PACE: the future of directed evolution

What if we could speed up evolution?

Evolution is a slow process that spans over millions of years, even the slightest changes have to be selected through thousands of generations until they are fixed in a population. Natural selection, the tool used by evolution to shape every single organic molecule and organism, acts from generation to generation, selecting the fittest.

Evolution has rendered us with a limited (albeit gigantic) set of biochemical tools we can use for biotechnological purposes. These proteins, RNAs, enzymes… have a certain effectiveness, speed and affinity among other characteristics, that we could not (at first sight) change. Given that our time playground only extends to a few decades, we need to find a way to improve nature’s biochemical tools in order to satisfy our biotechnological needs involving energy, pollution, food, medicine and many others.


Figure 1. Directed evolution cycles in the laboratory to characterize and create biocatalysts. Source: (Roderer et al., 2009, Chimia)

Rational design was the first step towards protein improvement, but it is considerably crippled due to our limited knowledge of protein structure and poor 3D-structure prediction capacity. Directed evolution appeared as a way to get pass this lack of understanding and obtain results without a previous complete knowledge. Although it clearly accelerates the evolution of the desired protein, a single round of mutation, gene expression, selection or screening, and replication typically requires days and frequent human intervention.

The more rounds you do, the better your protein will be, but there is clearly a time limitation. To do 50 rounds you would need roughly a year. Therefore usually only a few rounds are made for every experiment, letting evolution play only a small role.

Here is where PACE can start a revolution. PACE or Phage Assisted Continuous Evolution can do up to 200 rounds of protein evolution over 8 days, generating proteins which exceed several hundred-fold the activity of the original protein.


Figure 2. Overview of the PACE system. Source: ( Esvelt et al., 2011, Nature)

In this system, Escherichia coli host cells continuously flow through a vessel or “lagoon” containing a replicating population of phage DNA vectors (SP or selection phage) encoding the gene of interest. Gene III (encoding pIII), which is required for infection, has been deleted from the phage and inserted into an “accessory plasmid” (AP) present in the E.coli host cells. By linking the production of pIII to the activity of the evolving gene in SP, it can be accomplished that only phage vectors able to induce sufficient pIII production once inside the E. coli, will be able to persist.

Additionally, a mutagenesis plasmid (MP) has been inserted in the E.coli host cells in order to increase the error rate during DNA replication. PACE requires no intervention during evolution and obviates the need to create DNA libraries, transform cells, extract genes, or perform DNA cloning steps during each round.

With the power to make proteins evolve so fast, a new range of possibilities opens up in the biotechnological world. Can we evolve common used enzymes to reach thousand-fold efficiencies? Does that mean patients would need doses a thousand times smaller? If we only need to produce a thousand times less, can we make bioreactors a thousand times smaller (home-size bioreactors)? Can we evolve enzymes to perform reactions that do not exist in nature? Can we evolve enzymes to degrade synthetic pollutants? Many crazy ideas could become a reality thanks to this new system.

Of course, being able to speed up evolution will also bring many questions: Where is the evolution limit for a given protein? Can any protein be improved?

Foto Marc Güell


Marc Güell (@marc_guell), biotechnology student at UAB, avid reader, curious scientist passionate about synthetic biology and systems biology and WhatIf’s collaborator.



1. Esvelt, K. (n.d.). Phage-Assisted Continuous Evolution. Retrieved from
2. Esvelt, K.M., Carlson, J.C. & Liu, D.R. (2011). A system for the continuous directed evolution
of biomolecules. Nature, 472(7344):499-503.
3. Roderer, K., & Kast, P. (2009). Evolutionary Cycles for Pericyclic Reactions – Or Why We Keep Mutating Mutases. CHIMIA International Journal for Chemistry, 63, 313-317.