Assessing the Antimicrobial Properties of Honey Protein Components through In Silico Comparative Peptide Composition and Distribution Analysis

The availability of reference proteomes for two honeybee species (Apis mellifera and Apis cerana cerana) opens the possibility of in silico studies of diverse properties of the selected protein fractions. The antimicrobial activity of honey is well established and related to its composition, including protein components. We have performed a comparative study on a selected fraction of the honey-related proteins, as well as other bee-secreted proteins, utilizing a publicly available database of established and verified peptides with antimicrobial properties. Using a high-performance sequence aligner (diamond), protein components with antimicrobial peptide sequences were identified and analyzed. The identified peptides were mapped on the available bee proteome sequences, as well as on model structures provided by the AlphaFold project. The results indicate a highly conserved localization of the identified sequences within a limited number of the protein components. Putative antimicrobial fragments also show high sequence-based similarity to the multiple peptides contained in the reference databases. For the 2 databases used, the lowest calculated percentage of similarity ranged from 30.1% to 32.9%, with a respective average of 88.5% and 79.3% for the Apis mellifera proteome. It was revealed that the antimicrobial peptides (AMPs) site is a single, well-defined domain with potentially conserved structural features. In the case of the examples studied in detail, the structural domain takes the form of the two β-sheets, stabilized by α-helices in one case, and a six-β-sheet-only domain localized in the C-terminal part of the sequence, respectively. Moreover, no significant differences were found in the composition of the antibacterial fraction of peptides that were identified in the proteomes of both species.