Even when the great potential of synthetic biology is acknowledged nowadays, the usage of genetically engineered organisms in large-scale industrial processes faces big hurdles. The natural evolvability of microbes results in the genetic instability of genetically engineered strains, which in turn leads to impaired productivity (Du et al., 2018). In selected cases the metabolism of a microorganism can be reengineered such that its growth becomes fully dependent on the synthesis of a specific (by) product, which is what we define as growth-coupled production.
Using this concept as a basis, the Molecular Microbial Physiology Group from the University of Amsterdam has been working on a set of algorithms and genetic engineering strategies that could potentially revolutionize biotechnology. If growth coupled production of a target compound can be achieved in a genetically modified microorganism, this would mean that genetic instability would no longer be an obstacle for the use of said genetically engineered organisms in a large-scale industrial process.
Forbidden FRUITS was born with that purpose. The project was formulated under the idea that, given a target compound, it is possible to find a set of reactions that can be added to the metabolic network of a microorganism in a growth-coupled fashion. This means that, in principle, Forbidden FRUITS would be able to identify a genetic engineering strategy to stably produce any compound of interest in any microorganism for which a metabolic model exists.
We decided to follow a modular approach for the development of our project. This means that each of the team members took care of one part of the algorithm making sure that in the end, each part could be merged with the rest to create Forbidden FRUITS. The different modules in which the algorithm would be developed were:
Forbidden FRUITS needs a complete database of chemical reactions in order to build the strategies that could be used for growth-coupling the production of a wide variety of target metabolites. The merger is in charge of retrieving the information from different public online databases and combining it in one big database for Forbidden FRUITS. This is important because some databases contain information that others do not; thus, merging the different databases gives a much more complete description of the reactions contained in it as well as a more varied set of reactions and compounds than a single database would give.
So in the first step, we provide Forbidden FRUITS with a complete universal database.
There are two steps in this module: the first one consists in developing an algorithm that will allow Forbidden FRUITS to take gene protein reaction association (GPRA) information into account when generating a strategy; the second one is to build a method that can transform a reaction-based strategy to a gene-based strategy. A transformation method based on GPR associations has been introduced in previous research (Machado et al., 2016, Centre of Biological Engineering, University of Minho, Braga, Portugal), which is going to be a scaffold for this project. This transformation method would also be useful to investigate the complex relation between genes and reactions of an organism, which is of huge importance in system biology research.
The second step is to incorporate the transformed network based on GPRA information to the database. So that Forbidden FRUITS can also use this information to build strategies.
The Forbidden FRUITS algorithm would output strategies for the production of non-native metabolites in a growth-coupled fashion. The advantage of these strategies is that the fluxes that are introduced or deleted for the production of the foreign metabolites do not compete with other reactions in the network, since the producing reaction is left as the only means of producing biomass. This implies that bacteria are forced to use this reaction to achieve growth, and it results in the likely release of a free target product. This can easily be achieved for primary metabolites which directly relate to growth, development or reproduction (e.g. amino acids or vitamins). However, from consultation with experts, we noticed that the industry has more interest in secondary metabolites. Therefore, the implementation of an algorithm that is capable of putting together an additional series of reactions that would enable the production of these “specialized” metabolites from an intermediate compound is what would enable the creation of a new type of strategies, called “cheap”-producing strategies.
This part of the project would make sure to incorporate a strategy-finding algorithm that would find cheap-producing strategies for secondary metabolites.
The Plumber is a linear problem which aims to find the minimal number of knockins and knockouts (a strategy) to couple the product to growth; it uses elementary flux modes and duality concepts to find possible genetic engineering strategies. To find strategies to couple a target product to growth, the elementary flux mode (path the flux will take) and a ‘coupling metabolite’ will be identified. This coupling metabolite is a metabolite upstream in the production pathway of the target product and is natively found in the metabolic network of the microorganism. By adding desired and undesired linear constraints to the problem, the minimal set of modifications which would make the pathway from the coupling metabolite to biomass production to be the only way the flux can go may be predicted.
Here, we will incorporate information of downstream metabolites information into Forbidden FRUITS. Essential metabolites upstream from the producing pathway of the target product will consume flux, which would have a direct effect on product yield. Here we will develop an algorithm based on linear programming which will find if those upstream essential metabolites could be produced downstream of the producing pathway. This way, a higher product yield would be achieved, since more flow would circulate through the producing pathway.
The different stages of the design process of Forbidden FRUITS are summarised in the following diagram, where the integration of the different modules comes together to create our algorithm.
As a proof of concept Forbidden FRUITS will be used to generate strategies for the production of Salicylic Acid, Lactate and Mannitol in Synechocystis PCC6803, the production of Salicylic acid in Escherichia coli and the production of lactate with Synechococcus UTEX 2973. This shows that Forbidden FRUITS can be indeed used to design a strategy for any product in any microbe.
The main purpose of the system is to make a controlled feedback system for growth of phototrophic cultures (e.g. cyanobacteria). The system has been developed to provide light to 48 wells, distributed in two 24-well plates, and to allow control of the light intensity provided to each well based on the optical density (OD730). While the culture is controlled, additional measurements can be made, such as: oxygen concentration or additional excitation like a saturation pulse. A series of experiments with the cyanobacterium Synechocystis PCC6803 will be carried out to test whether the conditions in the newly developed cultivator allow these bacteria to achieve exponential growth and to investigate which conditions might limit their growth in the 24-well cultivator.
[1]Du, W., Caicedo Burbano, P., Hellingwerf, K.J., Branco dos Santos, F. (2018). Challenges in the Application of Synthetic Biology Toward Synthesis of Commodity Products by Cyanobacteria via “Direct Conversion.” In X. S. W. Zhang (Ed.), Synthetic Biology of Cyanobacteria (pp. 3–26). Springer Nature Singapore Pte Ltd.
[2]Machado D, Herrgård MJ, Rocha I. Stoichiometric Representation of Gene–Protein–Reaction Associations Leverages Constraint-Based Analysis from Reaction to Gene-Level Phenotype Prediction. Patil KR, editor. PLOS Comput Biol. 2016;12: e1005140.
