In silico prediction of peptide variants from chia (S. hispanica L.) with antimicrobial, antibiofilm, and antioxidant potential

Abstract

Plants are known as a rich source of bioactive peptides, and a variety of plant peptides have been studied as potential alternatives to conventional antimicrobial, antibiofilm, and antioxidant agents in food products to prolong their shelf-life, which could pose potential health risks for consumers. Regardless of their high functional potential, no plant peptides are currently used in the food industry for these purposes. In this study, it is performed the selection and optimization of peptides that are not currently reported in any database, derived from a chia peptide fraction. Computer-aided tools were used to identify multifunctional peptides with antimicrobial, antibiofilm, and antioxidant potential. Two peptide sequences (YACLKVK and KLKKNL) showing the highest probability scores for antimicrobial activity were identified from a total of 1067 de novo sequences in a chia peptide fraction (F<1 kDa). Subsequently, the peptides YACLKVK and KLKKNL were used to create scrambled libraries containing permutations of these sequences to explore the antibiofilm potential of different amino acid arrangements. The peptide variants with the highest probability scores for antibiofilm activity were subjected to optimization for the improvement of their activity. Finally, the optimized sequences were analyzed to determine the presence of antioxidant fragments. This computational approach could be a solution for the screening of a large number of peptides with more than one function, allowing the development of multifunctional peptides as alternatives to traditional food preservatives.

Keywords: Antibiofilm; Antimicrobial; Antioxidant; Chia peptide; Machine learning; Multifunctional.