(288e) A Genome-Scale Metabolic Model of Anabaena Sp. ATCC 33047 and Its Application to Design Strategies to Overproduce Nylon Monomers

Large scale biofuel/chemical production using cyanobacteria is hindered due to poor growth rates of model cyanobacteria and demand for fresh water and nitrogen supplementation requirements. Anabaena sp. ATCC 33047 is a fast growing, marine and nitrogen fixing cyanobacteria that can overcome these challenges. However, it is less studied and its metabolism is poorly understood, making it presently unfit as a production host. In order to deepen the understanding of this organism we constructed a genome scale metabolic model, iAnC892, for this organism by pooling together biochemical information from multiple databases including MetaCyc, ModelSEED, KEGG and a recently published metabolic model for the close relative Anabaena sp. PCC 7120 (Anabaena 7120). iAnC892 includes 953 unique reactions accounting for the annotation of 892 genes. It also captures the diazotrophic life cycle of Anabaena 33047 by accounting for both the vegetative and heterocyst cell types. Reaction content comparison with GSM of Anabaena 7120 revealed that there are 109 reactions including uptake hydrogenase, pyruvate decarboxylase, and pyruvate-formate lyase unique to iAnC892. These reactions have the potential to affect the flux distribution under dark fermentative conditions. iAnC892 facilitated the evaluation of energy production pathways in the heterocyst through cell specific deactivation of light reactions and glucose-6-phosphate metabolizing pathways. The analysis revealed the importance of light dependent electron transport in generating ATP and NADPH at the required ratio for optimal N₂ fixation. To demonstrate the usefulness of the model for metabolic engineering, it was used alongside the strain design algorithm, OptForce to identify strategies to overproduce valerolactam and caprolactam. The model was able to recapitulate several of the experimentally successful strategies that overproduced precursors of valerolactam (lysine and 5-aminopentanoate) and caprolactam (succinate).