| Summary: | Improvements in fermentation processes, coupled with the rising demand for biologies, are requiring that new ways in which therapeutics are recovered and purified be identified. Column-based approaches are constrained by the need for pre-treatment which leads to decreases in process yield and throughput. By contrast batch adsorption has the potential to handle high solids concentrations and can eliminate the need for pre-treatment. It may offer a viable, robust and simple means to improved process yields and throughput and this forms the central theme to this work. The thesis presents a scale-down evaluation of the use of both column and batch approaches based upon the adsorption of an antibody fragment (Fab') using ion exchange media. The particular Fab' used throughout this study was found to bind to cell debris and therefore removal of cell debris had a significant impact on the total process yield. The recovery of the Fab' from unclarified feedstocks, through the use of packed bed columns and stirred tank batch adsorption, was investigated, as was the impact of feedstock ionic strength and solids concentration. The use of homogenisation and periplasmic heat lysis as distinct product release routes were also investigated and compared via a series of Windows of Operation. Isotherm data was used to develop an empirical model capable of predicting process performance. From this the effect of cell disruption and feedstock properties, as well as other factors such as the length of the lysis procedure and the number of adsorption cycles, were modelled and the data used to generate Windows of Operation, identifying feasible operating conditions. The limitations of using packed bed columns for the recovery of Fab' from an unclarified E. coli feedstock are discussed. Batch adsorption was found to offer a viable alternative to packed bed chromatography. An analysis of the commercial and regulatory aspects of the use of batch adsorption as an alternative to conventional packed bed adsorption concludes the thesis.
|
|---|
