Predicted protein-protein interactions between sugar beet root maggot trypsins and sugar beet Kunitz trypsin...

Sugar beet (Beta vulgaris ssp. vulgaris), SB, is one of two plants from which sugar is broadly extracted, annually accounting for 55% of U.S. sugar ($1B, U.S.), and 35% global raw sugar ($4.6B). The sugar beet cDNA DV501688, encoding a serine threonine protease inhibitor (BvSTI) was shown to have increased relative transcript abundance in sugar beet during a resistant reaction to the Dipteran sugar beet root maggot, Tetanops myopaeformis SBRM, pest. BLASTp searches of the sugar beet proteome identified 21 additional proteins, each having homology to Kunitz trypsin inhibitors (KTIs) with BvSTI annotated as BvKTI1. Prior deep learning analyses predicted that BvKTI1 bound each of the 9 T. myopaeformis trypsins in the predicted binding loop region located between β-strands 4 and 5 of KTI proteins. The identification of 21 additional BvKTI proteins suggested the plant-insect pest interaction may be more complex, requiring further analysis. The deep learning analysis presented here, using AlphaFold, suggeststhe different BvKTIs may not be equally effective in binding the T. myopaeformis trypsins. BvKTI13 is predicted to bind all 9 T. myopaeformis trypsins with high confidence interface predicted template modelling (ipTM) scores between 0.77 and 0.82. In contrast, BvKTI14, BvKTI20, and BvKTI22 are predicted to not bind any of them (ipTM scores < 0.6). The remaining BvKTIs are predicted to have intermediate capabilities to bind T. myopaeformis trypsins, having predicted binding for some and lacking predicted binding for the others. However, the AlphaFold-based results are predictive and not scientific justification/evidence.