(110b) Directed Evolution Strategies for Escaping Local Fitness Maxima

Directed evolution is a powerful tool for enhancing or creating protein function without the existing knowledge of its structure. However, proteins exhibit a high degree of sign epistasis, implying a rugged fitness landscape containing many local maxima in fitness. Most directed evolution studies exclusively use positive selection (i.e. selection for improved variants). Present approaches for bypassing local fitness maxima are limited. One can expand the initial variation in sequences to attempt a variety of evolutionary pathways, use neutral drift to expand the diversity in a manner enriched for function, or simply recombine evolved sequences. However, these techniques rely on chance to escape a fitness peak since they do not traverse the valleys of the fitness landscape but instead rely on single-step hopping from one peak to another. Here we compare the efficacy of four evolutionary strategies for bypassing local maxima:  positive selection, neutral selection, negative selection, and oscillating positive and negative selection.  All four strategies end with a final positive selection step.  Specifically, we evolved variants of β-lactamase, an antibiotic resistance gene, and make use of a synthetic gene circuit that that kills cells possessing β-lactamase activity above or below a desired threshold to perform the neutral and negative selections. The implications of our results for applied directed evolution and for natural evolutionary mechanisms, particularly of antibiotic resistance, will be discussed.