Synthetic biology is being used to develop production cell factories by constructively importing pathways from other organisms into industrial microorganisms. Our work is focusing on a retrosynthetic biology approach to the production of therapeutics with the goal of developing an in situ drug delivery device in host cells.
Retrosynthesis, a concept originally proposed for synthetic chemistry, iteratively applies reversed biotransformations (i.e. reversed enzymes-catalyzed reactions) starting from a target product in order to reach precursors that are endogenous to the chassis. In our method, substrates, products and reactions are coded into molecular signatures and metabolic maps are represented as annotated hypergraphs.
The complexity of the retrosynthetic enumeration of all feasible biosynthetic hyperpaths for a given compound, a problem that has been limiting so far the adoption of retrosynthesis into the manufacturing pipeline, can be efficiently addressed in our approach by varying the specificity of the molecular signature. Our approach also enables candidate pathways to be ranked, to determine which ones are best to engineer. The ranking function is based on several criteria such as inhibitory effects, cytotoxicity of heterologous metabolites, host compatibility (codon usage, homology). Furthermore, we use several in-house machine learning predictive tools (the MolSig package) in order to estimate structure-activity relationships for enzyme activity and reaction efficiency at each step of the identified pathways.
We thus present a unified framework of all of the aforementioned techniques involved in the design of heterologous biosynthetic pathways through a retrosynthetic biology approach in the reaction signature space. This approach enables the flexible design of industrial microorganisms for the efficient production of chemical compounds of interest. We will outline examples of production of antibiotics, antitumor, and other therapeutic agents in various bacterial and eukaryotic hosts.