Directed evolution as a tool in Synthetic Biology
Detailed and systematic characterisation is the traditional scientific tool used to understand the function and mechanism of biological systems and their components. Although undoubtedly very successful, characterisation alone may not be sufficient to give us a complete understanding of any particular system or component.
Despite biology’s immense diversity, in some cases biology has provided us with a single answer. One such example is the storage and propagation of chemical information, where DNA and RNA are the only genetic materials, and the genetic code is universal. My view is that biology has not (and cannot have) explored all possible solutions to any given problem. Simply put, as an evolutionary process, biology is extensive but not thorough.
Synthetic Biology seeks an alternative approach in which well-characterised parts are assembled to reconstitute biological function. This ‘bottom-up’ approach can yield novel insights at all levels, be it at level of individual parts, of the system as a whole, or of more general principles that emerge in biology.
Nevertheless, the understanding required to re-engineer components for novel function is limiting and generally not available. Directed evolution bridges that gap in our current understanding.
Directed evolution, implemented through sequential rounds of sequence diversification and purifying selection, allows an original biopolymer (be it protein or nucleic acid) to be systematically modified towards the desired function. Crucially, in principle, sequence diversity can be introduced without any knowledge of the underlying component or mechanism of action – although in practice, all available information on the system is used to target diversity, to maximise coverage of the search space.
Methodologies for protein and nucleic acid directed evolution
Selection and screening are central to all directed evolution methodologies. The goal of selection is to create a strong link between phenotype and genotype, such that isolation of functional molecules, or molecules with the desired function, results in the co-isolation of their respective genetic information.
A number of versatile selection platforms have been developed, differing in how selective pressure can be introduced and modulated. Our goal is to adapt existing and develop novel selection platforms, establishing a technological toolbox for the directed evolution of natural and synthetic biopolymers – whether in vitro, ex vivo or in vivo.
Xenobiology and genetic orthogonality
The development of synthetic genetic materials (xenobiotic nucleic acids or XNAs) by systematically engineering DNA polymerases  is a clear example of the power of directed evolution for synthetic biology, and the first step towards developing an organism based on a synthetic genetic material.
Directed evolution of DNA polymerases for XNA synthesis was achieved through selection, using compartmentalised self-tagging (CST), and high throughput screening . Together with XNA reverse transcriptases, which were rationally designed, it was possible to demonstrate that a number of synthetic nucleic acids can store information and are viable genetic materials. This approach not only enabled the development of the first synthetic genetic materials, but also identified a novel region in the DNA polymerase involved in substrate recognition and discrimination . In addition, development of synthetic genetic systems allows exploration of the boundary conditions of chemical information storage .
We are currently interested in extending our existing selection platforms to isolate XNA replicases (XNA ->XNA) and to assemble systems that can be used to bring XNA genetic elements to a bacterial cell. That would extend the Central Dogma and alter the topology of information transfer in biology creating a genetic enclave inaccessible to natural organisms but that can co-exist with the natural system – an orthogonal system.
The storage of chemical information is not the only biological process that can be re-engineered to create orthogonality; there are other routes to altering the topology of information transfer in biology. The conversion of information from RNA to protein, via the genetic code is another process accessible to re-engineering. The genetic code, bar minor exceptions, is universal – an RNA message will give rise to the same protein in most living organisms.
A viable alternative genetic code can expand the chemical functionality of life and create organisms unable to exchange information with biology. Even if the informational molecules are unchanged (i.e. DNA and RNA), the content of the information remains inaccessible because of the different code; semantically orthogonal.
Aminoacyl-tRNA synthases (aaRS) are the gatekeepers of the genetic code through charging precise sets of tRNA with specific amino acids. Modification of the genetic code by aaRS engineering has been achieved and is being extensively explored. We are interested in investigating how evolvable tRNA synthetases are and whether the systematic engineering of aminoacyl-tRNA synthetases is a feasible route towards rewriting the genetic code.
For synthetic biology to deliver on its potential as a disruptive technology, revolutionising our chemical, pharmaceutical and material industries, it must incorporate biosafety at its core – to protect the environment and to ensure public acceptance of the technology. Multiple redundant safeguards must be developed to address known, foreseeable and unknown potential risks, ensuring that our environment cannot come to harm – minimising the ecological risk of genetically engineered organisms and the informational risk of the information encoded in the organism.
Both genetic and semantic orthogonality can be routes towards increased biosafety. Genetic information stored in an XNA that cannot be accessed by natural enzymes and that requires synthetic precursors for its maintenance, e.g. hexitol nucleic acids (HNA), would establish a biosafety “dead man’s trigger” – the function encoded in XNA is not accessible to nature and is lost in the absence of a constant supply of precursors. Thus, escaped organisms (or information) would be rapidly removed from the environment.
Similarly, life operating under a different genetic code cannot exchange information with natural organisms: genetic material from natural organisms will not encode functional proteins in a modified organism and vice-versa. A recoded auxotroph, which would depend on the external supply of an essential compound for cell survival, would therefore provide containment for the organism and for the information encoded in it.
Azole-containing microcins can be generated by the post-translational modification of peptides and are widespread in bacteria and archaea. They have been implicated in a diverse range of biological activities, including antimicrobial, anti-tumor and anti-malarial compounds.
We are currently characterising available microcin synthetases with a view towards establishing them as platforms for the directed evolution of novel bioactive compounds.
 Pinheiro, V.B., Taylor A.I., Cozens, C., Abramov, M., Renders, M., Zhang, S., Chaput, J., Wengel, J., Peak-Chew, S-Y., McLaughlin, S.H., Herdewijn, P. and Holliger, P., Synthetic genetic polymers capable of heredity and evolution. Science, 336, 341 (2012).
 Cozens, C., Pinheiro, V.B., Vaisman, A., Woodgate, R. and Holliger, P., A short adaptive path from DNA to RNA polymerases. PNAS, 109, 8067 (2012).
 Pinheiro, V.B.*, Loakes, D. and Holliger, P., Synthetic polymers and their potential as genetic materials. Bioessays, 35, 113 (2013). *corresponding author.