In-silico selection of peptides for the recognition of imidacloprid

Abstract

The sensitive detection of pesticides using low-cost receptors designed from peptides can widen their uses in the environmental surveillance for emerging pollutants. In-silico selection of peptides can help accelerate the design of receptor sequence banks for a given target of interest. In this work, we started from Lymnaea stagnalis acetylcholine-binding protein Q55R mutant receptor-imidacloprid complex, available in the PDB databank, to select three primary short peptides (YSP09, DMR12, WQW13 respectively having 9, 12 and 13 amino acids (AA) in length) from the pesticide interacting zones with the A, B and C chains of the nicotinic receptor. Using molecular docking and molecular dynamics (MD) simulations, we showed that the three peptides can form complexes with the target imidacloprid, having energies close to that obtained from a reference RNR12 peptide. Combination of these peptides allowed preparing a new set of longer peptides (YSM21, PSM22, PSW31 and WQA34) that have higher stability and affinity as shown by the MM-PBSA calculations. In particular, the WQA34 peptide displayed an average binding free energy of –6.44±0.27 kcal/mol, which is three times higher than that of the reference RNR12 peptide (–2.29±0.25 kcal/mol) and formed a stable complex with imidacloprid. Furthermore, the dissociation constants (Kd), calculated from the binding free energy, showed that WQA32 (40 μM) has three orders of magnitude lower Kd than the reference RNR12 peptide (3.4 × 104 μM). Docking and RMSD scores showed that the WQA34 peptide is potentially selective to the target imidacloprid with respect to acetamiprid and clothianidin. Therefore, this peptide can be used in wet-lab experiments to prepare a biosensor to selectively detect imidacloprid.