Engineering a novel high-throughput screen for directed evolution of enzyme stability.

Directed evolution techniques have revolutionised strategies for enzyme development and improvement. Large libraries, containing typically > 1000 protein mutants, are pro duced and screened for desired properties (e.g. novel activity or improved stability). The correct application of selection pressure depends on the accuracy and robustness of the screen: hence direct screens are preferred to indirect screens. Protein stability is often screened indirectly by monitoring protein aggregation or resi dual activity after incubation at elevated temperatures. Both screens rely on irreversible inactivation of the protein upon unfolding. The assumption is made that positive 'hits' result from variants resistant to denaturation under the test conditions and do not take1 spontaneous protein refolding into account. We have established an affordable, high-throughput, direct screen for protein stability that can be automated. Building on previous work Edgell ct. al. (2003) Biochemistry. 42. 7587-7593 we have extended and optimised the technique for application entirely in the higher throughput niicroplate format. Pure target proteins are unfolded in niicrowells by serial addition of chemical denaturants such as urea or guanidine hydrochloride. The unfolding curves are recorded by measuring the changes in tryptophan fluorescence under equilibrium conditions. The midpoint of unfolding determined for each protein variant (CY2) is an indication of its stability. This screen was developed and tested using bovine and equine cytochrome r as well as preparations of fat-stabilised BSA. This screen requires high-throughput protein purification and suitable buffer conditions for stability analysis of each mutant. A multi-step process route, amenable to automation. has been developed that involves cell culturing. protein purification with metal affinity resin, buffer exchange by dialysis and unfolding analysis, all in a high-throughput format. factors taken into account are maximum cell density of micro-scale cell-culturing. target protein title, resin binding capacity and minimum final protein concentration required for stability screening. This route was developed using wild-type IIis-tagged transketola (E. coli tkt gene) over-expressed in E. coli IM107 pQR791. Transketolase mutants. D381A and Y440A. with decreased dimer stability were also designed as model systems for the stability analy sis. The reduced stability was clearly observed. The route needs more optimisation to yield sufficient transketolase required for accurate1 screening. Proper implementation of the process route will allow for screening of a thousand mutants per day for improved stability. The stabilities of proteins with 'hard to screen activities' and reversible folding can be improved more readily using this route.