Optimization of fermentation conditions and enzymatic properties of extracellular lipase produced by Serratia marcescens derived from food waste

A strain designated C41802, capable of high lipase production, was isolated from a food waste environment. Based on morphological characteristics, physiological and biochemical tests, and phylogenetic analysis, the strain was identified as Serratia marcescens . To optimize lipase production by C41802, fermentation conditions were refined using an Artificial Neural Network (ANN) combined with a Genetic Algorithm (GA), establishing a GA-ANN model. The enzymatic properties of the extracellular crude lipase were then characterized under these optimal conditions. Results demonstrated that the ANN-GA model surpassed the Response Surface Methodology with Central Composite Design (RSM-CCD) in predicting optimal fermentation conditions and maximum enzyme yield, with a relative error of merely 0.67% for the ANN-GA model compared to 2.54% for RSM-CCD. Optimal fermentation conditions, as determined by the GA-ANN model, included: olive oil at 10 g/L, glucose at 8 g/L, tryptone at 3.1 g/L, K₂HPO₄ at 2 g/L, an inoculum size of 3.5%, cultivation time of 26.2 hours, temperature at 30°C, and initial pH at 5, achieving a lipase yield of 613.39 U/mL, which is 6.45 times higher than before optimization. Under these optimized conditions, the extracellular crude lipase produced by S. marcescens C41802 exhibited an optimal temperature of 70°C and pH of 7. After incubation at 50°C for 12 hours, the enzyme retained 52.56% of its relative activity, indicating substantial potential for industrial applications.