Molecular Simulation Of Ligand-gated Chloride Channels And Design Of Target Drugs

Ligand-gated chloride channels(LGCCs)are important targets of pesticides,mainly including ?-aminobutyric acid-gated chloride channels(GABACls),glutamate-gated chloride channels(Glu Cls),and histamine-gated chloride channels(His Cls)in insects.However,traditional insecticides targeting at these targets often have some disadvantages,such as high toxicity to mammals,environmental unfriendliness,and insect resistance.Thus,it is necessary to continuously design and develop new insecticides with novel structures and different mechanisms of action.In this thesis,Drosophila Melanogaster GABACl,Glu Cl and His Cl were modeled to explore the mechanism of action and difference between these three targets and isoxazolines at the molecular level,and to predict the key amino acids for their interactions.We expect to find novel and efficient insecticide lead compounds by virtual screening.The main contents of this thesis are as follows:1.Homology modeling and molecular dynamics simulation.Using the crystal structure of human GABA receptor or glycine receptor with high sequence consistency as templates,the three-dimensional structures of Drosophila melanogaster GABACl,Glu Cl and His Cl were constructed by homology modeling.Molecular dynamics simulation was used to optimize the structures and energies of these protein models.Finally,stable and reliable three-dimensional structure models were obtained.2.Molecular docking.Taking the above model proteins as receptors and the isoxazoline insecticides with known activities as ligands,the molecular docking was conducted to find out the potential binding sites to explore the differences of action mechanisms of isoxazolines in these three receptors.The results showed that fluralaner,a representative compound,could bind to Drosophila Melanogaster GABACl and Glu Cl simultaneously,and it had a stronger ability to inhibit GABACl and little effect on His Cl.The amino acids that played key roles in GABACl were predicted to be Ile218,Leu222,Leu250,Met256,Gly277,Phe280,Val281 and Ala285.3.In silico mutagenesis.Based on the above protein models and the results of molecular docking,the approach of in silico mutagenesis was used to construct the Drosophila melanogaster GABACl mutants.After energy optimization,fluralaner was then docked into these mutant models.The key amino acids involved in the interaction between fluralaner and GABACl were predicted by comparing docking scores before and after amino acid mutations.The results indicated that Phe280 might be the key amino acid binding to fluralaner.The residues Leu222,Met256 and Ala285 were also likely to be key amino acids.4.Pharmacophore model and virtual screening.Firstly,the pharmacophore models of ten isoxazolines were constructed,and the optimal pharmacophore model containing three hydrogen-bond receptors,five hydrophobic centers and a negatively charged center was obtained,which could perfectly match these isoxazoline molecules.The enrichment factor of the optimal model was 15.4,indicating that the model had good ability to significantly screening active compounds from database.Then,the pharmacophore-and docking-based virtual screening were performed on Zinc database.Six candidate compounds that are expected to be novel and highly effective insecticides were finally found in combination with graphics inspection.