In a previous blog, we mentioned the theory of Anfinsen that the structure of a protein is determined solely by the amino acid sequence and that this property enables proteins to be refolded in vitro. If this were universally true then many or all proteins would be produced in E. coli as insoluble aggregates, known as inclusion bodies, and these aggregates would be isolated, solubilised in denaturant and then refolded. However, this remains a somewhat niche activity although when the process works it can be a much simpler way to obtain large quantities of protein. How do you decide when it is worth attempting to refold a protein when there is no literature precedent? As a rough ‘rule of thumb’, smaller (~<30 kDa) secreted proteins that don’t have too complex a disulphide bonding pattern, and are not too heavily glycosylated, probably have a moderate chance of being refolded.
Once a decision has been made to attempt refolding, a number of things are worth having in mind. Ideally, there should be an assay for the protein to measure the correct re-formation of the native structure. Mere solubility is not a good surrogate for correct protein folding.
One of the strongest proponents of refolding strategies was Rainer Rudolph who has written numerous papers and reviews describing the approach. We suggest that a reasonable place to start would be a screen (Tobbell et al), possibly designed statistically to examine a number of different parameters at the same time. Parameters to consider are things such as pH, concentration and type of reductant for disulphide bond containing proteins, the presence of additives, salt and protein concentration amongst others. If this screen doesn’t generate any active and therefore correctly folded protein, in any of the conditions, it might be preferable to stop and explore other strategies. Knowing when to stop is as important as knowing when and how to tackle re-folding in E.coli. An example of where we took an alternative strategy to express a poorly soluble protein in E. coli can be found in our case study page.