Since scientists began directing the evolution of proteins to obtain various desirable outcomes, the tools and techniques used to accomplish these goals have themselves evolved. One recent development, for example, is phage-assisted contin­uous evolution (PACE), which can generate desired protein variants in a fraction of the time it takes using traditional stepwise evolution methods.

A PERFECT MATCH: To evolve a strong binding affinity between a protein of interest (POI) and a desired target, the gene for the POI (fused to an RNA polymerase subunit) is first encoded into the genome of a bacteriophage lacking a gene (gene III) critical for robust infection of bacteria. These POI-containing viruses are then cultured with E. coli that contain gene III as well as the POI’s desired target (above).© GEORGE RETSECKInteraction between the POI and target results in recruitment of the E. coli RNA polymerase to the gene...

And now PACE, too, has evolved. Early examples of PACE were largely used to evolve DNA-binding proteins, which was all well and good, says Greg Weiss of the University of California, Irvine, but the latest incarnation of the technique—protein-binding PACE—is “easily the coolest demonstration of PACE to date.” The ability to evolve novel protein-protein interactions, Weiss explains, “brings the technology into the realm of . . . the therapeutics industry, diagnostics development—a whole bunch of fields.”

The basic principle behind any PACE approach, says Harvard’s David Liu, who first developed the technique in 2011, is that bacteriophage viruses and the E. coli they infect must be engineered so that the virus’s survival depends on the particular interaction the researchers are trying to evolve. In the case of protein-binding PACE, the protein of interest (POI), which is expressed by the virus, must part­ner with the target protein—expressed in the bacterium—to drive expression of an essential virus gene. Put simply, the viruses need to quickly evolve the POI’s ability to bind to the target, or die.

Liu and colleagues, in partnership with Monsanto, have used protein-binding PACE to generate a new variant of a bacterial toxin that binds to a receptor in the insect pest Trichoplusia ni, also known as the cabbage looper, and in so doing have created an insecticide hundreds of times more potent than the wild-type toxin, to which many pests have become resistant. Engineering the toxin-receptor system took a lot of work, says Liu, but “mercifully, all of that development greatly benefits future applications . . . without requiring us to reinvent the wheel.” (Nature, 533:58-63, 2016)

Cell-surface display A library of bacteria or yeast contains variants of a protein of interest (POI) displayed on cell surfaces. Cells with target-bound POIs are isolated and the POI gene is further mutated to improve binding affinity. Relatively easy
Protein-binding PACE Bacteriophages lacking an essential gene (gene III) are engineered to express the POI. E. coli are engineered to contain gene III and the POI’s target. Upon bacteriophage infection of E. coli, only viruses producing a robust POI-target interaction generate gene III and thus viable virions. Necessitates large quantities of media—approximately 6 liters/day for 3 weeks—together with complex plumbing to enable correct culture media flow rates  


Interested in reading more?

Magaizne Cover

Become a Member of

Receive full access to digital editions of The Scientist, as well as TS Digest, feature stories, more than 35 years of archives, and much more!
Already a member?