FACS: a high throughput method for protein export and engineering

Production of recombinant proteins is typically done in bacteria, i.e. E. coli, since they can be grown quickly, inexpensively and can be used to express a variety of proteins. However, obstacles, such as the presence of cytoplasmic proteases and low periplasmic protein yields, exist that prevent E. coli from being more widely used for protein production. By using the methods of protein engineering coupled with highthroughput screening by Fluorescence Activated Cell Sorting (FACS), recombinant protein yields in E. coli may be increased which would lead to the greater utilization of E. coli as a recombinant protein production host. The recently discovered Twin Arginine Translocation (Tat) system allows the export of fully-folded proteins across the inner cell membrane of E. coli. The Tat system now allows protein engineers to take advantage of the available protein folding machinery in the cytoplasm, expanding the range of recombinant proteins that can be produced in bacteria. However, since the efficiency and yield of export via Tat is low, we sought to examine whether mutations within the mature part of the protein can result in increased yield of periplasmic proteins. Using directed evolution and FACS, the export of the anti-digoxin 26-10 scFv was increased in an oxidizing strain of E. coli. One isolated clone showed increased export of 26-10 to the periplasm. In vitro analysis of the thermodynamic stability and folding kinetics indicated that the mutant scFv exhibits faster folding kinetics, and this is likely to be responsible for the improved export to the periplasmic space. In addition, in order further increase the export of the 26-10 scFv through the Tat pathway, we examined the effect of the co-expression of certain chaperone and Tat translocon proteins that were previously shown to partially relieve the saturation of the Tat pathway. Co-expression of the Tat translocon proteins, TatA, TatB and TatC, resulted in an additional 2.4-fold increase in cell fluorescence yielding an overall cell fluorescence increase of 5-fold as compared to the parental 26-10 scFv. The effect of media, pH and temperature and the use of different oxidizing strains of E. coli on cell fluorescence and growth were also investigated but did not yield any improvement on cell fluorescence. Finally, flow cytometry was utilized in an attempt to develop a genetic system for the better understanding of the Clp system and for studying how amino acid substitutions might affect the mechanical stability of protein substrates. The Clp system recognizes proteins tagged for degradation, unfolds them by applying a mechanical force to the protein and degrades them into smaller peptides. A FACS screen was employed to screen a genomic library for genes that can act as negative regulators of ClpXP degradation. Unfortunately, no such genes could be isolated. In addition, since the system applies a mechanical force to unfold proteins, the Clp system could be used as an assay to engineer more stable proteins against mechanical denaturation. Using GFP tagged with the SsrA peptide as a fluorescent reporter in a FACS assay, the screening of several mutant libraries indicated that no mutations could stabilize GFP against degradation by ClpXP.