Predictable Protein Production

Specific proteins including biopharmaceuticals (=protein-based medicines), protein-based specific probes, and enzymes, constitute a growth market. Their complexity allows these agents and reagents to be highly specific, and thereby, for instance, kill diseased cells whilst leaving healthy cells intact. In addition, the fact that they are composed of the 20 natural amino acids offers an important aspect of safety; they essentially consist of natural compounds that can be degraded safely. The complexity of these products is well served by the complex production environment of living cells. This comes at a price however: Living cells are difficult to manage with simple methodologies, and difficult to understand with the traditional molecule or whole cell focused biologies. Protein production by living cells is unpredictable and rather inefficient, making biopharmaceuticals more expensive than they should be, to the exent that some cannot be produced economically. Systems Biology is a recently amplified scientific discipline studying how the complex intracellular networking of molecules controls the functioning of whole living cells. The UK is among the word leaders in this Systems Biology, with its 6 government (BBSRC+EPSRC) funded Centres for Integrative Systems Biology and three such Doctoral Training Centres. One of these research centres, the MCISB at the University of Manchester, which also hosts one of the DTCs, has put together a complete tool set enabling the understanding of the growth of baker's yeast, which is one of the most famous and useful living cells (bread, beer and wine). Manchester is also home to the Centre of Excellence in BioPharmaceuticals (COEBP), in which the production of biopharmaceuticals by mammalian cells is being studied. This research project does the obvious: it brings together the MCISB and the COEBP. It does this in the context of the Biotechnology Research Industry Club (BRIC) in which academics and industrialists meet to discuss how science and engineering can be implemented to lead to work that is better for the public, through better industrial processes. The idea is to throw the new Systems Biology at the problem of understanding the living cells whilst producing proteins. This will first be done for the better known yeast cells, now producing proteins rather than beer, but immediately thereafter the same approaches will be implemented in mammalian cell lines that are known to produce less protein but in better shape. If successful, this will give Britain a world leading role in the production of biopharmaceuticals through the mathematical-modelling of the production processes that occur in the cell factories. Enabling maximal efficiency and control, the bioprocess industry associating with BRIC will become highly versatile and quick in designing and optimizing processes for a great variety of biopharmaceuticals. This could not now be accomplished in other countries. This increased versatility may lead to a substantial augmentation of bioprocessing in the UK process industry. The project is extremely challenging and interesting scientifically. One thing is that for the mammalian cells, the available information is limited: The structure of their DNA is not precisely known. Modelling methods will therefore have to be used that reckon with the many possible DNA structures. This requires substantial computer power and adeptness of the programmer. Another challenge is to measure precisely the chemical activity of the protein producing cells and use the results to deduce which parts of their networks they are using to produce the protein. This should enable us to predict their maximum efficiency and then perhaps to direct them to improved performance. Because Life is subtle, yet another challenge is to cultivate the cells in such a way that they 'feel cool' when they are producing 'hot' protein for us, thereby preventing them from resisting doing their job well.

Systems biology tools have been developed, collated, and implemented in the quantitative understanding of the growth of baker's yeast. In a different research grouping, CHO (Chinese Hamster Ovary) cells have been shown to be limited in producing antibody fragments by the metabolic composition of the growth medium that was supposedly optimal. This project will put the two research settings together to build a systems biology based platform for predictable protein production. It will first develop the yeast-growth systems-biology platform so as to include the ability to make protein production by yeast more predictable. The platform will be readied to be ported to Pichia pastoris. The analogous systems biology platform for predictive protein production will then be built for the industrially important mammalian cell line CHO. Here the complete experimental/theoretical suite of activities developed for the yeast case will be implemented. This will again involve quantitative endo- and exo-metabolomics/flux balance analysis, genome-based metabolic network reconstruction, dynamic modelling, detection of where the limitations of protein production reside (e.g. amino acid/tRNA metabolism, glycosylation, secretion; or in a number of these at the same time). The earlier work in the CHO research grouping has shown most of this to be doable at sufficient accuracy. The uncertainties about cell-line genome sequences will be approached by family modelling based on known mammalian genomes. Auxiliary activities with the human HEK293 cell line (for which we have the pathway reconstruction) will be undertaken for validation. Experiments will be carried out throughput to validate and iteratively improve the modelling. This will include testing of predicted yield and engineerability of protein production in the cell lines. The project will be one of the first to address comprehensively the challenges of making systems biology work for an industrial context.

This research will benefit researchers in industry by presenting deep insight in the extent to which Systems Biology has grown from potential to capability. Particularly the group of BRIC companies following this research will become acquainted with the strengths (and still remaining limitations) of systems biology. They will also witness experimental and modelling possibilities that have recently come into action in the participants' laboratories, which may have applications in areas beyond systems biology but relevant for the industries. These will include the metabolomics of cell cultures, but also calculations of production and consumption fluxes. These company scientists will hereby become experts in what systems biology can do and this will enable them to assess and formulate much better how, why and where to implement systems biology for the benefit of their companies. At a second tier, the companies themselves will benefit from the research, in at least two ways. The first is direct, i.e. in providing avenues for improving their protein production processes. Although the research in this project is precompetitive and open to all BRIC members and ultimately to the whole world for inspection of the results, the amount of additional detail the BRIC members will see will enable them to copy parts of the research into their own protein production systems and thereby increase the predictability and most probably yield, stability and general performance of the latter. In addition to this we expect a significant image effect. By associating with this research the companies will demonstrate that they no longer wish to rely on trial and error processes when defining the ways in which they make substances that are used by healthy humans and patients. They show that they try to and will ultimately make their processes rational and controllable. There is an important future aspect of quality control, safety, and even of FDA and EMEA driven regulation. At some stage, industrial production processes for substances to be used in humans, may need to become as defined as chemical synthesis processes. This already relates to the benefits for policy makers, within companies but also within public bodies funding fundamental and applied research and formulating the reasons for funding research and training: not only the products but also the research projects leading to new products may be helped to become more defined by systems biology implementation; and this project may help in achieving this. It will show how the functioning of a biological production system can be researched rationally. It can thereby help to identify where funding may rationally be positioned. For instance, when biology and biotechnology become more predictive and less stamp collecting, it may become easier to defend its funding. The research should help BRIC and other companies improve their protein production processes. This should increase the profitability of these companies, but also reduce the prices of their products. The latter aspects may well become an extremely important issue, as many more biopharmaceuticals are being discovered that will work well but only for a small fraction of the patient population. This feature makes the drugs much more expensive per patient, to the extent that the National Health Service is already unable to admit the use of certain such drugs. The reductions in cost, as well as the production versatility being increased through increased predictability, may help alleviate this problem by reducing the price and enabling the same culture system to produce more biopharmaceuticals, the one after the other. This could potentially help a great many small cohorts of patients, totalling large numbers of people. The impact will be enhanced by the appointment of an informal governing board, of volunteer members of BRIC.