Improving eicosapentaenoic acid yield by constructing an enzyme-constrained model of Schizochytrium

Eicosapentaenoic acid (EPA), an essential omega-3 polyunsaturated fatty acid (PUFA), cannot be produced by the human body. It finds applications in various fields, including health foods, pharmaceuticals, and animal feed. Schizochytrium sp., an oleaginous microorganism rich in PUFAs, holds promise for EPA production. However, its complex metabolic network limits its full potential. To address this challenge, we constructed a detailed enzyme-constrained model eciCY1170_DHA of Schizochytrium's metabolism. This model incorporated 1083 genes, 5236 reactions, and 2989 metabolites. We then used this model to simulate fermentation conditions and validated our predictions through experiments. Our research results indicate that the optimal nitrogen source absorption rate is 0.7692 mmol/gDW/h, and the optimal oxygen absorption rate is 3.41 mmol/gDW/h. When both absorption rates are higher or lower than the optimal values, they will hinder EPA production. Conclusions drawn from simulations and experiments show that by adjusting the nitrogen source concentration, Ventilation volume, and agitation speed, the maximum yield of EPA reached 1.09 g/L in a 5 L fermenter. By analyzing protein requirements, we identified 20 potential targets predicted to enhance EPA production. This study provides new ideas for the optimization of medium composition and strain modification for the industrial production of EPA using Schizochytrium.