Broadening the Scope of Phage-Assisted Continuous Evolution

Abstract

Continuous directed evolution methods allow the key steps of evolution—gene diversification, selection, and replication—to proceed in the laboratory with minimal researcher intervention. As a result, continuous evolution can find solutions much more quickly than using traditional discrete evolution methods. Continuous evolution also enables the deep exploration of many evolutionary trajectories, permitting access to solutions that require many steps through sequence space. In the past decade, phage-assisted continuous evolution (PACE) has emerged as a particularly potent technique for evolution of useful molecules. PACE combines the short generation times of bacteriophage with high rates of mutagenesis to achieve evolution on rapid and practical timescales. Further, PACE selections are highly programmable, allowing selection for a wide range of desired characteristics. This work focuses on efforts to expand the capabilities and scope of PACE evolutions, with an eye to practical utilities and unfulfilled needs. In Chapter 1, I discuss the historical advances that have expanded continuous evolution from its earliest days as an experimental curiosity to its present state as a powerful and flexible strategy for generating tailor-made biomolecules, focusing particularly on historical innovations that laid the foundations for phage-assisted continuous evolution. I comment on novel technologies and their potential impact upon the future of directed evolution. In Chapter 2, I discuss the application of protein-protein binding selections in PACE to evolve an agricultural biopesticide to address resistance in the crop pest Pectinophora gossypiella, the Pink Bollworm. Cry toxins, derived from Bacillus thuringiensis, are potent and highly selective insecticides widely used in agriculture. In Pink Bollworm, resistance has arisen through disruption or loss of the gene encoding a receptor protein which is key to infectivity of Cry toxins. To counteract downregulation of the receptor, we increased the affinity of a Cry toxin for its canonical receptor. Following 776 hours of phage-assisted continuous evolution, improved toxins show up to 14-fold increases in affinity and achieve single-nanomolar affinities for the receptor. Preliminary bioassay data indicates improved toxicity on susceptible insects. In Chapter 3, I describe a system for the phage-assisted continuous evolution of protein-protein interactions in the E. coli periplasmic space. Periplasmic PACE (pPACE) enables continuous evolution to take place in the oxidizing and disulfide-compatible environment of the periplasm. I first applied pPACE to rapidly evolve novel noncovalent and covalent interactions between subunits of homodimeric YibK knottin protein and to correct a binding-defective mutant of the anti-GCN4 Ω-graft antibody. I developed an intein-mediated system for regulating periplasmic export at the host cell level in order to select for soluble periplasmic expression in pPACE, leading to a ten-fold increase in soluble expression of the anti-GCN4 Ω-graft antibody and correction of a binding defect within 96 hours of evolution. Finally, I evolved disulfide-containing trastuzumab scFv antibody variants with improved binding to a Her2-like peptide and improved soluble expression. I conclude that pPACE can rapidly optimize proteins containing disulfide bonds, broadening the applicability of continuous laboratory evolution. Together, these results expand the utility of phage-assisted continuous evolution, both in the application of established selection to tackle the challenge of Cry toxin resistance in Pink Bollworm and in the development of a novel PACE selection for protein-protein interactions under disulfide-compatible conditions. My findings broaden the scope of phage-assisted continuous evolution.