Linking natural product synthesis to cell growth with gcFront Laurence Legon 1,2,3 , Christophe Corre 1,3 , Declan G. Bates 2,3 , Ahmad A. Mannan 2,3 1 School of Life Sciences, University of Warwick, UK, 2 School of Engineering, University of Warwick, UK, 3 Warwick Integrative Synthetic Biology, University of Warwick, UK Nature has the ability to synthesise a wide range of valuable natural products, like biofuels, pharmaceuticals and chemical building blocks. However, natural organisms have evolved to only synthesise as much product as is useful to them in their native environment, which limits their utility for industrial biosynthesis. Genome engineering can be used to optimise an organism for biosynthesis, but this is a slow, labour-intensive process. Furthermore, during cell culture, natural product synthesis usually drains the cell’s resources, leaving less resources available for cell growth. Fast growth is selected for during cell culture, which means that during an industrial synthesis process there will typically be a selective pressure for mutations that interfere with natural product synthesis, which can result in lower yields. A promising methodology for addressing these challenges is growth coupling- where metabolism is engineered so rapid growth can only occur if the compound of interest is synthesised. By doing so, an indirect selective pressure is created for natural product synthesis, as the cell will have to improve its biosynthesis in order to evolve to grow more rapidly. This selective pressure facilitates the use of evolutionary approaches to improve biosynthesis, and also renders biosynthesis evolutionarily robust, as mutants with poor synthesis will grow slowly, and so will be selected against. A number of algorithms analyse genome-scale metabolic models to find gene or reaction knockouts that will enforce growth coupling, but these typically focus on finding a single strategy that maximises product titre, and neglect to consider coupling strength (how dependent growth is upon product synthesis); an important consideration, as strategies that enforce weak coupling may only create a weak selective pressure for product synthesis. Furthermore, genome-scale metabolic models can contain thousands of genes, creating a huge number of potential combinations of knockouts that can take an intractably long time to explore. Here, a software package called gcFront was developed to use a multiobjective genetic algorithm to identify a range of strategies that balance product titre, cell growth rate, and coupling strength, giving users the freedom to select the strategy that best suits their needs, and also features a graphical user interface to aid users in this selection. Furthermore, gcFront is fast- it can discriminate between knockouts that do not cause coupling to favour those which are closer to being coupled, which speeds the algorithm’s convergence onto coupling strategies and contributes to its superior speed when compared to several competing algorithms. Thus, gcFront should prove a useful tool for the engineering of microbes for natural product biosynthesis.
P41F
© The Author(s), 2022
Made with FlippingBook Learn more on our blog