| Neonicotinoid insecticides(NNIs)are a class of broad-spectrum insecticides that act on insect acetylcholine receptors(nAChR)and have become an important part of the global insecticide market due to their high efficiency and low toxicity.NNIs are mainly used in agroecosystems for foliar sprays,soil and seed treatments at crop planting,and they have a highly lethal effect on a variety of economically important crop pests such as aphids,whiteflies and leafhoppers.However,NNIs are often blindly increased during crop planting to improve their control effect,which also exacerbates the resistance on target organisms and the toxicity on non-target organisms,such as neurotoxicity in fish,immune toxicity in bees,etc.Besides,NNIs are easily enriched in water,soil and other environmental media due to their small molecular weight and highwater solubility,and about 2%to 20%of NNIs can be absorbed and spread throughout roots,stems,leaves,flowers,fruits and other parts of the plant.The residual NNIs in the above media would contaminate the food chain with the "water-soil-plant" system and enter human tissues,eventually posing a threat to human health.Epidemiological studies have pointed out that NNIs have a strong correlation with carcinogenic effects,developmental disorders,and teratogenic effects in humans.Although the ecological environmental problems of NNIs have been recognized,and the risk assessment as well as the development of substitutes to NNIs have been carried out accordingly,the problems of resistance on target organisms and toxicity on nontarget organisms of NNIs have not been effectively addressed,and the assessment results showed that the latest registered NNIs do not have optimal toxicological characteristics.Therefore,it is an important subject to construct an all-round vertical assessment and control system for resistance on target organisms,toxicity on non-target organisms,and human health risk under NNIs exposure,as well as develop new green pesticide substitutes that meet the drug research and development standards.In this thesis,the estimation programs interface(EPI),ecological structure activity relationships(ECOSAR),pesticides properties(PPDB)databases,molecular docking,molecular dynamics,TOPKAT,acute and chronic reference dose assessment,toxic unit and hazard index methods,were used to construct the ecological environment risk assessment and management system under NNIs exposure,and the suggestions for the management and application of NNIs in different regions and crops were put forward.In addition.the three-dimensional quantitative structure-activity relationship(3DQSAR)models of the integrated risk for NNIs were modified by toxicity regression equation,fuzzy integrated evaluation method,and comprehensive integration weighting method,and the NNIs substitutes with bi-directional biotoxicity selectivity and human health risk-friendliness were designed.Finally,based on Gaussian 09,PaDEL-Descriptor and Chembiodraw software,density functional theory(DFT)and time-dependent density functional theory(TDDFT),machine learning method associated with normalization,Pearson correlation coefficient,random forest(RF)and decision tree(DT).the molecular synthesizability of NNIs substitutes were predicted.For exposure risk assessment and control of NNIs,the targets and non-target organisms in farmland,and national drinking water sources were taken as examples.A bi-directional selective risk identification and control system for resistance and crossresistance on target organisms,toxicity and combined toxicity on non-target organisms under NNIs exposure in farmland ecosystems,and a combined toxicity and oral toxicity assessment system under NNIs and their transformation products exposure in national drinking water were constructed.The synergist resistance control scheme to effectively inhibit cross-resistance and combined toxicity was developed,and the regulation effect of the ideal scheme was up to 50%and 200%,respectively.The priority control lists for NNIs application in food and vegetable crops were identified,and suggestions on the control and application of NNIs and their transformation products in drinking water were proposed.For design and screening of green substitutes to NNIs,the farmland ecosystem in planting area and the sensitive population near planting area were taken as examples.In source prevention,the bi-directional biological toxicity selectivity and human health risk friendliness 3D-QSAR models were constructed based on the total selectivity indexes of animals,plants and microorganisms and the integrated evaluation indexes ofcarcinogenicity,mutagenicity,skin sensitization and skin irritation.One NNIs substitute with increased toxicity on target organisms(13.69%)and reduced toxicity on non-target organisms(70.20%),and two NNIs substitutes with low risk of multiple toxicities(carcinogenicity,mutagenicity,skin sensitization,skin irritation)were designed.In process control,a target mutant modification scheme with bi-directional toxicity selectivity under multiple scenarios was proposed,and the combined improvement effect of ligand-receptor modification was up to more than 30%.A regulatory scheme of external environmental factors to reduce the oxidative damage caused by NNIs residues in leaves and promote the degradation of NNIs by microorganisms in different soil types(acidic,neutral,and alkaline)was developed,the degradation efficiency increased by 61.90%,45.50%,and 26.82%.In terminal treatment,based on the mechanism of interaction between NNIs,substitutes and human acetylcholine receptor α4β2 subtype,a reasonable dietary recommendation to effectively alleviate the risk of carcinogenicity and mutagenicity in humans was proposed,and the ideal improvement effect of this recommendation was about 300%.For synthesizability prediction of NNIs and their substitutes,the theoretical design of NNIs substitutes was taken as an example.The positive and unlabeled(PU)machine learning prediction and verification models for molecular synthesizability of NNIs substitutes were constructed based on bagging-RF and bagging-DT algorithms.Three NNIs substitutes(UN-1.UN-2.UN-3)labeled as positive" samples were selected,and the target of reducing the number of "unlabeled" samples by 95.89%was achieved.The study breaks through the bottleneck in the synthetic development of cis-NNIs and fills the gap in the field of applying machine learning methods to predict the synthesizability of NNIs substitutes.This thesis aims to innovatively develop a three-dimensional risk assessment and control system for resistance on target organisms,toxicity on non-target organisms and human health risk,and effectively propose a coupling scheme for the modification and synthesizability prediction of NNIs substitutes,to form a complete set of technology model of "source prevention,process control,end treatment" for agricultural source pollution prevention and control,and providing new ideas for achieving green agricultural development in China. |
