A persistent goal of industrial biotechnology is to produce recombinant protein at reduced costs, while maintaining high product quality and yield. A high proportion of production costs are incurred due to downstream processing. One route to reduce these processing costs is to achieve secretion of proteins by the expression organism, an objective that has also been shown to increase product yield and quality. One route to achieve this is to re-engineer the bacterial flagella type III secretion system (FT3SS). This organelle provides motility to the cell through motor and filament structures - however in the context of biotechnology, the FT3SS provides a high capacity one step route for protein export from the cytoplasm to the extracellular space - exporting approximately 1000 protein subunits per minute - a rate which rivals other extracellular secretion systems in a range of biotechnologically relevant organisms. This project aimed to optimise an E. coli strain for directed secretion of proteins from the cytoplasm directly to the extracellular environment, through a modified FT3SS. The flagella secretion system served as a platform technology for the secretion of a range of proteins. Initially native protein secretion through the modified secretion apparatus was investigated, using a flagellin monomer variant. Once established, directed recombinant protein secretion of an antibody fragment (CH2) through the modified FT3SS secretion apparatus was achieved with the aid of a secretion construct, which harboured recombinant cargo protein and was also comprised of elements of the native flagellin to aid secretion (the untranslated regions and secretion signal) and tags for purification. The secretion construct is modular and therefore amenable to alteration of the cargo protein - and also modification of the secretion construct for improved secretion. Strain optimisation for improved secretion was also implemented. With the generation of a wealth of strain improvements and secretion construct variants, a high-throughput assay was required to quantify secretion output of combinations of these secretion tools. An enzyme based secretion assay was developed to screen the multitude of strains and secretion plasmids in an efficient and accurate manner. This resulted in identification of a secretion strain and construct which resulted in high capacity secretion. When utilised together a 25 fold increase in the secretion capacity was observed. Finally expression of a 45kDa human collagen I a chain was investigated as extrusion of collagen from the FT3SS was an attractive possibility, while secretion was not achieved, strain improvements resulted in an increase in intracellular concentration of this large therapeutically relevant protein. The yield of secreted protein was measured throughout and it is possible that with further improvements to strains and fermentation procedures that the FT3SS platform could be competitive in the field of industrial biotechnology.