Extraction of polysaccharides from Camellia oleifera leaves by dual enzymes combined with deep eutectic solvents screened by ANN and COSMO-RS

Camellia oleifera leaves were byproduct of the C. oleifera industry which was rich in polysaccharides. Deep eutectic solvent-dual enzyme system (DES-dEAE) was established to achieve the simultaneous hydrolysis reaction of dual enzymes and DES extraction. Currently, there was a lack of research methods that simultaneously deal with the relationship between DES and enzymes and the effect of DES on the polysaccharide extraction rate. In the study, artificial neural network (ANN) and conductor-like screening model for real solvents (COSMO-RS) were used to screen the most suitable DES by predicting the pH and activity coefficient of DESs. Choline chloride and sorbitol constituted the best DES and were used to optimize the process conditions of DES-dEAE by response surface method (RSM). The polysaccharide yield was verified to be 21.46 mg/g with the optimal process conditions: liquid-solid ratio of 33:1, enzyme ratio of 3:1, extraction time of 15 min and water content of 40 %. The C. oleifera polysaccharides obtained by water, DES, and DES-dEAE extractions were observed by scanning electron microscopy and cell wall damage effect of DES-dEAE was confirmed microscopically. The study optimized the complex polysaccharide extraction system using bioinformatics methods, contributing new ideas to the development of simulated extraction.