An optimized growth medium for increased recombinant protein secretion titer via the type III secretion system

Background Protein secretion in bacteria is an attractive strategy for heterologous protein production because it retains the high titers and tractability of bacterial hosts while simplifying downstream processing. Traditional intracellular production strategies require cell lysis and separation of the protein product from the chemically similar cellular contents, often a multi-step process that can include an expensive refolding step. The type III secretion system of Salmonella enterica Typhimurium transports proteins from the cytoplasm to the extracellular environment in a single step and is thus a promising solution for protein secretion in bacteria. Product titer is sensitive to extracellular environmental conditions, however, and T3SS regulation is integrated with essential cellular functions. Instead of attempting to untangle a complex web of regulatory input, we took an "outside-in" approach to elucidate the effect of growth medium components on secretion titer. Results We dissected the individual and combined effects of carbon sources, buffers, and salts in a rich nutrient base on secretion titer. Carbon sources alone decreased secretion titer, secretion titer increased with salt concentration, and the combination of a carbon source, buffer, and high salt concentration had a synergistic effect on secretion titer. Transcriptional activity measured by flow cytometry showed that medium composition affected secretion system activity, and prolonged secretion system activation correlated strongly with increased secretion titer. We found that an optimal combination of glycerol, phosphate, and sodium chloride provided at least a fourfold increase in secretion titer for a variety of proteins. Further, the increase in secretion titer provided by the optimized medium was additive with strain enhancements. Conclusions We leveraged the sensitivity of the type III secretion system to the extracellular environment to increase heterologous protein secretion titer. Our results suggest that maximizing secretion titer via the type III secretion system is not as simple as maximizing secreted protein expression-one must also optimize secretion system activity. This work advances the type III secretion system as a platform for heterologous protein secretion in bacteria and will form a basis for future engineering efforts.