Integrating upstream host cell line selection and development with improved downstream bioprocessing

Many of the new drugs currently under development are based upon proteins rather than traditional small molecules (e.g. antibiotics). One of the type of protein molecules that is particularly challenging to make are antibodies e.g. herceptin. These protein drugs are produced for the treatment of diseases such as cancer by mammalian cells kept in culture under defined conditions. One problem with this is that the cells we use to make such proteins secrete not only the target protein into the medium in which the cells grow, but other proteins from the cell as well, called host cell proteins (HCPs). Further, cell breakage during fermentation or downstream handling (e.g. centrifugation) can result in the release of intracellular protein material. To complicate things further, what these HCPs are (the contaminants) and how they change throughout cell fermentation and with target products is not known. What this means is that the target drug must be purified from the rest of the material in the medium before it is deemed safe for use and this is referred to as downstream bioprocessing. Downstream bioprocessing is now a major part (>40%) of the total cost of manufacturing such drugs and as such improvements in this area would be of major benefit to both manufacturers and the end patient. We aim to begin addressing this lack of knowledge with respect to the HCPs in the medium during the culturing of Chinese hamster ovary cells engineered to express a recombinant monoclonal antibody. Specifically, we will determine the HCP profile throughout culture and throughout a standard template purification procedure for a monoclonal antibody. We intend to use this information to then remove the most troublesome HCPs by cell engineering approaches and determine what effect this has on product yield, cell growth and subsequent purification procedures. Ultimately we envisage that this information will allow us to redesign downstream purification procedures either to remove or to integrate better current chromatographic steps which are expensive and time consuming. This information is of high industrial relevance since the production of commercially valuable proteins (e.g. antibodies) can be hindered, and the cost dramatically escalated, as a result of the multiple chromatographic steps currently required to purify the target protein to acceptable levels. A better understanding of the HCP profile and how this influences downstream processing is very important as it is expected that with an increasing number of protein 'drugs' being developed we will lack the capability of producing large enough amounts to meet the required demand at a cost which can be affordable for the majority. Hence these products remain prohibitively expensive to many, but very effective, medicines.

Over the last 20 years recombinant protein yields from in vitro cultured mammalian cells have at times exceeded 5 g/L, however there have been fewer major advances in the downstream bioprocessing (DSP) of proteins produced in this manner. In DSP, removal of host cell protein (HCP) is a major goal, however there have been few public reports focussed upon identifying the HCP complement from industrially relevant cells. This is surprising as the rational application/manipulation of DSP approaches would be enhanced by knowledge of the principal contaminants, whether these change/accumulate during fermentation/recovery, which HCPs specific steps remove and, if specific types of product influence the HCP profile. Such insights would aid the design of novel/improved DSP approaches and inform upstream strategies for improved DSP. Here we will test the hypothesis that identification/characterization of the major CHO HCPs will allow the design of more efficient, or alternative, purification strategies and the rational selection and/or engineering of hosts to limit HCP levels. We will characterise the links between fermentation length and HCP accumulation in null and producer cell lines. Any interactions between HCPs and the target protein and the influence this has on product yield/ease of DSP, the fate of HCPs throughout DSP, and the effect removal of specific HCPs by RNAi has on DSP will be assessed. The direct outcomes will be (1) a CHO HCP profile for a model antibody and an understanding of how this changes/accumulates during fermentation, (2) knowledge as to whether the target protein changes the HCP profile and the ease with which these contaminants are removed, (3) an understanding of the HCPs removed throughout the template DSP, (4) methods of monitoring/measuring HCPs, (5) determination of the effects of eliminating specific HCPs on cell phenotype and subsequent DSP, and (6) the design of alternative processes to remove HCPs via either up- or down-stream approaches.