culture and batch fermentation. A test protein, GFP, is already produced at levels of approximately 0.5 g protein per litre culture.

We will develop the strain for two related purposes.
1. The strain will be developed for recombinant protein production on both laboratory and industrial scales, with the key point that protein product can be purified directly from the culture medium. tatAdCd genes will be integrated into the chromosome for even more efficient production and we will create a subset of strains that also express a thiol oxidase, Erv1. This catalyses disulphide bond formation in the cytosol without harming E. coli cells, and the Tat pathway can export folded, disulphide-bonded proteins. These strains will offer unique advantages for production and purification of high-quality proteins by both academic and industrial groups.

2. The same strains will form the basis of a new technique for identifying new therapeutic proteins e.g. antibodies using surface display. A library of proteins will be expressed behind an uncleavable Tat signal peptide, which directs export and display on the periplasmic face of the inner membrane. TatAdC-expressing cells contain large holes in the outer membrane and added, labelled target protein can thus reach and interact with high-affinity partners. Cells expressing the protein (e.g. antibody fragment) can then be isolated for further study. This system offers a powerful alternative to current phage display systems.

A. Groups benefitting from this research.

This project will be of direct benefit to a wide cross section of academic and industrial groups.

The first part of the work, involving the novel TatAdCd-based protein secretion platform, offers major advantages over other techniques and will be useable for a wide range of proteins.

(i). The impact on industry could be very significant: the Sec protein export pathway currently underpins a multi-billion dollar industry, and this Tat-based system offers real advantages over current protein export strategies. Given the size and importance of the therapeutic protein market, the work will have a major impact if adopted for production of even a small proportion of recombinant therapeutic proteins. We have recently shown that the Tat system can export proteins at levels required by industry (a key paper was only accepted only a few weeks ago) and we believe that interest in the Tat system will now will rise sharply. Further interest is certain once we publish data showing that the TatAdCd system exports proteins to the culture medium, and CyDisCo-Tat cells can export disulphide-bonded proteins.

(ii). The impact on academic research should also be high. A vast range of research projects require highly-purified protein in large (multi-mg) amounts; examples include biophysical studies, protein-protein interaction work, structural studies (especially crystallisation and NMR work). The TatAdCd expression system produces high levels of protein even in shake flask systems, and the exported protein can be purified from the culture medium with ease. We will undertake to assist groups to adopt these secretion techniques throughout the project.

The second phase of the work, involving a novel surface display system, will be of interest to a different range of groups and the research will again have significant impact in both fundamental and applied research. There is an urgent need to assay for new, high-affinity therapeutic proteins and there is little doubt that this Tat-based export/display technique will be able to display proteins that Sec-reliant systems cannot. We would therefore expect this system to be of direct benefit to a range of industrial groups interested in identifying new therapeutic proteins in microbial expression systems. Academic groups will use the system for a range of protein-protein interaction studies, some directed towards applied aims and some for fundamental studies.

B. Net impact of the work

Given the cross-section of groups who will be able to exploit these techniques, we predict direct benefits within the biotechnology industry in the form of:

(i). New reagents identified and produced.

(ii). More cost-effective production procedures for existing biopharmaceuticals.

The actual level of benefit could be very high: the therapeutic protein market is currently worth approximately $100 billion per year, and recombinant antibodies - a major focus in this project - are the fastest growing product sector.

The benefits to academic groups are harder to predict but they could again be significant: phage display has been a major experimental tool and this work will develop variants that should offer important advantages for many studies.

C. Benefits to UCB

UCB are partners in this research and they have expressed a keen interest in the use of Tat to produce recombinant proteins. This project is expected to benefit UCB through the identification of new protein products during the test phases, and the availability of new approaches to both protein production and protein display. However, it is important to stress that the technologies will be made available to other academic and industrial groups.